UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A formal total synthesis of 9-isocyanopupukeanane Winter, Manfred 1980

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1980_A1 W56.pdf [ 6.74MB ]
Metadata
JSON: 831-1.0060782.json
JSON-LD: 831-1.0060782-ld.json
RDF/XML (Pretty): 831-1.0060782-rdf.xml
RDF/JSON: 831-1.0060782-rdf.json
Turtle: 831-1.0060782-turtle.txt
N-Triples: 831-1.0060782-rdf-ntriples.txt
Original Record: 831-1.0060782-source.json
Full Text
831-1.0060782-fulltext.txt
Citation
831-1.0060782.ris

Full Text

A FORMAL TOTAL SYNTHESIS OF 9-ISOCYANOPUPUKEANANE by MANFRED WINTER Dipl. Chen., Universitaet Karlsruhe, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF . DOCTOR OF PHILOSOPHY \ in THE FACULTY OF GRADUATE STUDIES (DEPARTMENT OF CHEMISTRY) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA AUGUST 1930 © MANFRED WINTER, 1980 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C olumbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and stud y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Depa rtment The U n i v e r s i t y o f B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 i i ABSTRACT A synthesis of (±)-9-pupukeanone (7_) is described. The key reaction was the establishment of the t r i cyc l i c carbon skeleton of 7_ by means of an in t ra-molecular Diels-Alder cycloaddition. The 1,3-cyclohexanedione 48 was obtained from ethyl methacrylate and ethyl 2-methylacetoacetate by a Michael addition followed by cyclization. Treatment of 48 with isobutyl alcohol in benzene furnished the enol ether 54_ which was alkylated with the tetrahydropyranyl ether of a-bromoethanol to give the ketone 55_. The lat ter was converted into the a-enone 57 by sequential treatment with l i thium aluminum hydride and ace-t i c acid in acetone. The lithium enolate of the compound 5_7 could be s i l y -lated to give the s i l y l enol ether 60_ and phosphorylated to give the enol phosphate 6_3. Selective tetrahydropyranyl ether cleavage in 6_3 with hydro-chloric acid in methanol provided the alcohol 64. A similar selective de-protection of 60_ to the alcohol 43_ could not be accomplished. The aldehyde 65_ was synthesized from the alcohol 64_ by a Moffatt-type oxidation of the la t ter . The n i t r i l e 76 was obtained from the aldehyde 65 in 2 steps via the corresponding oxime. Reaction of 65_with vinyl l i thium or vinyl magnesium bromide furnished the a l l y l i c alcohol 69_ which, upon reflux in xylene, smoothly cyclized to the Diels-Alder adduct T8. The a l l y l i c alcohol 69_ could not be oxidized to the a-enone 71_. The enol phosphate 78_ was converted into the keto alcohol _18.w"ith sodium ethoxide in ethanol. Treatment of either the alcohol 78 or 1_8 with 2,2-dimethyl-l ,3-propanediol in benzene provided the ketal alcohol 86_ which could be cleanly oxidized to the ketal ketone 8_7 with pyridinium dichromate. Addition of isopropenyllithium to 87 generated the ter t iary alcohol 90_. The diene 20_ was obtained from 90_ by simultaneous dehydration and deketalization with boron t r i f luor ide etherate. Hydrogenation i i i of the compound 20 wi th i r id ium black as ca ta lys t furnished a sample of 9-pupukeanone (7_). The alcohol 78 was oxidized wi th pyridinium dichromate to the ketone 96_, which, upon reaction with 2,4,6-tr i isopropylbenzene-sulfonylhydrazide, provided the tr isylhydrazone 97. The compound 97 could not be induced to decompose to a vinyl l i t h ium intermediate by treatment wi th bu ty l ! i t h ium. 71 R=0 iv V TABLE OF CONTENTS Page Abstract i i Table of Contents v List of Tables vi List of Figures v i i Acknowledgements . v i i i Abbreviations ix I INTRODUCTION 1. Synthetic Chemistry - Status and Significance 1 2. 9- and 2-Isocyanopupukeanane - Isolat ion, Structure Elucidation 4 3. Mollusks, Sponges, Natural Isocyano Compounds 6 4. Previous Syntheses of 9-Isocyanopupukeanane (5_) and 9-Pupukeanone' (7_) 7 I I DISCUSSION 1. Synthetic Planning - Strategy and Tactics or Retrosynthetic Analysis and Synthon 15 2. Envisaged Route 16 3. Preparation of the Dienol Derivatives 60_ and 63_ 27 4. Synthesis of the A l l y l i c Alcohol 69 41 5. Synthesis of the a-Enone73 56 6. Conversion of the Aldehyde 65_ to the N i t r i l e 76 59 7. Diels-Alder Cyclization of the A l ly l i c Alcohol 69_ and Synthesis of 9-Pupukeanone (J) 66 8. Conversion of the Diels-Alder Adduct 78. into the Trisylhydrazone 97 87 I I I EXPERIMENTAL SECTION 95 IV BIBLIOGRAPHY 123 LIST OF TABLES Page Table I : Selected Spectral Data of the Compounds 60 and 63 71 v i i LIST OF FIGURES Paje Figure 1: Retrosynthetic Analysis of 9-Isbcyano-pupukeanane 17 Figure 2: 6 Proton Resonance of the Compound 57_ 34 Figure 3: Nmr Vinyl Proton Resonances of the Two Diastereomers of the A l l y l i c Alcohol 69_ 50 Figure 4: 0CH? Resonance in the Nmr Spectrum of 87 82 vi i i ACKNOWLEDGEMENTS My thanks, above a l l , go to my wife for her understanding and patience during these studies and for her help in drawing al l the diagrams. I t was a pleasure to work under the direction of Professor Edward Piers whom I sincerely thank on this occasion for al l his advice, interest and encourage-ment during the course of these studies. Financial assistance in the form of a scholarship from the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. I thank Professor David Dolphin who read this thesis prior to submission, Mr. Edward Ruediger and Mr. Max Burmeister for proofreading parts of i t , and Dr. Norman Lewis for suggest-ions during the wr i t ing. Last, but not least, I thank the staf f of the various services in the department for their ef f ic ient co-operation. tx ABBREVIATIONS The following abbreviations are used in this thesis: HOAc = Acetic Acid n-BuLi = n-Butyllithium CDIm = N,N - Carbonyldiimidazole NCS = N - Chiorosuccinimide DIBAH = Diisobutylaluminum Hydride DMF = Dimethylformamide DMSO = Dimethyl Sulfoxide glc = gas - l iquid chromatography HMPA = Hexamethylphosphoramide i r = infrared LDA = Lithium Diisopropylamide nmr = nuclear magnetic resonance PCC = Pyridinium Chlorochromate PDC = Pyridinium Dichromate PPA «= Polyphosphoric Acid pyr = Pyridine r t = room temperature THF c Tetrahydrofuran THP c Tetrahydropyranyl TMEDA c Tetramethylethylenediamine t i c = thin-layer chromatography £-TsOH = para - Toluenesulfonic Acid p-TsCl = para - Toluenesulfonyl Chloride TosMIC = Tosylmethylisocyanide TFA = Trifluoroacetic Acid 1 I . INTRODUCTION 1.. Synthetic Chemistry - Status and Significance In chemical science, synthesis, part icularly organic synthesis, is an important and expanding area. I t has far reaching implications in the l i f e of nearly everybody today: the ready avai labi l i ty of a large variety of synthetic drugs is a cornerstone of contemporary medicine; signif icant improvements in agriculture are made possible through the extensive use of fe r t i l i ze rs as well as weed and insect-destroying agents; synthetic fibres are commonplace in modern tex t i le materials. Clearly, the significance of synthetic chemistry in our society is considerable, and in al l l ikelihood on the increase. The development of alternative sources of energy l ike fuel cells and air batteries wi l l mean an increasing technological application of organic synthesis. A pertinent case is the search for and preparation of effective catalysts for electrochemical reduction of oxygen by, for instance, covalently linked metal-porphyrin dimers. ^ A large number of research establishments a l l over the world are engaged in syntheses of natural products, as evidenced by many publications in this 2 3 area. ' Both old, well-established procedures and relat ively recently developed synthetic methods are advantageously ut i l ized in the search for ef f ic ient routes. The use of newly discovered transformations is often part icularly of interest with regard to their compatibility with the presence of sensitive groups, since intermediates in a synthetic sequence usually are of multifunctional nature. New procedures are frequently tested only on simple molecules with l i t t l e additional functionality leaving an information gap on this important point. The successful synthesis of relat ively complicated molecules such as 2 gibberell ic acid H a prominent and, as a plant growth regulator, biologically signif icant diterpene, vinblastine (2), a bisindole alkaloid used as antitumor agent and among the most used chemotherapeutic agents for the treatment of leukaemia and Hodgkin's disease, chlorophyll-a {3}, a magnesiumporphyrin complex involved in photosynthesis, and reserpin ^ (4_), another indole alkaloid which is useful in the treatment of hypertensive disorders,give testimony to the accomplishments and advances that have been and are being made in this area. 3 MeO 'OCO OMe OMe OMe An ingenious chemist, the late Professor R.B. Woodward wrote in 1956: "The synthesis of substances occurring in Nature, perhaps in greater measure than act iv i t ies in any other area of organic chemistry, provides a measure of the condition and powers of the science. For synthetic objectives are seldom i f ever taken by chance, nor w i l l the most painstaking, or inspired, purely observational act iv i t ies suff ice. Synthesis must always be carried out by plan, and the synthetic f ront ier can be defined only in terms of the degree to which rea l is t ic planning is possible, u t i l i z ing a l l of the intel lectual and physical tools available." All compounds of natural origin which an organic chemist sets out to synthesize have already been synthesized before him by Nature. As another great chemist, Sir D.H.R. Barton, points out, she normally does i t with ultimate, albeit a 9 different kind of efficiency: "From molecular biology we know that Nature is capable of carrying out syntheses which are thousands of steps in length with a yield and an optical specif ic i ty of 100% in each step. The best industrial syntheses may attain average yields of over 90% for a 30 - 40 step synthesis. I t is clear, there-fore, that the synthesis of natural products remains an objective of great sc ient i f ic value and of great social and economic significance." The status and relevance of synthetic organic chemistry is well reflected in these two ci tat ions. 4 2. 9- and 2-Isocyanopupukeanane - Isolation, Structure Elucidation In 1975, Scheuer et al_. ^ reported the isolation of a new sesquiterpene from a nudibranch, f_hyl]_id_ia_ varicosa, Lamarck 1801, and from i ts prey, a sponge, Hymen_i^ cjdo_n_S£. This natural product possesses a new t r i cyc l i c carbon skeleton and a rare isocyano functionali ty. The authors named i t 9-isocyanopupukeanane (5_) after the location, Pupukea, near which the sponge and mollusk were collected. I R=NC R - = h 6 H NC 8 H H H NHCHO 10 H NH 2 II N(Me)C(S)NHC6H5 H The IUPAC nomenclature of the carbon framework is 1,3-dimethyl-5-endo-3 7 11 isopropyltricyclo[4.3.1.0 ' ] decane. In a subsequent paper , Scheuer and co-workers also reported the isolation of the corresponding 2-exo-isocyano isomer J5. Early in the investigation of the crude extracts i t was recognized, through nmr spectra, that a binary mixture of isoni t r i les with identical carbon skeletons was at hand. However, a separation of the two isomers i n i t i a l l y proved very d i f f i c u l t and upon degradation of the natural mixture, loss of the 2-isomer occurred and only 9-pupukeanone (7_) was obtained. I t was subsequently observed that the pure crystall ine 2-isocyano isomer 6_ was very unreactive. For example, defunctionalization to the parent hydrocarbon 8_ and hydrolysis to the formamide £ were the only two reactions which could be achieved. Reduction of either compound 6_ or 9_with l i thium aluminum hydride proved to be unsuccessful. Hydrolysis of the formamide 9^  to the 5 7 X = 0 Y= H 2 11 X = H 2 Y = 0 amine J_0 proceeded in very low y ie ld . The low reactivi ty of these compounds is not surprising since C(2) constitutes a very hindered bis-neopentyl carbon atom. Moreover, the amine 1_0 was v i r tua l ly unextractable into aqueous acid from organic solvents and reacted with N-chlorosuccinimide to give not the corresponding chloramine, but a complex mixture of products in poor y ie ld . Therefore, when the natural mixture was subjected to the sequence of reactions (1) 6N hydrochloric acid ( — y amine), (2) N-chlorosuccinimide ( — • chloramine), (3) sodium ethoxide in ethanol (—»- imine) and (4) 10% sulfuric acid ( — • ketone), i t appears highly l ike ly that no 2-ketone Jj_ was ever obtained, but only the degradation product ]_ of the 9-isocyano isomer 5_. The structures of both isomers of isocyanopupukeanane were confirmed by X-ray di f f ract ion analysis. The crystal l ine 2-isomer 6_ was eventually obtained by fractional crystal l izat ion of the mixture at refrigerator temp-erature. I t is a stable compound. The oi ly 9-isomer 5_, which is rather unstable and slowly isomerizes to the corresponding cyano derivative, was converted into the crystal l ine phenyl thiourea derivative 1_2 for structure determination. The six-membered rings in this tricyclodecane are in a boat conformation, imposed by the bicyclo[2.2.2]octane substructure (C(l)-C(3), C(6),C(7)-C(10)). The five-membered ring adopts an envelope conformation with C(3)-C(6) being approximately planar. The absolute configuration of 6 9-isocyanopupukeanane was deduced as (1R,3S,5R,6S,7S,9R). 3. Mollusks, Sponges, Natural Isocyano Compounds Nudibranchs, also commonly called sea-slugs, are soft-bodied marine mollusks lacking the hard and protective calcareous shell with which other mollusks such as clams, mussels and snails protect themselves against predators. Quite frequently nudibranchs are also endowed with bright, conspicuous colours. They are often associated with sessile marine inverte-brates, e.g. sponges or coelenterates in a symbiotic or grazer-prey relat ion-12 ship. Other marine mollusks void of the protection of a shell yet apparently 13 suffering from only very few predators are sea hares. Responsibility for this dichotomy l ies with the toxic nature of these animals, a fact already 14 1 known to the Romans. The chemical nature of these toxins has been uncovered and they were identi f ied as aplysiatoxin and debromoaplysiatoxin, two macrocyclic lactone ethers. The la t ter compound has also been found in the 16 17 blue-green alga Lyngbya_ £raci_li_s. As i t has been demonstrated that most and possibly a l l of the sea hares' secondary metabolites originate in their algal d iet , i t is l ike ly that these two compounds are not biosynthesized by the animals themselves. The mixture of 9- and 2-isocyanopupukeanane is known to be lethal to 18 f ish and crustaceans , both potential predators of P_hyll_id_ia_ v_. The natural products are thus assumed to be the defensive allomone of this animal. Obtained from i ts prey, a sponge, they are the active constituents of i t s mucous skin secretion. The rat io of the two isomeric isocyanides was highly variable from one sponge collection to another. Individual sponge specimens were found to occasionally produce nearly pure single isomers. An isocyano functionality is a rare occurrence in natural product chemistry. The majority of naturally occurring isoni t r i les have been isolated 7 from marine sponges, and they a l l appear to be of terpenoid biogenesis. Altogether, eight have been reported to date, seven sesquiterpenes and 19-23 one diterpene. In addition, a C ^Q diisocyano compound, which is probably not terpenoid in nature, has been isolated. Frequently the corresponding formamido and isothiocyanato derivatives can be isolated in conjunction with the ison i t r i les . For some sponges, at least, these formamides are thought to be the biogenetic precursors of the 20 24 isocyanides. ' In general, most of the terpenes discovered in sponges possess unique structural features, unparalleled in terrestr ia l sources. Isocyanides isolated from such land-based organisms are very few, only two appear to be known today and both are encountered as microbial metabolic A * 25-27 products. 4. Previous Syntheses of 9-Isocyanopupukeanane (5j and 9-Pupukeanone (7_) 9-Isocyanopupukeanane (5_), the biogenesis of which is unknown, represents the f i r s t known sesquiterpene with this particular type of t r i cyc l i c carbon skeleton. This fact , together with the presence of the seldom-encountered isocyano functionality makes the natural product an attractive synthetic target and i t is not much of a surprise that i t should catch the attention of synthetically active chemists. Indeed, whereas at the outset of this work in 1977 no synthesis of these compounds was known,up to now two total syntheses of S_ 28»29^ Q n e Q ^ ^ e . j s o m e r 30a ^ Q n e Q ^ 9-pupukeanone {]_) 3 ^ have been published. 3 7 A variety of ways can be proposed to construct the tr icyclo [4.3.1.0 1 ] -decane skeleton of the compound 7_. As outlined in section I I .2 of this thesis, our approach incorporated as the key step for constructing the carbon frame-work an intramolecular Diels-Alder cycloaddition, similar to that used by 8 Yamamoto et In contrast, the key reaction in the synthesis of 29 Corey et al_. , to set up the carbon skeleton was carbon-carbon bond formation by means of an intramolecular alkylation in a cis-hydrindanesystem. A short outline of the two syntheses wi l l be given, starting with the work of Yamamoto et al_. The a l l y l i c alcohol J_3, prepared by reduction of 3,5-dimethyl-2-cyclohexen-1-one, was heated in ethyl vinyl ether in the presence of mercuric acetate to yield the aldehyde 1_4 as product of the Claisen rearrangement. Reaction of T4_with vinylmagnesium bromide and protection of the resultant a l l y l i c alcohol as tetrahydropyranyl ether furnished the diene J_5 which was oxidized to the a,6-unsaturated ketone 16^with the chromium trioxide-dipyridine complex. Sequential treatment of 1_6 with l i thium diisopropylamide and chlorotrimethyl-silane produced the s i l y l enol ether 17, the desired compound for the in t ra-molecular Diels-Alder cycloaddition. The t r i cyc l i c keto alcohol 18 was 9 M e 3 S i°Y\ 0 T H P L160°c 1 8 i. HO OH ii. NCS,Me2S; Et3N 20 i. > - L i ii. H 30 + iii. CH3S02CI. NEt3 1 9 H 2 lr i. NH20H ii. TifDIBAH iii. CH3C02CHO iv. £-TsCl,pyr 21 R=NH2 5. R= NC 10 obtained by thermolysis of 1_7, followed by acid hydrolysis. The compound 18^was ketalized with ethylene glycol and the cyclopentanol oxidized to provide the cyclopentanone 1_9. Transformation of the lat ter compound into the endo-isopropyl compound 1_ could be achieved by the following sequence: (1) reaction of 19_with isopropenyllithium, (2) acid hydrolysis of the ethylene ketal funct ional i ty, (3) ter t iary alcohol dehydration with methanesulfonyl chloride-triethylamine to give the dienone 20_ and (4) highly stereoselective catalytic hydrogenation using iridium black as catalyst. Another four synthetic steps converted 9-pupukeanone {]_) into 9-isocyanopupukeanane (5_): (1) reaction of 7_ with hydroxylamine to the corresponding oxime, (2) reduction of this material with low-valent titanium followed immediately by di isobutyl-aluminum hydride (DIBAH) which provided the amine 21_, (3) transformation of 21_ to the i son i t r i l e 5_ by _p-toluenesul fonyl chloride-pyridine mediated dehydration of the corresponding formamide. In comparison, Corey's synthesis was realized via an entirely dif ferent route. Conjugate addition of isopropylmagnesium chloride to methyl 3-(4-methoxy-3-methylphenyl) propenoate (22) and acid-catalyzed cyclization of the resulting ester provided the hydrindanone 23_. The n i t r i l e 24_, as a mixture of stereo-isomers, was obtained by reaction of 23_with potassium tert-butoxide and tosylmethylisocyanide (TosMIC) with subsequent acidi f icat ion. The compound 24 was elaborated to the ester 25_ by base hydrolysis, ester i f icat ion and a methylation with l i thium diisopropylamide and methyl iodide. Treatment of 25_ with boron tribromide resulted in cleavage of both the methyl ester and the methyl ether. Catalytic hydrogenation of the resulting phenolic acid produced the 6-lactone 26_ having the desired cis-fused hydrindane ring system. By l ithium aluminum hydride reduction of 25, treatment of the resultant di ol with one equivalent of £-toluenesulfonyl chloride in pyridine and subsequent n CH=CHC02CH3 i. >-MgCl. Cu+ % ^ ^ 0 C H 3 i i . P P A 21 i-BuOK TosMIC H 3C. ,C02CH3 OCH3 2 5 i. K0H -H 20 2 i i . CH 2 N 2 i i i . CH3I.LDA OCH3 2U pyridinium chlorochromate oxidation the keto tosylate 27_ was prepared, which was to serve as substrate for the key internal alkylation reaction. This could be achieved by reaction of 27_with potassium tert-butoxide in te r t -butanol, yielding the t r i cyc l i c ketone 7_. Elaboration of this material to the desired isocyanide 5_ was done via a route very similar to that followed by Yamomoto et al_.: (1) conversion of 7_ into the oxime with hydroxylamine, (2) stereoselective catalytic hydrogenation of the oxime to the amine 21 and f ina l l y (3) dehydration of the formamide derived from 21_ with methanesulfonyl chloride in pyridine. 30a In a subsequent paper , Corey also reported the conversion of the intermediate 6-lactone 26^  to 2-isocyanopupukeanane {§) in a sequence involving an intramolecular aldol reaction. In addition, very recently a formal 12 , C 0 0 OTs i. BBr 3 ' i i ' i. LJAIH4 H3C I ii. £-TsCl V. 23. ii. H 2 CH 3 iii. PCC 2 5 21 P CH 3 ieri-BuOK ieri-BuOH 7 synthesis of 9-isocyanopupukeanane (5) has been accomplished by G.A. Schiehser pupukeanane framework was assembled using an intramolecular Diels-Alder approach. In this case, however, the dienophilic unit was conjugated to a carbonyl group which resulted in substantially milder conditions for cyclization. The dicyclopropyl adduct 29_ was prepared from the diene 28 by sequential treatment with diiodomethane-zinc and ethyl diazoacetate-copper. Acid - catalyzed double cyclopropane ring opening of 29 followed by ketaliza-tion of the intermittent a-enone produced the compound 30. Upon d i s t i l l a t i on i t eliminated methanol. Lithium aluminum hydride reduction of the diene thus obtained to the corresponding alcohol and oxidation of the la t ter gave the aldehyde 31_. The internal Diels-Alder substrate 32^  was synthesized from the compound 31 via addition of vinylmagnesium bromide to 31_ followed by oxidation of the resultant alcohol. Reflux of this material in benzene furnished the compound 33 as the product of the internal Diels-Alder and J.D. White 30b with the preparation of 9-pupukeanone (_7). Again the 13 )Me 13. 33 i. CH 2I 2 Zn(Ag) MeOv ii. N2CHC02Et h > ^ Cu OMe Et02C / C 6 H 6 2 9 32 MeOv ,OMe i. HCl ii. HC(OMe)3 C02Me 30 i. A ii. LiAlH* iii. pyr SO3 Me2SO.Et3N OMe i. - ^ M g B r ii. pyr -S03 Et3N. Me2SO CHO 31 i. J^U ii. Me2SO,A 20 1 cyclization. Addition of isopropenyllithium to the compound 33 followed by simultaneous ketone deprotection and dehydration of the ter t iary alcohol intermediate provided the diene 20_. Lastly, synthesis of 9-pupukeanone (7) was accomplished by catalytic hydrogenation of 20. 14 As the ketone 7_ represents a suitable compound for correlation with previous syntheses i t s preparation became the aim of our endeavour. 15 I I . DISCUSSION 1. Synthetic Planning - Strategy and Tactics or Retrosynthetic Analysis  and Synthon The increasing importance of chemical synthesis simultaneously brought a higher awareness of the problems involved in planning and successfully executing a natural product synthesis. The la t ter can be the experimental ver i f icat ion of a detailed plan. Guided by a synthetic strategy i t can also be, in i ts most important phases, the exploration of the react iv i ty of com-31 pounds with regard to their use as intermediates in a synthetic sequence. This search for new ways of structure modifications within a synthetic concept, receiving i t s stimulation and specific impulses from the target structure, part icularly applies to the synthesis of complex molecules such 32 33 as vitamin B^2 ' • The increased knowledge about the place of synthetic planning in conjunction with the development of many new methods in synthesis rendered i t necessary to systematize the subject and elaborate on ways of teaching i t . A number of books dealing with the strategy and tactics of 34 35 planning an organic synthesis have been published. ' In "Designing 35 Organic Synthesis" , S. Warren gives detailed instructions, i l lustrated by relat ively simple examples, toward this goal. By means of disconnections (breaking bonds) and Functional Group Interconversions (FGI, substituting one functional group for another) the target molecule is intel lectual ly broken down into consecutively simpler fragments, to the stage of easily available starting material(s). Appropriately this operation is called retrosynthetic analysis. Using suitable reagents or synthetic equivalents corresponding to the necessary synthons (generalized fragment produced by a disconnection) each intermediate is convertible to i ts retro-synthetic precursor, at least in theory, thereby allowing for the synthesis of the desired compound. 16 The ongoing systematic development of a synthetic methodology has been in i t ia ted mainly by E.J. Corey. He recognized, in 1967, that " i t is more than an intriguing theoretical exercise to set down in a generalized form the process by which a synthetic chemist devises an original but valid route to a complicated structure. Such an ef for t is a prerequisite to a deeper comprehension of synthesis and the methodologies that are fundamental to i t " . In further logical developments synthetic planning with the aid of computers 37 is an often discussed topic today. Its practical side is in a rather early experimental stage and individual opinions on i ts usefulness d i f fe r . 