INTRAMOLECULAR DIELS-ALDER REACTIONS OF SULPHONYL-SUBSTITUTED TRIENES
A Thesis Presented by
Andrew Marsh
In Partial Fulfilment of the Requirements
for the Award of the Degree of
DOCTOR OF PHILOSOPHY
OF THE
UNIVERSITY OF LONDON
Harwood Laboratory
Department of Chemistry
Imperial College of Science, Technology and Medicine
London SW7 2AY September 1991 1
Abstract
This thesis is divided into two parts. The first part describes the development of methodology for the intramolecular Diels-Alder (IMDA) reaction of sulphonyl-substituted trienes. A brief introduction to the the area of the IMDA reaction is provided. This provides the basis for the rational development of a new class of IMDA substrate. A series of E- and Z-sulphonyl substituted deca- and undecatrienes was synthesised and their cyclisation behaviour under thermal conditions examined. Cyclisation of (IE, 7E, 9E)-l-phenylsulphonyl-l,7,9-undecatriene gave a 6:1 mixture of cis- and trans-fused bicyclic sulphones. Contrastingly, cyclisation of the corresponding (Z, £, E)- sulphonyltriene yielded a 3:1 mixture of trans- and cis-fused isomers. These results added weight to the postulate that a suitable dienophile could exert a sereocontrolling influence over the outcome of the IMDA reaction. The lower homologues of these trienes were found to provide similar results. The reaction of a series of a-methylated trienes was carried out to examine the degree of stereocontrol exerted by the phenylsulphone group over the [4+2] cycloaddition. Characterisation of the bicyclic sulphones derived from these reactions is supported by X-ray crystallographic evidence and supplemented by chemical correlation. The second part begins with a review of approaches towards the total synthesis of the CD ring of vitamin D 3 and its analogues. This introduces the use of the IMDA reaction as a concise strategy towards the construction of a vitamin D 3 CD ring synthon in a stereocontrolled fashion. The synthesis of a suitable model substrate and its cyclisation
via a novel IMDA route is described. With the success of this reaction, an
enantiomerically pure triene was synthesised. Cyclisation under thermal conditions led to
material suitable for elaboration to the natural product, vitamin D 3. The diastereomeric bicyclic tetrahydroindene adducts were epoxidised to allow their separation and
characterisation using spectroscopic techniques. Contents
Page
Abstract 1 Acknowledgements 3 Abbreviations ^ Stereochemical Notation and Compound Numbering 8
PART I 1.0 Introduction 9 2 .0 Results and Discussion Part I 22
PART II 3 .0 Review: Approaches to the Total Synthesis of the CD Ring System of Vitamin D 3 and its Metabolites 57 4 .0 Results and Discussion Part II 112
5 .0 Experimental:
Part I 149 PartE 211
6 .0 AppendixI: X-ray crystallographic data 263 Appendix II: Tables 277 Appendix III: nOe Experiments and nmr Spectra 280
7 .0 References 288 8.0 Corrigenda 298 3
Acknowledgements
I would like to express my gratitude to Dr Donald Craig for his boundless enthusiasm in his support, help and friendship during my studies. The Old Building in the Chemistry Department has been a stimulating and enjoyable place to work and learn.
The friendship of many colleagues in the Harwood, Whiffen, Barton and Perkin laboratories has made my time at Imperial College all the more enjoyable. Particular thanks are due to Jim Anderson, Steve Smith, Alison Smith and David Rainford for advice, entertainment and friendship. Thanks are due to the technical staff for providing the following services. Foremost, Mr Dick Sheppard and Mr Paul Hammerton for high-field nmr spectra; Dr D
J Williams and Ms A M Z Slawin for X-ray structure determinations; Mr J N Bilton and Mr G P Tucker for mass spectroscopic measurements; Mr K Jones and Miss H O’Callaghan for elemental analyses. Mr E Poggiolli and colleagues are thanked for the sundry services that ensure a pleasant and efficient laboratory. The SERC Mass Spectroscopy Service at the University College of Swansea deserve special mention for
such swift service. Many thanks are also due to Don, Jim, Smiffy, Biffa and Hartmuth for 500 MHz
nmr spectra at a moments notice. Dr P Grice is gratefully acknowleged for help with
molecular modelling as is Dr R Munasingh for my endeavours with the HPLC. Thanks
to my proofreaders: Don, Martin Clasby, John Reader and Neil Press. I acknowledge the SERC for providing funding during this course of work.
I would like to thank my parents for their love and support during my education. Finally my special thanks are reserved for Jill, for her patience and love which made this so
enjoyable. 4
To Jill 5
Abbreviations
Ac Acetyl
At Aromatic
Bn Benzyl br. Broad bp Boiling point n-Bu rt-Butyl f-Bu re/t-Butyl cat. Catalytic (amount)
Cl Chemical ionisation CSA 10-Camphorsulphonic acid A Heat DCM Dichloromethane
d. e. Diastereomeric excess DIBAL-H Diisobutylaluminium hydride DMAP 4-(Dimethylamino)pyridine DME Dimethoxyethane
DMF Dimethylformamide
DMPU Dimethylpropyleneurea (l,3-dimethyl-3,4,5,6-tetrahydro-2(l//)- pyrimidone
DMSO Dimethylsulphoxide
e. e. Enantiomeric excess El Electron impact eq Equivalent(s) Et Ethyl Ether Diethyl ether 6
FAB Fast atom bombardment hu Light hr Hour(s) HMPA Hexamethylphosphoramide HPLC High performance liquid chromatography
IPA Isopropyl alcohol ir Infrared LDA Lithium diisopropylamide m Multiplet m Meta m-cbpa meta-Chloroperbenzoic acid min Minute(s) MoOPH Oxodiperoxymolybdenum(pyridine)hexamethylphosphoramide
mp Melting point Me Methyl Ms Methanesulphonyl NaHMDS Sodium hexamethyldisilazide
NMO N-Methylmorpholine-N-oxide
nmr Nuclear magnetic resonance
nOe Nuclear Overhauser effect
0 Ortho
P Para PCC Pyridinium chlorochromate PDC Pyridinium dichromate Ph Phenyl
iPr iso-Propyl
Py Pyridine q Quartet rt Room temperature s Singlet
t Triplet TBAF Tetra-rt-butylammonium fluoride TBS rm-Butyldimethylsilyl TBDPS tert-B utyldiphenylsilyl Tf Trifluoromethanesulphonyl THF Tetrahydrofuran THP T etrahydro-2//-pyran
tic Thin layer chromatography TES Triethylsilyl TMS Trimethylsilyl TPAP Tetra-n-propylammonium perruthenate Ts p-Toluenesulphonyl
UV ultraviolet 8
Stereochemical Notation and Compound Numbering.
Throughout this thesis, the graphical representation of stereochemistry is in accord with the conventions proposed by Maehr .1 Thus, solid and broken wedges denote absolute configuration and solid and broken parallel-sided lines denote racemates. For the former, greater narrowing of both solid and broken wedges indicates increasing distance from the viewer.r°"i r0vi
single enantiomer racemate
In sections 1 ,2 ,4 and the experimental section, 6 , compounds adopt the systematic (IUPAC) numbering system. In section 3 however, compounds are named according to accepted steroid nomenclature since this is the convention used when
describing fragments of vitamin D 3 and its homologues. In the experimental section nmr assignments follow the systematic numbering
scheme for the compound. 9
1.0 Introduction.
Our understanding of the intramolecular Diels-Alder reaction (IMDA) has advanced greatly since its discovery nearly thirty years ago .2 This powerful strategy has found application within the arena of synthetic organic chemistry largely due to the often predictable fashion in whch four or more (if linking chain substituents are present) stereocentres may be constructed. The reaction has been the subject of several books^ and reviews4 within the past ten years, which has resulted in extensive coverage of the literature and the ideas developed in it. Whilst much research has been reported on the
IMDA reaction, there are still areas which remain poorly understood. One such fundamental area is the relationship between dienophile substitution and stereoselection exhibited by the cyclisation reaction. These factors may be expected to be closely linked. Relevant research however, has often led to the conclusion that dienophile geometry plays a secondary role in determining stereochemical preferences in intramolecular cycloadditions .5 For thermal processes, the development of a reaction whose stereochemical outcome is a direct and predictable consequence of dienophile geometry is an important and attractive goal.
In order to understand the concepts that lie behind achievement of this goal, it is necessary to consider previous research. Many natural products contain bicyclo[4.3.0]-
or bicyclo[4.4.0]-fused rings and as a result, the construction of these systems in a
controlled manner has been the subject of frequent investigations utilising this reaction.
1.1. Background: Thermal IMDA Cyclisations of Simple Trienes.
The study of simple trienes without substituents in the linking chain has emerged
as a powerful tool in learning about the stereochemical outcome of the IMDA reaction. 10
Useful insight into the origins of stereocontrol in the cycloaddition process has undoubtedly been gained from work carried out towards natural product synthesis; however, attendant functionality may obscure underlying trends. It is more instructive to consider simple trienes which have been designed to elucidate information about cycloadduct stereochemistry.
1.1.1 The thermal and Lewis-acid catalysed cycloadditions of a number of representative trienes have been studied. Substrates leading to both bicyclo[4.3.0]^ and
bicyclo[4.4.0 ]7 systems were considered (Scheme 1).
R1
1 r ' s CO2CH3, R2 = H 94% 49 : 51
2 R1 = H, R2 = CO2CH3 90% 49 : 51
55 : 45 Scheme 1 11
The conclusions drawn from these studies were: (i) product selectivity is independent of dienophile stereochemistry; (ii) secondary orbital interactions do not control the stereochemical course of thermal IMDA reactions. It has been argued 4 (a) that the small difference in selectivities between these reactions may be due to non-bonded interactions between the dienophile and terminal diene substituent in the emio-transition state (Scheme 2). If this is so, the effect is small and it should be noted that the yield is ca. 70% in both cases.
H
c h 3o 2 c
Scheme 2 1 ,1 ,2 1,3,8-Nonatrienes may cyclise via two alternative transition states to form trans-ox cis -fused bicyclo[4.3.0] systems. Moderate selectivity in favour of the cis-fused isomer was observed for the cyclisation of unsubstituted nonatriene 3 (Scheme
3)8
3 73 : 27 Scheme 3 When the dienophile was activated with a terminal ester function 9 (Scheme 4), cycloaddition occurred with a preference for trans -ring fusion. The addition of an isopropyl substituent caused a very slight increase in ironj-selectivity .10 As may be seen from the results illustrated, dienophile geometry had a marginal effect on the stereoselectivity of the cycloaddition for both ring systems. 12
Scheme 4 By way of contrast, 1,3,8-nonatrienes with a more strongly electron-withdrawing substituent on the dienophile, such as nitrotriene 4 cyclised under milder conditions and with enhanced stereoselectivity (Scheme 5 ).11
Scheme 5 13
The E, E, E-isomer 4a cyclised to give a majority of the rrans-fused indene. This was rationalised on the basis that the asynchronous nature 12 of the IMDA reaction results in the transition state resembling a five-membered, rather than a nine-membered ring. Since trans-1,2-susbstituted cyclopentanes are more thermodynamically stable than their m-counterparts, the observed product is to be expected. Notably the Z, E, E-isomer 4b gave &ca. 1:1 mixture of products. Due to the mild conditions of this cyclisation, endo- effects, such as secondary orbital overlap become a significant factor in determining the stererochemical outcome of the reaction. The transition state leads to a cw-fused product. In opposition to this, the asynchronous nature of the IMDA reaction demands that the rrarts-fused isomer be formed. The result of these two opposing factors was a non-selective reaction. Interestingly, triene 4a has been reported to undergo a silica-gel catalysed IMDA reaction which led to a single diastereoisomer in high yield (Scheme 6).^
SiC>2, hexane
25°C, 4 h
H 85% overa
E H N02
Scheme 6
The conclusions drawn from the studies on trienes substituted with a moderately
electron-withdrawing group were essentially the same as those for trienes leading to
bicyclo[4.4.0] systems. 14
1 .1 .3 It has become clear that the Alder endo- rule' is of little use in predicting the outcome of thermal IMDA reactions .4 When deciding which transition states will predominate in the reaction, steric interactions play a large role in determining the outcome. In order to help decide which of these non-bonded interactions may be of particular importance, the concept of the 'concerted but non-synchronous’ reaction pathway has been advanced .12 This model proposes that transition-state bonding is at a more advanced stage between the two termini possessing the closest-matched coefficients of the frontier molecular orbitals .14 So, when considering the thermal cyclisation of nitrotriene 4a (§1.1.2), C-2 and C -6 will be at a more advanced stage of bonding earlier along the reaction coordinate than C-l and C-9. This means that the transition state more closely resembles a 5-membered than a 9-membered array. This explains the dominance of the rrans-fused product, since the most stable pseudo-five- membered transition state possesses diene and dienophile in an anti disposition.
Decatrienes terminally-substituted with an electron-withdrawing group9, ^ cyclise such that the linking chain adopts the most stable chair-like conformation resulting from transition state’s pseudo-six-membered ring character.4 Hence substituents in the linking chain prefer to adopt a more stable equatorial disposition and often exert a stereocontrolling influence. In considering the cyclisations of ester-substituted 1,3,8-nonatrienes and 1,3,9- decatrienes the lack of stereocontrol exhibited by the dienophile geometry is surprising. The reason is almost certainly that the flat, sp2-hybridised nature of this electron- withdrawing group resulted in little non-bonded interaction with the diene system. It was postulated therefore, that one method of enabling the the dienophile to exert a stereocontrolling influence would be to make it no longer sp2-hybridised and increase its
effective three-dimensional size. Replacement of the ester function by a phenylsulphonyl group would not grossly change the established electronics of the
system (pKa a-position of phenylsulphone ~29 vs. ester -24), but the tetrahedrally- 15
disposed sulphone oxygen atoms may be in a position to interact with the diene system in a non-bonded fashion. These steric interactions may be sufficiently large in one of the transition states as to influence the stereochemical outcome of the reaction. The phenylsulphonyl group is known to exhibit an anisotropic effect in nmr spectroscopy , 15 which, it was expected would aid in characterisation of the cycloadducts. A bonus of utilising sulphonyl-substituted materials is that they arc often crystalline, a further benefit in the identification of adducts.
1.2 The Vinylic Sulphone Moiety
The sulphone group has established itself at the centre of much useful synthetic methodology.16 Its utility derives from: (i) ease of introduction; (ii) diverse reactivity; (iii) comparative ease of removal. In particular, vinyl sulphones have been shown to be versatile intermediates for cycloaddition reactions .17 Vinyl sulphones have served as convenient equivalents for
ethylene, acetylene and ketene in the context of [4+2] cycloadditions .15 The vinyl
sulphone group is usually introduced at a point in the synthesis immediately prior to that
in which they will be used. This strategy holds two advantages: firstly, it is undesirable
to carry a reactive functional group through many synthetic steps. Secondly, it enables
many analogous compounds to be derived from a small number of intermediates. Whilst
simple reductive desulphonylation is a useful and often desirable mode of sulphone
reactivity, the opportunities offered by this group are varied and as a result it has been
dubbed a 'chemical chameleon ’.15 1.2.1 In keeping with previous research into the IMDA cyclisation of terminally-substituted tricnes, we chose to study the reactions of two homologous series. Series a, n = 1 (Scheme 7) would lead to bicyclo[4.3.0] products, whereas the higher homologous series b, n = 2 would give bicyclo[4.4.0] adducts upon [4+2] cycloaddition. Trienes to be studied would further be classified according to the geometry of the dienophile. The cyclisation of both E- and Z-vinyl sulphonyl- substituted trienes would be studied. a series b series Scheme 7 The product bicycles would be expected to be of value in synthesis since the sulphone group is capable of undergoing many subsequent transformations .16 Scheme 8 illustrates some possible reactions. Sn 2' reaction SOzPh "Co CO reductive cleavage isomerisation elimination [O] oo displacement I alkylation R RQo '^ 'S O jP h Scheme 8 1.3 Vinyl Sulphones in [4+2] Cycloaddition Reactions When this research was initiated, vinyl sulphones were recognised as dienophilic partners to a variety of homo- and heterodienes in [4+2] cycloaddition reactions.1^ Sulphonyl-substituted dienes have also been recognised as 4it participants in cycloaddition reactions and several examples of their use as such have been reported .18 Most research has been directed towards the former however, and it is instructive to illustrate the reactivity of the vinyl sulphone group with some pertinent examples. The intermolecular Diels-Alder reaction of vinyl sulphone with cyclopentadiene and 1,3- cyclohexadiene has been studied (Scheme 9 ).19 18 * * ^ S 02Ph exo 22 : 78 endo ^ ^ S O jP h exo 19 : 81 endo Scheme 9 Clearly, the sulphone group had a preference for minimising non-bonded interactions with the tetrahedral sp^-hybridised carbon atoms in the incipient bridge. This was achieved by alignment endo with respect to the diene system. Several examples of intramolecular Diels-Alder reactions in which the vinyl sulphone group acted as dienophile have been reported. Benzocyclobutane 5 reacted via an o-quinodimethane intermediate in an IMDA process to give 6, a steroidal precursor Scheme 10 The dr-ring fusion obtained from this reaction is in sharp contrast to other o- quinodimethane IMDA processes .21 These usually result in trans -fused cycloadducts being formed in the majority via a less sterically encumbered exo transition state. It has been reported 22 that sulphone 7 undergoes IMDA reaction to give two products: 8 derived from an exo transition state, 9; 11 from the endo transition state 10, in the ratio 19 5.6 : 1 (R = n-Bu) and 3.4 : 1 (R = Ph) respectively (Scheme 11). Analogous sulphide and sulphoxide trienes failed to exhibit reactivity towards either thermal or Lewis-acid catalysed [4+2] cycloaddition under the conditions studied. Scheme 11 The observed stereoselectivities were rationalised entirely on the basis of unfavourable non-bonded interactions experienced by the hydrogen atoms depicted in transition state 10 (Scheme 11). In a synthesis of forskolin, an intramolecular Diels-Alder reaction of an acetylenic sulphone was used to construct part of the decalin fragment .23 Alcohol 12 was reacted with acetylenic acid 13 to give lactone 14 in 72% yield. The initially formed ester cyclised under these mild conditions to furnish the desired cycloadduct (Scheme 12). 20 Scheme 12 A related IMDA reaction has been carried out using vinyl sulphone 15.24 The dienophile is again doubly 'activated', and this was found to be a prerequisite for cycloaddition to take place (Scheme 13) leading only to trans-fused isomer 16. O Scheme 13 21 The stereocontrolling element in this reaction is clearly not the sulphone, but the carbonyl oxygen, which must suffer considerable non-bonded interactions in the exo transition state leading to the unobserved product. 1.4 Summary It was hoped to demonstrate that trienes of the type proposed in § 1.2.1 (Scheme 7) would undergo [4+2] IMDA cyclisation in a controlled fashion such that the geometry of the dienophile would determine the stereochemical outcome of the reaction. It was expected that rational design of the cyclisation substrates would aid in elucidating underlying trends as well as in unambiguous identification of cycloadducts. 22 2.0 Results and Discussion: Part I The first of the Results and Discussion chapters deals with the cyclisation of simple trienes as outlined in the Introduction (§ 1.2.1)25. The second part (§ 4.0) concerns the extension of this methodology to more demanding substrates as applied to the synthesis of a vitamin D 3 CD ring synthon. 2.1 Strategy for the synthesis of cyclisation substrates Since our objective was to study the IMDA reactivity of a number of similar sulphonyl-substituted trienes, the central theme of the strategy was to incorporate the reactive sulphone group into a common intermediate at a late stage in the synthesis. Given the ability of .the sulphone group to stabilize an adjacent carbanion,1^ (fc) this strategy was realized by coupling a nucleophilic sulphonyl species 20 with an electrophilic diene unit (Scheme 14). Thus the four required unsubstituted trienes 17- 18 would be available from just two parent aldehydes 19. 0 19 + 17a n = 1, R1 = S02Ph. R2 * H; 17b n = 2, R1 = SC^Ph, R2 = H; 18a n = 1, R1 = H, R2 = SC^Ph; 18bn = 2,R1 =H,R2 = S02Ph. R2 20 Scheme 14 23 2.2.1 The requisite dienals 19a and 19b were synthesised in good yield as summarised in Scheme 15. PhSOj (iii) x n JOn 1 \ OY OTHP I----- 21 X = OH, Y = H 24a 52% from 21a p*T 22 X = Br, Y = H 24b 61% from 21b 00 L— 23 X = Br, Y =THP (iv) OCOPh (vh) 19a 69% from 24a 19b 60% from 24b Reagents and conditions: (i) 48% aq HBr, benzene, reflux; (ii) 3,4-dihydro-2//- pyran, cat. H+, CH 2CI2, rt; (iii) PhS02Na, DMSO, rt; (iv) n-BuLi, THF, - 78°C, then CH3CH=CHCHO, -78°C, then PhCOCl, -78 to 0°C; (v)6 % Na(Hg), 3:1 THF-MeOH, -20°C; (vi) cat. H+, MeOH, rt; (vii) (COCl) 2, DMSO, CH2CI2, -60°C, then add 27, then add Et 3N, -60°C to rt. Scheme 15 24 Treatment of diols 21 with concentrated aqueous HBr in benzene under reflux with azeotropic removal of water 26 gave bromoalcohols 22. Protection of crude 22 as their THP ethers27 23 followed by reaction with phenylsulphinate anion in DMSO 28 gave sulphones 24 after chromatography. A small amount (< 5%) of product resulting from O-alkylation was always formed, but chromatographic separation of this less polar material was straightforward. When carried out with DMF as solvent ,28 this reaction was found to require more forcing conditions and greater amounts of O-alkylation were observed. Addition of phenylsulphinate anion was more direct than the method utilised initially ,25 which involved displacement of the bromide with sodium thiophenoxide and subsequent S-oxidation. The introduction of the diene function was next carried out, using Julia chemistry .29 Deprotonation of sulphones 24 using n-butyllithium in THF and reaction with crotonaldehyde followed by benzoyl chloride gave benzoates 25 as a 1:1 diastereomeric mixtures upon simple extractive workup. Reductive elimination of 25 using buffered sodium amalgam in THF-methanol at low temperature gave the dienyl THP ethers 26 after chromatography in high overall yields from sulphones 24. Analysis of the dienes 26 by *H nmr showed the presence of both E- and Z-isomers at the newly-formed double bond in the ratio of ca. 6:1. Removal of the THP protecting group 27 and Swem oxidation 28 of the resulting alcohols 27 gave sensitive dienals 19 in excellent yield. In order to remove the undesired E, Z-isomer, alcohols 27 were converted to the crystalline 3,5-dinitrobenzoate esters 28. Recrystallisation twice from ether-petrol enabled purification of 28a to 12:1 (58% yield) and 28b to 20:1 (52% yield) E, E and £, Z dienes respectively. Saponification of 28 proceeded in excellent yield, without altering the established E, E and E, Z ratio (Scheme 16). Another approach to this problem would have been to use stereospecific cuprate methodology .31 It is intriguiging to note however, that the usual starting material for this route is commercially available sorbic acid, which possesses a ca. 6:1 mixture of E, 25 E and Z, E isomers32. Purification by crystallisation 33 has also been used to improve this ratio prior to the cuprate step. (ü) 27a £,£:£,Z* 12:1 (98%) 27b £,£:£,Z= 20:1 (95%) OH Reagents and conditions: (i) 3,5-dinitrobenzoyl chloride, DMAP (cat.), Et3N, DCM; (ii) 10% aqueous KOH, THF. Scheme 16 Not only did the crystallinity of 28 allow isomeric purification, but its storage and handling properties were superior to those of alcohols 27. 2.2.2 Synthesis of (E, E,E)-a-unsubstituted trienes With access to a plentiful supply of dienals 19 ensured, the synthesis of E, E, E- trienes 17 was straightforward (Scheme 17). The reaction of dienals 19 with the lithio- anion derived from (phenyl sulphonyl)methane followed by low temperature proton quench gave a 1:1 mixture of p-hydroxysulphones 29 in excellent yield. Treatment of a crude mixture of 29 with methanesulphonyl chloride in the presence of excess triethylamine 34 gave trienes 17 in high overall yield as a ca. 97:3 mixture of E- and Z- A],2 isomers. 26 (ü) 19 17a: 82% from 19a 17b: 69% from 19b Reagents and conditions: (i) Add 19 to PhS0 2 CH2Li, THF, -78°C, then 10% v/v AcOH-THF, -78°C; (ii) CH3S0 2 C1, Et3N, CH2CI2, -6°C. Scheme 17 2.2.3. Synthesis of (Z,£ ,£)-a-unsubstituted trienes The synthesis of (Z, E, £)-a-unsubstituted trienes, 18 (Scheme 14) proved to present a far more serious synthetic challenge. Unsubstituted Z- and E-vinyl sulphones are almost invariably inseparable, even using HPLC techniques. a-Selenenylation 35 of a saturated sulphone followed by oxidation and sy/i-elimination was discounted since the E-isomer would almost certainly have resulted. A method for introducing the phenyl sulphonyl group to the same dienal precursor, 19, was required. 2 .2 .3 .1 The modified Wadsworth-Emmons reagents reported by Still 36 appeared promising. By replacing the customary diethyl or dimethyl groups on the phosphonate with bis (2,2,2-trifluoroethyl) moieties and using a potassium counterion with a cosolvent such as 18-crown-6, Z:£ ratios of ca. 10:1 have been reported in the olefination of long-chain aldehydes giving Z-a,p-unsaturated esters. The phenylsulphonyl analogue 30 of Still's carbomethoxyphosphonate was synthesised in reasonable yield as shown in Scheme 18. 27 (0, (ii) 44% PhS02‘ II O 30 Reagents and conditions: (i) PCI5,70°C; (ii) CF3CH2OH, iPr2NEt, PhH. Scheme 18 Olefination of hcptanal, considered a suitable model compound, under a variety of experimental conditions gave, at best a 3:1 E:Z ratio of vinyl sulphones in excellent yield. This compared unfavourably with the amount of Z-isomer obtained using a Peterson-type reagent ,37 (ca.. 2:1 Z: E) this particular phosphonate was not investigated further. 2 .2 .3 .2 The use of alkyl phosphoramidates has been reported as a route to geometrically pure alkenes .38 The intermediate alkoxides (Scheme 19) did not eliminate under standard Wadsworth-Emmons conditions, but could be isolated and the diastereoisomers 31 and 32 separated by crystallisation. Each diastereoisomer could then be eliminated stereospecifically to give the E- (from threo isomer 31) or Z-alkene (from erythro isomer 32) only, as required. H H R1 = alkyl, allyl 31 threo 32 erythro (iü) Reagents and conditions: (i) n-BuLi, THF, -78°C; (ii) R^CH2CHO, -78°C to rt; (iii) heat. Scheme 19 28 We sought to apply this methodology to the synthesis of Z-vinyl sulphones. Phenylsulphone analogue 33 was synthesised by sulphonylation of bis (NJJ- dimethylamino)methanephosphoramidate 34 (Scheme 20). o (0,(ii) O II II (Me2N)2P. (Me2N)2P, S02Ph 98% based on PhSC^F 34 33 Reagents and conditions: (i) n-BuLi, THF, *78°C; (ii) PhS02F, -78°C to rt Scheme 20 When the reaction of phosphoramidate 33 was carried out with heptanal however, the only isolable product was 35 (Scheme 21) o PhSOa ii S 0 2Ph (i). (ü) (Me2N)2P. (Me2N)2P ii O 33 35 Reagents and conditions: (i) n-BuLi, THF, -78°C; (ii) CH3(CH2)5CHO, -78°C to rt. Scheme 21 Thus it would appear that under the reaction conditions, the proton a to the sulphone was sufficiently labile for the elimination of the elements of water to take place. The same result was obtained by quenching the reaction mixture with 10% acetic acid in THF at -40°C. It was decided that this methodology was not applicable to the system used here and a different method was sought for the introduction of a Z-vinyl sulphone. 2 .2 .3 .3 The methods tried thus far had all used a nucleophilic sulphone portion. By reversing the sense of reactivity, and using an electrophilic sulphone moiety further possibilities became apparent for achieving the desired result (Scheme 22). 29 C-S ---- —»* "+S02Ph" S02Ph M Scheme 22 Z-l-Iodo-l-octene 36 was generated using a Wittig reaction modified according to Stork 39 (Scheme 23). Halogen-metal excange at -95°C with rerr-butyllithium thus provided the Z-l-lithio species ,40 37 which was exposed to phenylsulphonyl fluoride .41 The only product incorporating a phenylsulphonyl group that could be isolated under a number of experimental conditions, was the £-isomer, 38 in low yield. 