2. Envisaged Route A t r i cyc l i c compound such as 9-isocyanopupukeanane (5_) permits quite a number of possible synthetic approaches. A compressed retrosynthetic analysis reflecting our original ideas is laid out in Figure 1. The major synthetic ef for t would be the establishment of the carbon skeleton,for the target com-pound is rather devoid of functional i ty. The only functional group is the isocyano moiety and i ts preparation was envisaged from a suitable precursor, such as the ketone 7_, using standard methodology in the last steps of the synthesis. Transformation of 9-pupukeanone (7_) into the corresponding oxime with hydroxylamine followed by lithium aluminum hydride reduction should provide the amine 21_ presumably together with i ts diastereomer. Reaction of 21_ with methyl formate and dehydration of the resulting formamide with phosphorous oxychloride, a common method of preparation for isoni t r i les from 38 amines , most l ike ly would give the target molecule, 9-isocyanopupukeanane (5J, This proposed partial sequence has not been tested in the laboratory. However, the f inal stages of both Corey's and Yamamoto's syntheses are closely related and appear to work well (cf. section 1.4). In fact , the reduction 17 54 R=isobutyl R'=H 55 R=isobutyl R = C2H4OTHP Figure 1: Retrosynthetic analysis of 9-isocyanopupukeanane (5_) 18 of the oxime derived from the ketone 7_ to the amine 21_ (Corey: catalyt ic hydrogenation; Yamamoto: Ti° - diisobutylaluminum hydride) proceeded with high stereoselectivity. Presumably the C(3) methyl group exerts a profound steric shielding effect upon the attack of the reducing agent on the carbon-nitrogen double bond. An attractive theoretical alternative to this 4-step conversion of the ketone 7_ into the ison i t r i le 5_ involves the reductive 39 animation of 7_ with ammonium acetate and sodium cyanoborohydride in methanol to afford the primary anime 21_, followed by another, phase-transfer 40 catalyzed transformation, the reaction of 21_with chloroform and strong base to give the ison i t r i le 5_. Thus the preparation of 5_ from 7_ could possibly be reduced to two steps. The la t ter reaction can also be accom-plished under non-phase-transfer conditions, but the yields are lower. In neutral medium, sodium cyanoborohydride reduces ketones only very slowly, whereas carbon-nitrogen double bonds, as in iminium ions, formed in this case from ammonia and the ketone 1_ are easily reduced. Reasonably expecting a similar stereochemistry of reduction to that of the oxime, this reductive amination should be a viable method of preparing the amine 21_. The next reaction would involve attack of electrophil ic dichlorocarbene, generated from chloroform and base by a-elimination of hydrochloric acid, on the free electron-pair of the primary amine and subsequent base-induced double B-elimination of hydrochloric acid to the target compound 5_. The synthesis is now reduced to the preparation of the t r i cyc l i c ketone 7_, 9-pupukeanone. We planned to set up the carbon framework of 7_ in one step from a monocyclic precursor by an intramolecular Diels-Alder cycloaddition. 19 To this end, the a , B-unsaturated ketone 40 was envisaged as a key inter-mediate which should give on thermolysis the adduct 99_. The enol s i l y l ether of 99_ would be an immediate precursor to the keto group of 7_, while the free carbonyl moiety in 99 would serve as a handle for the introduction of the desired endo-isopropyl group at C(5). Thus 99_ would be expected to react with * 42 isopropylmagnesium bromide to provide the endo ter t iary alcohol 100. Dehydration of the la t ter substance with phosphorous oxychloride and 43 pyridine, or a similar procedure, could furnish the endocyclic olefin 101. Enol s i l y l ether hydrolysis by standard methods, such as tetra-n-butylam-44 monium fluoride , to give the compound 102 followed by catalyt ic hydrogena-45 tion which would undoubtedly proceed in the desired stereochemical direction , should then result in the formation of 9-pupukeanone (7_). How could one obtain the a , e-unsaturated ketone 40? To establish the quaternary center at C(5) of the 1,3-cyclohexadiene nucleus, an alkylation reaction introducing a side chain modifiable to an a-enone appeared to be feasible. This in turn necessitated the presence of an activating group on C(4) to be transformed afterwards to the C(3) - C(4) olef inic linkage. The second double bond of the diene is part of an enol ether moiety and should be easily available Reaction of ketones with Srignard reagents possessing a d i - or t r i -substituted nucleophilic carbon often results in enolate formation rather than addition.41 A good alternative would then be the use of isopropyl-l i th ium, which possesses higher nucleophilic power at low temperatures. 20 from the parent ketone by s i ly lat ion of i t s enolate anion with ter t -buty l -46 47 5i dimethylchlorosilane. Through use of a well-known reaction sequence, ' 1,2-hydride reduction of a B-alkoxy a, 6-unsaturated ketone followed by acid-catalysed hydrolytic rearrangement of the intermediate a l l y l i c alcohol to an a-enone (Scheme!) these desired transformations should be realizable from the A^OR LiAlH^ A^-OR H+ A^p 0 OH Scheme 1 structural ly simple compound 54. The la t ter substance could certainly be 48 prepared painlessly from 4,6-dimethyl-l,3-cyclohexadione (48), the pre-49 paration of which is described in the l i terature. In short, starting from the diketone 48, the following sequence represents an attract ive route to the compound 40: (a) conversion of 4J3 into the isobutyl enol ether 5^  with isobutyl 48 alcohol and £-toluenesulfonic acid, Cb) alkylation of 54_ at C(5) with a suitably functionalized alkylating agent to produce, for example,the compound 55_, (c) hydride reduction and acid hydrolysis to provide a 2-cyclo-hexenone (eg. 57), (d) conversion of the cyclohexenone to the corresponding enol ether (eg. 60), _ For reasons of simplicity the numbering of carbon atoms refers to the substituted 1,3-cyclohexadiene system as shown in the structural formula 40. I t is not necessarily in accordance with the proper nomenclature of the particular compound under consideration. * (e) modification of the C(5) functionalized substituent to give the a-enone 40. Two points in this suggestion require further comment: i . The ketone 54_ possesses two possible sites of proton abstraction by a strong, non-nucleophilic base, since both the a 1 and y proton are rendered acidic by the carbonyl group. Shown below are the two possible enolate anions that could be produced, 54a and 54b. Under conditions of kinetic control, preferential removal of the a ' proton ^8a,50 t Q a f -p 0 r c j t n e dienolate anion 54b should be possible, i i . The derivatized substituent introduced in the alkylation reaction [step (6), above] should contain functionality stable to acid, base and lithium aluminum hydride. I t should conveniently and under mild conditions be con-vert ible into the four-carbon Cf^ COChNC^ moiety. To this end a few electrophiles come to mind. One which would put the f u l l four-carbon chain with the double bond in place is 4-bromobutene. The derived substituent would be stable to the reagents mentioned above. However, 5Aa M = counterion 54b * See footnote, page 20. 22 the a l l y l i c oxidation, 3 4 — • 40, necessary after enol ether formation [step (d), above] would constitute a potential source of problems, for the cyclohexadiene system along with i t s attendant a l l y l i c positions would undoubtedly be quite susceptible to oxidation. A possible three-carbon alkylating agent is 2-methoxy-3-51 bromopropene , a disguised 1-bromoacetone equivalent. The derived substituent would not be expected to be stable to acid and would presumably give upon treatment with acid [step (c) , above] the diketone 35_. The la t ter substance could be a viable intermediate only i f two conditions are met. F i rs t l y , the methyl group attached to the side-chain carbonyl carbon would have to be con-vert ible to a vinyl group without interference from the cyclohexenone system. This could be feasible by the method of Gras o c using s-trioxan and N-methylanilinium tri f luoroacetate. Secondly, the derivative 36 thus obtained would have to be convertible into the desired dienol ether 40. This would appear possible by means of a hindered base such as lithium diisopropylamide or 53 l ithium 2,2,6,6-tetramethylpiperidide since the methylene protons adjacent to the carbonyl group of the substituent are on a ster ical ly shielded neopentyl-type carbon atom. For a two-carbon alkylation several electrophiles are available. One which shares with the just discussed 2-methoxy-3-bromopropene the fact that no oxidation at 54 C(2') would be necessary is 2-methylene-l,3-dithiane. I ts electrophil ic properties, shown by successful reactions 54 55 with carbarn"ons * are due to the capability of sulfur 56 to stabi l ize an adjacent negative charge. Reaction with a nucleophileproduces an intermediate carbanion that can be trapped with a suitable electrophile to give a compound 57 of the general structure 37a. Seebach et_ al_. have shown that dimethylformamide (DMF) can be used to attach an aldehyde function to a nucleo-l i phi l i e center. Recently, Meyers et a l . ^ developed another convenient reagent for such a direct formylation, 2-(N-methyl-N-formylami no)pyridine. The dienolate anion 54b, upon reaction with 2-methylene-l,3-dithiane and dimethylformamide should give the inter-mediate 37_ which then might be transformable with methylenetriphenylphosphorane into the olefin 38. Even though the aldehyde functionality of 37_ is quite 59 hindered, the Wittig reaction between this substance and a reactive yl ide might work. As dithianes can also be hydrolysed by acid another potential problem with this sequence might arise in the acid hydrolysis step necessary after l i thium aluminum reduction of the compound 38. Assuming successful synthesis of the a-enone 39_, formation of the corresponding dienol ether should not meet with d i f f i c u l t i e s , and dithiane hydrolysis with 55 60 mercury ( I I ) salts ' would f ina l ly give the a-enone 40, the proposed substrate for the internal Diels-Alder reaction. 21 A second, more readily available two-carbon alkylating agent is ethylene oxide. As epoxides are capable of nucleophilic ring opening i t s reaction with the dienolate 54b could generate the alcohol 41_ after hydrolysis. Being a gas ethylene oxide is more d i f f i c u l t to apply as a measured reagent, and the i n i t i a l alkoxide 41a could combine with excess epoxide. Furthermore, i t * might easily attack internally electrophil ic C(4) giving closure to the tetrahydrofuran derivative 42^which might undergo further transformations after hydrolysis. 42 R = c o u n t e r i o n M R = H 41 a R = c o unter ion This problem, however, is conveniently avoidable by use of a third two-carbon alkylating agent, an O-protected derivative of 2-bromoethanol. This alcohol protecting group has to meet four requirements. Its introduction into 2-bromoethanol would have to be carried out in neutral or acidic medium, for basic conditions would certainly result in ethylene oxide formation. I t has to tolerate a reductive and acidic environment during the formation of a 2-cyclo-hexenone, e.g. 55_ to 57. I t has to be removable without * See footnote, page 20. 26 cleavage of the dienol ether moiety such as in 60_, and i t has to be stable toward strong base such as that used for enolate anion formation. One common method for alcohol protection under acid catalysis is formation of the corresponding tetrahydropyranyl ether. ^ Another popular alcohol protecting group introduced under neutral conditions 62 is the tert-butyldimethylsi lyl ether. Both meet the con-ditions one and four mentioned above. Whether or not they would f u l f i l the requirements two and three would have to be tested by laboratory experiments. Assuming suff icient acid s tab i l i t y , successful selective hydrolysis would yield the primary alcohol 43_. Oxidation of this compound by one of a 63 variety of available methods and reaction of the resultant aldehyde with a vinyl-anionic compound, such as vinyl l i th ium, 27 can reasonably be expected to produce the a l l y l i c alcohol 44. I ts oxidation with manganese dioxide or another suitable oxidant to the a-enone 40 is al l that is then necessary to obtain the substrate for the envisaged internal [4+2] cycloaddition. Because of i ts ease of preparation and the prospects of a relat ively straightforward and trouble-free transformation to an a , e-unsaturated ketone group, the tetrahydropyranyl ether of 2-bromoethanol ^ was chosen as alkylating agent. 3. Preparation of the Dienol Derivatives 60_and 63_ As outlined in section I I . 2 , the isobutyl enol ether 54_ possesses a suitable structure for eventual transformation into a dienol ether of type 48a 60 or 6_3. Its l i terature preparation ut i l izes the conversion of 4-methyl-1,3-cyclohexanedione (45) into a mixture of the two isomeric 3-isobutoxy-2-cyclohexen-l-ones 46 and 47, their separation by column chrom-atography, and methylation of the r ight isomer 46 to give 3-isobutoxy-4,6-dimethyl-2-cyclohexen-l-one (54). Two features, the necessary separation of isomers coupled to the preponderance of the wrong isomer 47 (46 : 47 = 1 : 2) in the product of the f i r s t step render this sequence unattractive. 49 In 1958 H. Stetter and U. Milbers reported the preparation of 4,6-dialkyl-1,3-cyclohexanediones by condensation of ethyl 2-alkylacetoacetates with ethyl 2-alkylacrylates. Thus, for example, ethyl methacrylate and ethyl 2-methylacetoacetate gave a 69% yield of 4,6-dimethyl-l,3-cyclohexanedione (48). This symmetrical diketone, 48, upon acid-catalysed reaction with an alcohol can form only one enol ether thereby effacing the disadvantages of the alternative preparation of the compound 54. Thus, i t was decided to use 28 48 as the starting material in our synthetic work. Addition of ethyl methacrylate to the sodium salt of ethyl 2-methyl-acetoacetate followed by base hydrolysis, decarboxylation and acidi f icat ion yielded the diketone 48 as a colourless, crystall ine solid in 61% y ie ld . I t could be recrystallized from water or ethyl acetate-hexanes, and sublimed at 95°C (0.1 Torr)_. Its infrared spectrum showed three characteristic 29 absorption bands in the carbonyl region, at 1735 and 1710 cm , attributed to the diketo form 43a, and a weaker band at 1604 cm - 1 , assigned to the enol form 48b. 6 4 2,2,5,5-Tetramethyl-1,3-cyclohexanedione (49_), where enolization is prevented, shows only two strong absorbances near. 1700 cnf^ whereas enolizable 1,3-cyclohexanediones such as the parent compound (50_), dimedone (51), the 4-methyl (5_2) and 2-methyl (53) derivatives exhibit a strong band near 1600 cm~^. R' 5Q R = H 52. R = H R = C H 3 51 R = C H 3 5 1 R = C H 3 R'=H The intensity of the 1604 cm" band of the diketone 48 was found to be dependent on concentration. Relative to the 1735 cm~^  band, i t decreased with increasing d i lu t ion. Apparently at higher concentrations the equilibrium 48a 48b shifted more to the side of the enol form. The existence, in solution, of such an equilibrium as well as i t s concentration dependence are 65 well-known for 1,3-diketones. For 1,3-cyclohexanediones the enol form is usually the predominant species even in di lute solution. In the present case, the i r spectrum of 48 leads to the conclusion that in a nonpolar solvent, in di lute solution, the equilibrium 48a^48b heavily favours the diketo form. The nmr spectrum of the compound 48_ clearly confirmed th is . I t exhibited a hydroxy! proton resonance at 6 8.12 ppm, two weak vinyl proton resonances at 6 5.44 and 5.38 ppm, and an AB-pair of doublets at 6 3.52 and 3.37 ppm, arising from the C(2) methylene protons. Two separate vinyl proton resonances were obtained because the two methyl substituents could be in a 30 cis or trans orientation to each other. All these resonances disappeared upon addition of D^O indicating a rapid equilibrium between the keto and enol tautomers of the isomers of the compound 48. The content of the diketo form could be estimated at 80% by comparing the intensity of the vinyl proton to the a methylene proton resonances. When the nmr spectrum of 48 was taken in very di lute solution no hydroxyl and vinyl proton resonances were observed. The low extent of enolization observed in the i r and nmr spectra of 4,6-dimethyl-l,3-cyclohexanedione (43) is surprising in l ight of the known propensity of 1,3-diketones to exist largely in the enol form. Examination of molecular models provided an explanation. In the cis-isomer of the enol 48b the two methyl substituents can be either pseudoequatorial or pseudoaxial. In the la t ter case a 1,3-diaxial interaction occurs whereas in the former case A ^ ' ^ - s t r a i n 6 6 between the C(3) hydroxyl substituent and the C(4) (1 Z) methyl group is found. In the trans configuration of 48b, A v ' -strain c i s - £ 8 b trans-48b does not exist. An equivalent to a 1,3-syn-axial interaction i s , however, introduced. Part of the strain inherent in the enol form can be relieved by adopting the diketo form 48a. In part icular, the diequatorial conformation of cis-48a w i l l be more favourable than either cis-48b, because of the absence f l 2) of A ' - s t ra in , or the diaxial conformation of cis-48a with i t s 1,3-diaxial interaction. Transition from the enol trans-48b to the diketo conformer trans-48a, on the other hand, does not lead to an appreciable change in strain 31 c i s -A8g trans -48a diequatorial diaxial energy since in both compounds a 1,3-syn-axial interaction prevails. The low content of the enol form 48b in a di lute solution of the compound 48 thus appears to be associated largely with i t s trans isomer, for the cis isomer would presumably exist to a high degree in the diequatorial conformation cis-48a. In more concentrated solutions where, via intermolecular hydrogen 65 bonding, dimeric and polymeric species are known , this association could provide enough stabi l izat ion energy to enhance formation of the enol form 48b. Through an established procedure (excess isobutyl alcohol, catalytic amount of p_-toluenesulfonic acid, refluxing benzene with continuous removal of water) the diketone 48 was cleanly and ef f ic ient ly converted into the enol 48 ether 54, a clear, colourless l iqu id . Formation of 3-isobutoxy-4,6-dimethyl-2-cyclohexen-l-one (54_) was confirmed by both i t s i r and nmr spectra. The i r spectrum displayed strong absorbances at 1660 and 1596 cm~\ 64 1 characteristic of a e-alkoxy a-enone moiety. The H nmr spectrum revealed the presence of two diastereomers since two dist inct vinyl proton resonances (6 5.31 and 5.26 ppm), two doublets due to the OCFL group (6 3.61 and 3.59 ppm) £8. 32 and four doublets due to the cyclohexenone methyl groups (6 1.26, 1.18, 1.14, 1.13 ppm) were obtained. Although the usual position for the resonance of ana proton of an a-enone system is close to 66 ppm, the electron-releasing B-alkoxy substituent caused a 0.7 ppm upfield sh i f t of this resonance. The vinyl proton of one of the diastereomers resonated as a doublet with J = 1.5 Hz, due either to a l l y l i c coupling or to coupling across the carbonyl group. On the basis of the intensities of the vinyl proton resonances the rat io of diastereomers was estimated to be about 1:1. The H^ nmr data of 54 showed good agreement with the published values except that the la t ter data did not seem to originate from an isomeric mixture. I t was not necessary at this or any later stage to attempt separation of the diastereomeric mixture of 54 into the pure isomers because both would ultimately yield the same product* The next step in the synthesis required the introduction of the second C(6) substituent. To this end, a-bromoethanol was converted into i ts tetrahydropyranyl ether ^ hy treatment with dihydropyran and jp_-toluene-sulfonic acid. The resulting tetrahydropyranyl ether was then used as 67a electrophile. Thus, alkylation of the kinetic l ithium enolate of 54 (54b, cf . section I I .2) at -78°C to ambient temperature afforded the desired product 55_ in 80% y ie ld . The reaction was slow even in the presence of hexamethylphosphoramide, presumably reflecting the formation of a crowded quaternary carbon atom. A small amount of unreacted starting material, 5-10%, 33 was usually recovered. The mass and H nmr spectra of the product 55_ con-firmed the introduction of the substituted alkyl group. In the former, the further peaks corresponding to loss of the OTHP and tetrahydropyranyl groups from the molecular ion. The presence of the tetrahydropyranyl ether was indicated in the nmr spectrum by a broad singlet at 6 4.50 ppm, arising from the methine proton adjacent to two oxygen atoms, and a multiplet at 8 3.97-3.22 ppm caused by six protons on three methylene carbons a to oxygen atoms. Some other features of the H^ nmr spectrum of 55_ were a sharp singlet at 6 1.08 ppm, due to the methyl group on the quaternary carbon C(6) , and a doublet at 6 0.97 ppm (J=6.5Hz) due to the isobutyl methyl groups. The fact that 55_ consisted of a mixture of stereoisomers was shown by the 270 MHz H^ nmr spectrum in which multiple methyl resonances and two dist inct doublets for the isobutyl methyl groups occurred. The i r spectrum of 55_ exhibited bands at 1658 and 1598 cm"^, verifying the integri ty of the e-alkoxy a-enone part of the molecule. The transformation of the 3-alkoxy-2-cyclohexen-l-one 55_ by hydride 47 50 reduction and acid hydrolysis ' to the 4,4,6-tr isubstituted cyclohexenone 57_ could be achieved without problems. Reduction of 55_with l ithium aluminum hydride in ether at 0°C followed by hydrolysis of the resultant product with a 1:1 mixture of acetic acid and acetone produced the desired a,3-unsaturated ketone 57_ in 80% y ie ld . No attempt was made to isolate and characterize the I i. L iA l rU f l l I base peak was due to loss of a -CH„CH?0THP moiety [m/e 324(M+)—> 196] with . 0 55 56 5Z 34 a l l y l i c alcohol 56_, which was involved as an intermediate in this reaction. The spectral data obtained from compound 5_7 were as expected. Retention of the tetrahydropyranyl ether group was clearly shown by a broad, one-proton singlet at 6 4.53 ppm of the nmr spectrum. Further features of the la t ter were resonances for the vinyl protons at 6 6.61, 6.59 and 5.91 ppm, character-i s t i c positions for the 8 and a protons, respectively, in 2-cyclohexen-l-ones, which confirmed the formation of the a-enone. The 6 5.91 ppm signal was, as expected for a cis o le f in ic , vicinal coupling, a doublet with J = 10 Hz. The 3 proton resonated at 6 6.61 and 6.59 ppm as a pair of overlapping double doublets (J=10,2Hz), giving the appearance of two t r i p le ts . In addition, weak signals immediately to lower f ie ld of, and in between, the two ' t r i p l e t s ' were observed (66.72,6.60ppm). Double resonance experiments helped to explain this pattern (Figure 2). Upon irradiat ion of the a proton (65.91ppm), the resonance a) b) c) d ) e) f ) Figure 2: & Proton resonance of 57_. a) normal run; i rradiat ion at b) 1.56 ppm c) 1.70 ppm d) 1.90 ppm e) 2.00 ppm f ) 5.83 ppm due to the 6 proton ^6.61,6.53ppm) collapsed to a mult iplet. Stepwise irradiat ion from 6 1.56 to 2.00 ppm produced signal sharpening and transforma-tion of the ' t r i p l e t s ' to two apparent doublets, the individual lines of each 35 being spaced apart by 2 Hz. Irradiation of the multiplet appearing between 6 2.1 - 1.4 ppm eliminated possible long-range coupling to the 3 proton. The observations described above and shown in Figure 2 may be rationalized by assuming that three of the four possible diastereomers of the compound 5_7 give rise to three dist inct 3 vinyl proton resonances. The chemical sh i f t of the 3 proton of the fourth diastereomer must be identical to that of one of the other three isomers. Each of the three resonances is sp l i t into a doublet (J=10Hz) by the adjacent a proton. Two of these doublets are coincidentally separated by 2 Hz, and by further long-range coupling of these protons to one other proton with a 2 Hz coupling constant a ' t r i p l e t ' w i l l be formed. The weaker resonances at 6 6.72 and 6 6.60 ppm are, in fact , the doublets of the third diastereomer (J=10Hz) which does not exhibit long-range coupling. Due to l ine broadening i t is par t ia l ly hidden in the fu l l y coupled spectrum. The fact that this long-range interaction gives rise to doublet formation suggests coupling to one proton of the C(5) methylene group. The relat ively r ig id cyclohexenone ring would provide for the necessary near-planarity of the W-type 68 coupling across four single bonds. A strong, broad absorbance band at 1680 cm"^  in the i r spectrum of 57_ was confirmatory evidence for the formation of an a-enone system. When the l ithium aluminum hydride reduction of 55_ was carried out at ambient temperature instead of at 0°G and with a larger excess of reducing agent, the desired product 57 was isolated in lower yield and a small amount (8%) of a non-ketonic compound was obtained after chromatography of the crude product. The i r spectrum of the non-ketonic material was devoid of any easily assignable absorption. The H^ nmr spectrum displayed a two-proton multiplet at 6 5.93 - 5.45 ppm, a broad one-proton singlet at 5 4.59 ppm, indicating an intact tetrahydropyranyl ether group, a multiplet between 6 4.01 and 3.03 36 ppm corresponding to seven protons and, among others, resonances at 6 1.09 -0.85 ppm caused by 12 protons. This last fact together with the high number of protons in the OCh^  region indicated the presence of the isobutyl ether group. A fragment in the mass spectrum arising from loss of a C^ Hg radical from the parent peak[m/e 310(M +)—• 253]further emphasized this deduction. These observations coupled with microanalytical data that pointed to a molecular formula of C-ig^Og could best be accommodated by assigning the structure 59_ to this compound. I t could conceivably have arisen from either nucleophilic attack at 0(3) of the intermediate 5J3 by excess hydride with elimination of an oxo-aluminum species or via a similar elimination in i t ia ted by the electron-releasing oxygen substituent on C(3), and subsequent neutraliza-tion of the resulting carbocation by hydride. With the properly functionalized cyclohexenone 57 at hand, the next envisaged step was i ts conversion into a diene via the formation of a dienol ether. The la t ter functionality would, hopefully, serve a dual purpose. F i rs t l y , i t would act as the diene moiety in the proposed internal Diels-Alder cycloaddition. Secondly, i t would preserve the keto