38 37 Scheme 23 Reagents and conditions: (i) [ICH2PPh3]I, NaHMDS, THF, rt; then add heptanal, -78°C to it; (ii) I-BuLi, THF, -95°C; (iii) PhS02 F, THF A number of variations were tried with this reaction, including inverse addition of the electrophile and additives such as boron trifluoride-etherate, to no avail. It then came to our attention that vinyl sulphones were accessible by the palladium (O)-catalysed cross-coupling of vinyl stannanes with alkyl and aryl sulphonyl chlorides .42 According to this report however, a 1:1 mixture of E and Z-vinyl stannanes yielded only the E-vinyl sulphone. It was postulated that the Z-vinyl sulphone was 30 formed initially, since palladium (O)-catalysed cross-couplings are usually stereo- integrous processes, but this then rearranged under the reaction conditions (THF under reflux). This process had been shown to occur in the palladium-catalysed cross coupling of acyl chlorides with Z-vinyl stannanes .43 2 .2 .3 .4 Previous work from this laboratory 44 and others 45 had shown that E- and Z-vinyl sulphoxides, unlike their sulphone counterparts are usually separable by chromatographic means. Thus it appeared that the desired Z, E, E-trienes may be available by chemospecific 5-oxidation of a Z-sulphoxide. By utilising a one-pot vinyl sulphoxide synthesis ,44 quantities of triene 39 were readily available (Scheme 24). Reagents and conditions: (i) (Me0 )2P(0 )Me, n-BuLi, THF, -78°C, then add PhS(0)iPr, -78°C, then add 19b, -78°C to rt; (ii) chromatography; (iii) see text for details. Scheme 24 A variety of mild methods for the oxidation of sulphoxides to sulphones were screened. These included peracetic acid,46 Oxone® 47 and its DCM-soluble derivative, tetra-n-butyl ammonium Oxone®,4^ ruthenium tetroxide ,49 sodium perborate ,50 and H2C>2-diphenyldiselenide .51 The most successful of those tried was tetra n-butyl ammonium Oxone®, which allowed ~15% conversion to the desired vinyl sulphone, 18b and re-isolation of ca. 80% of the unconsumed starting material. However, 31 reproducible results were difficult to obtain. The major byproducts in the case of the other oxidants was tentatively assigned on the basis of ]H nmr and mass spectral evidence as the internal double bond epoxide. The strategy of chemoselective 5- oxidation was also hampered by the fact that Z-vinyl sulphoxides are known to undergo oxidation more slowly than their E-isomers .52 2 .2 .3 .5 The solution to this problem lay in protecting the electron-rich diene function against oxidation by forming the [4+1] cheletropic cycloadduct with sulphur dioxide .53 Heating 39 with SO 2 under pressure gave the dihydrothiophene S^S-dioxide 40 in 51% isolated yield (80% based on 39 consumed). Reagents and conditions: (i) SO2, 70°C; (ii) CH3CO3H, DCM, 0°C; (iii) PhMe, 80°C. Scheme 25 Oxidation of 40 to the sulphone 41, now proceeded cleanly with peracetic acid and gentle thermolysis in toluene revealed Z, E, E-triene 18b in 36% overall yield from sulphoxide 39 (Scheme 25). Hence sulphonyl triene 18b was available in 21% overall yield from dienal 19b. 32 A closely-related strategy was pursued with the homologous a series in which the diene fragment was protected prior to introduction of the dienophile portion of the target substrate (Scheme 26). (0 ¿ 0 42 ( ii) (iv) 43 X = S(0)Ph •c 46 X - S02Ph Reagents and conditions: (i) SO2. MeOH, 84°C; (ii) (COCl) 2, DMSO, THF, -78°C, then add 42, -78°C to -35°C, then add Et 3N, -786C to rt, then add (Me0 )2P(0)CHLiS(0)Ph, -78°C to rt; (iii) CH 3CO3H, DCM, 0°C; (iv) PhMe, 92°C Scheme 26 Heating a 6:1 E, E: E, Z mixture of dienyl THP ethers 26a with S 02 -methanol under pressure gave dihydrothiophene 5,5-dioxide 42 as a single diastereoisomer in 46% isolated yield. It is notable that this reaction simultaneously effected deprotection of the THP group, protection of the diene and removal of the minor Z, £-isomer. The remaining unprotected diene alcohol 27a was isolated in 42% yield as a 3:1 mixture of 33 £, £ and £, Z isomers, illustrating the slower reaction of the latter towards [4+1] cycloaddition. Swem oxidation of alcohol 42 under standard conditions 30 resulted in a 22% yield of aldehyde 45, with apparent decomposition of much material on work-up. TPAP Oxidation 54 of 42 on a small scale allowed isolation of aldehyde 45 ( 68 %) together with an unidentified by-product. The route finally adopted utilised a modified 55 Swem procedure. The oxidation was carried out in THF and then a solution of the pre formed sulphinylphosphonate reagent 44 was added in situ. This reaction allowed isolation of the desired Z-sulphoxidc 43 (39%) together with the unwanted £-isomer 44 (23%) and a small amount of aldehyde 45 (9%). Oxidation of Z-sulphoxide 43 as for the higher homologue proceeded cleanly to give sulphone 46. Gentle thermolysis of 46 in toluene gave Z, £, £-sulphonyl triene 18a in good overall yield. Whilst the expedient of using a protected diene provided material to carry out the cycloaddition studies, this was considered to be a rather unsatisfactory strategy to employ in synthesis. The dienes used here were disubstituted, leading to relatively unencumbered dihydrothiophene 5,5-dioxides. If tri- or tetra-substituted dienes were present in the molecule and required protection, the chcletropic reaction with SO 2 would undoubtedly become less favourable. Even if the products formed, the reverse reaction would be expected to occur more readily. Indeed, whilst stable at room temperature for long enough to allow isolation and purification, it was found necessary for long periods to store the thiophene 5,5-dioxides under argon at -18°C. 2 .2 .3 .6 A report has recently emerged 56 detailing the preparation of Z-vinyl sulphones in a stereospecific fashion. Alkynyl sulphones have been reduced using a mixture of copper (II) tetrafluoroborate and diethylmethylsilane (Scheme 27). A copper (II) hydride intermediate has been proposed as the active reducing species. HSiEt2Me/Cu(BF4)2 (94%) S i — S02Ph ‘PiOH /“\ SOzPh DSiEt2Me + Cu(BF4)2 [DCu+ *BF4] -----i - = S02Ph D/H « 32/64 D D(H) D Cu*BF4* w , w / S02Ph • S02Ph Scheme Deuterium incorporation studies employing diethylmethyldeuteriosilane have shown that the copper (II) hydride species appears to undergo a regioselective syn addition to the alkyne. The subsequent deuterium scrambling is thought to arise from two possilble termination pathways, i.e. coupling with further DCu(II)X or protonolysis by ROH. No further reduction of the vinyl sulphone was detected. It is also interesting to note that a related copper (I) hydride reagent 57 will also effect the stereospecific reduction of alkynes to Z-alkenes, although as yet there has been no report of its use with alkynyl sulphones. The requisite alkynyl sulphones 47 have also become more readily available using the recently disclosed 58 elimination reaction of a p-phenylsulphonylvinyl- phosphonate (Scheme 28). 35 PhS02 o (i) NaH, THF, rt, 30 min PhS02 R ------► \ = / 49 MR (ii) (Et0)2P(0)Cl, 2-15 h OP{0)(OEt)2 48 I-BuOK, THF, R = alkyl, aryl -786C (52 -87%) PhS02— ■ i i i i i'h h R 47 Scheme 28 In order to make the desired Z, E, E-sulphonyltrienes, the (J-oxosulphone 48 would be readily available from dienals 17 or 18. Addition of the lithio anion of (phenylsulphonyl)methane, then oxidation of the resultant p-hydroxysulphones 29 would furnish 48 (R = (CH 2)5CH=CHCH=CHCH3). Phosphorylation and elimination would then be expected to proceed as shown giving the required alkynyl sulphone. It is expected that reduction of the acetylenic sulpone using the copper (II) tetrafluoroborate and diethylmethylsilane system (vide supra) would effect stereo- and chemospecific reduction to the Z-vinyl sulphone 18b. This technology would appear to make Z, £, E-sulphonyltrienes much more readily available and their use as synthetic intermediates more attractive. 36 2.3 IMDA reactions of a-unsubstituted E, E, E-and Z, E, E- trienes The thermal IMDA reactions of E, E, E-sulphonyltrienes 17 and Z, E, E- sulphonyl trienes 18 were studied initially. The cyclisation reactions were carried out in a resealable Fischer-Porter tube in dry degassed toluene or xylene under an atmosphere of argon. Care was taken to ensure anaerobic conditions, since the presence of oxygen, even in trace amounts resulted in the formation of benzaldehyde on prolonged heating. lH Nmr of the crude reaction mixture was used to assay the ratio of cycloadducts formed for all thermal IMDA reactions studied. 2 .3 .1 E, E, E-Sulphonyltriene 17b was examined first, as a 6:1 mixture of A7 9 E, E and Z, E isomers. Heating a dilute solution of 17b at 173°C for 96 h gave a 6:1 ratio of two products. The ratio of cycloadducts was determined by ]H nmr analysis of the crude reaction mixture. Integration of the newly-formed vinyl signals was found to be the most accurate assay for achieving this comparison. Fractional crystallisation allowed isolation of pure samples of each product. X-ray crystallography (Figure 1; Appendix 1) showed the major isomer to be 50, possessing a m -ring fusion .25 00 Consideration of the possible transition states for the cyclisation of E,Z, E-triene 51, the minor contaminant in 17b showed that it may only cyclise to give a cis-fused isomer, 52 (Scheme 29).59 H 51: E, Z, £-triene 52 Scheme 29 37 It was believed,25 (a> therefore, that both the E, E, E-triene 17b and its minor contaminant, 51 had cyclised stereospecifically to give one cycloadduct each. However, X-ray crystallography (Figure 2; Appendix 1) allowed unequivocal assignment of the minor product as 53, possessing a trans-ring fusion. Clearly this could not have arisen from minor E, Z, E-triene 51, but must have been formed from cyclisation of 17b (Scheme 30). Since the IMDA cyclisations of a 6:1, or 86:14 mixture of E, E, E- and E, Z, E-trienes led to a 92% isolated yield of cycloadducts in a 6:1 ratio, it follows that some E, Z, E-triene, 51 must have been converted to the observed products. JH Nmr analysis of the reaction mixture at 50% conversion showed the same 6:1 ratio of cycloadducts as had been observed at complete reaction. This result effectively eliminated the possibility of a short-lived diastereomeric adduct. Scheme 30 In a further control experiment, cis-bicycle 50 was resubjected to IMDA conditions (175°C, 72 h) to give a 95% recovery of starting 50, with no evidence of other isomers having been formed. These observations suggested that E, Z, E-trienc 51 was not cyclising under the IMDA conditions, but underwent isomerisation to the E, E, E-isomer. The cyclisation of E, Z, E-trienes under thermal conditions is known to occur ,61 which would imply that the isomerisation 62 of E, Z, E-triene 51 takes place faster than cyclisation under these conditions. 38 In an attempt to aid understanding of the proposed isomerisation, a 6:1 mixture of £, E- and £, Z-dienols was protected as the rerr-butyldiphenylsilyl ethers, 54 and subjected to the IMDA conditions (Scheme 31). Interestingly, the isomeric ratio at ¿6 remained unchanged as evidenced by nmr. Oi) C 27b R = H 5« 54 R = TBDPS Reagents and conditions: (i) TBDPSC1, Et3N, DMAP, DCM; (ii) PhMe, 172°C, 96 h. Scheme 31 Together, these results would suggest: (a) If the 6:1 £ ,£ :£ , Z diene ratio is a thermodynamic mixture, then isomerisation of £, Z, £-triene 51 to £, £, £-triene 17b and subsequent cycloaddition is much faster than cycloaddition of the £, Z, £-triene itself under these conditions, and / or: (b) The vinyl sulphone group is a pre-requisite for the observed isomerisations to take place. It has been reported that Z, E-diene systems may apparently isomerise under similar thermal conditions when the IMDA reaction has been particularly demanding .63 By way of contrast, Boeckman has reported that no such isomerisation was observed in the IMDA reaction of a Z, Z, Z-triene .61 It should be noted however, that both of these diene systems were more highly substituted than those under consideration here and that they also contained extra electron-withdrawing substituents. Apparently therefore, a 39 sufficiently activated substrate might be expected to cyclize faster than it would isomerise. Two features of the nmr spectrum of cycloadducts 50 and 53 are worthy of note. Firstly, the alkene protons in c/i-isomer 50 resonated as a two-proton singlet, whereas those in the trans-isomer were well resolved one proton multiplets. Secondly, the CH3 group in cis isomer 50 gave rise to a doublet at 0.95 ppm, 0.26 ppm upfield from the corresponding doublet in the spectrum of 53. The second effect is attributed to the relatively deshielded environment experienced by the methyl group in trans-isomer 53. This arises due to its 1,2 -syn relationship with the phenylsulphonyl group. This feature appeared to be a general trend in the JH nmr spectra of the cycloadducts studied, which greatly aided the assignment of the crude mixtures from cyclisations. 2 .3 .2 The next substrate to be studied was Z, E, E-triene 18b. This was used as a 95:5 mixture of Z, E, E- and Z, Z, E-isomers. Thermolysis of 18b in toluene at 165-175°C for 146 h led to a 3:1 mixture of cycloadducts as evidenced by nmr analysis of the crude mixture. Purification by chromatography gave a 92% combined yield of 55 and 56 (Scheme 32) in a ratio of 3:1. ( ii) H Reagents and conditions: (i) 165-175°C, PhMe, 146 h; (ii) f-BuOK, THF, rt Scheme 32 40 Crystallisation gave pure samples of each isomer and the structure of the major, trans fused cycloadduct was confirmed unambiguously by X-ray crystallography (Appendix 1 : Figure 3). The structural assignment of the minor isomer rests on nmr evidence and the partial conversion of 56 to the as-fused isomer 50 on exposure to potassium tert- butoxide in THF. The structure of 50 had been determined unambiguously by X-ray crystallography (Figure 1; Appendix 1) Clearly, this ratio of cycloadducts differs dramatically from that observed in the cyclisation of E, E, E-triene 17b. It is also quite different from those results reported1* 10 for the cyclisation of 1,7,9-decatrienes possessing weakly electron-withdrawing groups at the 1-position (§1.1.1). These trienes typically gave ca. 1:1 mixtures of cis- and imni-fused adducts regardless of dienophile geometry. The as-selective nature of the E, £, £-triene 17b may be attributed to unfavourable non-bonded interactions between the bulky tetrahedral sulphone group and the diene in the endo transition state leading to trans-ring fusion (Scheme 33). Scheme 33 41 In similar fashion, examination of the possible transition states of 18b leading to 55 and 56 (Scheme 34) show that in this case the endo transition state possesses unfavourable non-bonded interactions between diene and dienophile. Scheme 34 Thus it would appear that simply changing the geometry of the dienophile changes the preference of this IMDA reaction from cis- to rra/w-selective. 2 . 3 .3 In order to investigate this hypothesis further, attention was turned to the lower homologous trienes 17a and 18a. Thermolysis of 17a, a 12:1 mixture of E, E, E- and £, Z, E-isomers gave a 1:1 mixture of cycloadducts 57 and 58 (Scheme 35). H H 57 58 Reagents and conditions: (i) 145°C, PhMe, 48 h Scheme 35 42 dr-Isomer 57 exhibited a 7% positive nOe between the C-4 methine and methyl protons as well as the characteristic upfield shift of the methyl doublet which possesses a 1,2- anti arrangement with the phenylsulphonyl group. The nOe was absent from the trans fused isomer 58. The structure of 58 was confirmed by X-ray crystallography (Figure 4; Appendix 1). Once again, the stereoselectivity of 17a is significantly different from that of related substrates possessing electron-withdrawing substrates other than the sulphone group. Reaction of methyl (2£, 7£, 9£)-methyl-2,7,9-dodecatrienoate gave a ca. 2:5 mixture of trans- and dr-fused products .6 Cyclisation of the nitro-substituted analogue of 17a gave a 1:9 mixture in favour of the trans -fused isomer .11 The apparent non selectivity of the sulphonyltriene reaction may be attributed to the asynchronous nature of intramolecular Diels-Alder reactions 12 (§ 1.1.3). It has been proposed that the trienes mentioned above, as well as 17a cyclise via transition states which resemble a five-membered ring more closely than than the alternative nine-membered array, trans- 1,2-Disubstituted cyclopentanes are more thermodynamically stable than their cis- counterparts, explaining the predominance of the trans-fused product in thé nitro- and estertriene cases. However, this effect is in opposition to the expected non-bonded interactions of the bulky tetrahedral sulphone, with the result that a 1:1 mixture of isomers is produced. This hypothesis was supported by the cyclisation of Z, £, £-triene 18a (Scheme 36). Thermolysis of a dilute solution of 18a gave rise to a 1:7 ratio of 59:60. The structure of trans -fused 60 was confirmed by X-ray crystallography (Figure 5; Appendix 1). In common with the previous dr-fused cycloadducts, 59 clearly showed a downfield resonance of the methyl group (1.43 ppm) compared with the rranr-fused isomer 60 (0.95 ppm). 