funct ional i ty, to be regenerated after cyclization. As i t was intended to eventually convert the 37 C(4) oxygen functionalized substituent to a four-carbon unit containing an a-enone (to serve as the dienophile moiety in the Diels-Alder reaction) the enol ether functionality should possess electron-releasing properties. For - O S i ^ ^OTHP 6Q O S i ^ Diels-Alder reaction O S i ^ 40 12 this purpose either alkyl or s i l y l enol ethers are usually employed. The tert-butyldimethylsi lyl derivative was chosen because i t is prepared under mild conditions by s i ly la t ion of the corresponding enolate anion. The l i thium di enol ate of 57, formed by adding the compound 57 to a solution of l i thium diisopropylamide in tetrahydrofuran, was allowed to react with tert-butyldimethylchlorosilane in the presence of hexamethylphosphoramide .0 i. LDA.THF ii. HMPA.dSi^ ^ O R 60 R=THP 43 R=H at 0°C and then at ambient temperature. The desired dienol ether 60 was thus obtained in a yield of 80%. Usual spectroscopic examination quickly established the identi ty of the compound 60. In i ts nmr spectrum, a six-proton singlet at <5 0.12 ppm and two singlets at <5 0.97 and 1.02 ppm, due to the tert-butyl and the C(5) methyl 38 substituents and together accounting for 12 protons, affirmed the incorporation of the t r i a l k y l s i l y l moiety. All other expected resonances for the s i l y l enol ether 60 were observed. The infrared spectrum of 60 showed absorbances at -1 69 1664 cm for the conjugated double bonds in the cyclohexadiene nucleus , and at 1255, 1031 and 841 cm~\ verifying the presence of the t r i a l k y l s i l y l * u 64,70 ether group. ' With no doubt about the structure of the 1,3-cyclohexadienol ether 60, elaboration of the C(5) ethano substituent to establish the dienophilic part of the molecule was started. The tetrahydropyranyl ether had to be hydrolyzed without destruction of the s i l y l enol ether funct ional i ty. Both protecting groups are labi le toward acid. However, i t was f e l t that preferential cleavage to the alcohol 43_ should be feasible because the tert-butyl dimethyl s i l y l enol ether represents a relat ively more stable functionality compared with alkyl enol ethers or the more common tr imethylsi ly l enol ethers. ^ Unfortu-nately, a l l attempts to cleanly achieve selective tetrahydropyranyl ether cleavage fa i led. For example, treatment of the compound 60_with glacial acetic acid, water and tetrahydrofuran (4:1:1) at room temperature for 5 hours l e f t more than 80% starting material. Prolonged reaction times had l i t t l e effect, while raising the temperature produced a multitude of products. With 0.1N hydrochloric acid in tetrahydrofuran and methanol at room temperature, after 23 hours, a mixture consisting of 16% starting material and four other 7° compounds was obtained. Acidic ion exchange resin (Dowex 50W-X8) in methanol at ambient temperature yielded, after 21 hours, a mixture of 48% starting material and two considerably more volat i le compounds. Dilute 67a perchloric acid in tetrahydrofuran (0.05%,1;1) , after 28 hours at room temperature, l e f t 78% starting material. Recently, Grieco reported the preparation and use of pyridinium p_-toluenesulfonate as catalyst for the introduction and removal of the 39 tetrahydropyranyl protecting group. / J This salt had l i t t l e effect at room temperature in ethanol, producing after 4Jg hours a mixture consisting of 73% starting material and several other compounds. At 60°C, in the same solvent, the major product obtained after 4 hours displayed no hydroxyl absorbance in the i r spectrum but did exhibit a carbonyl band at 1718 cm" ^ . The mass spectrum of this product showed the molecular ion at m/e 168. These two pieces of information supported the tentative assignment of the structure 62 to this material. As both the s i l y l enol ether and the tetrahydropyranyl ether groups are acid-sensitive i t could possibly have arisen by simultaneous 60 61 62 hydrolysis of both ethers to give the intermediate 61_, followed by in t ra-74 o molecular cycl ization. Boron t r i f luor ide etherate in ether at 0 C, upon reaction with 60^  for 2 hours, produced as crude product a mixture consisting mainly of starting material as well as several other compounds. Treatment of the compound 6Q with the same reagent in ether for 4 hours at room temperature resulted in part ial conversion to one major and several minor products. Investigation of this mixture by i r spectroscopy did not indicate much alcohol formation, therefore use of this reagent was not further explored. I t became increasingly apparent that preferential tetrahydropyranyl ether cleavage in 60_ was d i f f i c u l t . To avoid this problem, a more easily removable alcohol protecting group or a dif ferent enol ether functionality could be used. The la t ter alternative was chosen. 40 In the reductive deoxygenation of ketones to olefins the appropriate enol phosphates or phosphorodiamidates are often employed as alkene precursors. * Less frequently these are used purely for the purpose of protecting a carbonyl function as an enol derivative. As phosphate esters or amidates can be hydro-75a 76 lyzed by acid or base * they are in principle a viable alternative for carbonyl group protection. Their usefulness in relation to the above problem was investigated. The l i thium enolate of the compound 57_, obtained by treating 57 with l i thium diisopropylamide, was phosphorylated with diethylphosphorochloridate in tetrahydrofuran-tetramethylethylene diamine. The reaction proceeded smoothly at room temperature and a 74% yield of the product 6_3 was realized. High resolution mass spectrometry established molecular composition of C-jgH^^ OgP for the compound 63. The presence of the enol phosphate group in 63_ was confirmed by both the H^ nmr and i r spectra. In the former, a four-proton multiplet at 6 4.28 - 3.98 ppm characterized the phosphate CCl^  group. Vicinal coupling to the methyl group and long-range coupling to phosphorus generated the multiplet resonance. Since Jp_g_cH = 7~^° H z » w n i c n i s close to the value of regular vicinal proton-proton coupling, this resonance was observed with nearly a l l compounds containing the enol phosphate moiety as a quintet- l ike mult iplet. Smaller long-range coupling to phosphorus in the order of 1Hz, was exhibited by the corresponding methyl protons of 63, 41 resonating as a double t r i p l e t at <5 1.30 ppm. The presence of the tetrahydropyranyl ether was indicated again by a broad singlet at 6 4.53 ppm. The vinyl protons gave rise to an AB-pair of doublets at 6 5.83 and 5.46 ppm. Relative to the s i l y l enol ether 60_ this represented both an overall downfield shi f t as well as an increased separation of the two doublets. In the i r spectrum of the compound 6_3 the intense bands at 1278 and 1030 cm"^  and a -1 64 weaker one at 1160 cm were characteristic for the diethyl phosphate ester. The conjugated diene moiety showed a medium intensity absorbance at 1675 cm" \ shifted to higher frequency by 11 cm"^  compared with the s i l y l enol ether 60_. In the 270 MHz nmr spectrum an AB-quartet at 6 2.29 and 2.05 ppm with J = 16 Hz was observed for the a l l y l i c methylene protons of 63_. Convinced of the identi ty of the enol phosphate 63_ the problem of selective tetrahydropyranyl ether hydrolysis was reinvestigated. 4. Synthesis of the A l l y l i c Alcohol 69 To obtain the necessary dienophile unit for the intended cycloaddition reaction the envisaged route incorporated the deprotection of the compound 63_ to the primary alcohol 64, oxidation of the lat ter to the aldehyde 65_, reaction of the compound 6_5 with vinylmagnesium bromide and, f i n a l l y , oxidation of the resultant a l l y l i c alcohol to the a, 3-unsaturated ketone 71. OP(OEt)2 [0] OP(OEt)2 ^ ^ M g B r IO] 71 0 0P(0Et)2 42 After i n i t i a l problems reminiscent of those with the s i l y l enol ether 60, the selective hydrolysis of the tetrahydropyranyl ether 63_ to the alcohol 64 could be achieved in high yield by using IN hydrochloric acid in methanol. The spectral data for the primary alcohol 64 were quite similar to those of i t s precursor, the compound 63. In the nmr spectrum of 64, the retention of the enol phosphate group was confirmed by a multiplet resonance at 6 4.38 -4.02 ppm and a double t r i p l e t at 6 1.36 ppm, arising from the 0CH2 and methyl protons, respectively, of the phosphorus ethoxy substituents. A broad singlet at s 2.53 ppm, exchangeable with D20, was due to the hydroxyl proton. In the i r spectrum, a strong absorption at 3443 cm - 1 , arising from the hydroxyl group, as well as bands at 1270, 1160 and 1038 cm"1 from the phosphate ester corrobo-rated the formation of the compound 64_. The effect of phosphorus long-range coupling to protons was clearly i l lustrated in the " 'H nmr spectrum o f f h e enol 31 phosphate 64_. Upon P decoupling the methyl and methylene proton resonances of the enol phosphate moiety changed to a t r i p l e t (J=7Hz) and two quartets (J=7Hz), separated by 0.8 Hz,respectively. Presumably the two ethoxy sub-stituents are not equivalent and give r ise, at least in the case of the methylene protons, to two distinguishable resonances. A broad two-proton doublet at 6 2.17 ppm, arising from the a l l y l i c methylene protons, collapsed to a broad singlet. The a l l y l i c methyl proton resonance, a broad singlet at 31 6 1.79 ppm, gave a sharp singlet with increased intensity when P coupling 43 was eliminated. The two la t ter cases involved homo-ally!ic phosphorus-proton coupling across f ive bonds. The C( l ' ) methylene protons of the C(5) 2*-hy-droxyethyl substituent exhibited geminal 0 I II coupling in the order of 14 Hz. Only the r ^ / O P ( O E t ) 2 center section of the multiple resonance . produced by further vicinal coupling of the OH C(l ) methylene protons to the adjacent C(21) methylene protons was observed, because the outer lines were buried under the adjacent methyl resonances. The interpretation indicated below was borne Jgem = U H z J v j c = 7 H z . : •. • i > / i f fl Ii i i I observed out by the 270 MHz H^ nmr spectrum. The AB-pair of doublets, which was due to geminal coupling, was pulled apart to the extent of 21 Hz. Further inter-action with the C(2') protons sp l i t each l ine into a t r i p l e t . The a l l y l i c methylene protons appeared in the 270 MHz H^ nmr spectrum as a double AB-31 quartet at 6 2.26, 2.15 ppm with geminal coupling of 16 Hz and P coupling of 4 Hz. Oxidation of a primary alcohol to an aldehyde is a very common operation in organic chemistry. Many different reagents reflecting the diverse require-ments dictated by the overall sensit iv i ty of the molecule have been developed 63 for this purpose , and the choice in each individual case i s , to a certain degree, an arbitrary one. The alcohol 64 possessed an acid and base labile phosphate ester group, therefore only f a i r l y neutral oxidizing agents were 44 worth considering. Among these Cr(VI) and dimethyl sulfoxide based reagents play an important role. hours and at ambient temperature for another hour produced the desired aldehyde 65_, but both the purity of product and the isolated yield were 78 disappointingly low. Pyridinium chlorochromate , a mildly acidic oxidizing agent, in dichloromethane at room temperature for 1.5 hours, with sodium acetate present as buffer, yielded only an unseparable mixture of products from the alcohol 64. From a second oxidation with pyridinium chlorochromate, without sodium acetate buffer and after 2.5 hours at ambient temperature one product could be isolated in 37% yield after puri f icat ion by preparative t i c . Glc analysis of this product showed a signif icantly shorter retention time than the starting material. In the i r spectrum of this material, a strong band at 1690 cm - 1 , a weaker one at 1648 cm - 1 and further absorbances at 1272, 1165 and 1030 cm - 1 indicated the formation of an ct-enone and the retention of the phosphate ester moiety, respectively. High resolution mass spectrometry indicated a molecular formula of 4^23^6^ ^ o r * ' 1 1 S c o r T 1 P o u n c ' ' ^ n e ^ H n m r spectrum reaffirmed that this compound was an enol phosphate. Both the a l l y l i c methyl and methylene proton resonances had undergone a downfield sh i f t , relative to the precursor alcohol 64, to <5 2.00 and 2.59 ppm, respectively. 31 The collapse of these two signals, upon P i r radiat ion, to two singlets con-firmed their assignment as due to the a l l y l i c methyl and methylene protons. Collins oxidation 77 of the alcohol 64 in dichloromethane at 0° for 0.5 65 45 In the oxP decoupled 'H nmr spectrum the OCF^  protons of the enol phosphate group appeared as two dist inct quartets, separated by 3.5 Hz. As in the case of the alcohol 64, non-equivalence of the two ethoxy substituents could be the reason for these two separate resonances. I f the assignment 66_ to this material is correct i t could also be due to cis-trans isomerism around the ring junction. A three-proton multiplet at 6 4.05 - 3.89 ppm indicated three protons in an a-position to an oxygen atom. On irradiat ion at 6 1.17 ppm this resonance changed to a sharp singlet at 6 3.96 ppm and two signals in close proximity at 6 4.00 and 4.02 ppm. The sharp singlet (63.96ppm) was found to be superimposable on a prominent single l ine of the multiplet in the fu l l y coupled spectrum when the decoupled and coupled spectra were compared with each other. Comparison of the coupling pattern of this 6 4.05 - 3.89 ppm resonance to that obtained in the same region of the spectrum for the compound 74_ (cf. section I I .5) revealed a close s imi lar i ty . The former multiplet resonance can be interpreted as being the same four-l ine resonance pattern as that obtained for the C(2') methylene protons of the compound 74_ admixed with a one-proton singlet due to the C(6) methine proton. No vinyl protons were observed in the H^ nmr spectrum of this material. All of these facts could be accommodated by tentatively assigning the structure 66 to this compound. The additional carbonyl conjugation accounted for the observed downfield sh i f t , in comparison with the alcohol 64, of the a l l y l i c methyl and methylene protons in the H^ nmr spectrum of'66_. A plausible mechanism IN 66 46 OP(OEt)2 for the formation of the a-enone 66_ would involve an acid-catalysed closure to a tetrahydrofuran derivative, followed by an a l l y l i c oxidation. Chromium (VI) complexes have been used to accomplish oxidation of vinyl ic methylene 79 carbon atoms. The identity of this material with the compound 66 cannot be def in i te , however, because there was another piece of confl ict ing evidence. In the i r spectrum, a hydroxyl absorbance at 3450 cm - 1 , and in the "'H nmr spectrum a decrease in the intensity of the 6 1.92 - 1.65 ppm multiplet resonance upon addition of D£0 was observed. Yet no structure incorporating an alcohol functionality could be thought of which reconciles a l l the other available evidence. Possibly the compound was very hygroscopic or, despite tedious puri f icat ion efforts involving chromatography and d i s t i l l a t i on an alcoholic impurity was retained. Treatment of the alcohol 64 with phosgene in benzene at ambient temper-ature overnight, followed by addition of dimethyl sulfoxide at 0°C and of 80 triethylamine after one hour at room temperature (Barton oxidation ) produced the aldehyde 65_ in 65% y ie ld . Some starting material was recovered as wel l . The use of highly toxic phosgene, however, constituted a drawback for routine preparations of 65 by this method. Another mild and well-established oxidation, based on dimethyl sulfoxide, 81 is that of Moffatt et al_.. By using a modified carbodiimide instead of dicyclohexylcarbodiimide, as i t is used in the original oxidation, the product 47 isolation was found to be greatly simplif ied. 82 In our work, 1-ethyl-3-(3 -dimethyl aminopropyl)carbodiimide hydrochloride (67) subsequently was found to be a convenient and reliable reagent in the oxidation of the alcohol 64 to the aldehyde 6_5 with dimethyl sulfoxide. Sequential addition of dimethyl sulfoxide, pyridine, t r i f luoroacet ic acid and the carbodiimide 67_ to a solution of the alcohol 64 in benzene at ambient temperature routinely generated the aldehyde 65 in yields of around 90%. After column chromatography, 70-80% yields of the aldehyde 65_ were obtained. The crude product, however, could be employed for further reactions for the oxidation proceeded in a very clean way. In the nmr spectrum of the product 65_ the aldehyde proton resonated as a t r i p l e t (J=2.5Hz) at 6 9.76 ppm. This small vicinal coupling was also present in a two-proton multiplet resonance at 6 2.36 ppm, due to the a methylene protons. In the 270 MHz spectrum, the la t te r , together with the a l l y l i c methylene protons of 65_, appeared at & 2.41 - 2.07 ppm as two over-lapping AB - quartets ( J g e m= l 6 >14Hz) , sp l i t further in the magnitude of 3 Hz due to coupling to the aldehyde proton and to phosphorus, respectively. The mass spectrum of the aldehyde 65_ featured a prominent loss of a ^2^3® t " r a 9 m e n t from the molecular ion [m/e 302(M +) :—• 259] to give the base peak at m/e 259. This predominant expulsion of the C(5) oxygen functionalized substituent in the mass spectrum was observed throughout the series of C(5) disubstituted OH 6 5 48 enol phosphates, and nearly always generated the base peak at m/e 259. The la t ter corresponded to the resonance stabilized a l l y l i c cation 68. The 6 8 R = P(0)(0Et)2 presence of an aldehyde functionality in the compound 65_ was further indicated by an i r absorption at 1718 cm \ The transformation of an aldehyde to an a l l y l i c alcohol requires a vinyl ic 83 nucleophile. H. Normant et al_. succeeded in 1954 in preparing vinyl ic Grignard reagents, including vinylmagnesium bromide. In contrast to a lky l -magnesium halides, the analogous alkenyl compounds can only be prepared in 41 tetrahydrofuran and are not formed in ether. The a l l y l i c alcohol 69_ could be prepared in high yield by adding the aldehyde 65_ to freshly prepared vinylmagnesium bromide in tetrahydrofuran at ambient temperature. On occasions, part icularly in small-scale reactions, 0 fl J ^ r j p < 0 E t , 2 J ^ O P ( O E t ) 2 f M = MgBr: THF f CHO M = Li : Et 20 W T ^ ^ 6 5 6 9 some aldehyde 65_ was recovered from the crude product. This problem could be eliminated by employing a standardized solution of vinyl l i thium in ether as nucleophilic reagent. The advantages of the la t ter l i thium reagent over the corresponding Grignard derivative are i ts higher react ivi ty at low temperature, resulting in substantially milder conditions. Moreover, external 49 addition of the nucleophile rather than in situ preparation, as in the case of the Grignard reagent, offers better control of the desired quantity of reagent. With vinyl l i thium in ether, the alcohol 69_ could routinely be synthesized from the aldehyde 65_ in high y ie ld . I t was formed as a mixture of diastereomers that could be separated into the individual isomers by column chromatography on F l o r i s i l . The H^ nmr spectra of both isomers of the compound 6_9 confirmed the formation of an a l l y l i c alcohol. A five-proton multiple resonance between <5 5-6 ppm was assigned to the vinyl protons. The resonance due to the a l l y l i c methine proton was hidden under the multiplet resonance of the phosphate ester OCH2 protons. In the more polar isomer of 69_ the resonance of these ethoxy methylene protons showed more lines than were usually seen for this resonance 31 in other compounds. Upon P decoupling two quartets, separated by 3 Hz, were obtained. Presumably the two ethoxy substituents on phosphorus are not equivalent and, at least with the methylene protons, gave rise to two dist inct resonances. In the 270 MHz H^ nmr spectrum, run on the mixture of diastere-omers, the a l l y l i c methine proton resonance appeared, separated from the OCH2 resonance, as a 'quartet' representing an overlapping double t r i p l e t . I t is formed when both coupling constants to the methylene and vinyl protons are of comparable magnitude (J=6Hz). A pair of AB - quartets at 6 2.15 - 2.35 ppm with J = 15 and 4 Hz, due to geminal coupling and coupling to phosphorus, was interpreted as due to the a l l y l i c methylene protons. The H^ nmr data of the individual stereoisomers were very similar to each other. The only major difference could be detected in the resonance patterns 1 of the vinyl protons and the C(l ) methylene protons. Whereas in the more polar isomer the la t ter gave rise to a doublet at 6 1.57 ppm with -= 6 Hz, and thus did not show geminal coupling, in the less polar isomer geminal 50 coupling was detected, in addition to vicinal coupling to the a l l y l i c methine proton. The value of J could not be determined, however, because the outer r gem resonance lines were buried under adjacent resonances. The vinyl proton resonances of both diastereomers of the compound 69_ are depicted in Figure 3. On the basis of the AB - quartet due to the diene protons being superimposed on the resonances of the terminal vinyl ic unit a l l resonance lines could be assigned. For the la t ter olef inic protons the following coupling constants were measured; J v i c > t r a n s = 17 Hz, J v i C j C i s = 10 and 10.5 Hz, J g e m - J a ] l y l i c = 1.5 and 1.7 Hz. The C(3') vinyl proton of the terminal alkene unit displayed a ) b) Figure 3: H^ Nmr vinyl proton resonances of the two diastereomers of the a l l y l i c alcohol 69_; a) less polar isomer, b) more polar isomer. HO 12* 3' 6 9 J ^ O P ( O E t ) 2 in both diastereomers a multiplet resonance pattern at <5 5.74 - 6.08 ppm, caused by di f ferent ial coupling to the cis and trans terminal methylene protons, and by coupling to the a l l y l i c methine proton (J c6Hz). The A - branch of the AB - quartet due to the 51 diene protons was superimposed on th is mu l t ip le t . The terminal v i n y l i c methylene protons appeared, for both diastereomers, at 6 5.00 - 5.32 ppm as four t r i p l e t s t ructures, two of which were overlapping with each other. This resonance of the terminal v iny l protons could be interpreted as being, in f ac t , overlapping t r i p l e doublets a r i s ing from geminal and a l l y l i c coupling of the same magnitude in addit ion to v i c i na l coupl ing. In the less polar isomer the C(4* , c i s ) proton resonated at 6 5.22 ppm (ddd,J=17,l.7,1.7Hz) the C(4 ' , t rans) proton at 6 5.04 ppm (ddd,J=10,l.7,1.7Hz). In the more polar isomer the corresponding resonances were at 6 5.23 ppm (J=17,l.5,1.5Hz) and at <5 5.07 ppm (J=10.5, l .5,1.5Hz). The dif ference in the chemical sh i f t of the B - branch of the diene proton AB - quartet was the major d i s t i n c t i on between the v inyl proton resonances of the two diastereomers. By v i r tue of the d i f fe ren t trans and c i s o l e f i n i c coupling constants each resonance could thus be assigned. The mu l t i p l i c i t i e s of these v iny l proton resonances were more c l ea r l y shown by scale expansion and ampl i f i cat ion of the nmr spectrum. In the i r spectrum of the alcohol 69 a strong hydroxyl absorbance near 3400 cm"1 and a weak band near 1640 cm"1 was found. The l a t t e r presumably arose from the terminal double bond. When the reaction of the aldehyde 65_with v iny l l i t h i um was quenched with t r imethy lch loros i lane, instead of aqueous sodium bicarbonate, the a l i y i i c t r ime thy l s i l y l ether 70_ was produced and i so la ted . I t was, however> unstable Q 7 * CHO OP(OEt)2 j . ^ L i ii. C l S i M e 3 Me3SiO OP(OEt)2 6 5 7 0 52 and readily lost the tr imethylsi ly l group. Corroboration for the structural assignment 70_ was provided by both the i r and nmr spectra of this product. In the former, a band at 845 cm~^, typical for t r i a l k y l s i l y l ethers and the absence of a hydroxyl absorbance indicated s i l y l ether formation. In the la t te r , a sharp singlet resonance at 5 0.03 ppm, integrating for 9 protons and due to the methyl groups on s i l icon, provided further support for successful incorporation of the tr imethylsi ly l group. The vinyl proton resonances were similar to those of the a l l y l i c alcohol 69. The presence of diastereomers was deduced from two dist inct resonances for the ter t iary methyl group on C(5), at 6 1.00 and 0.98 ppm. The resonance from the ter t iary a l l y l i c proton was again buried under the prominent OC^ multiplet. Lastly, the signif icant loss of a methyl group from the parent ion in the mass spectrum of the com-pound 70_ correlated well with the other spectral evidence for the formation of the s i l y l ether 70. As no signif icant loss of a methyl group from the parent ion was observed with most of the other compounds in this synthesis, except those containing a t r i a l k y l s i l y l group, i t is highly l ike ly that the methyl group was eliminated from the tr imethylsi ly l ether moiety. ^,86 At this stage of the synthesis only an a l l y l i c oxidation was necessary to accomplish synthesis of the a, 6-unsaturated ketone 71_, the envisaged sub-strate for the key internal cyclization reaction. 