43 H E H PhS02 7 60 Reagents and conditions: (i) 165°C, PhMe, 60h. Scheme 36 This showed how the two effects discussed above were now operating in a cooperative fashion to produce a preponderance of the trans-fustd isomer 60. Scheme 37 demonstrates how non-bonded interactions in the endo transition state clearly disfavour formation of the riî-fused diastereoisomer 59. e n d o disfavoured 59 Scheme 37 2.3.4 Summarising the results obtained from the IMDA cyclisation of a- unsubstituted sulphonyltrienes, it is clear that the dienophile geometry has a pronounced effect on the stereochemical course of the reaction. The key difference between these substrates and those previously investigated is the sp3-hybridized nature of the electron- 44 withdrawing group. Thermal IMDA reactions in which dienophile geometry has had little effect on the stereochemical outcome of the reaction have utilised sp2-hybridised electron withdrawing groups. Whilst other effects, such as the asynchronous nature of the IMDA reaction also play a large part in determining product selectivity, the concept of a dienophile geometry-controlled reaction was clearly worthy of further investigation. 45 2.4 Substrates containing a-methyl dienophiles As noted in § 1.0, substituents on both diene and dienophile may alter the stereochemical outcome and rate of an IMDA reaction. The next, logical step in examining this novel class of IMDA substrate was to see to what extent substitution affected the hypothesis outlined in § 2.2. A substituent a to the dienophilic electron withdrawing group would certainly add to the non-bonded interactions in the transition state. It was expected that this would lead to a retardation in chemical reactivity (rate of reaction) of this class of triene. The presence of two sp3-hybridised substituents on the terminus of the dienophile would help define to what extent the bulky tetrahedral sulphone group is controlling the outcome of these IMDA reactions. Thus, a-methyl substituted trienes 60-61 were chosen as suitable substrates (Scheme 38). 60a n =1 60b n = 2 Scheme 38 2.4.1 Synthesis of a-substituted trienes Dienals 19 were reacted with the lithioanion of (phenylsulphonyl)ethane 64 to give p-hydroxysulphones as ca. 1:1 mixtures of erythro- and r/zreo-diastereoiomers 62 and 63 which were separated by chromatography (Scheme 39). Treatment of the purified eryr/zro-diastereoisomers 62 with n-butyllithium followed by tosyl chloride gave tosylates 64, contaminated with unreacted p-hydroxysulphones 62. 1, 10- 46 phenanthroline was used as an indicator to ensure only one equivalent of base was added. Exposure of tosylates 64 to potassium rerr-butoxide caused E2 elimination 65 of the elements of tosic acid, giving isomerically pure E, E, £-trienes 61. 61a: 11% from 19a 60a: 20% from 19a 61b: 15% from 19b 60b: 21% from 19b Ph Reagents and conditions: (i) Add 19 to PhS 02CHLiCH3, THF, -78°C, then 10% v/v AcOH-THF, -78°C; (ii) chromatography (Si02)'. (»») «-BuLi, THF, -78°C, then TsCl, -78°C to rt; (iv) /-BuOK, THF, -20°C. Scheme 39 Trienes prepared in this way were identical with those prepared by a-Iithiation of unsubstituted analogues 17 followed by reaction with iodomethane .66 In similar fashion, r/ireo-diastereoisomers 63 were reacted with tosyl chloride to generate tosylates 65. Interestingly, tosylate 65a crystallised, enabling X*ray crystallographic analysis to 47 confirm the Mreo-stereochemistry and hence the E2 nature of the elimination reaction (Figure 6, Appendix 1). Base mediated elimination of tosylates 65 provided Z, E, E- trienes together with a small amount (< 8 %) of £, E, E-isomers 60. Fortuitously, the E- and Z- isomers of a-substituted vinyl sulphones are separable by careful chromatography, or HPLC, enabling the isolation of isomerically pure trienes. In accordance with previous re p o rts ,th e base catalysed elimination of the i/ireo-benzoyloxy sulphones was found to be non-stereospecific, leading to a ça. 1:1 mixture of E, E, E- and Z, E, E-trienes 60b and 61b. Although separation of these isomers was possible (vide supra) the use of E2 tosylate elimination was found preferable on a preparative scale. 2.4.2 Thermal cyclisation of a-methyl undecatrienes As expected, trienes 60 and 61 cyclised more slowly than their a-unsubstituted counterparts. When subjected to thermolysis in toluene at 175°C for 120 h, E, E, E- triene 61a gave a 2:3 ratio of cycloadducts in moderate yield (Scheme 40). 175°C, PhMe, 120 h Scheme 40 The nmr spectrum of cis-fused isomer 66 exhibited a methyl doublet ca. 0.25 ppm upfield from the corresponding signal in the spectrum of mwr-fused isomer 67, which was consistent with previous observations. The structure of 67 was proven unambiguously by X-ray crystallography (Figure 7, Appendix 1). Confirmation of the 48 assignment of structure 66 to the minor isomer from this reaction came from the méthylation of cycloadduct 57. When treated with n-butyllithium at -78°C, followed by iodomethane, 57 gave a 54:46 mixture of bicycles 66 and 68 in 94% yield (Appendix 2: Table 2, Entry 1). The major product was shown to be identical to 66 obtained from the IMDA reaction of triene 60a. The rranj-fused isomer from this IMDA reaction was also correlated with one of the méthylation products of cycloadduct 58 (Appendix 2: Table 2, Entry 2). This result showed that although more sluggish than the IMDA cyclisation of its unsubstituted counterpart 17a, E, £, E-triene 60a reacted to give a similar ratio of cis- and trans-fused cycloadducts, indicating that the bulky sulphone group was once again contributing to the observed stereoselection to a large degree. The rôle that the sulphone group was playing was confirmed by the cyclisation of Z, E, E-undecatriene 60a. Thermolysis provided a 1:8 ratio of cycloadducts 68 and 69 in excellent yield (Scheme 41). Reagents and conditions: (i) 190°C, xylene, 60 h. Scheme 41 The structures of both isomers rested on nmr evidence and chemical correlation, cis- Fused isomer 68 exhibited a methyl doublet ca. 0.59 ppm downfield from the corresponding signal in the nmr of /rtmi-fused isomer 69. Cycloadduct 68 was also identical with one of the products obtained from the méthylation of 57 (Appendix 2: Table 2, Entry 1). The rra/w-fused bicycle 69 was correlated with the product obtained from the méthylation of rrans-fused isomer 58 (Appendix 2: Table 2, Entry 2). The 49 asynchronous nature of the IMDA reaction and the steric demands placed on the transition state worked in concert to produce, as with the unsubstituted triene 18a, the rranr-fused isomer as the major product 2.4.3 Thermal cyclisation of a-methyl dodecatrienes These substrates reacted sluggishly under the thermolysis conditions employed. Indeed, prolonged reaction times were required in order to obtain significant conversion of starting materials, resulting in considerable decomposition and low isolated yields of cycloadducts. The use of the methyl doublets in assigning the ratios of cycloadducts present in crude reaction mixtures proved invaluable in analysing these processes. E, E, £-dodecatriene 61b cyclised with moderate selectivity but low yield in favour of the c/s-isomer (Scheme 42). These cycloadducts were identified by comparison with two methylation products 70 and 71 (Appendix 2: Table 2, Entry 3; Table 2 Entry 4 respectively). Reagents and conditions: (i) 170-178°C, xylene, 312 h Scheme 42 The final IMDA reaction in this series was that of Z, £, E-dodecatriene 60b (Scheme 43). This substrate cyclised with low frans-selectivity to give a 1:2 mixture of 72 and 73. 50 Reagents and conditions: (i) 190°C, xylene, 120 h Scheme 43 The structure of rra/ts-fused isomer 73 was determined unambiguously by X-ray crystallography (Figure 8 , Appendix 1), whilst the cw-fused isomer 72 correlated with one isomer from the methylation of adduct 50 (Appendix 2: Table 2, Entry 3). When bicyclo[4.4.0] adducts 50 and 53 were subjected to methylation as shown in Table 2, Entries 3 and 4, (Appendix 2) products due to aromatic ring methylation were observed (74,75 and 76). This result may be due in part to the hindered nature of the 4-position, a to the phenylsulphonyl group. 51 2.5 Substrates containing an a-thiomethyl substituted dienophile It was clear that the sulphone group was indeed exerting a strong stereocontrolling effect on the outcome of the IMDA reaction. The possibility of replacing the a*methyl group by other more synthetically useful functionality was appealing, since it appeared the sulphonyl group would retain control over the cycloadduct ratio in the IMDA reaction. Triene, 77 containing an a-thiomethyl moiety would act as a ketene equivalent 67 because hydrolysis of the cycloadduct 78 (Scheme 44) would result in formation of the formal product of an IMDA using a ketene as a dienophile. Further possibilities, such as acid-catalysed alkylation 68 of these cycloadducts could also be investigated. S02Ph 77 hydrolysis t O Scheme 44 2 .5 .1 Triene 77 was synthesised from aldehyde 19b by the two-step sequence shown in Scheme 45. 52 Reagents and conditions: (i) PI1SO2CH2SCH3, n-BuLi, THF, -78 C, then add 19b; AcOH-THF (10% v/v), -78°C; (ii) MsCl, Et 3N, DCM, 0°C. Scheme 45 The lithio anion of (thiomethylphenylsulphonyl)methane 69 was reacted with dienal 19b and quenched at low temperature with AcOH-THF. The resultant hydroxysulphones were then converted to triene 77 by mesylation with in situ elimination in very good overall yield. 2 .5 .2 Thermolysis of a dilute toluene solution of 77 at 175°C for 48 h did not produce a clean reaction. Analysis of the crude reaction mixture by !H nmr revealed the presence of several methyl doublets, but much useful information was obscured by decomposition products. Purification of the reaction mixture yielded a 1:1 mixture of what was tentatively assigned as cycloadducts 80 and 81 (Scheme 46). 53 Reagents and conditions: (i)175°C, PhMe, 48 h. Scheme 46 It would appear that under the reaction conditions the sensitive thioether function undergoes decomposition to unidentified byproducts. Conceivably, milder conditions such as high-pressure mediated reaction may enable cleaner conversion of this class of triene to cycloadducts suitable for further investigation. 54 2.6 Synthetic transformations of IMDA cycloadducts Alkylation of the phenylsulphonyl cycloadducts has been described (§ 2.4). Alkylation with electrophiles other than iodomethane and acylation reactions have been shown to be less facile .70 The presence of a double bond in the products suggested other possible transformations: (i) epoxidation; (ii) hydroboration; (iii) catalytic hydrogenation. The first of these reactions was performed on cis-fused bicyclo[4.4.0] adduct 50 using m-cpba to give a separable mixture of two epoxides. Chromatography gave the two crystalline epoxides, 82 and 83 (Scheme 47). The structure of the minor, less polar epoxide 82 was inferred from !H nmr and lH-!H COSY evidence. Rigorous proof of this structure came from X-ray crystallography (Figure 9, Appendix 1). The a- stereochemistry of epoxide 83 followed by deduction and was consistent with its *H nmr spectrum. 50 82 83 Reagents and conditions: (i) m-cpba, DCM, 27 h. Scheme 47 The stereoselective nature of this epoxidation undoubtedly arose from the concave shape of alkene 50, presenting the convex face as less sterically encumbered. 55 By way of contrast, hydroboration 71 of 50 with borane-dimethylsulphide complex did not exhibit a great deal of selectivity. A mixture of regio- and diastereoisomers resulted, none of which could be unambiguously characterised. Catalytic hydrogenation 72 of 50 proceeded cleanly giving crystalline 84 in good yield (Scheme 48). (i) 81% 84 H (ii) 97% -ipPhS02 85 Reagents and conditions: (i) H2, Pd-C (10%), EtOAc, 24 h; (ii)H2, Pd-C (10%), EtOAc, 4 h. Scheme 48 In similar fashion, catalytic hydrogenation of indene 58 gave 85 in high yield. Rigid chair-like sulphones such as 85 may be of interest in studying the conformational stability of sulphone anions .73 56 Part II 57 3.0 Approaches to the Total Synthesis of the CD Ring System of Vitamin D3 and its Metabolites. 3.1.0 Introduction Vitamin D3 (Cholecalciferol), its metabolites and related compounds such as Vitamin D2 (Ergocalciferol) have been the subject of considerable synthetic research effort for over 50 years. Consequently a large number of reviews covering many aspects of this work have appeared: (i) biological and biochemical research ;1 (ii) semi-synthetic methods ;2 (iii) total synthesis .3 Lythgoe's, Inhoffen's and DeLuca's seminal contributions to all three fields have been extensively covered 2 and more recently several comprehensive reviews 3 have been published covering (ii) and (iii), albeit in journals which organic chemists may find rather obscure. In order to provide a broad introduction to the field of the total synthesis of the CD rings of vitamin D 3, some repetition of this work has been necessary. Emphasis has been placed on the diverse strategies employed towards total synthesis, particularly during the last twelve years. Some of the more recent research has been concerned with the construction of ring A synthons 4 and particularly methodology for coupling them to a pre-formed CD ring system. This work has not been covered here. A brief risumd of the biochemistry involved with these targets is offered to place the synthetic work in the context in which it deserves to be viewed. Note on nomenclature: chole- or suffix-3 e.g. D3 denotes material derived from cholesterol (occurring in humans); ergo- or suffix -2 denotes material derived from ergosterol (occurring in plants). 58 3.1.1 Biochemistry Cholecalciferol 1 is produced in human skin ,5 by the action of ultraviolet radiation in sunlight on 7-dehydrocholesterol or 'procholecalciferol', 2 , followed by the isomerisations outlined in Scheme 1. 59 Electrocyclic ring opening of 2 in 6ir, conrotatory fashion leads to 3 (precalciferob) which undergoes a [l, 7]-antarafacial hydrogen shift to give 4. Rotation about the C -6 C-7 bond then provides cholecalciferol, vitamin D 3. Although this compound is biologically active, DeLuca et al. 6 carried out radiolabelling studies which elucidated its subsequent transformations in vivo. Initially, 1 is transformed in the liver to 25- hydroxyvitamin D 3, 5 (Scheme 2). In turn, the fate of 5 depends on exact physiological conditions. In a healthy human being however, 5 is hydroxylated on the A ring to give 6, la, 25-dihydroxyvitamin D 3. This compound exhibits greater biological potency than either 4 or 5 in, for example, calcium ion mobilisation. Scheme 2 la, 25-dihydroxyvitamin D 3 has been found 7 to stimulate calcium absorption and the synthesis of a calcium binding protein by the intestine. More recently ,7 it has 60 been found that la, 25-(OH)2D3 receptors are commonplace throughout many different cell types in the body. This has lead to the postulate that this compound is responsible for regulating intracellular levels of calcium ions, la, 25-(OH)2D3 is considered to be the hormonally active form of vitamin D. It is believed to be involved in the normal growth and maturation of certain cells .8 Thus, analogues of 6 may be useful as drugs in the treatment of certain cancers and skin disorders (e.g. psoriasis),9 which explains the continued interest in these molecules as synthetic targets for medicinal chemists. 3.1.2 Degradation fragments Much early synthetic work 8 centred on the degradation of vitamins D for structural elucidation purposes. Two key fragments which arose are now often chosen as targets for total synthesis, since it has been shown by several researchers that they may be easily elaborated into both natural vitamins and their unnatural analogues. Schemes 3 and 4 depict the disconnections leading to these fragments, the 'Windaus- Grundmann ketone’, 7, derived from D 3 and the ’Inhoffen-Lythgoe diol', 8 , from D 2. Scheme 3 61 3.2.0 Synthetic Strategies. The strategies employed to approach the vitamin D 3 CD ring system may be broadly divided into five areas: 3.2.1 Elaboration of a general steroidal CD ring fragment; 3 .2 .2 Formation of a highly functionalised cyclopentanoid unit, followed by introduction of the 6-ring 3 .2 .3 Biomimetic (squalene type) cyclisations; 3 .2 .4 Intermolecular Diels-Alder approaches; 3 .2 .5 Intramolecular Diels-Alder approaches. Each area will be considered in turn, showing the thinking involved with individual approaches, as well as the merits of the overall strategy. 3.2.1 Elaboration of a general steroidal CD ring fragment. An initially attractive strategy for vitamin D 3 synthesis has been to take a readily available CD ring fragment and add the steroidal sidechain in a stereoselective fashion. 62 Inevitably there have been many attempts towards this goal, often not directly connected with vitamin D3, but aiming towards a more general steroid synthesis. For the sake of brevity, this section is confined to those methods leading to D 3 and its metabolites, although other routes 11 may be applicable to this end as well. 3 .2 .1 .1 The key CD fragment often encountered as the starting material for these syntheses is optically active acid 9 (Scheme 5). Its optimised synthesis has been described in considerable detail by Hajos and co-workers .12 This enabled 9 to be used as a common precursor for analogues of vitamin D 3, as well as other steroidal targets. Baggiolini et a/.13 were able to elaborate it in linear fashion to the 25-hydroxy Windaus- Grundmann ketone, a degradation product of 25-hydroxyvitamin D 3. Hydrogenation to the rranr-fused hydrindane was achieved stereospecifically, followed by reduction of the carbonyl function to the corresponding alcohol. Treatment of the acid with excess methyllithium, then mesylation and elimination furnished unsaturated ketone 10. Hydrogenation and equilibration to the more thermodynamically stable C -8 a-epimer gave a saturated ketone which was oxidised under Baeyer-Williger conditions to ester 11. Removal of the rer/-butyl function and oxidation gave a ketone at C-17. Saponification of the C -8 ester and Wittig reaction gave a 96:4 ratio of 17Z : 17E isomers 12. Acetylation was followed by a diethylaluminium chloride-catalysed ene reaction with ethyl propiolate. The 17Z-isomer reacted faster, allowing isolation of 13 as virtually a single product. This ene reaction, also developed independently by Dauben et a/.,14 is thought to proceed via the transition state depicted in Scheme 6. Thus the relative stereochemistry of C-17 and C-20 was ensured. 63 0 ‘Bu 4 steps ? H HO 12 11 [2,3] ene reaction 4 steps Scheme 5 Hydrogenation of the product, 13 proceeded from the less hindered a-face and DIBAL- H reduction gave saturated aldehyde 14. Wittig reaction followed by oxymercuration/ 64 demercuration and PCC oxidation finally gave the 25-hydroxylated Windaus- Grundmann ketone 15. Scheme 6 3 .2 .1 .2 In the synthesis of 1 a, 25S, 26-trihydroxyvitamin D 3 16 (Scheme 7) an interesting intermolecular [3+2] cycloaddition was utilised to install the asymmetric centre at C-25. Wovkulich et a/.15 had expected to use the C-20 centre to control the it- face selectivity in the key reaction. OH Scheme 7 Z-Nitrone 17 was synthesised in four steps from the Inhoffen-Lythgoe diol. Reaction with methyl methacrylate produced a 36:45:7:12 mixture of diastereoisomeric isoxazolidines 18 - 21 in high yield (Scheme 8 ). After separation of desired isomer 18 the unwanted cycloaddition products were recycled to give 18 in 71% yield after four cycles. Subsequent manipulation of 18 lead to the desired 25S, 26-dihydroxy Windaus- Grundmann ketone which was coupled with the A ring using established methodology 16 to give 16. 65 + 20 S,R (7%) +21 R,S (12%) Scheme 8 The desired jt-face selectivity directed by the C-20 stereocentre had failed to emerge to any significant degree. In order to improve upon this, a derivative containing a removable directing group (OAc), 22, was then synthesised using a similar route and treated with methyl methacrylate at 50°C (Scheme 9). This gave an 81:19 mixture of the desired S, S and unwanted /?, R diastereoisomers 23 and 24 respectively. Removal of the acetoxy group by standard methods and exposure of the 25S, 26-diol gave material which was then convened through to la, 25S, 26-trihydroxyvitamin D 3, 16. 66 AcO 81 : 19 Scheme 9 3 .2 .1 .3 An alternative synthesis of the key intermediate 28 has been reported ,17 starting from readily available1^ ) enedione 25. Reduction of 25 with tert-butyl copper/ DIBAL-H complex (Scheme 10) gave an aluminium enolate which was trapped with bromine to generate 26 in 57% yield. Reduction with a bulky aluminium hydride reagent gave a bromohydrin which was closed to epoxide 27. Standard transformations then gave alcohol 28 in good yield. o O (i) *BuCu, DIBAL-H THF, HMPA o (ii) Br2 f H 25 Br 26 (iii) (lBuO)3LiAlH (iv) KH, HMPA (v) PDC (vi) Wittig (vii) L1AIH4 Scheme 10 67 3 .2 .1 .4 The synthesis of Vitamin D 4, 29 (Scheme 11) by Lythgoe et a/.18 is Des-AB-ergost- 8 -ene, 30, obtained by degradative means was converted (Scheme 12) to sulphide 31. Reduction of the chloride, assisted by an episulphonium intermediate, followed by oxidation and épimérisation gave sulphone 32. The preferred route involved taking the initially formed hydroxysulphide and subjecting it to benzoylation and S-oxidation. This benzoyloxysulphone was then eliminated, giving a vinyl sulphone which rearranged in situ to the more thermodynamically stable allylic sulphone 33. Hydrogenation then furnished crystalline 32. This key intermediate was then metallated, coupled with aldehyde 34 and the oxy-anion trapped with TMS-C1 (Scheme 13). Reductive elimination gave only the desired product, vitamin D 4. 68 Scheme 12 (i) LDA, THF, -78°C 29 (ii) 34 (iii) TMS-C1 O When the oxy-anion was trapped with benzoyl chloride and one diastereoisomer was subjected to reductive elimination, two products emerged in a 1:1 ratio. One was the desired, 5E, 7Z-vitamin D4, whereas the other was the 5£, 7£-isomer, 35 (Scheme 14). 69 When the corresponding acetates were reductively eliminated however, the result was identical to that obtained with the silyl ethers. 3.2.2 Highly Functionalised Cyclopentanoid Units 3.2.2.1 Strategies based around formation of a highly functionalised cyclopentanoid fragment have proven to be effective in approaching the problem of establishing the relative stereochemistry at C-20, C-17 and C-13 of vitamin D3. Whilst cyclopentanones such as 36 cannot easily be regio- and stereoselectively alkylated at C- 2, this problem may be circumvented by the use of conformationally rigid bicylo[2.2.1 ]heptane derivatives such as 37 (Scheme 15). Scheme 15 This type of strategy has also been employed by Corey et al. in the synthesis of, for example PGF 2a 19- Another Corey strategy 20 for the synthesis of prostaglandins, the 70 conjugate addition approach, has also found application in the vitamin D 3 field (§3.2.2.6 ). 3 .2 .2 .2 A synthesis of the Inhoffen-Lythgoe diol utilising the first of these strategies was achieved by Trost et a l? 1 following the retrosynthesis outlined in Scheme 16. Bicyclo[2.2.1]heptanone derivative 43 is readily available, and was converted to ketone 42 using sulphone-based anion methodology to construct the oxygenated sidechain. Méthylation of this rigid system proceeded from the less hindered endo-f&cc to give 41 in >30:1 endo\exo ratio. Thus, at an early stage in the synthesis all four stereocentres in the target, C*13, C-14, C-17 and C-20 had been 41 Scheme 16 71 Baeyer-Williger oxidation occured with regiospecific transfer of the more reactive allylic carbon to give a bridged bicyclic hydroxyacid. This rearranged upon treatment with pTsOH in benzene to a bicyclic lactone. Reduction and protection of the primary alcohol gave 40 (R' = H). Conversion to its vinyl ether followed by a [2,3]-sigmatropic Claisen rearrangement under flash vacuum pyrolytic conditions gave the correctly functionalised D ring portion 39. Cyclisation and removal of the methyl ketone moiety from 38 gave the Inhoffen-Lythgoe diol, 8 . This work was initially carried out on the racemic series, but was repeated using enantiomerically pure material to give an efficient and elegant asymmetric synthesis of key intermediates towards vitamin D 3. 3 .2 .2 .3 Independent work carried out by Grieco et a l 23 utilised a broadly similar strategy to construct a different target, (+)-de-AB-cholest-ll-en-9-one, 44, a known 24 synthetic precursor to tachysterol 3 and precalciferol 3, both structurally close to vitamin D3. Scheme 17 shows the retrosynthetic thinking applied to this target. The synthesis of key bridged bicycle 50 was achieved in a similar fashion to that used by Trost et al..21 Baeyer-Williger oxidation of 50 lead to the sensitive hydroxyacid 49 which was rearranged to lactone 48 by exposure to BF 3.Et2 0 . Wittig olefination installed the cholestyl sidechain. A Claisen rearrangement of vinyl ether 47 lead to 46, which possessed the correct relative and absolute stereochemistries at C-20, C-17, C-14 and C-13 of the target molecule. Functional group interconversion then gave ketoaldehyde 45 which was cyclised under basic conditions to give known enone 44. 72 Scheme 17 The detraction from these and other approaches which have constructed the four key stereocentres at an early stage in the synthesis is the necessary linearity of the subsequent steps required to elaborate these attractive intermediates to the desired targets. 3.2.2.4 Shimizu et a/.25 have also utilised bicyclo[2.2.1]heptanones to synthesise conformationally rigid systems. These may be cleaved to give highly functionalised vitamin D 'D' ring synthons. This approach differs retrosynthetically from those described previously. Cyclopcntanone 51 was formed by the reductive cleavage of bicyclo[2.2.1 ]heptanone 52, shown in Scheme 18. 73 54 53 Scheme 18 Acid 52 (R = H) could be alkylated diastereoselectively on the endo -face of the norbomyl framework, presumably aided by coordination of the counterion to the endo- oxygen functionality. 52 was readily accessed by saponification of the lactone 53, which in turn came from iodolactonisation of 54. Although acid 54 has been known for some time26, the synthesis of either antipode has only recently been described by Evans et al..21 This methodology enabled (+)-55 to be accessed in good yield and high enantiomeric excess. The acid 55 was methylated (Scheme 19) to give (+)-56 (R = H). Esterification afforded 56 (R = Me) which was then reductively cleaved to yield (+)-57. This was converted into the known 28 vitamin D3 synthon (-)-58 by routine synthetic transformations. 74 (i) LDA H (ii) Mel ro 2c o (+)-54 (+)-55 (+)-56 C02Me OH 6 steps O O (-)-58 (+)*57 Scheme 19 3 .2 .2 .5 The conversion of optically active cyclopentanone 58 to give 62, an intermediate in Lythgoe's synthesis 29 of vitamin D 3, has been carried out by Ficini et ai} 0 (Scheme 20). Alkylation of the thermodynamic enolate of of (-)-58, followed by protodesilylation and ring closure gave enone 59 (R = H). It was necessary to carboxylate 59 to ensure that catalytic hydrogenation proceeded from the least stoically encumbered face to give an 85:15 mixture of cis- and rrani-fused hydrindanes. 60 was separated and four routine steps via ester 61 then secured the known unsaturated ester 62. 75 Scheme 20 Ficini28 used a very different strategy to prepare 58. [2+2] Cycloaddition of ynamine 63 with enone 64 (Scheme 21) gave cyclobutene derivative 65. This enamine was hydrolysed at different rates, according to the conditions used. Previous work 31 had outlined methodology for the selective (thermodynamic vs kinetic) hydrolysis of enamines and its application to systems suitable for elaboration to vitamin D 3 appeared feasible. Hydrolysis of 65 under thermodynamic conditions led to 66a, whereas hydrolysis of 65 under kinetic protonation conditions led to 66b. This methodology was succesful in both cases for R 1 = H. However, when R 1 = Me, as required for Vitamin D 3, only the thermodynamic protonation and hydrolysis behaved as expected. Fortuitously, this lead to the required ^-stereochemistry at C-20. Thus, hydrolysis of enamine 67 gave only 68 (Scheme 22). 76 Scheme 21 Scheme 22 Resolution of 68 followed by diazomethane esterification, ketalisation and reduction gave 69. Formation of the phosphoramidate at the primary hydroxyl function, reductive elimination and deprotection achieved the conversion to optically active 70a (equivalent 77 to (*)-58). The isomer leading to the 25-hydroxy derivative, 70b, was processed differently to allow epoxidation of the olefin for subsequent conversion to the secondary hydroxyl group. 3 .2 .2 .6 The need to produce a general steroid 'D' ring synthon complete with a correctly functionalised sidechain has brought about other approaches to the problem. One solution to alkylating C-20 in an enantiospecific fashion has been to block one side of an acyclic system using functionality on an adjacent ring .32 Cyclopentanoid systems such as 71 have been synthesised in several steps from (+)-camphor. Alkylation of the anion derived from 71 with isohexyl iodide (Scheme 23) provided 72 and its diastereoisomer with the cholestyl sidechain in a ratio of 95:5 (no yield cited). Scheme 23 78 The same workers also reported *2 that lactone 73, derived by the same means, may be alkylated in >80% yield with complete diastereoselectivity (Scheme 24). Scheme 24 The reason postulated for the observed diastcreoselection was the avoidance of 1,3- diaxial interactions present in the alternative diastereoisomer. 3 .2 .2 .7 As intimated in §3.2.2.1, conjugate addition methods have enjoyed a degree of success in the synthesis of vitamin D 3 targets. Two key C-C bonds are constructed in one pot, hence defining the relative stereochemistry at three centres in a controlled fashion. De-AB-cholestan-9-one has been synthesised utilising dithianylidene anion methodology.** Treatment of the lithio-anion of74 (Scheme 25) with 2- methylcyclopent-2-enone gave an anion which was C-alkylated with allyl bromide. Standard transformations were used to convert 75 to 76. Hydroboration and oxidation gave a keto-alcohol which was elaborated using classical steroid technology*4 to de-AB- cholest-8-en-9-one. Hydrogenation of this material gave a mixture of cis- and trans- ring fused hydrindanones which were equilibrated via the p-keto ester to give the desired trans-fused isomer, 77. 79 one pot E:Z 74 O 4 steps steps O H O Scheme 25 Haynes et a/.35 used p-sulphonyl enones to trap the anion formed in the alkylation of the cyclopentenone, enabling a more highly functionalised product to be formed. The p- sulphonyl group was important for two reasons: (i) it enhanced the reactivity of the electrophile, (ii) its subsequent loss ensured the alkylation would be irreversible. The use of an allylic phosphine oxide reagent as the nucleophile allowed a subsequent Homer-Emmons reaction to be used to extend the sidechain without further functional group interconversion. Addition of the lithio-anion of phosphine oxide 78 to 79 (Scheme 26) formed an intermediate enolate 80 which was alkylated with p-sulphonyl enone 81. Loss of the sulphonyl group thus gave 82 (£:Z ratio 97:3; 76% yield). Hydrogenation and base-mediated ring closure furnished hydrindenone 83. Catalytic hydrogenation of this compound at high pressure gave a 95:5 transxis ratio of ring fused compounds. These optimised conditions relied heavily on the p-sidechain to ensure that the a-face was more accessible to the catalyst surface. This is in contrast to other workers ,12 who have utilised a C -8 carboxyl function to hydrogen bond to the C-9 carbonyl, purportedly mimicking a steroidal B ring to confer conformational rigidity. 