53 Unfortunately, no ef f ic ient oxidizing agent could be found. Manganese 87 dioxide which has had a long history of being a mild reagent to effectuate a l l y l i c oxidations was the f i r s t one to be t r ied . With freshly prepared oxide and in dif ferent solvents, such as benzene, chloroform-carbon tetrachloride, ether and dichloromethane, no a-enone 7J_ could be obtained. For example, reaction of 12 mg of the a l l y l i c alcohol 69_with 120 mg of manganese dioxide on carbon in benzene at room temperature for 20 hours furnished 9 mg of a crude product. I ts i r spectrum s t i l l exhibited a hydroxyl absorbance, in addition to a weak band near 1720 cm~^. Examination by t i c confirmed the presence of start ing material. The act iv i ty of this preparation of manganese dioxide was checked by allowing i t to react with 2-cyclohexenol in benzene at room temperature. After one hour a strong carbonyl absorption, corre-sponding to an a, 6-unsaturated ketone, had developed in the i r spectrum of the crude product, confirming that the oxidizing agent was active. Similarly, the alcohol 69 was treated with a ten-fold excess of manganese dioxide, 88 b prepared by a dif ferent procedure , in chloroform-carbon tetrachloride at room temperature for 8 hours. After the suspension had been f i l te red through F lo r is i l and the solvent evaporated, only 55% recovery of material could be realized. The i r spectrum of the crude product showed absorbances from a hydroxyl group together with two bands in the carbonyl region of the spectrum, at 1725 and 1680 cm~^. The lat ter could possibly have been associated with the a-enone functionality of the desired product 71_. However, t i c examination revealed several compounds in the crude mixture and no puri f icat ion was 88c attempted. Manganese dioxide, prepared by s t i l l another procedure , was investigated in a similar manner for i ts ab i l i t y to bring about oxidation of 69_ to 7J_. Neither reaction in benzene nor in ether or dichloromethane was satisfactory, since only l i t t l e material could be recovered and the i r 54 spectrum of the crude product s t i l l showed a hydroxyl absorbance, together with a band near 1725 cm~^. Next, another mild oxidizing agent, si lver 89 carbonate on Celite ' was employed. Reaction of the alcohol 69_with a ten-fold excess of freshly prepared reagent, in benzene at ambient tempera-ture for 2 hours, yielded only starting material. The same reagent, after 3 hours in refluxing benzene, converted the a l l y l i c alcohol 69_ to a crude product, the i r spectrum of which did not show a carbonyl absorption but neither did there appear to be much starting material l e f t . The oxidation of 69 under conditions similar to those described for the Moffatt-type oxidation of the alcohol 64 to the aldehyde 65_ produced at least three different products, as shown by t ic examination of the crude mixture. The i r spectrum of the la t ter exhibited absorptions at 1670 and 1715 cm~^  but did not indicate retention of any starting material. When the alcohol 69_ was treated with 1.7 equivalents of pyridinium chlorochromate in dichloromethane at room temperature for 2.5 hours, the crude product obtained did not show a carbonyl absorption in the i r spectrum. The combination of 90 dimethyl sulfoxide with oxalyl chloride is known to be an ef f ic ient oxidizing agent. However, i t fai led to achieve the desired conversion of 69 into the a-enone 71_. For example, reaction of the a l l y l i c alcohol 69 with the dimethyl sulfoxide-oxalyl chloride system at -60° for 45 minutes, followed by addition of triethylamine and hydrolysis of the reaction mixture after 35 minutes at ambient temperature yielded only starting material. When this oxidation was performed at -15°C under otherwise similar conditions, a mixture of compounds was obtained. The i r spectrum of this crude product showed both hydroxyl and two carbonyl absorbances. Treatment of the alcohol 91 69 with 2,3-dicyano-5,6-dichlorobenzoquinone ( 1 . 5 - 2 equivalents) in benzene at ambient temperature for 16 hours produced a crude product in 55 only 52% y ie ld . Its i r spectrum possessed a hydroxyl absorbance and two carbonyl bands at 1680 and 1720 cm" 1. Examination of the mixture by t i c showed some starting material l e f t together with a few unidentified products. 92 Barium manganate, a recently developed reagent that also allows alcohol oxidation in the presence of unsaturation, was in at least one case found 93 superior , in terms of yield and purity of product, to manganese dioxide and Jones reagent. Similar observations regarding the oxidation of a primary propargyl alcohol with manganese dioxide, pyridinium dichromate and 94 barium manganate were made in our laboratory. In the present case, however, attempted oxidation of the a l l y l i c alcohol 69_ with a ten-fold excess of barium manganate at room temperature for 22 hours in dichloromethane furnished only starting material. The reaction of 6_9 with chromyl chloride 95 and pyridine proved to be equally unsuccessful, since, upon addition of 69 to a mixture of chromyl chloride (1.2 equivalents) and excess pyridine in dichloromethane at -78°C (1.5 hours), then 0.5 hours at room temperature, only starting material was recovered. Pyridinium dichromate ^ , another reagent added by Corey et al_. to the collection of Cr (VI) based oxidizing agents, was investigated las t , but was also found to be unsatisfactory from a synthetic point of view. When, for instance, the a l l y l i c alcohol 69 was treated with 1,4 equivalents of pyridinium dichromate in dimethylformamide at Q°C for 7.5 hours, 50% material could be recovered after work-up. I t was found to consist of a mixture of start ing material, one major and several minor products by t i c and glc examination. An absorbance at 1675 cm"1 in the i r spectrum of this material, and new resonances in i t s nmr spectrum at 5 2.73, 2.53 ppm (AB-quartet, J=15Hz, CH2C0) and 6 6.0 - 6.4 ppm (B protons of a terminal carbonyl conjugated double bond) lead to the conclusion that some desired a-enone 71_ had been formed. However, due to the limited 56 avai lab i l i ty of starting material for this oxidation, i t could not be further investigated. Final ly, no more further efforts to accomplish the oxidation of the a l l y l i c alcohol 69_ to the enone 7J_ were made with the discovery, that the a l l y l i c alcohol i t s e l f smoothly cyclized to the desired product. 5. Synthesis of the a-Enone 7_3 In the course of this synthetic work, i t was investigated whether an alcohol protecting group other than the tetrahydropyranyl ether would offer any advantages. As outlined in section I I .2 of this thesis, the conversion to the corresponding tert-butyldimethylsi lyl ether is a common method of alcohol protection. This derivative was chosen for our purposes because the necessary alkylating agent (vide in f ra ) , the tert-butyldimethylsi lyl derivative of a-bromoethanol, could be prepared under neutral conditions and f luoride ion offered the possibi l i ty of a non-acidic cleavage. The transformation of a-bromoethanol into the corresponding te r t -butyldimethylsilyl compound was cleanly achieved by treating the alcohol with tert-butyldimethylchlorosilane in dimethylformamide in the presence of imidazole at room temperature. Treatment of the l i thium enolate of the ketone 54 with the alkylating agent thus prepared, under conditions similar to those used for the preparation of the tetrahydropyranyl ether 55_ (tetrahydrofuran-hexamethylphosphoramide, -78° to room temperature) allowed the synthesis of the compound 72 in good y ie ld . The spectral data for the compound 72 confirmed the introduction of the oxygen functionalized substituent. In the nmr spectrum, two sharp singlets at s 0.88 ppm (9H)and 5 0.04 ppm (6H) were attributed to the tert-butyl and silylmethyl protons, respectively. Absorbances at 1253, 1098 and 835 cm""' 57 54 7 2 in the i r spectrum of 72_ further emphasized the presence of the te r t -butyl dimethyl s i l y l ether group. All the other nmr and i r data were in f u l l agreement with the proposed structure for the compound 72_. The reduction of the ketone 72_with l ithium aluminum hydride in ether at 0° followed by treatment of the crude product with acid furnished the product 73_ toqether with some starting material. In comparison with the tetrahydropyranyl ether 55_, the s i ly l ether 72^  was found to be less reactive in this reaction sequence since, after a longer reaction time there was s t i l l some starting material recovered. Therefore, in a subsequent reaction, the reduction was carried out at 0 C i n i t i a l l y , then at ambient temperature, in order to convert a l l starting material. After acid hydrolysis, a crude yield 5 8 of only 50% was obtained, however. The formation of the desired compound 73_ was indicated by g lc examination of the crude product but a second, more vo l a t i l e compound was present to an even higher percentage. The l a t t e r was iso lated and pur i f i ed by column chromatography and short-path d i s t i l l a t i o n to give a colourless l i qu id in 29% y i e l d . On the basis of spectral and microanalyt ica l data, the structure 74^was assigned to th i s mater ia l . For example, examination of i t s nmr and i r spectra revealed the retention of the isobutoxy enol ether func t i ona l i t y . In the former, a one-proton resonance, two doublets (J=4Hz) at 6 4.51, 4.49 ppm and assigned to the v iny l proton, resonances at 6 3.9 - 3.3 ppm (4H) a r i s ing from methylene protons adjacent to an oxygen atom and methyl proton resonances at 6 1.14 -0.90 integrat ing for 12 protons were found. For the OCh^  protons, two d i s t i n c t sets of s ignals were obtained, two doublets at 6 3.41, 3.44 ppm (J=6Hz), due to the isobutyl methylene protons, and a four - l i ne resonance at 6 3.81 - 3.97 ppm (2H) which was assigned to the C(2') methylene protons. Furthermore, another one-proton resonance, two doublets at 6 4.19, 4.17 ppm ( J c 4Hz) indicated the presence of a methine proton a to an oxygen atom and adjacent to a carbon with one proton. This resonance, the isobutyl OCh^  signal and the v iny l proton resonance each showed as a double signal because of the presence of stereoisomers, further underlined by two separate doublet resonances for the isobutyl methyl groups. No resonances a r i s ing from a t r i a l k y l s i l y l group could be detected. In the i r spectrum of 74 the only band in the double bond region was an absorbance at 1654 cm" 1 , due to the enol ether system. The mass spectrum of the compound 74 indicated a prominent loss of a C 4 H 7 fragment from the molecular i on , and a less intense [f^-C^H^O] fragment. A l l the information given above could best be accommodated by ascr ib ing the structure 74 to th is mater ia l . 59 In contrast to the related 2-cyclohexen-l-one 5_7 the B vinyl proton of the a-enone _73 did not exhibit long-range coupling. Its resonance was obtained at 6 6.67, 6.65 ppm, as two doublets (J=10Hz) because of the presence of stereoisomers. Two singlet resonances, at 60.90 ppm (9H) and 6 0.04 ppm (6H) were attributed to the tert-butyl and the methyl substituents, respectively, on s i l icon. The i r spectrum of 73_ showed the a-enone absorption at 1678 cm - 1 . All further attempts to selectively convert the compound 72_ into the a-enone 73_ met, unfortunately, with fa i lure. Carrying out, as in the original reaction, the reduction and hydrolysis at 0°C s t i l l provided a mixture of both compounds 73_ and 74_. The proportion of the desired a-enone 73_ was usually found to be 60-70% by subjecting the crude products to glc examination. This divergency discouraged further exploration along this route. 6, Conversion of the Aldehyde 65 to the N i t r i le 76 Sparked by the inab i l i t y to accomplish effective oxidation of the a l l y l i c alcohol 69 to the a, B-unsaturated ketone 71_,an alternative preparation of the enone 71_ was sought. I t is well known that ketones can be prepared from 41 m.triles by reaction of the lat ter with carbon nucleophiles followed by a usually readily occurring hydrolysis of the resultant imine. Such a sequence would not involve an oxidation step since the n i t r i l e carbon atom already is at the desired oxidation level. In relation to the synthesis under discussion a possible conversion of the n i t r i l e 76 to the a-enone 7]_ would circumvent the problems with the a l l y l i c oxidation of the alcohol 69 to the a, B-unsaturated ketone 71_. Therefore the n i t r i l e 76_ was synthesized in two steps from the aldehyde 65, with the oxime 75_ as intermediate. 97 Treatment with hydroxylamine hydrochloride in aqueous pyridine readily converted the aldehyde 65 to the oxime 75 in 82% y ie ld . In the 60 1 CN 76 nmr spectrum of the compound 75_ a one-proton t r i p l e t at 6 7.45 ppm (J=6.5Hz, CH=N) and a broad singlet at 6 6.81 ppm (OH) corroborated the formation of the oxime moiety. Two dist inct AB-quartets for the vinyl protons and two sharp singlets for the ter t iary methyl group on C(5) indicated the presence of stereoisomers around the carbon-nitrogen double bond. In the i r spectrum AOpVt)2 NH 20HHCI J^-OPtOEE^ C D | m J^OP(OEt) 2 H 20-pyr C H2C'2 CHO CH=N0H CN 65 75 76 a strong hydroxyl absorbance at 3290 cm~^  reaffirmed the presence of a hydroxyl group. The carbon-nitrogen double bond absorption band either coincided with the diene absorption or was too weak to be. detected. A number of reagents are known for the dehydration of oximes, and for the conversion of 75_ to 76_ good results were obtained by using N,N -98 carbonyldiinidazole (CDIm), a versatile reagent developed by Staab. I t is part icularly convenient to use for only easily removable byproducts (carbon dioxide, imidazole) are formed. The n i t r i l e 7j5 could be cleanly synthesized by N,N -carbonyldiimidazole induced elimination of the elements of water from the oxime 75_ in refluxing dichloromethane. A weak i r 61 absorption at 2245 cm - 1 confirmed the presence of a n i t r i l e group in the product thus obtained. In the nmr spectrum of 76_ the a methylene protons [C(l ) protons] resonated as a singlet at 6 2.39 ppm. In addition, a l l the other expected resonances for the dienol phosphate 7_6 were observed. The n i t r i l e 76_ could be obtained in 58% overall yield after chromatography from the tetrahydropyranyl ether 63_. 0P(0Et)2 U steps ^ ^ o S l O E t k 5 8 % With the prerequisite n i t r i l e 76_ secured,its potential use as precursor to the desired a-enone 71_ was investigated. Whereas n i t r i l es react, as expected for an electrophile,with alkyl Grignard reagents to form ketones, the corresponding reaction with alkenylmagnesium halides is usually not a 41 synthetically useful method of preparing a-enones (Scheme 2). Most often, condensation products derived from the intermediate unstable magnesio imines R R'= alkyl R R-CN + R-MgX • ^C=NMgX - C=0 R^ i K R = alkyl, aryl R'=vinyl X=CI, Br,l condensation products Scheme 2 99 are obtained instead. With the more reactive vinyl l i thium compounds successful n i t r i l e addition appears to have been generally limited to the 99b rather special case of benzonitrile. This is presumably due to the high susceptibi l i ty of alkyl n i t r i l es to a metalation as well as to a certain 62 propensity of the intermediate N-lithioimine to either add to excess n i t r i l e or react with a second mole of the l ithium reagent to form dilithioamides (Scheme 3). Thus conversion of a lky ln i t r i les to ketones with organolithium reagents appears to be somewhat problematic. On the other hand, the reaction of carboxylic acids with alkenyllithium compounds to give a , e-unsaturated ketones is known. ^ ' \=N-C=NLi Y R'Li j - C - N L i 2 R' Scheme 3 The addition of vinyl l i thium in ether to a solution of the n i t r i l e 76_ in dimethoxyethane at -7S°C, followed by keeping the reaction mixture at 0°C for 2 hours produced, after hydrolysis with IN hydrochloric acid in methanol, a mixture consisting of about 60% starting material and three products. The i r spectrum of this crude product showed a band near 1680 cm"^  and by glc examination the formation of about 20% desired product appeared possible. I t seemed that the react iv i ty of the n i t r i l e carbon atom was not high enough under these conditions. Raising the reaction temperature would be one possibi l i ty to achieve higher react iv i ty but possible interference from the electrophil ic phosphate ester group would then also be more l i ke ly . Therefore, either an increase in the nucleophilicity of the lithium reagent or in the electrophi l ic i ty of the n i t r i l e group appeared more desirable. In both instances, however, unwanted side reactions,such as a metalation, might occur more easily as wel l . Increasing the react iv i ty of organolithium compounds by adding alkoxides or amine bases is a well-known concept in 63 organic chemistry. 5 , L " Higher electrophi l ic i ty of the n i t r i l e carbon atom can be achieved by increasing the electron-withdrawing property of the nitrogen atom. In the condensation reaction of n i t r i l es to ketones with 102 phosphonium ylides i t was discovered that added anhydrous lithium salt functioned as a n i t r i l e activating agent by complexation of the lithium cation with the nitrogen atom. In such an intermediate of the general structure 1J_ the electrophil ic react iv i ty of the n i t r i l e carbon atom would R—CE N—Li X" R = alkyl. aryl X=l. CIO* 77 be enhanced by the increased electron-withdrawing property of the nitrogen atom. When the n i t r i l e _76 was treated with vinyl l i thium in ether for 3 hours at 0°C, then for 30 minutes at ambient temperature, in the presence of 1.8 equivalents of anhydrous l ithium iodide, in dimethoxyethane, no starting material could be detected in the crude product after acid hydrolysis. Recovery of material was, however, only 53%. Glc examination of the crude product showed the formation of at least three different products. The i r spectrum did not indicate much a-enone formation, and no attempt was ;-.-,ade to identify the products. A reaction run under similar conditions with cyclo-hexylacetonitrile and vinyl l i thium yielded a gummy, polymeric crude product. When the reaction temperature was lowered to -78°C (30 minutes) then 0° for 1.5 hours a mixture of the starting material, cyclohexylacetonitrile, and unidentif ied, non-ketonic products was obtained. Treatment of the same n i t r i l e with vinyl l i thium for 3 hours at room temperature in benzene, without l i thium salt being present, only yielded a polymeric material as wel l . With the apparent inab i l i ty to accomplish the synthesis of the a , 6-unsaturated ketone 64 71 from the n i t r i l e 76_ and vinyl! i thium attention was focused on another possible vinyl ic nucleophile. In recent years the chemistry and use of organoaluminum compounds for synthetically useful transformations has been a f ie ld of increasing interest. Due to the ionic character of the aluminum-carbon bond these compounds react with electrophiles in a manner of Grignard reagents, albeit the efficacy of these reactions often is less satisfactory. Mixed alkylalkenylalanes and Scheme 4 alanates preponderantly transfer the alkenyl groups (Scheme 4). The reaction of alkynylalanes with ter t iary alkyl halides and secondary sulfonates is remarkable in so far as i t allows cross-coupling of a highly substituted carbon center with an alkynyl group in good y ie ld . In natural product synthesis alanes have been employed to conjugatively introduce both alkenyl and alkynyl 104 groups into a-enones. Quite frequently addition reactions of alanes are 105 catalysed by nickel complexes, such as in the reaction with a-enones and ketones ^ and in the coupling reaction with aryl and alkenyl halides. Often use of the corresponding alanates does not require the presence of a catalyst. At elevated temperature benzonitrile was found to undergo ketone formation with tr imethyl- , tr iphenyl- and mixed methylchloroalanes. ^ 108 Later, Mole et_ al_. employed nickel acetylacetonate [ N i ( a c a c i a s catalyst to considerably mitigate the reaction conditions and extend the reaction to a lky ln i t r i les as wel l . Satisfactory reaction took place at ambient temperature between a number of n i t r i l es and trimethylalane to give methyl ketones after hydrolysis (Scheme 5). These observations combined with the established preference for transfer of the unsaturated ligand in mixed C = C y \ M R 2 E R= a l k y l E= e l e c t r o p h i l e 65. R-CN + Me3Al R= alkyl. aryl Ni(acac)2 H 30 + 0 II C R CH 3 Scheme 5 alkylalkenylalanes prompted a closer look at the feas ib i l i ty of achieving the addition of a vinyl group to the n i t r i l e 76_ with a vinylalane reagent. Diethyl vinyla lane > u y was prepared by addition of diethylchloroalane in toluene (obtained from Aldrich Chemical Co.) to a suspension of v inyl -l i thium in benzene at room temperature. An aliquot of the resulting reagent solution was transferred to a solution of cyclohexylacetonitrile in benzene at 0°C containing a catalyt ic amount of nickel acetylacetonate. The tempera-ture was raised to 20 C, after 16 hours a l i t t l e more catalyst, dissolved in benzene, was added and the temperature raised again, to 35°C. Acidic work-up after 4 hours allowed the isolation of 90% of unchanged start ing material. In a second,slightly modified reaction, diethylchloroalane in toluene was added to a solution of vinyl l i thium in ether at 0°C, then at room temperature, the solvent was removed under high vacuum and benzene was added to the residue. Addition of an aliquot of the reagent thus produced to a solution of cyclo-hexylacetonitrile in benzene at room temperature, followed by a catalyt ic 0 76 71 66 amount of nickel acetylacetonate in benzene and s t i r r ing of the reaction mixture at room temperature for 64 hours with additional catalyst being added after 20 hours furnished, after acidic hydrolysis, a crude product that consisted by glc and i r examination mainly of starting material. I t is not necessarily surprising that no transfer of the ethyl substituent of the alane reagent occurred, for i t probably is less reactive than a methyl sub-stituent and vinyl or halide groups decrease the react iv i ty of adjacent alkyl carbon-aluminum bonds. After the fai lure of these reactions,attempts to synthesize the a , 6-unsaturated ketone 71_ were discontinued. 7. Diels-Alder Cyclization of the A l ly l i c Alcohol 69 and Synthesis of 9-Pupukeanone (7_) The intramolecular version of the well-established Diels-Alder cyclo-addition reaction ^ ° has received widespread attention in the last few years. ^ A number of syntheses of polycyclic compounds ^ 2 and natural 113 products have incorporated this stereoselective and versatile modifica-t ion, including a heteroanalogous variation in a study on the synthesis of cannabinoids. An a , B-unsaturated ketone serving as the diene in a Diels-Alder cyclization with inverse electron-demand was transformed to a t r i cyc l i c 114 dihydropyran derivative. Equally well-known is the profound influence of added Lewis-acids on the ease and regiochemical course of the cyclo-115 addition. This effect does not seem to have been made use of in the intramolecular version so far. 28 In Yamamoto's synthesis of 9-isocyanopupukeanane the internal cyclization was achieved on the 0-protected a l l y l i c alcohol derivative 17. to give, after acid hydrolysis of the i n i t i a l adduct, the keto-alcohol J_8. As i t became apparent at the time that the oxidation of the a l l y l i c alcohol 69 to the a-enone 7_1_, the proposed substrate for the intramolecular cyclo-67 .OSiMe3 j . A ii. H 3 0 + 1 . THPO-addition, would not be as straightforward as anticipated, the compound 69_ i t se l f was investigated as possible substrate for the Diels-Alder reaction. fi fi OP(OEt)2 10] A^OP(OEt) 2 The a l l y l i c alcohol 69_ smoothly cycl i zed to the desired t r i cycl o-3 7 [4.3.1.0 ' ]decane functionalized carbon skeleton in high yield upon reflux in xylene. Careful spectral and chromatographic examination of the product gave no indication of i t being a mixture of regioisomers. Only the t r i cyc l i c alcohol 78_ was formed. I t appears that with a rather inactivated dienophilic uni t , such as in 69, the internal cyclization proceeds via the less strained transit ion state 69a whereas in the alternative 69b, leading to the regioisomer 79_ of the alcohol 78_, considerably higher angle distort ion must occur in order to get good orbital overlap for cyclization to the twistane derivative 79. The activation energy associated with 69b must be considerably higher than that of the transit ion state 69a. In addition to the angle distort ion st ra in, in the transit ion state 69b of one of the two diastereomers of the a l l y l i c alcohol 69_ a destabilizing 1,3-diaxial interaction between the 68 6 9 a 6 9 b o n e d i a s t e r e o m e r hydroxyl substituent and the C(l) methyl group would occur. The closely 28 related system 80_ of Yamamoto and similar cycloadditions to the t r i cyc l i c adducts a ' c 81_ and 82_ further support the assignment of the structure 78 to the product of the internal cycloaddition. A last piece of evidence that the cyclization of 69_ proceeded with only the formation of the desired alcohol 78^was provided by the oxidation of the la t ter to the corresponding cyclopentanone derivative 96_ (cf_. section I I . 8) which furnished one stereo-69 chemically pure compound. O S i M e 3 O S i ^ 80 81 82 0 OP(OEt) 2 PDC 21 9 6 Considering the dienophilic part of the a l l y l i c alcohol 69 as being polar-ized as depicted below, because of the inductive electron-releasing effect of the alkyl substituent, the indicated polarization pattern of the diene unit would further assist the formation of the desired adduct 78. The C(l) methyl substituent of the diene, known to be electron releasing, would assist this polarization pattern. The C(2) oxygen substituent, however, would favour C(l) to rather be electron-rich than electron-deficient since i t usually exercises a strong mesomeric electron-releasing ef fect , as shown above. This character-i s t i c property of an oxygen substituent at the 2-position of a 1,3-butadiene is HO £9. 70 responsible for the observed regioselectivity of the cycloaddition between 2-trimethylsiloxy-l,3-butadiene and methyl vinyl ketone where only the para-isomer, resulting from cycloaddition in the indicated sense, was o b t a i n e d . ^ Me 3 Si(K^6- 0 Me^iO-^^-^ Ambiguity in the regiochemical outcome of such a cycloaddition arose when an additional methyl substituent at C(l) was introduced. For instance, both possible regioisomeric products were produced upon reaction of 1-methyl-2-trimethylsiloxy-1,3-butadiene with an a-enone ^ (Scheme 6). With enol R^R Me 3SiCK^j ° X ° M e3 s'0 t R=CH2CH2C(SCH2CH2S)CH2CH2 R'= C(0)OC(Me2)OC(0) Scheme 6 phosphates, the electron-releasing properties of the oxygen substituent would presumably be diminished in comparison with enol s i l y l ethers because in addition to the mesomeric electron-releasing effect of the oxygen an inductive electron-withdrawing effect of the phosphate ester group could be expected. Thus, in a 1,2-disubstituted 1,3-butadiene system of type 83 the Q increased electron-withdrawing properties of the - v x^/OP(OEt)2 C(2) oxygen substituent would assist the property j i ^ y of the CO) methyl group to promote the indicated 6-diene polarization pattern and, as a consequence, 83 assist in the formation of the desired alcohol 78. 