80 DIBAL-H reduction of the carbonyl in the presence of 2,6-di-fm-butyl-4-methylphenol gave mostly the p-hydroxy compound. Standard transformations with the hydroxyl moiety unprotected then completed the synthesis of 84. Although this synthesis involved racemic material, the authors expected that by using an enantiomerically pure phosphine oxide with differentiated alkyl groups, the key C-C bond formation would transfer chirality to C-13, C-17 and C-20. 81 3.2.2.7 Vitamin D3 has been synthesised3^ utilising an 8 -phenylsulphonyl derivative, 85. This key intermediate was used by Fukumoto et al. in a Julia oleflnation of the type developed by Lythgoe and co-workers .18 Sulphone 85 was derived 37 in linear fashion from diketone 86 (Scheme.27). Routine transformations gave enol ether 87 which was cleaved ozonolytically to 88 . Homologation and ring closure gave 85. Scheme 27 This was metallated (Scheme 28), reacted with aldehyde 89 and the intermediate hydroxysulphones trapped with acetyl chloride. Reductive elimination and desilylation Scheme 28 82 3 .2 .2 .9 Ketone 90, a potential vitamin D 3 synthon 38 was synthesised via a highly functionalised cyclopentanone, 92. The retrosynthesis (Scheme 29) shows the Scheme 29 1,4-Addition of an ¿soamyl copper species generated in situ to enedione 95 secured the C-13, C-17 and C-20 stereocentres in the final product . Derivatisation of the cyclohexyl ketone in 94 as its enol acetate followed by ozonolytic cleavage produced 93. Functional group interconversion gave 92 which now included the correct cholestyl sidechain. A series of routine synthetic transformations culminating in a Homer-Wittig 83 reaction to effect one-carbon homologation, followed by lactonisation gave 91. Known methodology then produced (-)-de-AB-8-oxacholesten-9-one, 90. 84 3.2.3.1 Cationic polyalkcne cyclisations play an important role in the biosynthesis of steroids ,39 and this has naturally inspired approaches to CD ring synthons. The use of C 2-symmetric chiral auxiliaries to transfer chirality has come to the fore in the field of asymmetric synthesis .40 The strategy employed by Johnson and co-workers 41 uses a C2-symmetric acetal to induce asymmetry in the highly selective biomimetic cyclisation shown in Scheme 30. Enyne 96 was prepared using known methodology 42 and its acid catalysed reaction with ( 2S, 4S)-pentanediol afforded 97. This was cyclised by exposure to T1CI4 via the transition state depicted. >TES TES YY' HO OH H* 0 * ^ 0 CO 97 Scheme 30 85 It has been argued43 that unfavourable interaction of the Lewis acid with the methyl group a to the oxygen lone pairs leads to the conformation shown being favoured for cyclisation. This led to the chiral auxilliaiy situated on the p-face in both of the trans fused products. Cycloadducts 98 and its diastereoisomer 99 (Scheme 30) were isolated in 82 and 9% yields respectively. Removal of the auxiliary, acetylation and partial hydrogenation gave 100 (Scheme 31). This was reacted with paraformaldehyde in the presence of BF3-Et20 in a Prins44 reaction to install the sidechain in a stcreospecific fashion. Catalytic hydrogenation and removal of the acetate thus gave Inhoffen-Lythgoe diol 101 in excellent yield. 100 101 Scheme 31 2 2 3 2 The second strategy described here uses the Sharpless asymmetric epoxidation reaction to introduce the stereocentre which ultimately determines the absolute stereochemistry of the key CD ring fusion. The two olefins from which epoxides 102 and 103 originated were synthesised using established procedures.45 Sharpless epoxidation using either (-)-diethyl tartrate (for 102) or (+)-diethyl tartrate (for 103) was employed to give the two enantiomerically pure epoxides depicted in Schemes 32 and 33. Cyclisation of the E-epoxy alcohol 102 in the presence of SnCLj gave a good yield of separable allene diols 104 and 105 in a 2.4:1 ratio (Scheme 32). 86 Scheme 32 Conversely, cyclisation of the Z-epoxy alcohol 103, under the same conditions gave aliéné diols 106 and 107 in a ratio of 5:1. OH 107 Scheme 33 87 Elaboration of the major product 106 used known methodology to complete the synthesis of the Inhoffen-Lythgoe diol. 3.2.4 Intermolecular Diels-Alder Approach. The total synthesis of the Inhoffen-Lythgoe diol has been reported by Grieco et al 44 using an unusual aqueous Diels-Alder reaction. The enantiomerically pure diene 108 was synthesised from /?-(-)-methyl 3-hydroxy-2-methylpropionate. Reaction of 108 with methacrolein in water for 16 h (Scheme 34) gave rise to an intermediate carboxylic acid in ca. 70% yield. Reduction gave the major, crystalline diol 109. The minor diol 110 was isolated in 15% yield from this sequence and its single crystal X-ray analysis lead to confirmation of the structures of 109 and 110. o 108 109 110 Scheme 34 The remote diastereoselection exhibited by this aqueous Diels-Alder reaction contrasted with the non-selective reaction of the corresponding methyl ester with neat methacrolein at 55°C, which lead to a 1:1 mixture of Diels-Alder adducts in low yield. The regio- and stereochemical control exhibited by aqueous Diels-Alder reactions has been attributed47 to the relative orientation of reactants held in an aggregate within a micelle. Indeed, micellar catalysis of Diels-Alder reactions has also been invoked 48 for non-aqueous protic solvents such as ethylene glycol. 88 Selective protection of diol 109 gave 111 which was then oxidatively opened, and re cycled to give cyclopentenone 112 (Scheme 35). Chirality transfer from C-16 to C-14 was carried out via a Claisen rearrangement, necessitating the replacement of the TBDPS group with the less sterically encumbered TBS function. Addition of methyl lithium, oxidation and énolisation of the resultant ketone gave 113. Formation of the required trans -fused hydrindane 114 was achieved by Lewis acid-induced aldol cyclisation of siiyi end ether 113, followed by base mediated elimination of the p-methoxy group and catalytic hydrogenation. ^ 4 steps Scheme 35 This ketone had already been transformed into the Lythgoe-Inhoffen diol by Ttost and co-workers21,hence potentially giving access to vitamin Dy 89 3.2.5 Intramolecular Diels-Alder Approach The Intramolecular Diels-Alder reaction has been widely studied in relation to the synthesis of 0 5 fragments of vitamin D and other steroidal targets .49 The reason for its continued popularity in this context is that the relative stereochemistry of three contiguous stereocentres in the CD nucleus may be established in one step. The central problem with this approach has been control of the stereochemistry of the ring fusion. Consequently, much effort has been directed towards understanding how this may be achieved. 3 .2 .5 .1 Roush and co-workers 50 have studied in some detail how suitable trienes may be functionalised and modified to maximise the selectivity for the desired products. He recognised that either of two basic types of triene might be utilised in an IMDA reaction to give the key rans-perhydroindane ring system 115 (Scheme 36). R Scheme 36 They chose to study the thermal cyclisation of trienes 116 of the type derived from disconnection A, in which the incipient angular methyl group originates from the dienophile. Activation of either diene or dienophile was thought to be necessary to achieve acceptable rates of cycloaddition. Limitations in this respect were apparent since it was known 50 that an sp2-hybridised carbon atom would not be tolerated in the linking chain. This restricted the study to trienes 118-121 shown in Scheme 37. 90 123 r\ / RR 3 ?P H H 124 125 Scheme 37 Triene R1 R2 R3 P 122 123 124 125 118 H H c h 3 THP 50 50 119 H H CHO Bn 22 38 : 34 120 CO2CH3 H CH-? Bn 40 : 30 17 : 13 121a H c q 2CH3_ c h 3 Bn 80 20 121b H c o 2c h 3 c h 3 THP 82 18 Table 1 The table clearly shows the dependence of the stereochemical outcome of these IMDA reactions on the dienophilic activating group. Whilst an unactivated dienophile, 118, resulted in a 1:1 mixture of cis- and trans -fused products, a terminally substituted, activated dienophile 120 gave a ca. 2:1 ratio. Roush concluded that his original, late (product-like) transition state based model 51*52 was an inadequate explanation. In particular, the cyclisation of 120 would be expected to have given a majority of trans fused product. Therefore a concerted, but nonsynchronous mechanism 53 was invoked in which bonding between the olefinic termini having the largest orbital coefficients is at a more advanced stage. Thus in Scheme 38, C-3 and C-7 are closer together than C-2 91 and C-10, which would compound unfavourable interactions between the diene and dienophile substituents. Scheme 38 Interestingly, Roush also noted a high degree of internal asymmetric induction in the cyclisation of 121. This was explained by the preference for the allylic substituent (OP) to adopt a conformation in which interaction with the Z-dienophilic activating group is minimised. Hence, in Scheme 39, transition state 126 is preferred over 127. 2 ps w2 ^ r1 yL/”\ 126 R1 ■ C02CH3, R2 - H 127 Scheme 39 3 .2 .5 .2 Early work in this area 54 reported that cyclisation of triene 128 led to a mixture of cycloadducts, with 129 as the major isomer (Schemc 40). The two minor isomers were uncharacterised. Scheme 40 Later studies presented a different picture .55 Thermolysis of 128 in benzene (190°C, 13 h) or toluene (170°C, 24 h) produced a 99% yield of cis- and trans-fused isomers in the ratio 7:3. It was argued that the carbonyl function could not have been providing the 92 dicnophilc with much lowering of the activation energy, due to the non-coplanarity of the two rc-systems in the transition state. Therefore ketal derivatives would be expected to cyclise at about the same temperature as 128. Acetals 130 cyclised in excellent yield to give, after hydrolysis, a ratio of ca. 30:70 cis and trans fused products (Scheme 41). (i) 170°C, 24h, 98% (ii) HC1/H20 130a R = Me 130b R = Et Scheme 41 The explanation offered for this significant reversal in diastereoselection was based upon the argument that a more stoically encumbered carbonyl derivative would be expected to cyclise via an exo transition state placing the ketal function further away from the diene system. This proved to be the case, but changing from a dimethyl to diethyl ketal failed to bring about the expected increase in trans-ring fusion. Independent studies 56 showed that derivatives 131 and 132 (Scheme 42) cyclised with increased rra/is-selectivity. 1 (i) 200°C, 15h, PhMe 3 1 (ii) IN HC1, DME, 23°C,9h 132 Scheme 42 93 3.2.5.3 Taber and co-workers57 had interpreted similar results in terms of non- synchronous, concerted bond formation53 in the IMDA reaction. Thus, in the cyclisations of nonatrienes leading to 6,5-fused ring systems, the terminii with the largest orbital coefficients would be at the most advanced stage of bonding. In the case of an acrolein type dienophile 133 (Scheme 43), this would lead to preferential m-ring fusion, 134, over 135 by virtue of the steric constraints once the nine-membered transition state had begun to form. Scheme 43 3 . 2 .5.4 These stereochemical studies of the IMDA reaction as applied to hydrindane systems were developed towards total syntheses of vitamin D 3 by two groups working independently. It was reported58 by Parker and Iqbal, that cyclisation of triene 138, gave the CD ring system with functionality suitably disposed to enable its conversion to vitamin D3. The synthesis of 138 used an Ireland-Claisen rearrangement to establish the desired relative stereochemistry at C-17 and C-20 [steroidal numbering] (Scheme 44). This produced a 4:1 mixture of erythro and threo isomers, which were separated, and then the erythro isomer 137 (42% from 136) reduced to give alcohol 138. Cyclisation of this substrate in benzene at 200°C for 18 h produced a 23% yield of adducts 139 and 140 in a ca. 1:1 ratio. Isomers 141 and 142 were not observed 94 (Scheme 45) due to unfavourable interactions in the transition states leading to these compounds. (Hi) 136 added to LDA / THF, 5 or6 stes from then add HMPA propanoic acid (iv) TBSC1 (viii) LÌALH4 Scheme 44 Separation was achieved by treatment with 3,5-dinitrobenzoyl chloride followed by argentation chromatography to give the trans- and ris-fused cycloadducts 139 and 140 in 37% isolated yield each (R = 3,5-dinitrobenzoyl). The structure of the trans -fused isomer, 139 was proven by comparison with material derived according to Lythgoe59 from ergocalciferol. The cw-fused isomer was compared to similarly derived material, with the exception that it had been epimerised at C-14. 95 Alternatively, 143 could be cyclised in benzene at 200°C for 6 h (Scheme 46) to give a quantitative yield of esters 142 and 143. Reduction with lithium aluminium hydride and separation as previously, gave 139 and 140 in 42% and 40% yield respectively. 96 CO2M0 ,C02Me ‘ ■\'"H 200°C, PhH 6 h H H 143 144 145 Scheme 46 It was argued that transition states C and D leading to the unobserved isomers 141 and 142 respectively show that the C-17 R-substituent suffers unfavourable 1,3-interactions with atoms in the linking chain. This effect is absent in those transition states leading to 139 and 140. It has been recognised 60 for some time that a C-3 diene substituent favours closure to a rranr-fused hydrindene over the m-fused isomer. This premise was investigated 61 for substrates leading to vitamin D 3 by studying the cyclisations of trienes 146a«d and 147b-d (Scheme 47). The construction and cyclisation of model compounds lacking the cholestyl sidechain (146a-d) was studied initially to develop the methodology required for these substrates. R R A H H Y 148a-c 149a-c 146a Y=H, R=H; 147a see 138,141; 146b Y=CH3, R=H; 147b R=cholestyl; 146c Y=Br, R=H; 147c R=cholestyl; 146d Y=CH2SPh, R=H. 147d R=cholestyl. Scheme 47 97 Substrate 148:149 Conditions 148:149 Conditions (Yd%); Ph or (Yd%); Ph or PhMe PhMe 146a 0.33 (<20) 200°C, 10 h, 0.70 (61) reflux, 48 h, DTBH DMA 146b 0.67 (50) 200°C, 18 h, 3.6 (80) reflux, 12 h, MB DMA 146c 0.70 (<20) 200°C, 6 h, MB 1.0 (70) reflux, 20 h, DMA 146d 0.50 (55) 200°C, 70h, 1.7 (78) reflux, 4 h, MB DMA 147a No direct comparison 147b 3.0 (62) 200°C, 18 h, 3.0 (93) reflux, 30 h, DTBH DMA 147c 1.3 (71) 200°C, 60 h, 1.8 (86 ) reflux, 12 h, MB DMA 147d 1.0 (50) 200°C, 18 h, 2.5 (80) reflux, 12 h, DTBH DMA Abbreviations MB: methylene blue, DTBH: di-tm-butylhydroquinone, DMA: N fl- dimethylaniline. Table 2 The conclusions drawn from this data were: (i) the presence of a Y substituent does indeed favour closure to a rranj-fused hydrindene; (ii) the presence of the C-17 alkyl sidechain [steroidal numbering] favours closure to a irarw-fused hydrindene; 98 (iii) the use of dimethylaniline as solvent favours trans-fused ring closure. Interestingly, effect (ii) was present in a hydrocarbon solvent, but absent when DMA was used. Effect (iii) has been noted by other workers62, although the dependence of ring junction stereochemistry was first reported by Parker et al. Importantly, Parker noted that these substituent effects do not appear to be additive in any simple fashion. 3 .2 .S .S A more convergent synthetic approach 63 is shown in Scheme 48. An anionic Claisen rearrangement was used to establish the relative stereochemistry at C-17 and C-20 prior to coupling with a preformed diene fragment Scheme 48 The rearrangement of 151 to 152 proceeded in good yield, giving an 8:1 ratio of erythro and threo acids. The erythro isomer was crystallised to purity and converted to 153 by protodesilylation. A reduction/oxidation sequence provided the aldehyde suitable for coupling to ( 3-methylpentadienyl)lithium to give 154, a 7:3 ratio of epimeric 99 alcohols, in 50% yield. Separation and cyclisation of the major isomer (Scheme 49) gave 155 and 156 in a 3:1 ratio. 160°C, 20 h Scheme 49 The (5-stereochemistry of the sidechain is attributed to allylic 1,3 (Atf) strain«* between the vinylic hydrogen and the C-17 substituent (Scheme 50). The origins of the assignment of the stereochemistry of cfj-isomer 156 are not clear however. The related study by Parker et al. suggests that 156 ought to possess a syn- 1,2 relationship between the C-17 substituent and the bridgehead methyl. The influence of the diene methyl-substituent^ on the outcome of this IMDA reaction was confirmed by the cyclisation of triene 157 to give a 1:1 mixture of trans- and m-fused isomers, 100 158 and 159 respectively (Scheme 51). Likewise, the replacement of the hydroxyl function in triene 154 resulted in a ratio of 3:1 of trans and cis ring fused products (no yield given). Scheme 51 This methodology has been extended 65 to a total synthesis of vitamin D 3. Scheme 52 101 The methyl group on the diene was now elaborated to an hydroxyethyl substituent and the mixture of trienes 160 was cyclised (Scheme 52) to give cycloadducts 161,162 and 163 in the ratio 60:20:20 (89% yield). Note that the cw-fused isomer 163 now has a 1,2-syn relationship between the C-17 sidechain and the bridgehead methyl, in accordance with the structures observed by Parker et al..58 The total synthesis of vitamin D 3 was completed 66 by the following sequence. Adducts 161 and 162 were converted to aldehyde 164 in 10 steps in reasonable overall yield. This was then coupled (Scheme 53) with an enantiomerically pure ring A synthon 165, to give, after solvolysis of the resulting cyclopropylcarbinols 166, vitamin D 3 in 45% yield from 164. Scheme 53 102 Although carried out on racemic material, the starting material for the initial Claisen rearrangement has also been made in enantiomerically pure form, via a Sharpless kinetic resolution thus making it formally an asymmetric total synthesis .67 3 .2 .5 .6 The final approach 68 considered here utilised an IMDA reaction in the synthesis of the 25-hydroxy Windaus-Grundmann ketone 167 (Scheme 55). An IMDA reaction between an o-quinodimethane 168 derived from thermal electrocyclic ring opening of a cyclobutane 169 and a suitable dienophile was used to effect closure to a trans-fused ring system. This approach had previously been used 69 in the synthesis of steroidal targets and had been recognised to give predominantly trans ring fusion. The synthesis of benzocyclobutane 169 proceeded directly from D-mannitol derivative 170 in six steps Thermolysis of 169 in o-dichlorobenzcne (Scheme 54) resulted in the formation of tricyclic compound 171 (90%). Presumably this remarkable diastereoselection arose from adverse steric interactions between the acetal moiety and the incipient benzene ring in the endo transition state leading to rir-ring fusion. The much less stencally encumbered transition state leading to the trans ring fusion is clearly depicted in Scheme 54. 103 Scheme 54 One of the drawbacks of the synthesis is the lengthy sequence required to elaborate 171 via cyclopentane 172 into the required target 167 (Scheme 55). Scheme 55 104 3.3.0 Conclusions The diversity of approaches to the synthesis of the CD ring system of vitamin D 3 and its metabolites is striking. The key disconnection between the A ring and the C ring, although not discussed in this review has likewise been the subject of several different approaches. Most of the strategies here, however have tended to use couplings favoured by Lythgoe et al., since the use of Homer-Wittig methodology has been well-proven in this respect. This may not have been the most expeditious route to couple some intermediates to ring A synthons however. This lead to overlong, linear functional group manipulation sequences in some cases. Indeed, whilst the Julia olefination strategy used by Fukumoto et al} 6 appears to be an attractive coupling procedure, the ease of availability of the required sulphone makes this route less viable. The trend towards the asymmetric synthesis of intermediates has been notable. Particularly worthy of mention are the cationic polyene cyclisations of Johnson et al. (§ 3.2.3.1) and also the remarkable intermolecular Diels-Alder reaction carried out by Grieco and co-workers (§ 3.2.4). Although much work has been carried out on the IMDA reaction of type A noted by Roush (§ 3.2.5.1), none have utilised disconnection B, leading to a triene with a 4-methyl substituted diene. The paucity of IMDA reactions carried out on enantiomerically pure trienes is also remarkable. With the exception of Wilson’s claimed asymmetric total synthesis (§ 3.2.S.5) the possibity of using the IMDA reactions described herein without resorting to subsequent resolution seems remote. In studying approaches utilising a cyclopentanoid intermediate, Haynes' sulphone-acceptor methodology has lead to rapid construction of potentially enantiomerically pure CD-ring synthons. In summary, much interesting methodology applicable to many areas of chemistry has arisen from the research reviewed here. With renewed interest in these targets as 105 potentially pharmacalogically active substrates, further developments towards production on a larger scale will be inevitable. 106 3.4.0 References for § 3.0 1 (a). Vitamin D: Molecular, Cellular and Chemical Endocrinology ; proceedings on the Seventh Workshop on Vitamin D, Rancho Mirage, CA; Norman, A.W.; Schaefer, K.; Grigoleit, H.-G.; Herrath, D. V. Eds.; Walter de Gruyter, Berlin, 1988. (b) . For a comprehensive account of reviews on vitamin D and its actions, see Steroids 1987 ,49, 1. (c) . DeLuca, H. F.; Schnoes, H. K. Ann. Rev. Biochem., 1983, 52, 411. (d) . Norman, A. W. Vitamin D: The Calcium Homeostatic Hormone ; Academic: New York, 1979. (e) . DeLuca, H. F.; Burmester, J.; Darwish, H.; Krisinger, J. Comprehensive Medicinal Chemistry ; Pergamon, New York, 1990; Vol. 3,1129. 2 (a). Georghiou, P. E. Chem. Soc Rev. 1980,6 , 83. (b) . Lythgoe, B. Chem. Soc. Rev. 1980,9,449 and references therein. (c) . Fieser, L. F.; Fieser, M. Steroids Reinhold: New York, 1959; pp 91-168. 3 (a). Wilson, S. R.; Yasmin, A. Stereospecific Syntheses of Vitamin D; In Studies in Natural Products Chemistry, Ur-Rahman, A.; Ed.; Elsevier, Amsterdam, 1991, in press. (b) . Kametani, T.; Furuyama, H. Med. Res. Rev., 1987,7, 147. (c) . Quinkert, G.; Ed. Synform 1985, 3, 41; Ibid 1986, 4, 31; Ibid 1987,5, 1. (d) . Pardo, R.; Santelli, M. Bull Soc. Chim. Fr. 1985, 98. 4 For leading references see Posner, G.H.; Kinter, C.M. J. Org. Chem. 1990,55, 3967. 5 Fraser, D. R. In Vitamins in Medicine-, Barber, B. M.; Bender, D. A. Eds.; 4th Edn.; Heinemann: London, 1980; 42. 107 6 . Blunt, J. W.; DeLuca, H. F.; Schnoes, H. K. Biochemistry 1968, 7, 3317. 7. See Dickinson, I. Nature 1987, 325, 18 and references therein. 8 . Miyaura, C ; Abe, E.; Kuribayashi, T.; Tanaka, H.; Kuono, K.; Nishii, Y.; Suda, T. Biochem. Biophys. Res. Commun. 1981, 102, 937. 9. Holick, M. F. Proc. Soc. Exp. Biol. Med. 1989, 191, 246. 10. (a). Windaus, A.; Grundmann, W. Justus Liebigs Ann. Chem. 1936, 524, 295. (b) . Heilbron, I. M.; Jones, R. N.; Samant, K. M.; Spring, F. S. J. Chem. Soc. 1936, 905. (c) . Inhoffen, H. H.; Quinkert, G.; Schütz, S.; Kampe, D.; Domagk, G. F. Chem. Ber. 1957, 90, 664. 11 (a). For leading references see Dauben, W. G.; Brookhart, T. J. Am. Chem. Soc. 1981, 103, 237. (b). For a review on steroid sidechain homologation, see Piatak, D. M.; Wicha, J. Chem Rev. 1978, 78, 199. 12 (a). Hajos, Z. G.; Panish, D. R. J. Org. Chem. 1974 ,39, 1612 (b). Micheli, R. A.; Hajos, Z. G. , Cohen, N.; Parrish, D.R.; Portland, L. A.; Sciamanna, W.; Scott, M. A.; Wehrli, P. A. J. Org. Chem. 1975, 40, 675. 13. Baggioloni, E. G.; Iacobelli, J. A.; Hennessy, B. M.; Uskokovic, M. R. J. Am. Chem. Soc. 1982, 104, 2945. 14. See reference 11 (a). 15. Wovkulich, P. M.; Barcelos, F.; Batcho, A. D.; Seremo, J. F.; Baggiolini, E. G.; Hennessy, B. M.; Uskokovic, M. R. Tetrahedron 1984, 40, 2283. 16. Partridge, J. J.; Shiney, S.-J.; Chadka, N. K.; Baggiolini, E. G.; Hennessy, B. M.; Uskokovic, M. R., Napoli, J. L.; Reinhardt, T. A.; Horst, R. L. Helv. Chim. Acta. 1981, 61, 2138. 17. Daniewski, A. R.; Kiegel, J. /. Org. Chem. 1988, 53, 5534. 108 18. Kocienski, P. J.; Lythgoe, B.; Ruston, S. 7. Chem. Soc., Perkin 1 1979, 1290 19. See Corey, E. J.; Cheng, X.-M. The Logic of Chemical Synthesis ; Wiley: New York, 1989, 255. 20. Corey, E. J.; Niimura, K.; Konishi, Y.; Hashimoto, S.; Hamada, Y. Tetrahedron Lett. 1986,27, 2199, 3566 21. Trost, B. M.; Bernstein, P. R.; Funfschilling, P. C. 7. Am. Chem. Soc. 1979, 101, 4378. 22. Grieco, P. A.; Pogonowski, C. S.; Burke, S. D.; Nishizawa, M.; Miyashita, M.; Maskai, Y.; Wang, C.-L. J.; Majetich, G. 7. Am. Chem. Soc. 1978,99, 4111. 23. Grieco, P. A.; Takisawa, T.; Moore, D. R. 7. Am. Chem. Soc. 1979, 101, 4380. 24. Littlewood, P. S.; Lythgoe, B.; Saksena, A. K. 7. Chem. Soc. C 1971, 2955. 25. Shimizu, I.; Matsuda, N.; Noguchi, Y.; Zako, Y.; Nagasawa, K. Tetrahedron Lett. 1990, 31, 4899. 26. Beckmann, S.! Geiger, H.î Schaber-Kiechle, M. Chem. Ber. 1959,92, 2419. 27. Evans, D. A.; Chapman, K. T.; Bisaha, J. 7. Am. Chem. Soc. 1988, 110, 1238. 28. Desmaële, D.; Ficini, J.; Guingat, A.; Kahn, P. Tetrahedron Lett. 1983, 24, 3079. 29. Kocienski, P. J.; Lythgoe, B.; Waterhouse, I. Tetrahedron Lett. 1979, 4419. 30. Desmaële, D.; Ficini, J.; Guingat, A.; Tonzin, A. M. Tetrahedron Lett. 1983,24, 3083. 109 31. Ficini, J.; Krief, A. Tetrahedron Lett. 1970, 3083. 32. Hutchinson, J. H.; Money, T. J. Chem. Soc., Chem. Commun. 1986, 288. 33. (a) Ziegler, F. E.; Lim, H. Tetrahedron Lett. 1983, 24, 1859. (b) For a different approach to CD synthons see Ziegler, F. E.; Lim., H. /. Org. Chem. 1984, 49, 3278. 34. Fujimoto, G. I. J. Am. Chem. Soc. 1951, 73, 1856. 35. Haynes, R. K.; Vonwiller, S. G; Hambley, T. W. J. Org. Chem 1989, 54, 5162. 36. Nemoto, H.; Kurobe, H.; Fukumoto, K.; Kametani, T. Heterocycles 1985, 23, 567. 37. Nemoto, H.; Kurobe, H.; Fukumoto, K.; Kametani, T. Chem. Lett. 1985, 259. 38. Nemoto, H.; Kurobe, H.; Fukumoto, K.; Kametani, T. Tetrahedron Lett. 1984 ,25, 4669. 39. Johnson, W. S. Acc. Chem. Res. 1968, 1, 1. 40. Whitesell, J. K. Chem. Rev 1989, 1581. 41. Johnson, W. S.; Elliott, J. D.; Hanson, G. J. J. Am. Chem. Soc. 1984, 106, 1138. 42. Gravestock, M. B.; Johnson, W. S.; McCarry, B. E.; Parry, R. J.; Ratcliffe, B. E. J. Am. Chem. Soc. 1978, 100, 4274 and references therein. 43. Bartlett, P. A.; Johnson, W. S.; Elliott, J. D. J. Am. Chem. Soc. 1983, 105, 2088. 44. Batcho, A. D.; Berger, D. E.; Davoust, S. G.; Wovkulich, P. M.; Uskokuvic, M. R. Helv. Chim. Acta. 1981, 1682. 110 45. Hatakeyama, S.; Numata, H.; Osanai, K.; Takano, S. J. Chem. Soc., Chem. Commun. 1989, 1893. 46. Brandes, E.; Grieco, P. A.; Gamer, P. J. Chem. Soc., Chem. Commun. 1988, 500. 47. Grieco, P. A.; Yoshida, K.; Gamer, P. J. Or g. Chem. 1983, 48, 3137. 48. Dunams, T.; Hoekstra, W.; Pentaleri, M; Liotta, D. Tetrahedron Lett. 1988, 29,3745. Example of molecular aggregation in ethylene glycol. 49. Kametani, T.; Nemoto, H. Tetrahedron 1981, 37, 3. 50. Roush, W. R.; Peseckis, S. M. J. Am. Chem. Soc. 1981, 103, 6696. 51. Roush, W. R. J. Org. Chem. 1979, 44, 4008. 52. Roush, W. R.; Ko, A. I.; Gillis, H. R. J. Org. Chem. 1980, 45, 4264. 53. (a) Boeckman, R. K., Jr.; Ko, S. S. J. Am. Chem. Soc. 1980, 102, 7146. (b) Houk, K. N J . Am. Chem. Soc. 1973, 95, 4092. (c) Mclver, J. W. Acc Chem. Res. 1974, 7, 72. 54. Bajorek, J. J. S.; Sutherland, J. K. J. Chem. Soc., Perkin 1 1975, 1559. 55. Jung, M. E.; Halwweg, K. M. Tetrahedron Lett. 1981,22, 3929. 56. Bal, S. A.; Helquist, P. Tetrahedron Lett. 1981,22, 3933. 57 Taber, D. F.; Cambell, C.; Gunn, B. P.; Chin, I.-C. Tetrahedron Lett. 1981, 5141. 58. Parker, K. A.; Iqbal, T. J. Org. Chem. 1982, 47, 337. 59 . Chapleo, C. B.; Hallet, P.; Lythgoe, B.; Waterhouse, I.; Wright, P. W. J. Chem. Soc., Perkin 1 1980, 102, 6353. 60. (a) Ichihara, A.; Kimura, R.; Yamada, S.; Sakamura, S. J, Am. Chem. Soc. 1980, 102, 6353. (b) Jung, M. E.; Halweg, K. M. Tetrahedron Lett. 1984, 25, 2121. (c) Kametani, T.; Matsumo, H.; Honda, T.; Fukumoto, K. Tetrahedron Lett. 1980,27, 4847. I ll (d) Rametani, T.; Matsumo, H.; Honda, T.; Fukumoto, K.; Nagai, M. Tetrahedron 1981, 37, 2555. (c) Boeckman, R. K., Jr.; Barta, T. E. J. Org.Chem. 1985, 50, 5421. 61. Parker, K.; Iqbal, T. I. J. Org. Chem. 1987, 52, 4369. 62. (a) Stembach, D. D.; Rossana, D. M. Tetrahedron Lett. 1982, 23, 303. (b) Williams, D. R.; Gaston, R. D.; Horton, I. B., Ill Tetrahedron Lett. 1985, 26, 1391. (c) Jung, M. E.; Gervay, J. J. Am. Chem. Soc. 1991, 113, 224. Rate increase in IMDA due to polar solvent explained in terms of polar transition state. 63. Wilson, S. R.; Haque, M. S. J. Org. Chem. 1982, 47, 5413. 64. Kozikowski, A. P.; Chen, Y. Y. Tetrahedron Lett. 1982, 2 3 ,2081. 65. Wilson, S R.; Haque, M. S. Tetrahedron Lett. 1984, 25, 3147. 66. Wilson, S. R.; Haque, M. S.; Venkatasen, A. M.; Zucker, P. A. Tetrahedron Lett. 1984,25, 3151. 67. For more recent modifications to this synthesis, see: Wilson, S. R.; Venkatasen, A. M.; Angelli-Szafran, C. E.; Yasmin, A. Tetrahedron Lett., 1991 , 32, 2339 and Rabat, M.; Riegel, J.; Cohen, N.; Toth, R.; Wovkulich, P. M.; Uskokuvic, M. R. Tetrahedron Lett. 1991, 32, 2343. 68 . Nemoto, H.; Ando, M.; Fukumoto, R. Tetrahedron Lett. 1990, 31, 6205. 69. For example, see Nemoto, H.; Nagai, M., Moizumi, M.; Roluzuki, R.; Fukumoto, R.; Rametani, T. J. Chem. Soc, Perkin 1 1989, 1639. 112 4.0 Results and Discussion Part II The preceding Results and Discussion chapter described the development of sulphone IMDA methodology. Successively more demanding substrates were cyclized in a study designed to test the limits of this chemistry. We sought an appropriately demanding natural product and the CD ring system of vitamin D 3 appeared to offer a particularly suitable opportunity to test the cyclisation. 4.1.1 Synthetic strategy towards vitamin D 3 The initial disconnection of vitamin D 3 followed an established route 74 (disconnection a) to give a sulphonyl-substituted indene ring system, I (Scheme 49). This was envisaged as being the product of a sulphonyl-substituted triene IMDA reaction. Clearly, there are substantial differences between this substrate, II and the trienes with which this methodology was developed: (i) there is an angular methyl group, which must arise from a 4-substituted 1,3- diene; . (ii) the presence of a substituent p to the diene had not previously been investigated. The internal diene substituent would, it was anticipated, retard the closure of the triene. However, it was recognised that this might be compensated for by the lack of terminal diene substituent, also the presence of the substituent p to the diene was expected to improve the rate of cyclisation. It has been recognised that substituents on the linking chain often control the stereochemical outcome of IMDA reactions to a large degree 4(a) 113 V III /?-(+)-pu!egone VII Scheme 49 The disconnection employed (Scheme 49) is of type B described by Roush7* (§ 3 2 5 1) and was hitherto unexplored in the context of vitamin D 3 synthesis. Indeed, 114 there are few examples of IMDA substrates containing a 4-substituted 1,3-diene (vide infra). This disconnection, as opposed to type A would ultimately also provide an 8 - sulphonyl indane fragment, suitable for coupling using modified 74 0>) Julia29 chemistry to an enantiomerically pure aldehyde HI, as had been previously demonstrated .74 (fi) Further rctrosynthetic analysis of in (Scheme 49) showed it to be derived from aldehyde IV, giving access, in principle to either E- or Z-dienophiles. At the outset, it was envisaged that the diene unit would arise from a methyl ketone V via Wadsworth- Emmons chemistry ,76 which in turn would come from acid VI. This retrosynthesis would enable Evans-enolate asymmetric alkylation methodology 77 to be applied to citronellic acid derivative Vn. Thus the asymmetric methyl group and sidechain would originate from enantiomerically pure fl-(+)-pulegone Vm . 