71 That enol phosphates are indeed weaker electron-releasing substituents than t r ia lky ls i loxy groups could be shown by comparing some spectral data of the two related enol derivatives 60 and 63_ (Table I ) . In the i r spectrum, the 6 0 R = C H 2 C H 2 O T H P 63 diene system of the s i l y l enol ether 60_ absorbed at lower wavenumber than that of the enol phosphate 63_, indicating a stronger electron-withdrawing effect of the phosphate ester group, relative to the t r ia lky ls i loxy sub-st i tuent, on the C(l) - C(2) double bond. The fact that in both compounds 60 and 63_ only one i r absorption band for the conjugated diene system was observed could either mean that the absorbance due to the less substitued C(3) - C(4) double bond was too weak to be detected, or the two absorbances were too close together and appeared as one band. The electronic effect of the C(2) substituent could be transmitted to the C(3) - C(4) double bond via the ir-system, thereby reducing the overall diene electron density in 63 in comparison with the compound 60. The nmr data of the olef inic protons of 60 and 63_ discussed below would further support th is . In the "'H nmr spectrum (6 scale values in ppm) the vinyl ic methyl group resonance of 60 i r ]H nmr 1 3 C nmr v ^ C c m " 1 ] vinyl H vinyl CH3 (sp2-C) silyl enol ether 60_ 1664 5.58 5.39 1.65 141(w) 125 135 lll(w) enol phos-phate 63 1675 5.46 5.83 1.75 138(w)* 121 135 117(w)* Table I : Selected spectral data for the compounds 60_ and 6_3 * Because of the presence of diastereomers a double resonance was obtained at 138.9, 138.5 ppm and at 117.3, 116.8 ppm; w = weak 72 appeared 0.10 ppm upfield of that of the compound 63. I f the C(2) phosphate substituent in 63 is indeed less electron-releasing than the corresponding siloxy group in 60 the resulting decreased electron-density at C(l) would have a deshielding effect on the vinyl methyl resonance, which is what was observed. The vinyl proton resonances of the enol phosphate 63_ were found at lower f ie ld than those of the s i l y l ether 60_, which could ref lect a decreased electron density of the diene system of the compound 63_ relative to 60. No assignment of an individual doublet of the AB-quartet to either the C(3) or C(4) vinyl proton could be made. 13 1 C Nmr spectroscopy is similar to H nmr spectroscopy in that the chemical sh i f t of a particular nucleus is sensitive to the local electron density around that nucleus. In general, an increase in electron density w i l l 13 result in an upfield sh i f t of the resonance being investigated. The C nmr spectra of the compounds 6_0 and 63_ were determined in deuterochloroform solution in the noise-decoupled mode of operation and the data given are in ppm downfield from tetramethylsi lane as internal reference standard. Of the four sp carbon resonances of each compound, two were considerably weaker than the other two. As fu l l y substituted carbon atoms usually give weaker signals than less substituted ones because of their longer relaxation times, the pair of weak signals were assigned to C(l) and C(2) of the diene system. The mesomeric electron-releasing property of the oxygen substituent induces a higher electron density at C(l) than at C(2), therefore, of the two weak resonances, the lowfield one was interpreted as being due to C(2), the other was assigned to C( l ) . In comparing the s i l y l enol ether 60_ to the enol phosphate 63 the resonance due to C(2) moved upf ield, that due to C(l) moved downfield, ref lecting a decreased electron-density at C(l) and a higher one at C(2) in 63 relative to 60. This was in fu l l agreement with what one would have expected for a decreased contribution of the mesomeric form 63a or 63b 73 to the overall electron distr ibut ion in the diene system of the enol phosphate 63_ as compared with the s i l y l ether 60. Thus i t can be said, in conclusion, that the C(2) phosphate substituent exerts a two-fold effect on the diene system. F i rs t l y , inductive withdrawal of electron-density makes the diene less electron-rich than in the s i l y l enol ether 60_. Secondly, the same property could, to some degree and in conjunction with the C(l) methyl substituent, induce a polarization pattern of the diene as shown with the structure 83_. A recent publication by Kienzle and Rosen lends 118 some support to this deduction. In an intermolecular Diels-Alder cycloaddition of dimethyl-1-vinylvinyl phospate with methyl vinyl ketone both possible regioisomeric adducts 84a and 84b were obtained in a rat io of 1:2. No definite structure assignment of the predominant isomer was possible since the isomeric mixture could not be separated into the individual com-pounds. As 2-trimethylsiloxy-1,3-butadiene reacts with methyl vinyl ketone with only the formation of the para-adduct corresponding to 84b ^ 6 (vide  supra), the phosphate ester group must have exerted a noticeable directing effect toward the regiospecifity of the cycloaddition contrary to that of 74 the corresponding trimethyl siloxy substituent. Evidence for successful, regioselective cyclization of the a l l y l i c alcohol 59_ to the cyclopentanol 7_8 was obtained from the nmr, i r and mass spectra of the compound 7J3. In the i r spectrum the olef in ic absorbance had shifted from 1675 cm"^  in 69_ to 1650 cm"^  in 78. In the nmr spectrum, the vinyl proton resonances had experienced the most d is t inct ly visible change. A complex pattern between 6 5 - 6 ppm in the precursor a l l y l i c alcohol 69_ had been substituted by two double doublets at 6 5.71 and 5.67 ppm. Coupling to the a l l y l i c proton and to phosphorus, and the presence of stereoisomeric alcohols in the solution accounted for this resonance pattern. The vinyl methyl resonance of the a l l y l i c alcohol 69_ had disappeared. Another double doublet, due to coupling to the vinyl proton and the adjacent Q C(6) proton was assigned to the a l l y l i c ,OP(OEt)2 proton. From the relative intensit ies of the two vinyl proton double doublets and the four dist inct singlets obtained for the C(l) and C(3) ter t iary methyl groups the rat io of diastereomers was 31 estimated as 1:1. P irradiat ion induced a collapse of the vinyl proton resonance to two overlapping doublets with J = 7.5 Hz. The collapse of the ethoxy CH2 and CH3 proton rescnanc.es to a quartet and t r i p l e t , respectively, confirmed the continued presence of the phosphate ester moiety in the compound 78. In contrast to the precursor alcohol 69_,the a l l y l i c proton in 78 did not undergo detectable homo-allylic 31 31 coupling to P for there was no change in i ts resonance upon P i r radiat ion. Presumably the orthogonal orientation of the double bond TT-orbitals to the a l l y l i c carbon-hydrogen bond, as shown in the accompanying diagram which 75 H<6~~0'>OP(0)(OEt)2 shows the relevant part of the molecule viewed from the top, precludes the necessary interaction between these orbi tals. Long-range coupling mediated by ir -orbi ta ls, such as a l l y l i c and homo-allylic coupling, is at a maximum 68 with the hyperconjugating bond being aligned with the u -orb i ta ls . Upon homonuclear decoupling of the vinyl proton the 6 2.72 ppm resonance converged to a doublet with J = 4 Hz, reconfirming i t s assignment as due to the a l l y l i c proton. The above coupling constant agrees well with values measured for the C(l) - CC7] interproton coupling in norbornane derivatives. In the 270 MHz 1H nmr spectrum, the OCH^  resonance appeared separated from that of the OCH proton on C(5). Two sets of signals for the C(5) proton were obtained for the two diastereomers of 78, a symmetrical, quintet- l ike multiplet at 6 4.36 ppm, the individual lines being separated by 5 Hz, and a multiplet at 5 4.12 - 4.06 ppm. On the basis of results from a variety of norbornanes, which consistently showed, for C(2) substituted or unsubstituted compounds, the exo proton to absorb at lower f i e ld than the endo one,^ the resonance at 6 4.36 ppm was ascribed to the endo isomer of the alcohol 7_8. The observed mul t ip l ic i ty is in l ine with this assignment. A large 10 Hz coupling con-stant for the exo-exo C(4) - C(5) proton interaction, predicted by the Karplus equation for a dihedral angle of near 0°, and two smaller, dimensionally simi-lar coupling constants for the corresponding e^-endo and the C(5, exo) -C(6) proton interaction of near 5 Hz result in a symmetrical resonance which is an overlapping ddd with J = 10, 5, 5, Hz. In the mass spectrum of 78_ the base peak has moved from m/e 259 in the precursor a l l y l i c alcohol 69 to m/e 246. 76 High resolution measurement showed that a fragment at m/e 259 did not correspond to a formula 2H20^4'3' o n e w n i c ' 1 w a s obtained from the a l l y l i c alcohol 69_ by expulsion of the C(5) side chain. From the a l l y l i c tr imethylsi ly l ether 7(3,the corresponding s i l y l protected Diels-Alder adduct 85_ could be obtained simi lar ly. However, with the observed ins tab i l i ty of the compound 70_ even towards chromatography on F l o r i s i l , a pure sample of 70 was d i f f i c u l t to obtain. I ts i r spectrum invariably showed a weak hydroxyl absorption. Under argon and with anhydrous solvent, 85_ was formed upon reflux of 70_ in xylene. As the precursor could only be obtained together with some a l l y l i c alcohol 69_ as contaminant, the product of this reaction was contaminated with the corresponding alcohol 78. With the a l l y l i c alcohol 69 smoothly undergoing cycloaddition to 78, there was no need to introduce a protecting group,and puri f icat ion of 85 was not attempted nor was this reaction further investigated. At this stage, with the sought-after t r i cyc l i c compound 78 secured, the proper functionali ty at C(5) and C(9) became the centre of attention. The 9-position did not need much elaboration for i t already represented a masked ketone group. As the introduction of the C(5) endo isopropyl substituent would presumably en ta i l , among others, a reduction step, prior conversion of the enol phosphate group to a more inert carbonyl protecting group was deemed necessary, A common method of protecting ketones is their transformation to 70 8 5 77 ketals, which are easy to prepare and stable to a variety of reagents. They are usually readily hydrolyzed by aqueous acid, or by other electrophil ic 119 species. The 5.5-dimethyl-l,3-dioxan derivative was chosen as protecting group. The enol phosphate 78_, upon treatment with sodium ethoxide in ethanol at room temperature cleanly transesterif ied to the keto alcohol 1_8. The same product could also be obtained by using potassium hydroxide in ethanol. Mechanistically reminiscent of carboxylic esters reacting with nucleophiles, this displacement proceeds with elimination of the best potential leaving group on phosphorus. The keto alcohol J_8 gave spectral data consistent 28 with i t s proposed structure, and in good agreement with l i terature values. In the i r spectrum of 1_8 an absorption at 1718 cm - 1 together with the absence of the typical phosphate ester pattern clearly showed complete transesteri f ication. This was confirmed by the nmr spectrum of 18 in which no resonances due to an ethoxy group could be detected. Additional "'H nmr data indicated a mixture of stereoisomers, such as the four singlets due to the ter t iary methyl groups on C ( l ) and C ( 3 ) , two doublets at 6 2.41 * and 2.37 ppm due to the C(8) methylene protons with J = 3 Hz and two dist inct multiplets from the C(5) methine proton. With the same argumentation as outlined for the precursor enol phosphate 78, the lower-field multiplet of the la t ter two resonances at 6 4.52 - 4.30 ppm was assigned to the endo isomer of the alcohol 1_8. I t gave the appearance of two overlapping t r ip le ts and was interpreted as ddd with J = 11, 5.5 and 5.5 Hz. The exo alcohol OCH resonance was not as well resolved. In stereochemically pure compounds (e.g. 20_ and 9-pupukeanone (7)) the C(8) a methylene protons resonate as one doublet near 6 2.4 ppm with J = 3 Hz. 78 The ketal 86_ could be synthesized from both the ketone 1J3 and direct ly from the enol phosphate 78_. Ketalization of J8_ was achieved by refluxing 120 the ketone in benzene with excess 2,2-dimethyl-l,3-propanediol in the presence of a catalyt ic amount of p_-toluenesulfonic acid and continuous removal of the water formed in the reaction. The same reaction using the enol phosphate 78 as substrate allowed the direct conversion of 7J} into the ketal 86_. One possible mechanism for the la t ter reaction would involve an acid-catalysed phosphate ester alcoholysis preceding the ketalization with the d io l . The product was obtained as a gummy, semi-solid material which could not be induced to crystal l ize. This presumably reflected the fact that i t was not a stereochemically pure compound but a mixture of diaster-eomers. Both the i r and nmr spectra of 86 indicated the formation of the ketal funct ional i ty. In the former, the absence of a carbonyl absorption was taken as indirect proof of successful ketalization. In the " 'H nmr spectrum, a 79 new, line-broadened AB - quartet at 6 3.67, 3.33 ppm with J g e m = 11 Hz, integrating for 4 protons and due to the methylene protons adjacent to oxygen, was direct evidence for the presence of the ketal group. Additional indirect evidence was the absence of the doublet near 6 2.4 ppm originating from the C(8) methylene protons in the spectrum of the ketone ]8_. The presence of a mixture of diastereomers was indicated by six ter t iary methyl group resonances and two multipiets for the C(5) methine proton. In the case of the endo hydroxyl isomer of 86 the C(5,exo) proton resonated as ddd at <5 4.26 ppm with J = 11, 5.5 and 5.5 Hz. The C(5,endo) proton resonance of the other diastereomer appeared at 6 3.97 ppm as a double doublet with J = 7 and 2.5 Hz. Molecular models of the compound 86 indicate a dihedral angle of close to 90° between the carbon [C(5)] - endo hydrogen bond and the carbon [C(6)] - hydrogen bond. Thus a very small, or even zero, coupling constant would be predicted for the interaction between the C(5,endo) proton and the C(6) proton. With only the di f ferent ial coupling to the C(4) methylene protons to be considered two doublets for the C(5,endo) proton resonance of the exo isomer of 86_ would be predicted, exactly what was observed. Pyridinium dichromate ^ in dichloromethane at room temperature cleanly oxidized the alcohol 86 to the cyclopentanone derivative 8_7. Being a stereochemically pure compound the ketone 87_ could be obtained as colourless crystals from hexanes. The formation of a cyclopentanone system was corroborated in the i r spectrum of 87_ by an absorption band at 1735 cm - 1 . In the mass spectrum the molecular ion showed at m/e 278, down by 2 mass units from that of the precursor 86_, In the H^ nmr spectrum only four ter-t iary methyl singlet resonances were detected indicating that the compound 87 is a stereochemically pure compound. The pattern observed for the 80 methylene proton resonance of the ketal moiety of the compound 87_ deserves special comment. At 100 MHz, two doublets at 6 3.72 and 3.66 ppm with J = 11.5 Hz, the A-branches of two AB - quartets, and two overlapping pairs of doublets at 6 3.38 and 3.35 ppm with J = 11.5 and 3 Hz representing the B-part of the AB - quartet were seen. Molecular model examination revealed that 87_ should exist largely in the conformation 87a for in the second conformer 87b a severe, repulsive interaction between the C(l) methyl group and the axial i i hydrogens at C(4 ) and C(6 ) of the 1,3-dioxan chain occurs. Therefore, i t appears reasonable to assume that the 'H nmr resonances originated from one single conformer, 87a. In cyclohexanes the axial protons normally resonate upfield from the equatorial ones, the difference being in the order of 0.1 -0.7 ppm. A similar relationship holds true for 1,3-dioxans and 1,3,5-trioxans. With cyclohexanes, at least, exceptions to this rule are possible. Partic-u lar ly , an adjacent equatorial methyl substituent causes a 0.3 - 0.5 ppm upfield sh i f t in the resonances of both the axial and equatorial a protons 81 whereas a neighbouring axial methyl substituent shields the equatorial a protons to the extent of about 0.2 ppm. 6 8 Thus with the r ight combination of numerical values,a reversal of the expected order of resonance for axial and equatorial protons is possible. Moreover, substituents in a 1,3-syn-axial configuration with respect to a proton such as the C(8) - C(9) bond relative to the axial C(4') and C(6') protons in 87a can induce a noticeably downfield sh i f t of that proton's signal by about 0.2 - 0.3 ppm. I t has to be kept in mind that these deductions have been arrived at with cyclohexane derivatives. The effect of substituting two carbon by two oxygen atoms is not being taken into account. However, the relative resonance position of axial and equatorial protons as well as the geminal coupling constants for i i the C(4 ) and C(6 ) protons in 1,3-dioxans are akin to cyclohexane compounds. With this apparent s imi lar i ty the effects discussed earl ier may well be operative in 1,3-dioxan-type ketals as well . Using the information just out-l ined, the observed resonance pattern for the OC^ protons in 87_ can be rationalized by assuming that (a) the C(4') methylene protons are magnetically not equivalent to those on C(6* ) and (b) the axial protons, by geminal coupling to the equatorial ones with J = 11.5 Hz give, rise to two doublets at 6 3.72, 3.66 ppm. Separated by 4 bonds in a near-planar configuration,the equatorial protons are ideally arranged for long-range W - coupling. A narrowing of the separation between the C(4') proton resonances from 7 Hz in the case of the axial to 3 Hz for the equatorial protons, equal in magnitude to the order of long-range coupling between the equatorial protons, generates two ' t r i p l e t ' structures with an apparent coupling constant of 3 Hz, exactly as observed. Three other pieces of evidence fu l l y support this interpretation. A computer simulation of this part of the ^H nmr spectrum using data extracted by above analysis is superimposable on the actual resonance (Figure 4). In the 82 Figure 4: a) OCH;? Resonance in the H nmr spectrum of 87. b) I ts computer simulation i " 270 MHz H nmr spectrum of the ketal 87_,the separation between the C(4 ) and 1 C(6 ) methylene proton resonances was increased. The axial protons gave rise to two separated doublets (A-branches of two AB - pairs of doublets). The equatorial protons showed as two 'doublets' (J=3Hz) with a t r i p l e t - l i k e 1 1 resonance (J=3Hz) in between. This pattern is formed when the C(4 ) and (6 ) equatorial protons resonate 8 Hz apart from each other. Geminal coupling to the axial protons with 11 Hz, to give the B - branches of the AB - quartet, combines with further long-range coupling of 3 Hz magnitude to produce the observed resonance pattern. I t represents two overlapping pairs of doublets. In 5,5-dimethyl-l ,3-dioxan-type ketals, such as 87_, another planar W - arrangement is possible between the axial C(4 ) or C(6 ) protons and the protons of the C(5 ) axial methyl group. I f there were long-range interaction between these protons,the mul t ip l ic i ty of the axial proton resonance would have to be d is t inc t ly dif ferent to what was observed in the OCH2 region. Moreover, one methyl group resonance would have to be a doublet. Al l ter t iary methyl resonances of 87_ showed as dist inct singlets, however, and indeed, a decoupling experiment with irradiat ion at 5 0.90 and 1.10 ppm, did not produce 83 any change in the OCh^  resonance pattern. A sl ight line-sharpening was observed for the lower f i e ld pair of doublets upon irradiat ion at 6 1.10 ppm, indicating a very small interaction between the axial protons and the methyl protons responsible for the 6 1.17 or 1.15 ppm singlet. In the compounds 86, 87, and 90_ the highest and lowest f i e ld methyl resonances appeared consistently at 6 0.73 ± 0.03 ppm and 6 1.16 ± 0.01 ppm. Analogous to 12} earl ier observations, and assuming l i t t l e influence of changes in the C(5) substituent on the resonances of the remote ketal methyl groups, the la t ter are thought to correlate to those resonances. In the case of the ketone 8_7 no dist inct ion could be made on this basis between the 6 1.17 and 1.15 ppm resonances as they are in very close proximity. However, comparison of signal intensities in a l l three compounds 86, 87 and 90_ suggested attr ibut ing the 6 1.17 ppm singlet to one of the ketal methyl substituents. This lowest f ie ld singlet usually showed the smallest signal height, next to that at highest f i e l d , and i t also exhibited the largest half-height width of a l l methyl resonances. Tertiary axial methyl substituents in cyclohexane derivatives are well known to undergo small (0.4 - 0.8 Hz) coupling to axial 68 hydrogens on the adjacent carbons via a planar W - path. This trans methyl-proton interaction usually is noticed by a broader methyl resonance compared to where such a planar arrangement is impossible. Thus, a small interaction between the axial C(4', 6') protons and the protons of the axial C(5') methyl group appears possible. In addition, an equatorial methyl substituent attached to a six-membered ring in a chair conformation was 122 found to resonate upfield from the same group axially positioned. These phenomena in relation to the result of the decoupling experiment described earl ier- reinforced the conclusion that Ca) in the 5,5-dimethyl-l,3-dioxan-type ketal 87_ an inversion of the normal order of equatorial-axial proton resonance occurs and (b) the lower f ie ld singlet of the two resonances 84 arising from the two C(5 ) methyl substituents is due to the axial C(5 ) substituents. No further resolution of the OCH^  coupling pattern could be achieved with a 400 MHz spectrum. The successful oxidation of 86 to 87_ rendered C(5) highly electrophi l ic, ready to provide an access to the necessary endo isopropyl substituent. The synthetic proposal provided for the reaction of 87 with isopropylmagnesium bromide or the equivalent l ithium reagent, dehydration of the resulting ter t iary alcohol to the endocyclic olefin 88_, hydrogenative conversion of 88 to the C(5) endo isopropyl substituted derivative and, f i n a l l y , unmasking of 0 * V i. > — L i i i . - H 2 0 1 2 88. 7_ the C(9) keto functionality to yield the objective of our e f for ts , 9-pupu-keanone (7_). In the event, isopropenyllithium rather than the fu l l y saturated derivative was chosen as nucleophilefor the transformation of the isopropyl-123 adduct 89_ to 9-pupukeanone (7_) had previously met with d i f f i cu l t i es . The reaction of isopropenyllithium with the ketone 87 in ether proceeded O without disturbances and a high yield of the a l l y l i c _ _ I O H alcohol 90 could be isolated. On the basis of 42 steric considerations the structure 90 was O H assigned to the product, the hydroxyl function 3SL being in the endo position. In the i r spectrum of 90 a strong hydroxyl absorbance at 3595 cm~V and an olef inic absorption band at 1638 cm~^  were conclusive proof of adduct formation. The H^ nmr spectrum further asserted the structural assignment 90. 85 P Li -78°C 87 30 Careful examination and comparison with earl ier C(5) diastereomeric compounds (e.g. 713, 1_8, 86_) uncovered no sign of the product being a mixture of stereoisomers. The terminal vinyl ic protons are non-equivalent and gave rise to a broad one-proton singlet at 6 4.90 ppm and another, one-proton t r i p l e t - l i k e resonance at 6 4.80 ppm. Scale expansion and amplification of the spectrum revealed the la t ter as an unresolved quintet- l ike signal. I t i s , in fact , an overlapping double quartet formed by geminal and a l l y l i c coupling of the same magnitude, 1.5 Hz. With the a l l y l i c interaction characteristics of propenes, cisoid coupling constants usually being larger than transoid ones, the 6 4.80 ppm signal was assumed to arise from the vinyl proton cis to the methyl substituent. In a double resonance experiment, i rradiat ion at the resonance position of the vinyl methyl group induced the transformation of both vinyl proton resonances into two sharp doublets with J = 1.5 Hz. This result affirmed the presence of a l l y l i c coupling for both vinyl protons, albeit unresolved in the case of the proton in a trans relationship to the methyl group, as well as geminal coupling, both in the order of 1.5 Hz. Decoupling of the olef in ic protons produced a noticeable intensity increase and sharpening of the vinyl methyl resonance. The OCH2 proton resonances of the ketal moiety in the compound 90_ appeared as a broad AB - quartet. Highly stereoselective addition of isopropenyllithium to 87_, most l ike ly from the exo face of the molecule,was deduced from the presence, 86 in the " 'H nmr spectrum, of only four sharp s inglets fo r the four t e r t i a r y methyl groups and only two resonances fo r the v iny l protons. Boron t r i f l u o r i d e etherate in ether at room temperature smoothly accomplished simultaneous deketal izat ion 1 2 ^ a and dehydration 1 2 4 b of the ketal alcohol 90 to the keto diene 20. The absence of a hydroxyl absorbance and a carbonyl absorption at 1721 cm - 1 in the i r spectrum of 20_, a one-proton s ing le t at 6 5.80 ppm and the charac ter is t i c doublet near 5 2.4 ppm (J=3Hz) fo r the C(8) a methylene protons in the "'H nmr spectrum corroborated BF3 • Et20 2P_ the s t ruc tura l assignment 20_. The spectral data f o r the compound 20 agreed well w i th published data. ^8,30b _ Hydrogenation of the diene 20_with i r id ium black as cata lyst has been 28 shown to proceed with high s te reose lec t i v i t y , and i t l a s t l y provided us with a sample of 9-pupukeanone (7). The ^H nmr spectrum of the product 1_ lr.H 2 EtOH 2 0 gave no ind icat ion of the continued presence of v inyl protons and a broad s ing le t resonance at 60.91 ppm was assigned to the isopropyl methyl groups and one of the two t e r t i a r y methyl subst i tuents. Comparison of spectral 87 data and coinjection of the product with an authentic sample 125 of 7 into three different glc columns ascertained the identity of the reduction product with 9-pupukeanone (7). In both previous syntheses of 9-isocyanopupukeanane (5),the keto functionality of 1_ was elaborated to the C(9) i son i t r i le group in 5_, therefore the present synthesis of 9-pupukeanone (.7) formally constitutes a total synthesis of the marine natural product 5_. 8. Conversion of the Diels-Alder Adduct 78 into the Trisylhydrazone 97 Prior to the synthesis of 9-pupukeanone (7_), as discussed in section I I . 7, a di f ferent approach to 7_ from the alcohol 78 was attempted. This section wi l l discuss this unsuccessful endeavour. In the last 15 years organocopper reagents have received widespread attention from organic chemists, part of the generally higher level of interest in the chemistry of organometallie compounds. They turned out to be 126 highly useful reagents for C-C bond formation, complementary to the well-established Grignard and organolithium reagents. The cross-coupling reaction between an organic halide and organocopper compounds has been developed to a general synthetic method and i t s potential for the synthesis of an endocyclic olefin such as 91_ was investigated in the course of this synthetic work. I f 91_ or the related C(5) isopropyl substituted endocyclic alkene 93 could be prepared, the transformation to the target compound 1_ by a reaction ii. reduction i. base 51 7 88 sequence involving hydrogenation or hydroboration was thought to be feasible. Synthesis of the alkene 93_ could be envisaged, on paper, from the ketone 87  via a nucleophilic displacement of the phosphate group in the intermediate enol phosphate 9j[. This proposal suffers the serious drawback of the coupling reaction 9j? to 93_, by means of an organocuprate,being a l i t t l e investigated 127 transformation. Blaszczak et a]_. found that unactivated enol phosphates give a good yield of coupled product only with the more reactive l i thium di(n-butyl]cuprate. Enol phosphates activated by being in the B position of an a,g-unsaturated ester can be substituted in high yield with the less 128 reactive l ithium dimethylcuprate. Successful conversion of the compound 92 into 93_ appeared thus quite doubtful and, weighed against the next possib i l i ty , this idea did not seem worth further investigation. I t is well known that l i thium diorganocuprates, the C-Cu(I) bonds of which show a simultaneous stabi l ization towards thermal decomposition and an increased nucleophi1icity of the carbon substituent as compared to uncomplexed 89 alkylcopper ( I ) reagents, react well with vinyl halides to the alkylated olef in (Scheme 7). On the other hand, the synthesis of a substituted , N / X C=C + LiCuR2 s c=cf L i C u v v .* V V t>=<!)2 *Rx -"" Scheme 7 alkene by coupling of an alkenylcopper reagent with an alkyl halide proceeds 130 cleanly with primary halides but with secondary halides success is errat ic and comes in any case only with l i thium cuprate complexes. 1 2 6 a The corresponding vinyl l i thium reagents do not offer any advantage as only 131 with primary halides useful yields of substitution products are obtained. Alkylation of vinyl halides to d i - and tr isubsti tuted olefins has also been 132a accomplished with Grignard reagents in hexamethylphosphoramide and by transit ion metal mediated bond formation. Thus, alkylation of the halides 129c 94a,b with lithium diisopropylcuprate or, possibly, related cuprate 133 reagents to the olef in 9J_ is conceivable. To circumvent possible complications the enol phosphate moiety might have to be transformed to another protecting group, such as a ketal. As such a route to 9-pupukeanone (7_) involving a cuprate mediated coupling reaction represented an attractive ° fi OP(OEt ) 2 p P ( 0 E t ) 2 L i C u ( i P r ) 2 9 4 a . b X = I . B r 91 !5_ X = L i 90 alternative to the actually realized sequence we directed our attention to the alkenyl halides 94a,b. Either the iodide or bromide should easily be available by electrophil ic interception of the vinyl l i thium compound 95_, the preparation of which was attempted. Ketones can be reduced to olefins by treating their p_-toluenesulfonyl-hydrazones (tosylhydrazones) with excess alkyl l i thium or l i thium diisopropyl-amide 134a followed by protonation of the intermediate vinyl l i thium derivative (Scheme 8). This has been made use of, for instance, to achieve 1,2-carbonyl transposition, Electrophiles other than the proton also trap the inter-mediate vinyl l i thium derivative and provide an entry to functionalized alkenes. ' I f these rather than the parent olefins are required the use of tosylhydrazones is problematic because the ortho aryl hydrogens are i \ J \ J - N - S 0 2 A r 2RLi L i k ^ N - N - S 0 2 A r I r Scheme 8 E = elect rophile j \ / N = N L i J acidic enough to protonate in situ the generated vinyl l i thium thereby reducing the yield of the substituted alkene. This problem is avoided by either using three or more equivalents of the alkyl l i thium reagent in tetramethylethylenediamine as cosolvent or by taking recourse to the correspond!'ng 2,4,6-triisopropy 1 benzenesulfonylhydrazones (trisylhydrazones). 136 Similar in behaviour to,and more reactive than,the analogous tosylhydrazones 91 they are the derivatives of choice for trapping the vinylcarbanion inter-mediates with externally added electrophiles. With the goal of synthesizing the necessary C(5) trisylhydrazone 9_7 in mind,the alcohol 78_was oxidized to the ketone 96 with pyridinium dichromate in dichloromethane. The compound 78 could be used as crude material without adverse effects on the yield of the oxidation. An intense absorption band at 1742 cm~^  in the i r spectrum of 96_, next to the typical enol phosphate absorbances,established the formation of the cyclopentanone derivative 96_. The H nmr spectrum showed, among others, the vinyl proton resonance at 6 5.78 ppm as a double doublet, and an AB - quartet with J = 18 Hz at 6 2.38, 2.12 ppm, due to the C(4) a methylene protons. The a methine proton resonance 31 was buried under this AB - quartet. P decoupling induced the expected changes, collapse of the vinyl proton resonance to a doublet (J=7 .5Hz) , of the OCH2 signal to a quartet and a sharpening of the t r i p l e t at 6 1.39 ppm. In the 270 MHz spectrum the a methylene proton resonance appeared separated from the C(6) methine signal. The la t ter was a double doublet at 6 2.28 ppm with J = 12 and 4 Hz, generated by coupling to the a l l y l i c proton ( 4 H z ) , a large C(6) proton - C(10, exo) proton and a nearly zero C(6) proton - C(10, endo) proton interaction. This pattern was reminiscent of what was observed for the C(5) proton-C(4) methylene protons coupling in the C(5) hydroxyl substituted compounds discussed ear l ier . A broadened t r i p l e t structure, in the 27Q MHz spectrum, at <5 1.77 ppm, the 3 resonance lines separated by 12 Hz, IS. 9 6 92 was attributed to the C(1Q} methylene protons. Spectral data and glc examination gave no indication of this product not being a stereochemically pure, single compound. The presence of the acid-labile enol phosphate group could have posed a problem in the following reaction, the preparation of the trisylhydra zone 97. Fortunately, the condensation reaction between the ketone 96_ and trisylhydrazide proceeded smoothly without requiring acid catalysis, and a high yield of the product 9_7 could be realized by just s t i r r ing the two components in tetrahydrofuran at ambient temperature. The compound 97 was obtained as an amorphous solid which could not be induced to crystal l ize. 31 All available spectral evidence corroborated the structural assignment 97_. In the i r spectrum of 97_ the carbonyl absorption present in the spectrum of the precursor 96 had disappeared, and at 1600 cm"1 an aromatic absorbance was found. In the nmr spectrum, a singlet at 6 7.22 ppm was assigned to the aromatic protons. The resonances of the two benzylic protons of the ortho isopropyl groups were buried under the multiplet resonance of the OCH^  protons. The benzylic proton of the para isopropyl substituent resonated as a multiplet at 6 2.94 ppm. Four singlets in the 270 MHz spectrum, at 6 1.06, 1.04, 1.02 and 1.00 ppm and due to the ter t iary methyl groups, indicated a mixture of stereoisomers around the carbon-nitrogen double bond. Further ver i f icat ion 93 of trisylhydrazone formation was provided by the mass spectrum of 97 which showed a molecular weight of 60.3 as well as loss of an isopropyl and of arylsulfony! and arylsulfonamidyl fragments from the molecular ion. Successful synthesis of the trisylhydrazone 9_7 nourished optimism about the ab i l i t y of obtaining the proposed endocyclic isopropyl substituted olef in 91 by the discussed sequence of reactions. Unfortunately, and quite unex-pectedly, i t proved impossible to bring on decomposition of the t r i s y l -hydrazone 9_7_ to the vinyl l i thium intermediate 95_. When the trisylhydrazone 97_ was treated with 2.2 equivalents of sec-butyl 1 ithium in n-hexane-tetra-methylethylenediamine (1:1) at -78°C (2.5 hours), then at room temperature for 45 minutes, addition of the electrophile ( ^ and work-up resulted in near quantitative recovery of starting material. A similar result was obtained with 3.2 equivalents of n-butyllithium at -78°C (2 hours), then 0.5 hours at 0°C, Reaction of the trisylhydrazone 97_ with n-butyllithium at -64°C to 0°C, followed by the addition of D^O as electrophile (see Experimental Section) produced a mixture consisting mainly of starting material and one product. Both were isolated by preparative t i c pur i f icat ion, and careful mass spectral analysis of the starting material did not reveal signs of deuterium incorpo-rat ion. The i r spectrum of the product, to which the structure 98 was assigned, showed absorbances at 1720, 1662 and 1600 cm-"' arising from the six-membered ring ketone, the carbon-nitrogen double bond and the aromatic r ing, respectively. fl D 94 The H nmr spectrum of the product 98 affirmed the retention of the trtsylhydrazone moiety and the disappearance of the enol phosphate group. In the high resolution mass spectrum of 98 two ions of equal intensity at m/e 473.2802 and m/e 472.2767 were shown. The f i r s t value agrees with incorpo-ration of one deuterium atom at the C(8) methylene group. The relat ively high intensity that was measured for the second ion was presumably either due to the prominent loss of one hydrogen atom or the compound was a mixture of mono and non-deuterated material of the structure 9S_. I t is not quite obvious why the trisylhydrazone 9_7 would be as reluctant as i t seems to be to undergo alky!l i thium induced decomposition to the vinyl l i thium inter-mediate 95, Unable to effectuate the desired transformation of 97 into the vinyl iodide 94a no further time and energy were expended into preparing the proposed endocyclic olef in 91_. 92 95 91 R=i-C3H7 95 I I I . EXPERIMENTAL SECTION General Information Melting points were determined with a Fisher-Johns melting point apparatus and are uncorrected. Dis t i l la t ion temperature (dist . temp.) refers to the mean air bath temperature during a short-path d i s t i l l a t i o n . Infrared ( i r ) spectra were recorded on Perkin Elmer model 710 or 71 OB infrared spectrophotometers. The proton nuclear magnetic resonance (^ H nmr) spectra were taken in deuterochloroform solution on Varian Associates Spectrometer models HA-100 or XL-100 (100 MHz), on a Bruker WP-80 (80 MHz) instrument and on a 270 MHz unit consisting of an Oxford instrument 63.4 KG superconducting magnet and a Nicolet 16K computer attached to a Bruker TT-23 console. Signal positions are given in parts per mil l ion (ppm) downfield from tetramethylsilane as internal standard. In cases of compounds containing t r i a l k y l s i l y l groups the resonance positions were determined relative to the chloroform signal. The mul t ip l i c i ty , number of protons, coupling constants and assignments are 13 given in parentheses. C nmr spectra were measured at 20.1 MHz on the Bruker WP-80 spectrometer. Analytical gas l iquid chromatography (glc) was performed on a Hewlett-Packard HP 5832 A gas chromatograph, using either a 6 f t . x 0.125 i n . , 5% 0V-210 on - Chrpmosorb W (100 - 120 mesh) or a 6 f t . x 0.125 i n . , 5% 0V-17 on Chromosorb W(100 - 120 mesh) column. For column chromatography either Si l ica Gel 60 (E. Merck, 70 - 230 mesh) or F lor is i l (J.T. Baker Chemical Co., 100 - 120 mesh) was u t i l i zed. Analytical thin-layer chromatography ( t i c ) was carried out on commercial, pre-coated Sil ica Gel plates with fluorescent indicator (Eastman Kodak, Sheet Type 13181). Visualization was effected either by iodine vapour staining or with short-wavelength u l t ra-v io let l igh t . Preparative t i c was carried out on 20 x 20 cm glass plates coated with 0.7 mm 96 of neutral Sil ica Gel GF^^ (Type 60) for Tic (E. Merck). Low resolution mass spectra were recorded with a Varian/MAT CH 4 B mass spectrometer, for high-resolution measurements a Kratos AEI MS 50 on MS 902 instrument was used. Microanalyses were performed by Mr. P. Borda, Microanalytical Laboratory, University of Bri t ish Columbia. All reactions involving air and moisture sensitive reagents were carried out under an atmosphere of dry nitrogen or argon using either oven or care-fu l l y flame-dried glassware. Liquid reagents or solutions were introduced into the reaction flask through a rubber septum with a syringe equipped with a hypodermic needle. Tetrahydrofuran and diethyl ether were freshly d is t i l l ed from lithium aluminum hydride or sodium benzophenone kety l ; benzene, pentane, hexane, dichloromethane and xylene from calcium hydride. Hexamethylphosphor-amide was d is t i l l ed from barium oxide; dimethyl sulfoxide, diisopropylamine, triethylamine and tetramethylethylenediamine from calcium hydride and kept over 4 H molecular sieves or a piece of calcium hydride. Dis t i l la t ion from magnesium ethoxide afforded dry ethanol. 67a Preparation of the Tetrahydropyranyl Ether of g-Bromoethanol To a solution of a-bromoethanol (3.5 ml, 50 mmol) in 40 ml of benzene containing a catalyt ic amount of p_-toluenesulfonic acid were added 4.6 ml (50 mmol) of dihydropyran. The reaction mixture was st i rred under nitrogen at room temperature for 2 hours, then f i l te red through basic alumina, act iv i ty I I I (Woelm). Solvent evaporation and d i s t i l l a t i on (dist . temp. 42°C/0.2 T o r r ) ( l i t . 6 7 a 94°C/14 Torr) of the residual l iquid yielded 10.34 g (99%) of pure product. i r ( f i lm) : u 1118, 1025 (C-0), 970, 895, 860 cm" 1; ]H nmr: 4.69 (brs, IH, 0CH0),4.16- 3.44 (m, 6H, 0CH2 and BrCH2), 1.97 - 1.48 (m, 6H, CH2); 97 mass spectrum; m/e 209 and 207 (M +, 100%). Anal, calcd. for C 7H 1 3Br0 2 C 40.21, H 6.27; found: C 40.26, H 6.24. Preparation of the tert-Butyldimethylsilyl Ether of g-Bromoethanol To a solution of imidazole (6.3 g, 92.5 mmol) and tert-butyldimethyl -chlorosilane (14.6 g, 97 mmol) in 20 ml of dry dimethylformamide were added 5.7 ml (80 mmol) of a-bromoethanol. After s t i r r ing the mixture at room temperature for six hours i t was poured into 100 ml of water and twice extracted with hexanes. The combined organic extract was washed with saturated aqueous sodium bicarbonate solution and with brine, then dried over anhydrous magnesium sulfate. Solvent evaporation and d i s t i l l a t i on (dist . temp. 83°C/12 Torr) of the residual colourless l iquid gave 17.2 g (90%) of the pure product. i r ( f i lm) : u m a v 1255 (SiMeJ, 1123 (C-0), 1098 (Si-0), 853 (Si-C) cm" 1 ; max c ]H nmr: 3.38 ( t , 2H, J=7Hz, 0CH2), 3.37 ( t , 2H, J=7Hz, BrCH2), 0.88 (s, 9H, te r t -bu ty l ) , 0.05 (s, 6H, CH3); mass spectrum: m/e 225 and 223 (M+ - CHg), 183 and 181 (M+ - ter t - buty l ) , 137 (100%). Exact mass calcd. for C?H16BrOSi (M +-CH 3 ;7 9Br): 223.0154; measured: 223.0152; calcd. for C 4 H 1 0 BrOSi (M+ -te r t -buty l ; 7 9 B r ) : 180.9684; measured: 180.9685. Preparation of 4,6-Dimethylcyclohexane-l,3-dione (48) To a vigorously s t i r red, freshly prepared solution of sodium ethoxide (0.25 moi) in 100 ml of anydrous ethanol were added dropwise 36 g(0.25 moi) of ethyl 2-methylacetoacetate (previously d i s t i l l ed ) . Dist i l led ethyl methacrylate (29 g,0.25 moi) was added to the resultant creamy-white slurry over a period of 15 - 20 minutes. The continuously st i r red reaction mixture was gently refluxed, after addition was complete, on a steam-bath for 2 hours. 98 To the resulting green-yellow solution was added dropwise, over a period of 25 minutes, a solution of 31 g of potassium hydroxide in 142 ml of water. Gentle reflux on a steam-bath was maintained for further 6 hours. The s t i l l hot solution was acidif ied with 4N hydrochloric acid against Litmus paper and most of the solvent removed in vacuo. The colourless, crystal l ine precipitate thus obtained was f i l te red off and a second crop of product isolated from the mother liquor by concentrating and cooling i t . The combined product was washed with water unti l a neutral eluant was obtained and dried in the desiccator to yield 21.6 g (62%) of the dione 4J3. An analytical sample was obtained by recrystal l ization from water; m.p. 112° -113°C ( l i t . 4 9 114°C). i r (CHC1,): u m = v 1735 and 1710 (carbonyl of the diketoform), 1604 (carbonyl, double bond of the enol form); H^ nmr: 8.12 (brs, OH), 5.44 and 5.38 (2 brs, vinyl H), 3.52 and 3.37 (A3-q, J=16Hz, C(2) methylene)(al1 signals: 2H,D20 exch.), 2.88 - 2.41 (sym. m, J=6.5Hz, ctH) and 2.32 - 1.80 (m, 3H both together), 1.40 - 1.09 (m, 7H); mass spectrum: m/e 140(M+, 100%), 112(M+ -CO). Exact mass calcd. for G Q H ^ O Q : 140.0837; measured: 140.0843. Anal. calcd. for CgH1 202: C 68.54, H 8.63; found: C 68.59, H 8.64. 48 a Preparation of 4,6-Dimethyl-3-isobutoxy-2-cyclohexen-l-one (54) O 99 A mixture of the dione 48 (20 g, 0.14 moi), d is t i l l ed isobutanol (35 ml, 0.378 moi) and a catalytic amount of rj-toluenesulfonic acid was refluxed for 4.5 hours in 130 ml of benzene with continuous removal of the water formed. After the mixture had cooled to ambient temperature i t was poured into saturated aqueous sodium bicarbonate solution which was then extracted with ether. The organic phase was washed with brine arid dried over anhydrous magnesium sulfate. Removal of the solvent followed by d is t i l l a t i on (d ist . temp. 75°C/0.1 Torr) of the residual l iquid furnished 26.5 g (95%) of the pure enol ether 54_ as a clear, colourless l iqu id ; Rp = 0.46 in 20% ethyl acetate in hexanes. i r ( f i lm) ; u m = v 1660 (O0) s 1596 (OC), 1220 (C-0); ]H nmr: 5.31 (d, J=1.5Hz, vinyl H) and 5.26 (s, vinyl H, IH both together), 3.61 and 3.59 (2d, J=6Hz, 2H, 0CH2), 2.90 - 1.50 (m, 5H, a ' r l , Y H, CH, CHg), 1.26, 1.18, 1.14, 1.13 (4d, J=7Hz, 6H, ring CHj), 1.00 (d, J=7Hz, 6H, isobutyl CH3); mass  spectrum: m/e 196 (M +). Exact mass calcd. for C-|2H20°2 : 196.1463; measured: 196.1466. Anal, calcd. for ^2H20°2: C 73.43, H 10.27; found: C 73.20, H 10.35. Alkylation of 54 to the Tetrahydropyranyl Ether 55 A solution of the ketone 54 (5 g, 25.5 mmol) in 5 ml of tetrahydrofuran was added to a st i rred solution of 26.8 mmol of l i thium diisopropylamide in 35 ml of tetrahydrofuran and 6.6 ml of hexamethylphosphoramide at -78°C. After 20 minutes the bath temperature was raised to 0°C and after 30 minutes at this temperature, 8 g (38 mmol) of the tetrahydropyranyl ether of a-bromoethanol, dissolved in 5 ml of tetrahydrofuran, were added. The reaction mixture was st irred at Q°C for 45 minutes, then at ambient temperature for 46 hours. I t was diluted with aqueous saturated sodium bicarbonate solution and twice TOO extracted with hexanes. The combined organic phase was washed with brine and dried over anhydrous magnesium sulfate. The solvent was evaporated and excess alkylating agent removed by fractional d i s t i l l a t i on (55°C water bath temperature/0.1 Torr) of the residual l iqu id. The d is t i l l a t i on residue was chromatographed on s i l i ca gel with 20% ethyl acetate in hexanes as eluant to afford, after solvent removal, 6.65 g (80.6%) of the pure ketone 55 as a viscous, l ight yellow o i l . From fractions preceding those containing the product, 251 mg (5%) of the starting material 54 could be recovered. An analytical sample of the product 55_ was obtained by short-path d i s t i l l a t i on (dist . temp. 175°C /0.15 Torr); Rp=0.36 in 20% ethyl acetate in hexanes. i r ( f i lm) : u a 1658 (C=0), 1598 (C=C), 1200 (=C-0), 1023 (C-0) cm"1; -—~ — ~~~ max ]H nmr: 5.17 (brs, IH, vinyl H), 4.50 (brs, IH, 0CH0), 3.97 - 3.68 (m, 2H, 0CH2), 3.62 - 3.22 (m, 0CH2) and 3.53 (d, J=6.5Hz, isobutyl 0CH2, 4H both together), 3.00 - 2.60 (m, IH, a l l y l i c H), 2.21 -1.51 (m, 11H, CH2, CH), 1.15 (d, J=7Hz, C(4) methyl) and 1.08 (s, ter t iary methyl, 6H both together), 0.97 (d, J=6.5Hz, 6H, isobutyl methyl); mass spectrum: m/e 324 (M +), 240, 223, 196 (100%). Exact mass calcd. for C | g H 3 2 0 4 : 324.2300; measured: 324.2294. Anal, calcd. for C ^ H ^ : C 70.33, H 9.94; found : C 70.49, H 9.92. Reduction - Hydrolysis of 55 to the a-Enone 57 101 To a suspension of 460 mg (12.2 mmol) of l i th ium aluminum hydride in 72 ml of anhydrous ether at 0°C was slowly added a solut ion of the 8-isobut-oxy a-enone 55_ (7.5 g, 23.1 mmol) in 17 ml of anhydrous ether. Af ter the reaction mixture had been s t i r r e d at 0°C fo r 2.5 hours, a saturated aqueous potassium carbonate solut ion was added dropwise with e f f i c i e n t s t i r r i n g u n t i l a granular, colourless prec ip i ta te separated. This was f i l t e r e d o f f and washed thoroughly wi th warm ether. The solvent was evaporated from the f i l t r a t e under reduced pressure and the product thus obtained dissolved in 100 ml of a mixture of acetic acid-acetone (1 :1 ) . The resu l t ing solut ion was s t i r r e d at room temperature for 1.5 hours, d i lu ted wi th 100 ml of water and twice extracted wi th hexanes. The combined hydrocarbon phase was washed with aqueous saturated sodium bicarbonate so lu t ion , then wi th brine and dried over anhydrous magnesium su l fa te . Evaporation of the solvent under reduced pressure and chromatography of the residual crude product on s i l i c a gel wi th 25% ethyl acetate in hexanes provided the pure a-enone 5J_ in 80% y i e l d (4.65 g). Short-path d i s t i l l a t i o n ( d i s t . temp. 135°C/0.2 Torr) furnished an analyt ica l sample of 57_; Rp=0.36 in 25% ethyl acetate in hexanes. i r ( f i lm) : u m a x 1680 (C=0); 1030 (C-0); ]H nmr : 6.72 - 6.57 (m, IH, @ vinyl H), 5.91 (d , J=10Hz, IH, a v inyl H), 4.53 (b rs , IH, 0CH0), 4.00 - 3.70 and 3.58 - 3.32 (2m, 2H each, 0CH 2), 2.76 - 2.36 (m, IH, a ' H), 2.12 - 1.40 (m, 10H, CH 2), 1.22 - 1.06 (m, 6H, CH 3); mass spectrum: m/e 252 (M + ) , 225, 169 (100%). Exact mass calcd. fo r CjgH 0 : 252.1725; measured: 252.1713. Anal, calcd. for C 1 5 .H 2 4 0 3 : C 71.39, H 9.59; found: C 71.09. H 9.62. Preparati on of the Compound 59 When the ketone 55_ (1.56 g , 4.8 mmol) was treated with l i t h ium 102 aluminum hydride (170 mg, 4.4 mmol) at ambient temperature in ether for 100 minutes and then following the same procedure as described for the preparation of the a-enone 57, 120 mg (8%) of the compound 59 could be isolated after chromatography. An analytical sample was obtained by short-path d i s t i l l a t i on (dist . temp. 138°C/0.13 Torr); Rp=0.57 in 20% ethyl acetate in hexanes. i r ( f i lm) : u 1195, 1074, 1022 (C-0) cm' 1 ; ]H nmr: 5.93 - 5.45 * max (m, 2H, vinyl H), 4.59 (brs, IH, 0CH0), 4.01-3.03 (m, 7H, 0CH2> a l l y l i c H), 2.16 - 1.28 (m,12H, CH2> CH), 1.09 - 0.85 (m, 12H, CH3); mass spectrum: m/e 310 (M +), 253 (M +-C 4H g). Anal, calcd. for C 1 9 H 3 4 0 3 C 73.50, H 11.04; found: C 73.60, H 11.00. Preparation of the tert-Butyldimethylsilyl Enol Ether 60_ To a solution of l ithium diisopropylamide (2.2 mmol) in 5 ml of tetra-hydrofuran at 0°C, which was prepared from 350 yl (2.5 mmol) of diisopropyl-amine and 1.27 ml (2.2 mmol, 1.76 M solution in hexane) of n-butyll i thium, was added 500 mg (1.99 mmol) of the a-enone 57_,dissolved in 2 ml of tet ra-hydrofuran. The solution was st irred at this temperature for 40 minutes, then 1.2 ml of hexamethylphosphoramide followed by 520 mg (3.45 mmol) of freshly sublimed (aspirator vacuum, ambient temperature) ter t -buty ld i -methyl chlorosi lane was added. St irr ing was continued at 0°C for 75 minutes and for another 9 hours at room temperature. The reaction mixture was diluted with saturated aqueous sodium bicarbonate solution and extracted with ether. The organic phase was twice washed with brine and dried over anydrous sodium sulfate. The solvent was removed in vacuo and the residual o i l chromatographed on F lor is i l with 20% ethyl acetate in hexanes as eluant to f ina l l y yield 581 mg (80%) of the pure s i l y l enol ether 60_. An analytical 103 sample was obtained by short-path d i s t i l l a t i on (dist . temp. 127°C/0.08 Torr); Rp=0.73 in 25% ethyl acetate in hexaaes. i r ( f i lm) : u m a x 1664 (C=C), 1255 (SiMe2), 1031, 872, 841 (Si-C), 783; ]H nmr: 5.58 and 5.39 (AB-q, J=10Hz, 2H, vinyl H), 4.52 (brs, IH, 0CH0), 4.00 - 3.24 (m, 4H, 0CH2)„2.17, 2.16 and 1.98, 1.97 (2 AB-q, J=16Hz, vinyl ic CH2), 1.78 - 1.25 (m, CH2) and 1.65 (s, v inyl ic CH3, 11H both together), 1.02 and 0.97 (2s, 12H, ter t iary CH3 and te r t -bu ty l ) , 0.12 (s, 6H, s i l y l CH3); mass spectrum: m/e 366 (M +), 281, 263 (100%). Exact mass calcd. for H 3 80 3Si : 366.2590; measured: 366.2566. Anal. calcd. for C 2 1H 3g0 3Si : C 68.80, H 10.45; found: C 58.60, H 10.45. Preparation of the Enol Phosphate 63_ O II 0P(0Et) 2 To a st i rred solution of l i thium diisopropylamide, prepared from 15.8 ml (28.5 mmol, 1.80M solution in hexane) of n-butyllithium and 4.3 ml (30 mmol) of diisopropylamine in 60 ml of tetrahydrofuran at -78°C were added 17.5 ml of tetramethylethylenediamine and a solution of 5.53 g (22 mmol) of the a-enone 57 in 10 ml of tetrahydrofuran. After 15 minutes at -78°C and 30 minutes at 0°C, 5.1 ml (35 mmol) of diethylphosphorochloridate were added. The reaction mixture was kept at 0°C for 30 minutes, then allowed to warm to room temperature. After 3 hours at the lat ter temperature i t was poured into saturated aqueous sodium bicarbonate solution. The resulting mixture was twice extracted with hexanes, the combined organic extract washed with 104 brine and dried over anydrous magnesium sulfate. Evaporation of the solvent under reduced pressure followed by chromatography of the residual o i l on s i l i ca gel with 50% ethyl acetate in hexanes as eluant yielded 6.29 g (74%) of the pure enol phosphate 63_ as a colourless o i l . An analytical sample was prepared by short-path d i s t i l l a t i on (dist . temp. 165°C/0.1 Torr); Rp= 0.34 in 25% ethyl acetate in hexanes. i r ( f i lm) : u m a j ( 1675 (C=C), 1278 (P=0), 1160 (P-O-Et), 1030 (C-0-P), 975; ]H nmr : 5.83 and 5.46 (AB-d, J=10Hz, 2H, vinyl H), 4.53 (brs, IH, 0CH0), 4.28 - 3.98 (m, 4H, ethoxy methylene), 3.96 - 3.26 (m, 4H, 0CH2), 2.50 - 1.44 (m, CH2) and 1.75 (brs, vinyl ic CH3, 11H both together), 1.30 (dt, J=7, 1Hz, 6H, ethoxy methyl), 1.02 (s, 3H, ter t iary CH3); mass spectrum: m/e 388 (M+), 343, 304, 299, 259 (100%). Exact mass calcd. for C i gH 3 30gP : 388.2015; measured: 388.2020. Anal, calcd. for ^gH^OgP : C 58.75, H 8.56; found: C 58.87, H 8.43. Hydrolysis of the Tetrahydropyranyl Ether 63 to the Alcohol 64 To a st i rred solution of the tetrahydropyranyl ether 63 (2.06 g, 5.3 mmol) in 92 ml of methanol were slowly added 15.5 ml of 1 N hydrochloric acid. The resulting mixture was st irred at room temperature for 1.5 hours, then diluted with 100 ml of water and twice extracted with ether. The combined organic phase was washed with aqueous, saturated sodium bicarbonate solution, then with brine and dried over anhydrous magnesium sulfate. Removal of the solvent under reduced pressure l e f t 1.58 g (98%) of the pure alcohol 64 as a colourless o i l . Short-path d i s t i l l a t i on (d ist . temp. 140°C/0.1 Torr) yielded an analytically pure product (93%); Rp=0.10 in 50% ethyl acetate in hexanes. i r ( f i lm) : u m 3 V 3443 (OH), 1673 (C=C), 1270 (P=0), 1160 (P-0-Et), 171 a X 1038 (br,C-0-P), 980; ]H nmr: 5.93 and 5.50 (AB-q, J=10Hz, 2H, vinyl H), 4.38 - 4.02 (m, 4H, ethoxy CH2), 3.73 ( t , J=7Hz, 2H, 0CH2), 2.53 (brs, IH, 105 OH, D20-exch.), 2.17 (brd, J=4Hz, 2H, vinyl ic CH2), 1.79 (brs, 3H, v inyl ic CH3), 1.65 and 1.60 (2t, J=7Hz, 2H, CHg), 1.36 (dt , J=7, 1.5 Hz, 6H, ethoxy CH3), 1.07 (s, 3H, ter t iary CH3); mass spectrum: m/e 304 (M+), 2.59 (100%). Exact mass calcd. for c i4 H 25°5 P : 3 0 4 - 1 4 3 9 " » measured: 304.1419. Anal. calcd. for C ^ ^ P : C 55.25, H 8.28; found: C 55.05, H 8.42. Oxidation of the Alcohol 64 to the Aldehyde 65 0 II OP(OEt)-CHO To a solution of the alcohol 64 (1.64 g, 5.4 mmol) in 95 ml of dry benzene under an atmosphere of argon were added consecutively 34 ml of dimethyl sulfoxide, 0.445 ml of pyridine (d is t i l led from calcium hydride) and 0.213 ml of tr i f luoroacetic acid (previously d i s t i l l e d ) . After 5 minutes-* 3.6 g of the carbodiimide were quickly added to the well st i rred reaction mixture. After 21 hours at room temperature the solution was cooled with cold water, 100 ml of cold water were added and the resulting two-phase system st irred vigorously for 15 minutes. The mixture was twice extracted with ether, the combined ethereal extract washed with saturated aqueous sodium bicarbonate solution, then twice with brine and dried over anhydrous sodium sulfate. Solvent evaporation afforded 1.50 g (91%) of the aldehyde 65. + After chromatography on s i l i ca gel with ethyl acetate as eluant 1.25 g (76%) of the product were obtained; Rp=0.29 in 50% ethyl acetate in hexanes. i r ( f i lm) : u m a x 1718 (C=0), 1672 (OC), 1273 (P=0), 1165 (P-0-Et), 106 1030 (br, C-O-P), 975; *H nmr: 9.76 ( t , J=?.5Hz, IH, aldehyde H), 6.02 and 5.65 (AB-q, J=10Hz, 2H, vinyl H), 4.34 - 4.03 (m, 4H, ethoxy CH2), 2.36 ( t , J=2.5Hz, 2H, CH2C=0), 2.26 - 2.13 (m, 2H, vinyl ic CH2), 1.75 (brs, 3H, vinyl ic CH3), 1.35 (dt, J=7, 1.5Hz, 6H, ethoxy CH3), 1.17 (s, 3H, ter t iary CH3); mass spectrum: m/e 302 ( M+ ) , 274, 259, 258 (100%). Exact mass calcd. for C ] 4 H 2 3 0 5 P : 302.1283; measured: 302.1296. * 1 - Ethyl-3-(3 -dimethyl aminopropyl)carbodiimide hydrochloride, obtained from Ott Chemical Company, Muskegon, Michigan. Prolonged exposure to the atmosphere should be avoided as the compound is very hygroscopic. + Partial decomposition occurred upon attempted d is t i l l a t i on of the aldehyde 65. The crude product normally was suff ic ient ly pure for further use. Oxidation of the Alcohol 64 with Pyridinium Chlorochromate To a solution of 144 mg (0.47 mmol) of the primary alcohol 64 in 2.5 ml of dry dichloromethane under an atmosphere of argon were added 150 mg (0.79 mmol) of pyridinium chlorochromate. The reaction mixture was st irred at room temperature for 2.5 hours, then f i l te red through F lor is i l with more solvent. The crude product obtained after solvent evaporation from the f i l t r a t e was purif ied by preparative t i c in 70% ethyl acetate in hexanes, and 53 mg (37% based on starting material) of a homogeneous material could be isolated. Short-path d i s t i l l a t i on gave a colourless o i l (d ist . temp. 145°C/0.08 Torr); Rp=0.30 in ethyl acetate. i r ( f i lm) : u v 3450 (OH), 1690 (C=0), 1648 (C=C), 1272 (P=0), 1165 met x (P-0-Et), 1030 (br, C-O-P), 975; ]H nmr: 4.47 - 4.14 (m, 4H, ethoxy CH2), 4.06- 3.89 (m, 3H, 0CH2, 0CH), 2.59 (d, J=4Hz, 2H, vinyl ic CH2), 2.00 (d,-J=2Hz, 3H, "vinylic CHg), 1 .92 - 1.65 (m, CH2, 2H after D20-exch.), 1.36 (dt , J=7, 1.5Hz, 6H, ethoxy CH3); 1.24 (s, 3H, ter t iary CH3); mass spectrum: m/e 107 318 (M +), 290, 275, 262, 247, 234, 206, 178 000%). Exact mass calcd. for C14H23°6P : 3 1 8 - 1 2 3 2 » measured: 318.1228. Preparation of the A l l y l i c Alcohol 69_ O II OP(OEt)2 a) By Reaction of 65 with Vinyllithium To a well st irred solution of the aldehyde 65_ (150 mg, 0.5 mmol) in 1 ml of ether at -78°C were slowly added 0.75 ml of an ethereal solution of * vinyl l i thium (1.1 M). The reaction was allowed to proceed for 8 hours at -78°C. The cooling bath was removed and after 30 minutes, 2 ml of an aqueous saturated sodium bicarbonate solution were added. The resultant two-phase system was st i rred for 10 minutes and twice extracted with ether. The combined organic extract was washed with brine and dried over anhydrous sodium sulfate. Removal of the solvent under reduced pressure and chroma-tography of the residual o i l on F lor is i l with 20% ethyl acetate in hexanes provided a total of 140 mg. (85%) of both diastereomers of the pure a l l y l i c alcohol 69. The f i r s t product containing fractions furnished pure less polar (lp) isomer,the last ones pure more polar (mp) isomer; Rp=0.28 and 0.21 in 40% ethyl acetate in hexanes. i r ( f i lm) : u m a x lp : 3410 (OH), 1675 (diene), 1642 ( i so l . C=C), 1270 (P=0), 1163 (P-0-Et), 1040 (br, C-0-P), 973; mp : 3400, 1675, 1640, 1262, 1160, 1040, 975; ]H nmr lp : 5.93 and 5.51 (AB-q, J=10Hz, diene vinyl H), 108 6.08 - 5.75 (m, C(3 ) vinyl H, 3H both together), 5.22 (ddd, J=17, 1.7, 1.7 Hz, cis_ vinyl H) and 5.04 (ddd, J=10, 1.7, 1.7Hz), trans vinyl H, 2H both together), 4.38 - 4.04 (m, 5H, 0CH2, OCH), 2.29 (br d, J=4Hz, 2H, vinyl ic CH2), 2.10 (brs, IH, OH, D20-exch.), 1.77 (br d, J=2Hz, 3H, vinyl ic CHg), 1.60 - 1.48 (m, 2H, C( l ' ) CHg), 1.37 (dt, J=7, 1.5Hz, 6H, ethoxy CH3), 1.10 (s, 3H, ter t iary CH3); mp: 5.91 and 5.62 (AB-q, J=10Hz), 6.07 - 5.74 (m), 5.23 (ddd, J=17, 1.5, 1.5Hz), 5.07 (ddd, J=10.5, 1.5, 1.5Hz), 4.34 - 4.04 (m), 2.25 (m), 1.77 (brs, 4H, vinyl ic CH3, OH) 1.57 (d, J=6Hz, 2H, C( l ' ) CH2), 1.35 (dt, J=7, 1.5Hz), 1.09 (s) ; mass spectrum: m/e 330 (M+), 260, 259 (100%). Exact mass calcd. for C 1 6.H 2 70 5P: 330.1596; measured: 330.1604. Anal, calcd. for C 1 6H 27°5 P : C 5 8 J 7 ' H 8 ' 2 4 ; f o u n d : c 58.04, hf 8.11. A solution of vinyl l i thium in ether was prepared from tetravinyl t in and n-butyll ithium. The act iv i ty was determined by the direct t i t ra t ion 84 method of Watson and Eastham. The reagent was stored in a freezer. b) By Reaction of 6_5 with Vinylmagnesium Bromide A solution of the aldehyde 65_ (73 mg, 0.24 mmol) in 1 ml of a mixture of tetrahydrofuran-ether (1:1) was added to the freshly prepared reagent (0.65 mmol in 2 ml of tetrahydrofuran) at room temperature. The reaction mixture was st irred at this temperature for 3.5 hours and for 0.5 hours at 55°C. Saturated aqueous ammonium chloride solution was added dropwise unti l a granular precipitate was obtained. I t was f i l te red off and washed well with warm chloroform. The f i l t r a t e was dried over anhydrous sodium sulfate, the solvent evaporated under reduced pressure and the residual crude product chromatographed on F lor is i l to yield 68 mg (86%) of the a l l y l i c alcohol 69 as a l ight yellow o i l . Tic examination and comparison of the H^ nmr spectra showed i t to be identical to the product obtained as described under a). 109 Preparation of the A l l y ! i c Trimethylsilyl Ether 70 A solution of 441 mg (1.40 mmol) of the aldehyde 65 in ether was treated with vinyl l i thium as described for the synthesis of the a l l y l i c alcohol 69_. * After 30 minutes at room temperature 1.5 ml of a solution of chlorotrimethyl-silane in ether was added instead of aqueous sodium bicarbonate. St i rr ing was continued for 1.5 hours at ambient temperature, after which time excess saturated aqueous sodium bicarbonate solution was added. The crude product was isolated in a way similar to that described for the preparation of the a l l y l i c alcohol 69. Rapid chromatography on F lor is i l provided, after solvent evaporation, 411 mg (70%) of the unstable tr imethylsi ly l ether 70 as a yellow o i l ; Rp=0.45 in 40% ethyl acetate in hexanes. i r ( f i lm) : u a 1675 (OC), 1275, 1253 (P=0, SiMe9), 1032 (br, C-O-P), 970, 845 (Si-C); ]H nmr: 6.00 - 5.45 and 5.20 - 4.85 (2 m, 5H, vinyl H), 4.37 - 3.94 (m, 5H, 0CH2, 0CH), 2.12 - 1.95 (m, 2H, v inyl ic CH2), 1.73 (brs, 3H, vinyl ic CH3), 1.62 - 1.45 (m, 2H, C(l )CH 2), 1.29 (dt , J=7, 1.5 Hz, 6H, ethoxy CH3), 1.00 and 0.98 (2s, 3H, ter t iary CH3), 0.03 (s, 9H, SiMe3); mass spectrum: m/e 402 (M +), 387 (M+ - CH3), 331, 259 (100%); Exact mass calcd. for C-jgH^OgPSi: 402.1991; measured: 402.1987. * A mixture of 0.6 ml (4.72 mmol) of chlorotrimethylsilane, 0.4 ml (2.9 mmol) of triethylamine and 1.5 ml of ether was prepared. The cloudy solution was centrifuged and 1.5 ml of the supernatant l iquid was withdrawn. Preparation of the tert-Butyl dimethyl si ly l Ether 72_ n o To a solution of l i thium diisopropylamide (55 mmol), prepared from 8.5 ml (60 mmol) of diisopropylamine and 33.5 ml (55 mmol, 1.64 M solution in hexane) of n-butyll i thium, in 100 ml of tetrahydrofuran at -78°C was added the ketone 54_ (9.8 g, 50 mmol), dissolved in 5 ml of tetrahydrofuran. After s t i r r ing for 15 minutes at -78°C and for 20 minutes at 0°C, 17.5 ml of hexamethylphosphoramide followed by 17.9 g (75 mmol) of l-bromo-2-tert-butyldimethylsiloxyethane were added. The reaction mixture was st irred at 0°C for one hour and at room temperature for 24 hours, then poured into aqueous saturated sodium bicarbonate solution. The resultant mixture was twice extracted with hexanes, the combined organic extract twice washed with brine and dried over anhydrous magnesium sulfate. After solvent evaporation and chromatography of the residual o i l on s i l i ca gel. with 20% ethyl acetate in hexanes as eluant, 15.5 g (88%) of the pure alkylated ketone 72_as a pale yellow o i l , together with 400 mg (4%) of the starting material were obtained; Rp=0.59 in 20% ethyl acetate in hexanes. i r ( f i lm) ; u m a v 1654 ( 0 0 ) , 1598 (OC), 1253 (SiMeJ, 1207 (=00 ) , " IMa X £' 1098, 835 (Si-C), 770; ]H nmr; 5.23 (brs, IH, vinyl H), 3.80 - 3.40 (m, 0CH2) and 3.55 (d, J=7Hz, isobutyl CH2, 4H both together), 3.01 - 2.78 (m, IH, a l l y l i c H), 2.18 - 1.37 (m, 5H, CH2> CH), 1.20 - 0.92 (m, 12H, ring CH3, isobutyl CH3), 0.88 (s, 9H, te r t -bu ty l ) , 0.04 (s, 6H, SiMe3); mass spectrum: m/e 354 (M +), 339 (M+-CH3), 297 (M+ - ter t -buty l ) , 241 (100%). Exact mass calcd for C 2 Q H 3 8 0 3 Si : 354.2590; measured: 354.2595. Anal. calcd. for C 20 H 38°3 S i : C 6 7 - 7 4 ' H 1 0 - 8 ° ; found: C 67.63, H 10.93. Reduction-Hydrolysis of 72_ to the a-Enone 73 A solution of the ketone 72 (270 mg, 0.76 mmol) in 2 ml of ether was m added slowly to a st i rred suspension of 16 mg (0.42 mmol) of l i thium aluminum hydride in 16 ml of ether at 0°C. The reaction was allowed to proceed at this temperature for 4 hours. Excess sodium sulfate decahydrate was added and s t i r r ing continued for another 25 minutes. The precipitate was f i l te red off and washed with ether. . The solvent was evaporated from the f i l t r a t e , under reduced pressure, and the residue thus obtained was taken up in 16 ml of a mixture of acetone and acetic acid (3:1). The solution was st irred at 0°C for 2 hours. I t was diluted with water and the resulting solution twice extracted with ether. The combined ethereal phase was twice washed with brine and dried over anhydrous magnesium sulfate. Solvent evaporation followed by preparative t i c purif ication of the crude product in 20% ethyl acetate in hexanes allowed the isolation of 165 mg (77%) of the pure a-enone 73_ and 30 mg of the starting material. An analytical sample of 73 was obtained by short-path d i s t i l l a t i on (dist . temp. 100°/0.09 Torr); Rp=0.64 in 20% ethyl acetate in hexanes. i r ( f i lm) : u a 1678 (OO), 1256 (SiMe,),- 1092, 835 (Si-C) cm"1; max c ]H nmr: 6.67 and 6.65 (2d, J=10Hz, IH, @ vinyl H), 5.87 (d, J=10Hz, IH, a vinyl H), 3.78 ( t , J=7Hz, 2H, 0CH2), 2.79 - 2.28 (m, IH, O ' H ) , 2.18 -1.25 (m, 4H, CH2), 1.14 (s, te r t . CH3) and 1.11 (d, J=6Hz, a'cH^, 6H both together), 0.90 (s, 9H, te r t -bu ty l ) , 0.04 (s, 6H, SiMe2); mass spectrum: m/e 267 (M+-CH3), 225 (H+ - ter t -buty l , 100%). Exact mass calcd. for C 1 5 H 2 7 0 2 Si(M+-CH3): 267.1780; measured: 267.1777; calcd. for C ^ H ^ S i (M+ - tert-butyl) : 225.1311; measured: 225.1308. Anal, calcd. for C , C H o n 0 o S i : C 68.03, 112 H 10.70; found: C 68.20, H 10.60. Preparation of the Ether 74_ To a suspension of l ithium aluminum hydride (376 mg, 10 mmol) in 230 ml of ether at 0°C was added a solution of 5 g (14.1 mmol) of the ketone 72^  in 10 ml of ether. The reaction mixture was st irred at 0° for 2.5 hours followed by 2 hours at room temperature. Work-up and subsequent acid hydrolysis was done as described for the preparation of the a-enone 73_. After column chromatography and short-path d i s t i l l a t i on of the residual product (dist . temp. 55°C/0.08 Torr), 909 mg (29%) of the ether 74 were obtained. i r ( f i lm) : u m a v 1654 (C=C), 1199 (=C-0), 1090, 1038 (C-0) cm" 1; ] H nmr: 4.51 and 4.49 (2d, j=4Hz, IH, vinyl H), 4.19 and 4.17 (2d, J=4Hz, IH, 0CH), 3.97 - 3.81 (m, 2H, C(Z ) methylene), 3.44 and 3.41 (2d, J=6Hz, 2H, isopropyl CH2), 2.66 - 2.27 (m, IH, CH-Me), 2.14 - 1.79 (m, 3H, CH, C( l ' ) CH2), 1.61 - 1.21 (m, 2H, CH2), 1.14 (s, tert.CH 3) and 1.10 (d, J=7Hz, second. CH3 of r ing, 6H both together), 0.96 and 0.94 (2d, J=7Hz, 6H, isobutyl CH3); mass  spectrum: m/e 224 (M +), 132, 180 (M+-.C2H40), 169 (100%, M+-C 4H y). Exact  mass: calcd. for C 1 4H 24°2: 2 2 4 - l 7 7 6 * > measured: 224..1769. Anal. calcd. for C14H24°2 : C 7 4 - 9 5 ' H 1 0 - 7 8 i found: C 75.13, H 10.66. Preparation of the Oxime 75_ A solution of hydroxylamine hydrochloride (235 mg, 3.4 mmol) in 1 ml of water was added to a solution of 782 mg (2.6 mmol) of the aldehyde 6_5 in 2.5 ml of pyridine. The reaction mixture was st irred at ambient temperature for 10Q minutes, water was added and the mixture twice extracted with ether. The combined organic phase was washed with 0.1 N hydrochloric acid, then 113 with brine and dried over anhydrous magnesium sulfate. The yellow oi l obtained after solvent removal under reduced pressure was purif ied by chromatography on s i l i ca gel with 70% ethyl acetate in hexanes as eluant, to yield 676 mg (82%) of the pure oxime 75_. I t showed Rp=0.33 in ethyl acetate. i r ( f i lm) ; u m a v 3290 (OH), 1672 (OC), 1255 (P=0), 1160 (P-O-Et), — — — — — max 1035 (br, C-O-P], 970 cm" 1; ]H nmr; 7.45 ( t , J=6.5Hz, IH, CH=N), 6.81 (brs, IH, OH), 5.97 and 5.52, 5.95 and 5.49 (2 AB-q, J=10Hz, 2H, vinyl H), 4.34 - 4.04 (m, 4H, 0CH9), 2.44 - 2.16 (m, 4H, vinyl ic CH , CH.ON), 1.75 (brs, 3H, vinyl ic CH3), 1.34 ( t , J=7Hz, 6H, ethoxy CH3), 1.09 and 1.06 (2s, 3H, te r t . CH3); mass spectrum: m/e 317 (M +), 299 (M +-H 20). Exact mass calcd. for C 1 4H 2 4N0 5P: 317.1392; measured: 317.1396. Anal, calcd. for C ] 4 H 2 4 N05P: C 52.99, H 7.62, N 4.41; found: C 53.09, H 7.65, M 4.32. Dehydration of the Oxime 75 to the N i t r i le 76_ A mixture of the oxime 75 (391 mg, 1.23 mmol) and 250 mg (1.54 mmol) of N,N -carbonyldiimidazole in 10 ml of dry dichloromethane was refluxed for 10 hours. I t was poured into aqueous saturated sodium bicarbonate solution and the resulting mixture extracted with ether. The organic layer was dried over anhydrous magnesium sulfate and the solvent evaporated. The crude product thus obtained was purif ied by column chromatography on s i l i ca gel with 50% ethyl acetate in hexanes as eluant to provide 323 mg (88%) of the pure n i t r i l e 76 as a colourless o i l . I t showed RF=0.48 in ethyl acetate. i r (film); u m a v 2245 (C=H\, 1675 (OC), 1275 (P=0), 1170 (P-0-Et), max 1055, 1027 CC-0-P), 971 cm"1; Vnmr: 6.04 and 5.56 (AB-q, OlOHz, 2H, 114 vinyl H), 4.37 - 4.05 (m, 4H, 0CH2), 2.39 Cs, 2H, CHgCN), 2.29 (d, J=4Hz, 2H, vinyl ic CHg), 1.81 (d,J=2Hz, 3H, vinyl ic CH3), 1,37 (dt, J=7, 1.5 Hz, 6H, ethoxy CH3), 1.24 (s, 3H, te r t . CH3); mass spectrum: m/e 299 (M+). Exact  mass calcd. for C 1 4H2 2N 0 4 P : 2 9 9 - 1 2 8 6 ; measured: 299.1284. Anal, calcd. for C 1 4H, 2N0 4P: C 56.18, H 7.41, N 4.68; found:C 55.89, H 7.44, N 4.46. Diels-Alder Cyclization of 69 to the Alcohol 73 )H A solution of 263 mg (0.79 mmol) of the a l l y l i c alcohol 69 in 100 ml of dry xylene was refluxed for 4.5 hours. The solvent was evaporated under reduced pressure. The residual o i ly product 78^  (260 mg, 99%) was pure by t i c and glc examination. Chromatography on Fieri si 1 with 40% ethyl acetate in hexanes afforded the analytically pure alcohol 78 in 89% (233 mg) y ie ld . I t showed Rp=0.14 in 40%, ethyl acetate in hexanes. i r ( f i lm) : u m a v 3410 (OH), 1650 (OC), 1270 (P=0), 1165 (P-0-Et), 1035 (br, C-0-P), 965 cm" 1; ]H nmr: 5.71 and 5.67 (2 dd, J=7.5, 2Hz, IH, vinyl H), 4,47 - 3.98 (m, 5H, 0CH2> 0CH), 2.72 (dd, J=7.5, 4Hz, a l l y l i c H) and 2,45 - 1.51 (m, C(4,10) CH2> C(6)CH, 6H both together), 1.38 (dt, J=7, 1.5Hz, 3H, ethoxy CH3),.1 .22 -1 .07 (m, C(2)CH2) and 1.17, 1.10, 1.00, 0.92 (4s, te r t . CH3, 14 H both together); mass spectrum: m/e 330 (M+), 274, 259, 246 (100%). Exact mass calcd. for CjgH^OgP: 330.1596; measured: 330.1601. Anal, calcd. for ^^27°5P: C 58.17, H 8.24; found: C 57.88, H 8.01. 115 Cyclization of 70 to the Si ly l Ether 85_ The same procedure as just described was followed. Under argon, from 210 mg (0.52 mmol) of the s i l y l ether 70, 200 mg (95%) of crude product were obtained. From "'H nmr and i r evidence i t consisted mainly of the s i l y l ether 85. i r ( f i lm) : u m 3 v 3430 (w, OH), 1650, 1255, 1165, 1035, 965, 845 (Si-C) cm" 1; "'H nmr: 5.95 - 5.6 (m, IH, vinyl H), 4.27 - 3.95 (m, 5H, 0CH2, 0CH), 2.75 - 1.45 (m, 6H, CH2, CH), 1.32 (dt, J=7, 1Hz, ethoxy CH3), 1.10, 1.04, 0.95, 0.93, 0.86, 0.84 (6s, te r t . CH3> 14H both together), 0.04 (s, SiMe3). 28 Preparation of the Keto Alcohol ]8_ To a solution of the enol phosphate 78 (93 mg, 0.28 mmol) in 15 ml of ethanol v/ere added 3 ml of an approximately 2 M solution of sodium ethoxide in ethanol. The mixture was st irred at room temperature for 15 hours, poured into di lute hydrochloric acid and the resulting solution twice extracted with ether. The combined ethereal extract was washed with aqueous saturated sodium bicarbonate solution, then with brine and dried over anhydrous magnesium sulfate. Removal of the solvent under reduced pressure and purif ication of the residual oi l by chromatography on s i l i ca gel with 40% ethyl acetate in hexanes as eluant afforded 53 mg (97%) of the pure product 18. The f i r s t and last product containing fractions contained the pure less polar and more polar isomers, respectively; Rp=0.30 and 0.22 in 40% ethyl acetate in hexanes. An analytical sample was obtained by short-path d i s t i l l a t i on (dist . temp. 137°C/0.08 Torr) of an aliquot of J_8. i r ( f i lm) : u m a v 3420 (OH), 1718 (C=0), 1255, 1035 (C-0) cm" 1; 1 Hnmr:* max 4.52 - 4.30 (ddd,J=ll, 5.5, 5.5Hz, C(5, exo)H), 4 . 2 2 - 4.08 (m, C(5, endo) H, IH both together), 2.41 and 2.37 (2d, J=3Hz, 2H, CH2 C=0), 2.29 - 1.20 (m, 9H, C(2, 4, 10) CH2, C(6,7) CH, OH), 1.12, 1.04, 0.97, 0.89 (4s, 6H, CH3); 116 mass spectrum: m/e 194 (M +), 176 (M+ - H,,0). Exact mass calcd. for C12H18°2 : 1 9 4 - 1 3 0 7 ' measured: 194.1305. o o * l i terature: 4.0 (br m, IH), 3.1(brs, IH, OH), 2.36(m, 2H), 1.08, 1.00, 0.92, 0.85. Prenaration of the Ketal 86 OH a) From the Ketone 13 and 2,2-Dimethyl-1,3-propanediol A mixture of the keto alcohol 18 (44 mg, 0.23 mmol), 2,2-dimethyl-l,3-propanediol (358 mg, 3.4 mmol) and a catalytic amount of £-toluenesu1fonic acid in 15 ml of benzene was refluxed for 23 hours with continous removal of the water formed in the reaction. Cooled to ambient temperature, the solution was poured into aqueous saturated sodium bicarbonate solution. The resulting mixture was twice extracted with ether, the combined organic extract washed with brine and dried over anhydrous sodium sulfate. The solvent was evaporated and the oi ly residue chromatographed on Flor is i l with 10% ethyl acetate in hexanes as eluant to provide 54 mg (84%) of the pure ketal alcohol 86. I t showed Rp=0.50 in 40% ethyl acetate in hexanes. i r (CHC1„): u m a v 3620 (OH), 1119, 1050, 1033 (C-0), 855 cm" 1; 1Hnmr: 4.26 (ddd, J = l l , 5.5, 5.5Hz, C(5, expjH), 3.97 (dd, J=7, 2.5Hz, C(5, endo) H, IH both together), 3.67 and 3.33 (AB-q, J=llHz, 4H, 0CH2), 2.23 - 1.25 (m, 10H, CH2, CH), 1.16, 1.02, 0.96, 0.95, 0.88, 0.72 (6s, 12H, CH3); mass  spectrum; m/e 280 (M +, 100%). Exact mass calcd, for c i7 H 28°3 : 2 8 0 - 2 ° 3 8 ; measured; 280.2Q34. Anal. calcd, for C 1 7H2 8 0 3 ; C 72.82, H 10.06; found: C 72.54, H 9.96. 117 b) From the Enol Phosphate 78 with 2,2-Dimethyl-l,3-propanediol A solution of the enol phosphate 78 (19 mg, 0.057 mmol) and 94 mg (0.9 mmol) of 2,2-dimethyl-l,3-propanediol in 3 ml of benzene containing a catalyt ic amount of £-toluenesulfonic acid was refluxed for 15 hours with continuous removal of the water. The same procedure as outlined under a) yielded 10 mg (62%) of the ketal alcohol 86. Spectroscopic, t i c and glc examination showed this product to be the same as that obtained by ketaliza-tion of the keto alcohol 1_8 as described under a). Oxidation of 86_ to the Ketone 87 To a solution of the alcohol 86_ (41 mg, 0.15 mmol) in 3 ml of dry * dichloromethane were added 85 mg (0.24 mmol) of pyridinium dichromate. After the reaction mixture had been st irred for 21 hours at room temperature i t was f i l te red through F lor is i l with more dichloromethane. The solvent was evapo-rated from the f i l t r a t e under reduced pressure to give 37 mg (92%) of the pure product 87_ as a colourless, crystal l ine sol id. To obtain an analytical sample, a small amount was recrystall ized from hexanes at freezer temperature; m.p. 150 - 152°C; Rp=0.66 in 40% ethyl acetate in hexanes. i r (CHC»3): u m a x 1735 ( 0 0 ) , 1113, 1100 cm" 1;  1ti nmr: 3.72 (d, O i l . 5 Hz) and 3.66 (d, J=11.5Hz, 2H both together, A-branch of two AB-q, 0CHo), 3.38 and 3.35 (2dd, O i l . 5 , 3Hz, 2H, B-branch of AB-q, 0CH2), 2.36 - 1.80 (m, 8H, C(4, 8, 10) CH2, C(6,7) CH), 1.10 - 1.00 (m, C(2) CH2) and 1.17, 1.15, 0.94, 0.75 (4s, CH3, 14H both together); mass spectrum: m/e 278 (M+, 100%). Exact mass calcd. for c-j7H26°3 : 278.1882; measured: 278.1884. Anal. calcd. for C-j 7H 2 g0 3: C 73.35, H 9.41; found: H 73.26, H 9.54. * Pyridinium dichromate was prepared from chromium trioxide and pyridine; m.p. 148°C ( l i t . 9 6 144 - 146°C). 118-Preparation of the Tertiary Alcohol 90 * A solution of isopropenyllithium (1.1 ml, 0.25 M in ether) was added slowly to a wel l-st i rred solution of 54 mg (0.2 mmol) of the ketone 87_ in 2 ml of ether at -78°C. The reaction was allowed to proceed at -78°C for 4 hours and at ambient temperature for 35 minutes. The mixture was diluted with water and twice extracted with ether. The combined organic extract was washed with brine and dried over anhydrous magnesium sulfate. The crystal l ine material obtained after solvent removal was chromatographed on F lor is i l with hexanes as eluant to give 55 mg (90%) of the pure alcohol 90. Recrystallization from hexanes at freezer temperature provided an analytical sample; m.p. 120 - 121°C; Rp=0.59 in 40% ethyl acetate in hexanes. i r (CHC1,): u m a 3595 (OH), 1638 (OC), 1126, 1098 cm" 1; ]H nmr: 4.90 o rTlaX (brs, IH, vinyl H), 4.80 (m, IH, vinyl H), 3.64 and 3.32 (AB-q, J=10.5Hz, 4H, 0CH2), 1.84 (brs, vinyl ic CH3) and 2.20 - 1.25 (m, CH2, CH, OH, 14 H both together), 1.15, 0.95, 0.92, 0.70 (4s, 12H, CH3); mass spectrum: m/e 320 (M +, 100%), 302 (M +-H 20). Exact mass calcd for C 2 Q H 3 2 0 ? : 320.2351 ; measured: 320.2352. Anal, calcd. for C 2 0 H 3 2 0 3 : C 74.96, H 10.06; found: C 75.20, H 10.00. * Isopropenyllithium in ether was prepared from 2-bromopropene and lithium 99a dispersion, containing 2% sodium,in ether. The act iv i ty was determined by 137 the double t i t ra t i on method of Gilman. 28 30b Preparation of the Keto Diene 20 ' To a solution of the ketal alcohol 90 (8 mg, 0.025 mmole) in 1.5 ml of 119 ether were added 46 yl of boron t r i f luor ide etherate. The mixture was st irred at ambient temperature for 8 hours. I t was poured into aqueous saturated sodium bicarbonate solution and the mixture twice extracted with ether. The combined organic extract was washed with brine, dried over anhydrous magnesium sulfate and the solvent removed under reduced pressure. Tic puri f icat ion of the crude product thus obtained in 5% ethyl acetate in hexanes provided 4.6 mg (84%) of the keto diene 20; Rp=0.36 in 5% ethyl acetate in hexanes. i r ( f i l m ) : * u m a v 1721 (OO), 1592 (OC) cm" 1; ]H nmr:* 5.80 (s, IH, max = CH), 4.96 and 4.90 (2 brs, 2H, = CH2), 2.89 - 2.68 (m, IH, C(6)H), 2.46 (d, J=3Hz, 2H, CH2O0), 1.91 (brs, 3H, vinyl ic CH3), 1.73 - 1.21 (m, CH2), 1.12 and 0.88 (2s, 3H each, CH^); mass spectrum: m/e 216 (M +). Exact mass calcd. for C 1 5 H 2 Q 0: 216.1514; measured: 216.1513. 28 30b * The spectral data for 20 were in good agreement with l i terature values. ' , x28,29,30b Hydrogenation of the Diene 20 to 9-Pupukeanone (7J  A solution of 4 mg of the keto diene 20 in 1.5 ml of ethanol was st i rred with. 12 mg iridium black as catalyst for 21 hours under an atmosphere of hydrogen at a pressure of 32 psi. The suspension was f i l te red through Celite with more ether. Solvent evaporation from the f i l t r a t e under reduced pressure yielded 3.5 mg of 9-pupukeanone {]_). When this product was co-injected with an authentic sample of 9-pupukeanone into three dif ferent glc columns (5% 0V-210, 5% 0V-17, 3% Carbowax on Chrom W,80-100 mesh) only one compound could be detected. i r (CHC13): 1718 cm'1 (00 ) cm" 1; h nmr: 2.33 (d, J=3Hz, 2H, CH 2O0), 1.60 - 1.13 (m, CH2, CH), 1.04 (s, 3H, te r t . CH3), 0.91 (brs, 9H, te r t . CH3, isopropyl CH3); mass spectrum: m/e 220 (M+), 177 (M+-C3 H 7). Exact mass 120 calcd. for C l g H 2 4 0: 220.1827; measured: 220.1831. The spectral data correlated well with those reported in the l i terature. Oxidation of the Alcohol 78 to the Ketone 96 To a st irred solution of the alcohol 7S_ (186 mg, 0.56 mmol) in 6 ml of dry dichloromethane were added 350 mg (1.0 mmol) of pyridinium dichromate. The mixture was st i rred at room temperature for 20 hours and f i l te red through Flor is i l with more dichloromethane. The solvent was evaporated from the f i l t r a t e and 175 mg (95%) of the chromatographically pure ketone 96 were obtained. Chromatography on Flor is i l with 20% ethyl acetate in hexanes provided 156 mg (85%) of the analytically pure product 96; Rp=0.22 in 40% ethyl acetate in hexanes. i r ( f i lm) : u m a v 1742 ( 0 0 ) , 1650 (OC), 1278 (P=0), 1185, 1165 (P-0-Et), 1055, 1025 (C-O-P), 960 cm" 1; ]H nmr: 5.78 (dd, J=7.5, 2Hz, IH, vinyl H), 4.38 - 4.07 (m, 4H, 0CH2), 2.74 (dd, J=7.5, 4Hz, IH, a l l y l i c H), 2.38 and 2.12 (AB-q, J=18Hz, 3H, CH2O0, CHO0), 1.94 - 1.64 (m, 2H, C(10) CH2), 1.53 - 1.28 (m) and 1.39 (br t , J"=7Hz, 8H both together, C(2) CH2, ethoxy CH3), 1.15, 1,13 (2s, 6H, CH3); mass spectrum: m/e 328 (M+), 259, 258, 246 (100%). Exact mass calcd. for C l g H 2 5 0 5 P: 328.1439; measured: 328.1439. Anal, calcd. for C - |6 H 2 5°5P : C 5 8 - 5 2 » H 7 -67; found: C 58.20, H 7.67. Preparation of the Trisylhydrazone 97_ To a solution of 10 mg (0.03 mmol) of the ketone 96_ in 0.6 ml of dry tetrahydrofuran were added 10 mg (0.034 mmol) of trisylhydrazide. After the 121 reaction mixture had been st i rred at ambient temperature for 6 hours, i t was f i l te red through F lor is i l with more solvent. The solvent was evaporated from the f i l t r a t e . The crude product thus obtained was purif ied by pre-parative t i c in 40% ethyl acetate in hexanes to yield 15 mg (81%) of the pure trisylhydrazone 9_7; Rp=0.36 in 40% ethyl acetate in hexanes. i r (CHC1J: u m a v 1650 (OC), 1600 (arom.), 1268, 1165, 1155, 1036, 965, j max cm" 1; V, nmr: 7.22 (s, 2H, aromatic H), 5.66 (br d, J=7Hz, IH, vinyl H), 4.40 - 4.05 (m, 6H, 0CH2, o-benzylic H), 2.94 (m, IH, p_-benzyl1c H), 2.62 -1.62 (m, 8H, C(2, 4, 10) CH2, a l l y l i c H, C(6) CH), 1.45-1.20 (m) and 1.29 (d, J=6.5Hz, 24H, ethoxy CH3, isopropyl CH3), 1.08, 1.06, 1.03 (3s, 6H, te r t . CH3); mass spectrum: m/e 608 (M+), 565 (M + -C 3 H 7 ) , 341 , 327. Exact mass calcd. for C o lH. oN o0 cPS: 608.3049; measured: 608.3036. Anal, calcd. for 31 49 2 6 ' C 31 H 49 N 2°6 P S : C 6 1 * 1 6 , H 8 , 1 1 ' N 4 , 6 0 ; f o u n d : C 61.08, H 8.06, N 4.50. Reaction of the Trisylhydrazone 97 with n-Butyllithium To a solution of the trisylhydrazone 97_ (11.5 mg, 0.02 mmol) in 0.2 ml of tetramethylethylenediamine and 0.2 ml of n-hexane were added, at -64°C, 45 yl (0.074 mmol, 1.65 M solution in hexane) of n-butyl1ithium. The mixture was st irred at -64°C for 2.5 hours, followed by 2 hours at 0°C. Deuterium oxide (0.5 ml) was added and after 5 minutes the mixture was diluted with ether and washed twice with brine. The organic phase was dried over anhydrous magnesium sulfate. Removal of the solvent under reduced pressure and purif icat ion of the residual crude product by preparative t i c in 40% ethyl acetate in hexanes allowed the isolation of two compounds. The more polar one (7 mg) was identif ied by t i c , nmr and mass spectral comparison as starting material. To the less polar one (2 mg) the structure 98 was assigned; Rr=0.53 in 40% ethyl acetate in hexanes. 122 i r (CHC13): u m a x 1720 (OO), 1662 (ON), 1600 (aromatic), 1383, 1327, 1164, 1155 (S0 2-N), 960 cm"1; ]H nmr: 7.20 (s, 2H, aromatic H), 4.22 (m, 2H, o-benzylic H), 3.10 - 1.51 (m, 11H, £-benzylic H, CH2, CH), 1.26 (d, J=7Hz) and 1.25 - 1.08 (m, 21H altogether, CH3), 0.84 (s, 3H, CH3); mass  spectrum: m/e 473 (M +), 472, 282 (100%). Exact mass calcd. for C 2 7H 3 gDN 2 03S: 473.2822; measured: 473.2802; calcd. for C ^ H ^ N ^ S : 472.2760; measured: 472.2767. 123 IV. BIBLIOGRAPHY J.P. Collman, M. Marocco, P. Denisevich, C. Koval, F.C. Anson, J. Electroanal. Chem. Interfacial Electrochem. 101, 117 (1979). Ref. 2 and 3 l i s t comprehensive treatises on the subject of natural products chemistry: a) K. Nakanishi et a]_. "Natural Products Chemistry" Academic Press, New York 1974, 2 Volumes; b) J. ApSimon, Ed. "The Total Synthesis of Natural Products", Wiley - Interscience, Vol. I + I I 1973, Vol. I l l 1977; c) I . Fleming "Selected Organic Syntheses", Wiley -Interscience 1973. "Stereoselective Synthesis of Natural Products", Proceedings of the Seventh Workshop Conference Hoechst, Schloss Reisensburg, 24-27 Septembe 1978, Excerpta Medica, Amsterdam-Oxford 1979. E.J. Corey, R.L. Danheiser, S. Chandrasekaran, P. Siret, G.E. Keck, J.-L. Gras, J. Am. Chem. Soc. 100, 8031 (1978); E.J. Corey, R.L. Danheiser, S. Chandrasekaran, G.E. Keck, P. Siret, B. Gopalan, S.D. Larsen, J.-L. Gras, ib id . 100, 8034 (1978). P. Mangeney, R.Z. Andriamialisoa, N. Langlois, Y. Langlois, P. Potier, J. Am. Chem. Soc. 101, 2243 (1979). R.B. Woodward, Pure Appl. Chem. 2, 383 (1961); R.B. Woodward, W.A. Ayer, J.M. Beaton, F. Bickelhaupt, R. Bonnett, P. Buchschacher, G.L. Closs, H. Dutler, J. Hannah, F.P. Hanck, S. I t6 , A. Langemann, E. LeGoff, W. Leimgruber, W. Lwowski, J. Sauer, Z. Valenta, H. Volz, J. Am. Chem  Soc. 82, 3800 (1960). R.B. Woodward, F.E. Bader, H. Bickel, A.J. Frey, R.W. Kierstead, Tetrahedron 2, 1 (1958). R.B. Woodward in A. Todd "Perspectives In Organic Chemistry", Wiley -Interscience, New York 1956, p. 155. D.H. R. Barton, Pure Appl. Chem. 49, 1241 (1977). B.J. Burreson, P.J. Scheuer, J. Finer, J. Clardy, J. Am. Chem. Soc. 97, 4763 (1975). M.R. Hagadone, B.J. Burreson, P.J. Scheuer, J.S. Finer, J. Clardy, Helv. Chim. Acta 62, 2484 (1979). L.G. Harris in "Current Topics in Comparative Pathobiology", Vol. I I , p. 213 f f , Acad. Press, N.Y. 1973. H. Bertsch, A.A. Smith, Veliger 13, 171 (1970). B.W. Halstead "Poisonous and Venomous Marine Animals of the World", Vol. I , p. 709, U.S. Government Printing Office, Washington, D.C. 1965. 124 15. Y. Kato, P.J. Scheuer, J. Am. Chem. Soc. 96, 2245 (1974); Pure Appl. Chem. 41, 1 (1975); ib id. 48, 29 (1976). 16. J.S. Mynderse, R.E. Moore, M. Kashiwagi, T.R. Norton, Science 196, 538 (1977). 17. M.D. Stal lard, D.J. Faulkner, Comp. Biochem. Physiol. B 49, 25-35, 37-41 (1974) . 18. R.E. Johannes, Veliger 5, 104 (1963). 19. A comprehensive review of the current state of marine natural products chemistry: P.J. Scheuer, Ed. "Marine Natural Products" "Chemical and Biological Perspectives", Vol. I , I I , Acad. Press N.Y. 1978; especially: Vol. I , Ch. 4 "Terpenoids from Marine Sponges" by L. Minale. 20. E. Fattorusso, S. Magno, L. Mayol, C. Santacroce, D. Sica, Tetrahedron 30, 3911 (1974), ib id . 31_, 269 (1975). 21. B. DiBlasio, E. Fattorusso, S. Magno, L. Mayol, C. Pedone, D. Sica, C. Santacroce, Tetrahedron 32, 473 (1976). 22. L. Minale, R. Riccio, G. Sodano, Tetrahedron 30, 1341 (1974). 23. B.J. Burreson, C. Christophersen, P.J. Scheuer, Tetrahedron 31, 2015 (1975) . 24. B.J. Burreson, P.J. Scheuer, C. Christopersen, J. Am. Chem. Soc. 97, 201 (1975); B.J. Burreson, P.J. Scheuer, Chem. Comm. 1035 (1974"TT~ 25. I. Hagedorn, H. Tonjes, Pharm. 12^ 567 (1957). 26. R.F. Vesonder, Lloydia, J. Nat. Prod. 42, 232 (1979). 27. M. Nobuhara, H. Taziraa, K. Shudo, A. I t a i , T. Okamoto, Y. I i taka, Chem. Pharm. Bul l . 24, 832 (1976). 28. H. Yamamoto, H.L. Sham, J. Am. Chem. Soc. 101, 1609 (1979). 29. E.J. Corey, M. Bekforouz, M. Ishiguro, J. Am. Chem. Soc. 101, 1608 (1979). 30. a) E. J. Corey, M. Ishiguro, Tetrah. Lett. 2745 (1979); b) B. A. Schiehser, J.D. White, J. Org. Chem. 45, 1864 (1980). 31. A. Eschenmoser, Naturwiss. 61, 513 (1974). 32. R.B. Woodward, Pure Appl. Chem. J_7, 519 (1968); 25, 283 (1971); 33, 145 (1973). 33. a) A. Eschenmoser, Quart. Rev. 24, 366 (1970); b) A. Eschenmoser, 23rd Int . Congr. Pure Appl. Chem., Vol. 2, p. 69, Butterworth, London 1971. 34. a) R.E. Ireland "Organic Synthesis", Prentice Hal l , Englewood C l i f f s , 1969; b) S. Turner "The Design of Organic Syntheses", Elsevier, Amsterdam, 1976; c) see also ref. 2c. 125 35. S. Warren "Designing Organic Syntheses", Wiley, Chichester, 1978. 36. E.J. Corey, Pure Appl. Chem. 14, 19 (1967). 37. a) E.J. Corey, Quart. Rev. 25, 455 (1971); b) S.R. Heller, G.W. Milne, R.J. Feldman, Science 195, 253 (1977); c) K. Heusler, Science 189, 609 (1975); d) G.J. Powers, R.L. Jones, G.A. Randall, M.H. Caruthers, J.H. van de Sande, H.G. Khorana, J. Am. Chem. Soc. 97, 875 (1975); e) J. Thesing, Naturwiss. 64, 601 (1977); f ) W.T. Wipke, W.J. Howe, Eds. "Computer-Assisted Organic Synthesis", ACS Symposium Series 61, Am. Chem. Soc, Washington, D.C, 1977; g) I. Ugi, J. Bauer, J. Brandt, J. Friedrich, J. Gasteiger, C. Jochum, W. Schubert, Ang. Chem. Int. Ed. Engl. 18, 111 (1979). 38. Review of chemistry of i son i t r i les : a) I . Ugi, Ed. " Ison i t r i le Chemistry" Acad. Press, New York, 1971; b) S.R. Sandler, W. Karo "Organic Functional Group Preparations", Vol. I l l , p. 179 f f , 1972, c) W. Foerst "Newer Methods of Preparative Organic Chemistry", Vol. IV, p. 37 f f , 1968; d) see also T. Mukaiyama et al_., Chem. Lett. 697 (1977) and references therein. 39. CF. Lane, Synthesis 135 (1975); see also G.L. Grunewald, D.E. Walters, T.R. Kroboth, J. Org. Chem. 43, 3478 (1978) and J. -M. Varlet, N. Collignon, P. Savignac, Synth. Comm. 8^ , 335 (1978). 40. a) CM. Starks, C. Liotta "Phase Transfer Catalysis", Acad. Press, New York, 1978; b) W.P. Weber, G.W. Gokel "Phase Transfer Catalysis in Orqanic Synthesis", Springer Verlag, 1977; c) G.W. Gokel, R.P. Widera, W.P. Weber, Org. Synth. 55, 96 (1976). 41. K. Niitzel "Organo-Magnesium Verbindungen" in Houben-Weyl, "Methoden der Organischen Chemie", Bd. XII I /2a, p. 47 - 258, Georg Thieme Verlag, 1973; H. Normant, Adv. Org. Synth. 2, 1 (1960). 42. Addition reactions to an exocyclic double bond in unhindered norbornane derivatives proceed mainly from the exo face of the molecule: H.C. Brown, J.H. Kawakami, J. Am. Chem. Soc. 92, 1990 (1970); F. Lauria, V. Vecchietti, W. Logemann, G. Tosolini, E. Dradi, Tetrahedron 25, 3989 (1969); CH. DePuy, P.R. Story, J. Am. Chem. Soc. 82, 627 (1960)7 R. Dulou, Y. Chretien-Bessiere, Bul l . Soc. Chim. Fr. 1362 (1959); H.O. House, "Modern Synthetic Reactions", W.A. Benjamin Inc., 1972, p.63. 43. 2-Isopropyl-4-methylbicyclo[2.2.1]heptan-2-61 yielded upon acid-catalysed dehydration the exocyclic o le f in : W.C. Agosta, S.Wolff, J. Org. Chem. 41, 2605 (1976); a dif ferent dehydration procedure could, however, give a different result. The cyclohexanol derivative obtained from isopropyl-l i thium and l-methyltricyclo[4.4.0.0 2,7] decan-8-one dehydrated to the endocyclic o le f in : CH. Heathcock, R.A. Badger, J.W. Patterson, J r . , J. Am. Chem. Soc. 89, 4133 (1967). 44. D.L. Fowler, W.V. Loebenstein, D.B. Pal l , C.A. Kraus, J. Am. Chem. Soc. 62, 1140 (1940); see also ref. 63. 126 45. Syn - additions, such as hydrogenation, to norbornene derivatives and similar strained alkenes preferentially occur from the least hindered face of the molecule, e.g. exo-addition in unhindered compounds: S.J. Cr is to l , R.T. Lalonde, J. Am. Chem. Soc. 81_, 1655 (1959); C.W.J. Jefford, Chimia 24, 357 (1970); H.C. Brown, J.H. Kawakami, S. Ikegami, J. Am. Chem. Soc. 92, 6914 (1970); T.G. Traylor, Acc. Chem. Res. 2_, 152 (1969); S.J. Cr is to l , T.C. Mor r i l l , R.A. Sanchez, J. Org. Chem. 3J_, 2719 (1966); J.K. S t i l l e , R.D. Hughes, ib id . 36, 340 (1971); R.T. Arnold, G.W. Eigenmann, J. Am. Chem. Soc. 81_, 3440 (1959). 46. a) G. Stork, P.F. Hudrlik, J. Am. Chem. Soc. 90, 4462 (1968); b) organo-si l icon compounds review: J.K. Rasmussen, Synthesis 91 (1977); P.F. Hudrlik, J. Organomet. Chem. Libr. _l, 127 (1976); E.W. Colvin, Chem. Soc. Rev. 7, 15 (1978); see also ref. 72, 86. 47. W.F. Gannon, H.0. House. Org. Synth. 40, 14 (1960). 48. a) A. de Groot, B.J.M. Jansen, Rec., J. Royal Neth. Chem. Soc. 95, 81 (1976); b) See also W.F. Gannon, H.0. House, Org. Synth. 40, 41 (1960). 49. H. Stetter, U. Mi Ibers, Chem. Ber. 91_, 374 (1958). 50. G. Stork, R.L. Danheiser, J. Org. Chem. 38, 1775 (1973). 51. R.M. Jacobson, R.A. Raths, J.H. McDonald, J. Org. Chem. 42, 2545 (1977). 52. J.L. Gras, Tetrah. Lett. 2111, 2955 (1978). 53. L.F. Fieser, M. Fieser, "Reagents for Organic Synthesis", Vol. 4, p. 310, Wiley-Interscience, 1974; R.A. Olofson, CM. Dougherty, J. Am. Chem. Soc. 95, 581, 582 (1973). 54. R.M. Carlson, P.M. Helmquist, Tetrah. Lett. 173 (1969). 55. Review on S,S-acetals: B.T. Groebel,. D. Seebach., Synthesis 357 (1977); see section 5.2.3, p. 392-394 and references therein for reaction of ketene-S,S-acetals and corresponding S-oxides with nucleophiles. 56. a) A. Streitwieser, . S.P. Ewig, J. Am. Chem. Soc. 97, 190 (1975); b) F. Bernardi, I.G. Csizmadia, A. Mangini, H.B. Schlegel, M.-H. Whangbo, S. Wolfe, J. Am. Chem. Soc. 97_, 2209 (1975). 57. D. Seebach, E.J. Corey, J. Org. Chem. 40, 231 (1975); D. Seebach, B. Seuring, H.-O. Kalinowski, W. Lubosch, B. Renger, Angew. Chem. Int . Ed. Engl. 16, 264 (1977). 58. A. I . Meyers, D. Comins, Synthesis 403 (1978). 59. A.W. Johnson, "Ylide Chemistry", Academic Press, 1966; A. Maercker, Org. React. T4, 270 (1965); S. Tr ipet t , Quart. Rev. 17, 406 (1963); U. Schollkopf, Newer Meth. Prep. Org. Chem. 3, 111 (T964); H.J. Bestmann ib id . 5, 1 (1968T 60. E.J. Corey, B.W. Erickson, J. Org. Chem. 36, 3553 (1971) and references therein. 127 61. CB. Reese in "Protective Groups in Organic Chemistry", J.F.W. McOmie, Ed., Plenum Press, 1973, Ch. 3; L.F. Fieser, M. Fieser, "Reagents for Organic Synthesis", Vol. 1, p.256, Wiley Interscience, 1967, and references therein. 62. E.J. Corey, A. Venkateswarlu, J. Am. Chem. Soc. 94, 6190 (1972). 63. J. March, "Advanced Organic Chemistry", McGraw H i l l , 2nd edition 1977, Ch. 19, p. 1073 f f . 64. K. Nakanishi, P.H. Solomon, "Infrared Absorption Spectroscopy", Holden-Day Inc., 1977} L.J. Bellamy, "The Infrared Spectra of Complex Molecules", Methuen, London, 1966. 65. N. Cyr, L.W. Reeves, Can. J. Chem. 43, 3057 (1965); A. Yogev, Y. Mazur, J. Org. Chem. 32, 2162 (1967). 66. F. Johnson, Chem. Rev. 68, 375 (1968). 67. a) G. Stork, J.F. Cohen, J. Am. Chem. Soc. 96, 5270 (1974); b) W.E. Parham, E.L. Anderson, J. Am. Chem. Soc. 70, 4187 (1948). 68. L.M. Jackman, S. Sternhell, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry", Pergamon Press, 2nd. edit ion, 1969. 69. R.T. O'Connor, L.A. Goldblatt, Analyt. Chem. 26, 1726 (1954). 70. G.M. Rubottom, M.I. Lopez, J. Org. Chem. 38, 2097 (1973). 71. E. Ackerman, Acta Chem. Scand. 1J_, 373 (1957). 72. Dowex 50W - X8: B.P. Mundy, R. Beier, Synth. Comm. 9, 271 (1979); Amberlyst H-15: A. Bongini, G. Cardil lo, M. Orena, S. Sandri, Synthesis 618 (1979). 73. P.A. Grieco, M. Miyashita, A. Yoshikoshi, J. Org. Chem. 42, 3772 (1977). 74. Boron t r i f luor ide etherate has been used to convert alcohols into the corresponding tetrahydropyranyl ethers: H. Alper, L. Dinkes, Synthesis 81 (1972). 75. a) R.E. Ireland, G. Pfister, Tetrah. Lett. 2145 (1969); R.E. Ireland, D.C. Muchmore, U. Hengartner, J. Am. Chem. Soc. 94, 5098 (1972); T. Mukaiyama, T. Kumamoto, T. Nagaoka, Tetrah. Lett. 5563 (1966); enol phospate review: F.W. Lichtenthaler, Chem. Rev. 61, 607 (1961); b) E. Piers, E.H. Ruediger, J. Chem. Soc. Chem. Comm. 166 (1979); R.E. Ireland, C.J. Kowalski, J.W. T i l ley , D.M. Walba, J. Org. Chem. 40, 990 (1975); E. Piers, M.B. Geraghty, R.D. Smil l ie, M. Soucy, Can. J. Chem. 53, 2849 (1975). 76. H.J. Schafer, Tetrah. Lett. 2913 (1979); F.H. Westheimer, Pure Appl. Chem. 49, 1059 (1977). 77. J.C. Coll ins, W.W. Hess, F.J. Frank, Tetrah. Lett. 3363 (1968); R. Ratc l i f fe, R. Rodehorst, J. Org. Chem. 35, 4000 (1970). 128 78. E.J. Corey, J.W. Suggs, Tetrah. Lett. 2647 (1975). 79. CrO,-pyridine: W.G. Dauben, M. Lorber, D.S. Fullerton, J. Org. Chem. 34, 3587 (1969); CrO -^HOAc: F.C. Whitmore, G.W. Pedlow, J. Am. Chem. Soc. 63, 758 (1941), L.F. Faiser, J. Sguszkovicz, J. Am. Chem. Soc. 70, 3352 (1948); Cr03-3,5-dimethylpyrazole: W.G. Salmond, M.A. Barta, J.L. Havens, J. Org. Chem. 43, 2057 (1978); see also ref. 28. 80. D.H.R. Barton, CP. Forbes, J. Chem. Soc. Perkin Trans. I 1614 (1975); D.H.R. Barton, B.J. Garner, R.H. Wightman, J. Chem. Soc. 1855 (1964). 81. K.E. Pfitzner, J.G. Moffatt, J. Am. Chem. Soc. 87_, 5661 , 5670 (1965); Moffatt, J. Org. Chem. 36, 1909 (1971). 82. For use of carbodiimides other than dicyclohexylcarbodiimide see: N. Finch, J.J. Fi t t , I.H.S. Hsu, J. Org. Chem. 40., 206 (1975); A.F. Cook, J.G. Moffatt, J. Am. Chem. Soc. 89, 2697 (1967). 83. H. Normant, Compt. rend. 239, 1510 (1954), Bul l . Soc. Chim. Fr. 728 (1957). 84. S.C. Watson, J.F. Eastham, J. Organometal Chem. 9_, 165 (1967); for a detailed descripton of the t i t ra t i on procedure, see M. Gall, H.0. House, Org. Synth. 52, 39 (1972). 85. D. Seyferth, M.A. Weiner, J. Am. Chem. Soc. 83, 3583 (1961). 86. H.0. House, L.J. Szuba, M. Gall , H.D. Olmstead, J. Org. Chem. 34, 2324 (1969). 87. A.J. Fatiadi, Synthesis 65, 133 (1976). 88. a) L.A. Carpino, J. Org. Chem. 35_, 3971 (1970); b) J. Attenburrow, A.F.B. Cameron, J.H. Chapman, R.M. Evans, B.A. Hems, A.B.A. Jensen, T. Walker, J. Chem. Soc. 1094 (1952); c) ref. 87, p. 67: from manganese dichloride and potassium permanganate. 89. M. Fieser, L. Fieser, "Reagents for Organic Synthesis", Wiley-Inter-science, 1969, Vol. 2, p. 363. 90. D. Swern, A.J. Mancuso, S.-L. Huang, J. Org. Chem. 43, 2480 (1978); see also: K. Omura, D. Swern, Tetrahedron 34, 1651 (1978). 91. D. Walker, J.D. Hiebert, Chem. Rev. 67_, 153 (1967); E.A. Braude, R.P. Linstead, K.R. Wooldridge, J. Chem. Soc. 3070 (1956). 92. H. Firouzabadi, E. Ghaderi, Tetrah. Lett. 839 (1978). 93. P.J. De Clercq, L.A. Van Royen, Synth. Comm. 9, 771 (1979). 9.4. E.H. Ruediger, private communication. 95. K.B. Sharpless, K. Akashi, J. Am. Chem. Soc. 97, 5927 (1975). 96. E.J. Corey, G. Schmidt, Tetrah. Lett. 399 (1979). 129 97. E. Vowinkel, J. Bartel, Chem. Ber. 107, 1221 (1974). 98. H.G. Foley, D.R. Dalton, J. Chem. Soc. Chem. Comm. 628 (1973); H.A. Staab, Angew. Chem. 74, 407 (1962). 99. a) E.A. Braude, E.A. Evans, J. Chem. Soc. 3333 (1955); b) H. Neumann, D. Seebach, Chem. Ber. I l l , 2785 (1978); c) review on organolithium compounds: B.J. Wakefield, "The Chemistry of Organolithium Compounds", Pergamon Press, 1974; D. Seebach, J. Organometal. Chem. Libr. 1_, 1 (1976). 100. M.J. Jorgenson, Org. React. 18, 1 (1970); J.C. Floyd, Tetrah. Lett. 2877 (1974). 101. A.W. Langer, "Polyamine-Chelated Alkali Metal Compounds", Am. Chem. Soc, Washington, D.C, 1974; M. Schlosser, J. Organometal. Chem. 8, 9 (1967). 102. R.G. Barnhardt, W.E. McEwen, J. Am. Chem. Soc. 89, 7009 (1967); for similar effects in the reaction of Grignard reagents with ketones see: M. Chastrette, R. Amouroux, J. Chem. Soc. Chem. .Comm. 470 (1970), Bull . Soc. Chim. Fr. 4348 (1970T 103. a) T. Mole, E.A. Jeffery, "Organo-Aluminum Compounds", Elsevier, 1972; b) G. Bruno, "The Use of Aluminum Alkyls in Organic Synthesis", Ethyl CorDoration, 1970, supplement 1973; c) H. Yamamoto, Angew. Chem. Int. Ed. Engl. 1_7, 169 (1978); d) E. Negishi, J. Organometal. Chem. Libr. 1, 93 (1976); E. Nigishi, S. Baba, J. Chem. Soc. Chem. Comm. 596 TT9T6). 104. Prostaglandins of E, F. series: J. Fried, CH. Lin, J.C. Sih, P. Dalven, G.F. Cooper, J. Am. Chem. Soc. 94, 4342 (1972); Brefeldin-A: P.A. Bart let t , F.R. Greene I I I , ib id . 100, 4858 (1978). 105. P.W. Coll ins, E.Z. Dajani, M.S. Brulm, CH. Brown, J.R. Palmer, R. Pappo, Tetrah. Lett. 4217 (1975). 106. J. Schwartz, D.B. Carr, R.T. Hansen, J. Am. Chem. Soc. 100, 2244 (1978). 107. S. Pasynkiewicz, W. Kuran, E. Soszynska, Roczniki Chem. 38, 1285 (1964); Chem. Abst. 62, 14555 d (1965). 108. T. Mole, E.A. Jeffery, A. Meisters, L. Bagnell, Aust. J. Chem. 27, 2577 (1974). 109. Chem. Abst. 60, 3008 d (1963). 110. 0. Diels, K. Alder, Just. Lieb. Ann. Chem. 460, 98 (1928). 111. W. Oppolzer, Angew. Chem. Int . Ed. Engl. ]6_, 10 (1977); R.G. Carlson, Ann. Rep. Med. Chem. 9, 270 (1974). 112. H. Greuter, H. Schmid, G. Frater, He!v. Chim. Acta 60, 1701 (1977); P.G. Sammes, R.A. Watt, J. Chem. Soc. Chem. Comm. 367 (1976); J.D. White, B.G. Sheldon, B.A. Solheim, J. Clardy, Tetrah. Lett. 5189 (1978). 130 113. a) Patchouli alcohol: F. Naf, G. Ohloff, Helv. Chim. Acta 57, 1868 (1974); b) Khusimone: G. Biichi, A. Hauser, J. Limacher, J. Org. Chem. 42, 3323 (1977); c) Norpatchoulenol: W. Oppolzer, R.L. Snowden, Tetrah. Lett. 3505 (1978); d) Dendrobine: W.R. Roush, J. Am. Chem. Soc. 100, 3599 (1978); e) Torreyol: D.F. Taber, B.P. Gunn, J. Am.  Chem. Soc. 101, 3992 (1979); f ) Eremophilane/valencane sesquiterpenes: F. Naf, R. Decorzant, W. Thommen, Helv. Chim. Acta 62, 114 (1979); g) CedrQl, Cedrene: A.G. Fa l l i s , E.G. Breithol le, J. Org. Chem. 43, 1964 (1978); h) Eudesmane sesquiterpenes: S.R. Wilson, D.T. Mao, J. Am. Chem. Soc. 100, 6289 (1978); i ) 9-, 2-Isocyanopupukeanane: ref. 28, 29, 30. 114. L.F. Tietze, G.v. Kiedrowski, K. Harms, W. Clegg, G. Sheldrick, Angew. Chem. Int . Ed. Engl. 1_9, 134 (1980). 115. P. Yates, P. Eaton, J. Am. Chem. Soc. 82, 4436 (1960); T. Inukai, T. Kojima, J. Org. Chem. 32, 869, 872 TT967); G. Stork, A.A. Hagedorn I I I , J. Am. Chem. Soc. 100, 3609 (1978). 116. M.E. Jung, CA. McCombs, Tetrah. Lett. 2935 (1976). 117. G.A. Mock, A.B. Holmes, R.A. Raphael, Tetrah. Lett. 4539 (1977). 118. F. Kienzle, P. Rosen, Helv. Chim. Acta 62, 442 (1979). 119. Me 3SiI: M.E. Jung, W.A. Andrus, P.L. Ornstein, Tetrah. Lett. 4175 (1977); Et 3 0+ BF», Ph3P+ BF4-: D.H.R. Barton, P.D. Magnus, G. Smith, G. Streckert, D. Zurr, J. Chem. Soc. Perk. Trans. I 542 (1972); D. H.R. Barton, P.D. Magnus, G. Smith, D. Zurr, J. Chem. Soc. Chem. Comm. 861 (1971); BF3 • Et 20: See ref. 124a. 120. M. Anteunis, Y. Rommelaere, Bul l . Soc. Chim. Beige 79, 523 (1970). 121. P.R. Jamieson, Ph.D. Thesis, Univ. of Bri t ish Columbia, 1979. 122. F. Johnson, N.A. Starkowski, W.D. Gurowitz, J. Am. Chem. Soc. 87, 3492 (1965); F. Johnson, N.A. Starkowsky, Tetrah. Lett. 1173 1796277 H. 0. House, T.M. Bare, J. Org. Chem. 33, 943 (1968). 123. H. Yamamoto, H.L. Sham, Am. Chem. Soc./Chem. Soc. Jap. Chemical Congress, Honolulu, Hawaii, April 1-6, 1979. 124. a) A. Bowers, L.C. Ibanez, H.J. Ringold, Tetrahedron 7, 138 (1959); b) H. Heymann, L.F. Fieser, J. Am. Chem. Soc. 73_, 5252 (1951 ) ; A.H. Nathan, J.A. Hogg, J. Am. Chem. Soc. 78, 6163 (1956). 125. We thank Prof. Yamamoto for kindly supplying us with a sample of 9-pupukeanone (7_). 126. a) G.H. Posner, Org. React. 22, 253 (1975); b) G.H. Posner, ib id . ]9_, 1 (1972); J.F. Normant, J. Organometal. Chem. Libr. 1_, 219"TT976); E. C Ashby, J.J. Lin, J. Org. Chem. 42, 2805 (1977). 127. L. Blaszczak, J. Winkler, S. O'Kuhn, Tetrah. Lett. 4405 (1976). 131 128. L. Weiler, F.W. Sum, Tetrah. Lett. 707 (1979), J. Chem. Soc. Chem. Comm. 985 (1978). 129. a) E.J. Corey, G.H. Posner, J. Am. Chem. Soc. 89, 3911 (1967), ib id . 90, 5615 (1968); b) G.M. Whitesides, W.F. Fischer, J.S. Fi l ippo, R.W. Bashe, H.0. House, J. Am. Chem. Soc. 9J_, 4871 (1969); c) A.T. Worm, J.H. Brewster, J. Org. Chem. 35, 1715 (1970); d) G. Buchi, J.A. Carlson, J. Am. Chem. Soc. 91, 6470 (1969). 130. F. Naf, P. Degen, Helv. Chim. Acta 54, 1939 (1971); G. Linstrumelle, J.K. Krieger, G.M. Whitesides, Org. Synth. 53_, 165 (1974); ref. 129b. 131. G. Linstrumelle, Tetrah. Lett. 3809 (1974); R. Lome, J. Mil lon, G. Linstrumelle, Synthesis 434 (1975); M. Schlosser, E. Hammer, Helv. Chim. Acta 57, 2547 (1974); R.H. Shapiro, M.F. Lipton, K.J. Kolonko, R.L. Buswell, L.A. Capuano, Tetrah. Lett. 1811 (1975); See also ref. 135b. 132. a) H. Normant, Angew. Chem. Int . Ed. Engl. 6, 1046 (1967), Bul l . Soc. Chim. Fr. 791 (1968); b] R.J.P. Corrin, J.P. Masse, J. Chem. Soc. Chem. Comm. 144 (1972); J.F. Normant, A. Commercon, G. Cahiez, J. V i l l ie ras , C.R. Acad. Sci. Paris C 278, 967 (1974); M. Tamura, J. Kochi, J. Am. Chem. Soc. 93, 1487 (1971), Synthesis 303 (1971). 133. [ RCuMgBr ] complex: M. Schlosser, G. Fouquet, Angew. Chem. Int. Ed. Engl. 1_3_, 82 (1974); t-butoxyalkylcuprate: G.H. Posner, CE. Whitten, Tetrah. Lett. 1815 (1973). 134. a) R.H. Shapiro, 0rg_. React. 23_, 405 (1975); R.H. Shapiro, K.J. Kolonko, J. Org. Chem. 43_, 1404 (1978); A.R. Chamberlin, E. L iot ta , F.T. Bond, Org. Synth. 58, no. 2069 (1977); b) L.A. Paquette, T.R. Bailey, W.E. Fristad, J. Org. Chem. 43_, 1620 (1978); S.Kano, T. Yokomatsu, T. Ono, S. Hibino, S. Shibuya, J. Chem. Soc. Chem. Comm. 414 (1978). 135. a) R.H. Shapiro, M.F. Lipton, K.J. Kolonko, R.L. Buswell, L.A. Capuan, Tetrah. Lett. 1811 (1975); b) F.T. Bond, J.E. Stemke, A.R. Chamberlin, Tetrah. Lett. 2947 (1976). 136. A.R. Chamberlin, J.E. Stemke, F.T. Bond, J. Org. Chem. 43_, 147 (1978). 137. H. Gilman, F.K. Cartledge, J. Organometal. Chem. 2, 447 (1964). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060782/manifest

Comment

Related Items