4,2 Related IMDA Cyclisations There have been several approaches to vitamin D 3 and its analogues using an IMDA strategy, described in detail in § 3.5. None utilised a 4-substituted diene unit however, preferring to introduce the angular methyl group as a substituent on the dienophile. IMDA cyclisations in which a 4-substituted diene is present are rare for bicyclo[4.4.0] systems as well as the bicyclo[4.3.0] systems under investigation. 4 2 1 In the IMDA cyclisation of substrate 86 (Scheme 50) a ca. 1:1 mixture of epimeric cycloadducts 87 was formed. The concern that 1,5-prototropic shifts along the methylated diene would thwart the reaction were unfounded .78 115 OTBS OTBS COjMeO 87 Similar concern had been expressed by Boeckman et al. in the cyclisation of Z, Z, Z- trienes, although the conclusion was drawn that even in highly substituted systems, [1 ^ 1-hydrogen shifts do not present limitations to desired cycloaddition processes .61 4 .2 .2 In the cyclisation of Z, Z, £-triene 88 however, it was proposed that rearrangement to Z, £, E-triene 89 preceded cyclisation to a 1:1 mixture of cycloadducts 90a and 90b, notably both m-fused (Scheme 51).63 90a 90b Scheme 51 116 The structure of 90b was subsequntly confirmed by its conversion to 8 -ep/-dendrobine. 4 .2 .3 The IMDA reaction of a mixture of trienes 91 and 92 led to a single product, 93 in high yield (Scheme 52).79 The authors raise the question of a Z to E isomerisation taking place in this reaction. 4 .2 .4 In the attempted IMDA reaction of a 3-thiophenylthiophene 5, 5-dioxide- derived triene,18 0>) the isomerisation of a terminal 1,3-diene has been observed, giving a methyl-substituted 1,3-diene. 117 4 .3 Development of a Model Cyclisation Substrate It is obvious from the foregoing examples that the stereochemical outcome of this type of IMDA cyclisation is far from clear. It was therefore decided to study the cyclisation of a suitable model compound before embarking on the synthesis of an enantiomerically pure substrate. Triene 94 (Scheme 53) appeared ideal since it possessed an isopropyl group which ought to be an effective model for the steric demands of the cholestyl sidechain. The proposed retrosynthesis follows a similar route to that suggested for the enantiomerically pure triene II (Scheme 49). Scheme 53 Disconnection of the sulphonyl unit reveals an aldehyde, which would originate from protected alcohol 95. The diene function was expected to be fabricated from methyl ketone 96, which in turn came from acid 97. Disconnection of an electrophilic three carbon synthon from 97 left the starting material as isovaleric acid. 4 .3 .1 The electrophilic synthon used was 2-(3-iodopropoxy)tetrahydro-2//- pyran 98. This was available in two steps 80 from 3-chloropropanol (Scheme 54). 118 Finkelstein81 reaction of the protected chloride99 effected transformation to the iodide in good yield. (i) 94% Cl I OTHP (ii) 67% 98 Reagents and conditions: (i) 3,4-dihydro-2//-pyran, CSA (cat.), DCM, (ii) Nal, acetone, reflux, dark, 48. Scheme 54 The next step, alkylation of isovaleric acid was problematic (Scheme 55). A mixed cation system 82 was found most suitable to accomplish the alkylation, but isolation of the resultant acid 100 was difficult. Reagents and conditions: (i) Add isovaleric acid to NaH, THF, ip^NH, rt; then add n-BuLi, 0°C; then add 98,0°C; (ii) CH 2N2, DCM, (iii) MeLi, THF, 0°C; add to rapidly stirred NH 4CI. Scheme 55 The presence of an acid labile protecting group 27 was necessary later in the synthesis, but meant that the aqueous phase could not be acidified in order to aid extraction of the product. Saturation with sodium chloride was used in an attempt to decrease the concentration of product in the aqueous phase, with limited success. The alkylation itself 119 was never satisfactory, since a large amount of unconsumed electrophile was always recovered. Addition of a cosolvent, DMPU ,83 did not appear to be beneficial and merely made purification more awkward. In order to aid characterisation of the product, the acid was esterified with diazomethane .84 to give methyl ester 101. Conversion of the crude acid 100 to methyl ketone 102 was reasonably efficient and there was no evidence of tertiary alcohol byproduct from this reaction .85 Wadsworth-Emmons76 reaction of ketone 102 with triethylphosphonoacetate failed under all conditions studied. This included deprotonation of the phosphonate by potassium hydride/18-crown-6 in DMF. The attempted addition of lithioallylsulphone 86 to 102 in the presence of boron trifluoride-etherate resulted in a moderate yield of the 1,3-addition product 103 (Scheme 56). This demonstrates the hindered nature of ketone 102. Reagents and conditions: (i) BF3.Et20, DME, -78°C; (ii) PhS02CH2CHCH, n-BuLi, DME, -78°C, add to (i). Scheme 56 Clearly, the combination of the poor alkylation step and lack of subsequent reactivity due to steric hinderance made this route a poor choice and was not investigated further. 120 4 .3 .2 Carbometallation route to 1,3-dienes. Hydromethylation of an alkyne may be accomplished with trimethylaluminium catalysed by zirconocene dichloride .87 The resultant alane intermediate has been reported 88 to undergo palladium( 0)-catalysed cross-coupling with vinyl bromide to give a 4-methyl substituted diene such as 104 (Scheme 57). It was hoped to use this methodology to construct the required dienol from alkynol 105. This would in turn be available from aldehyde 106 using a homologation and elimination sequence. 104 Scheme 57 4 3 2.1 a-Alkylation of a nitrile is considerably more efficient than of a carboxylic acid.89 Isovaleronitrile was deprotonated with LDA and added to the electrophile at -78°C, giving nitrile 107. Inverse addition was necessary to prevent dialkylation. Despite the supposed ease of alkylation, scale-up reduced yields by 10-15%, even when a cosolvent, DMPU was added to the anion to increase its reactivity. Reduction of the nitrile function with DIBAL-H 90 at -78°C gave the aldehyde 106 in good yield. This THP-protected compound appeared rather unstable, however. Attempted homologation using triphenylphosphine / carbon tetrabromide9* failed to produce satisfactory results. A similar observation when preformed iodomethylenetriphenylphosphorane 89 was employed to give vinyl iodide 108 led to the conclusion that aldehyde 106 was too unstable to use in this sequence (Scheme 58). 121 (i) CN 35-50% Reagents and conditions: (i) Add isovaleronitrile to LDA, -78°C, THF; then add to I(CH)20THP; (ii) DIBAL-H, -78°C, PhMe; H20-THF (10% v/v), -78°C; (iii) [ICH2PPh3]I, NaHMDS, THF, rt; cool to -78°C then add 106 Scheme 58 4 .3 .2 .2 A vast improvement in the stability of the aldehyde and efficiency of the subsequent homologation was found simply by exchanging the THP protecting group for the acid-stable TBDPS27 function. (Scheme 59). Deprotection to alcohol 109 and reprotection to give silyl ether 110 proceeded in excellent yield. Reduction to aldehyde 111 proceeded cleanly, without evidence of decomposition. Wittig reaction using iodomethylenetriphenylphosphorane gave an 80% yield of isomerically pure Z-vinyl iodide 112. 122 (iii) 90% Reagents and conditions: (i) MeOH, CSA (cat.); (ii) TBDPSC1, E 13N, DMAP (cat.), DCM, 0°C; (iii) DIBAL-H, PhMc, -78°C, H20-THF (10% v/v), -786C; (iv) [ICH2PPh3)I. NaHMDS, THF, rt; cool to -78°C, then add 111. Scheme 59 Elimination of a vinyl bromide to an alkyne is known 92 to proceed smoothly at low temperature using two equivalents of base. Vinyl iodide 112 was subjected to elimination using potassium rerr-butoxide, furnishing alkyne 113 in excellent yield. Reagents and conditions: (i) /-BuOK, THF, -78°C, rt; (ii) TBAF, THF, rt. Scheme 60 There was no evidence of this process occurring in situ during the Wittig reaction. Deprotection with TBAF 27 gave alkynol 105 in near-quantitative yield (Scheme 60). 123 4 .3 .2 .3 The Negishi carbometallation procedure 88 using in situ palladium (0) catalysed cross coupling procedure was attempted next (Scheme 61). Reagents and conditions Z1CP2CI2 (0.4 eq), Me3Al, C1(CH2)2C1, then add 105,16 h; then ZnCl 2.Et20, CHB1CH2, Pd(PPh3)4 (cat.). Scheme 61 Despite much experimentation, only minor quantities of the desired diene 104 could be isolated, the major product being vinyl compound 114, which resulted from proton quench of the intermediate vinylalane species. This reactivity problem of the intermediate alane/zincate was circumvented by trapping it with iodine9^ and then reversing the sense of the Pd(0)-catalysed cross-coupling .94 Thus, reaction of vinyl magnesium bromide with vinyl iodide 115 in the presence of palladium (0) gave the desired dienol 104 in very good overall yield, contaminated with minor amounts (<5%) of 114 (Scheme 62). Reagents and conditions: (i) Z1CP2CI2 (0.4 eq), Me3Al, C1(CH2)2C1, then add 105, 16 h, then add I2, THF, -40°C; (ii) Pd(PPh3)4, PhMe, then add CH2=CHMgBr, THF; (iii) (COCl)2. DMSO, DCM, -60°C; then add 104; then add Et 3N, -60°C to it. Scheme 62 124 It is noteworthy that the initial carbometallation procedure is known 95 to proceed in improved yields if the alcohol function is unprotected, since this reagent system is not compatible with rerf-butyldiphenylsilylethers. This now gave access to the key dienal 116, obtained by Swem oxidation 30 of alcohol 104 in 85% yield. In principle, both Z- and E-vinyl sulphones are available from this dienal. The E, E, E-triene was chosen initially for two reasons: (i) it was felt that the presence of the 4-methyl group and linking-chain substituent would be such as to alter the pattern of stereocontrol observed with the simple trienes and hence E, E, E-triene might cyclise with rro/u-selectivity. (ii) It was much more readily available than the Z-isomer. With regard to (ii), several unsuccessful attempts were made to protect dienol 104 via cheletropic reaction with SO 2.53 Reaction of dienal 116 with the lithioanion derived from (phenylsulphonyl)methane gave p-hydroxysulphones 117. Reagents and conditions: (i) PhS02CH2Li, THF, -78°C; AcOH-THF (10% v/v), -78°C; (ii) MsCl, Et3N. DCM, -6°C. Scheme 63 125 Mesylation and in situ elimination furnished triene 94 in good overall yield with high isomeric purity (A 1>2 E:Z = 97:3) (Scheme 63). 4.3.3 IMDA reactions of model substrate Thermal IMDA reactions of E, E, E-triene 94 were initially in carried out in ds- toluene in a sealed nmr tube. This enabled the progress of the reaction to be monitored by lH nmr. Heating triene 94 at 180°C in a Woods' metal bath for 21 h initially consumed ca. one-third of the starting material (Scheme 64). After a further 23 h, the reaction was two-thirds complete and no triene was evident after a further 12 h. At this stage there appeared to be two cycloadducts 118 and 119 in a ca. 1:1 ratio as evidenced by new vinylic signals and the presence of only two methyl singlets at 1.21 ppm and 0.95 ppm in the nmr spectrum of the crude mixture. Reagents and conditions: (i) 180°C, ds-PhMe, 57 h, sealed nmr tube. Scheme 64 The reaction was scaled up to 0.62 mmol and conducted in a rescalabic Fischer-Porter tube, the type used previously. Thermolysis of a dilute toluene solution at 180°C for 68 h in a Woods' metal bath produced only ca. 5% conversion to the expected cycloadducts. The original nmr tube cyclisation was repeated to verify the result, and after thermolysis in dg-toluene for 68 h, a 1:1 mixture of cycloadducts resulted in 66% isolated yield. A control experiment was performed, heating 94 in toluene in a sealed nmr tube at 170°C 126 for 69 h, which led to ca. two-thirds conversion to a 1:1 mixture of products. The slightly lower temperature was thought to account for the slower rate of reaction. This experiment effectively discounted the possibility that an impurity in the dg-toluene had catalysed the reaction. In an attempt to investigate the possible pressure-dependence of the cycloaddition, triene 94 was subjected96 to a pressure of 10 kbar at 30°C for 72 h.97 This resulted only in the quantitative recovery of starting material. Although this experiment did not disprove any pressure-dependence of the system, it strongly suggested the observed rate of increase for nmr tube cyclisations was not pressure related.98 The problem of scale-up was solved by performing the reaction in a Carius tube sealed under vacuum and heated in a purpose-built Carius oven at 220°C for 48 h. These conditions produced an 85% yield of cycloadducts in the same 1:1 ratio as observed previously, as well as a third, uncharacterised isomer (< 5% by nmr). Purification of the resultant oil by HPLC removed the minor cycloadduct but was unable to effect separation of the two major diastereoisomers. The oily mixture crystallised after 6 weeks at -18°C from ethanol-water, however as yet, fractional crystallisation of the solid has been unsuccessful. Assignment of the stereochemistry of the products has been made, based on several independent pieces of evidence: (i) the chemical shift of the methyl singlets assigned to the angular methyl groups suggested a 1:1 mixture of cis- and trans-fused isomers had been formed.99 (ii) Molecular mechanics100 and Karplus-equation101 derived coupling constants for two of the four possible IMDA cycloadducts and their hydrogenated relatives showed good correlation with experimental values. (iii) nOe Experiments on the hydrogenated products, whilst not conclusive, 127 strongly suggested the assigned structures were correct. (iv) more extensive nOe and COSY studies on pure samples of derivatives of the cycloadducts from the related enantiomerically pure triene confirmed the assigned structures (vide infra, §4.4.9) (v) consideration of the possible transition states (Scheme 65). This last piece of evidence shows that there are clear preferences exhibited for TS A, the desired rrans-fused isomer over that leading to the other /rani-fused isomer TS B. In the case of the two transition states leading to the two dr-fused isomers, there is a less clear- cut difference, although TS C appears to be less sterically encumbered than TS D. Together these pieces of information would appear to suggest that the /rani-fused isomer possesses the correct relative stereochemistry for a vitamin D 3 model and the other isomer from this cycloaddition is dr-fused with the relative stereochemistry depicted in Scheme 65. The internal methyl group would appear to influence the possible transition states, but without carrying out the IMDA reaction of the Z, E, £-triene for comparison, it is not possible to decide unequivocally what effect the phenylsulphonyl group has. 128 favoured disfavoured i m i i favoured PhSO, H disfavoured TS D Scheme 65 129 4.3.4 Derivatisation of IMDA cycloadducts The non-separability of the IMDA cycloadducts 118 and 119 by flash chromatography, HPLC or, latterly, fractional crystallisation was not entirely unexpected. It was thought however, that the different stereochemical environments of the cis- and trans-fused isomers would provide for some degree of chcmosclectivity in subsequent derivatisations and that such derivatives may prove separable by standard techniques. Hydrogenation72 of the indene double bond was initially carried out, providing an excellent yield of a 1:1 mixture of trans- and cts-fused isomers, 120 and 121 respectively (Scheme 65). 118 119 121 Reagents and codifions: (i) H2, 10% Pd-C, EtOAc, 24 h Scheme 65 Unfortunately, these indanes remained resistant to all chromatographic purification techniques attempted. It should be noted however, that one technique not tried, but known to have effected separation on similar systems, was argentation chromatography." Removal of the sp2 hybridised olefinic carbon atoms had a beneficial effect on several resonances in the JH nmr spectrum. In particular those assigned to H- \ cis (3.20 ppm) and H -l irons (2.96 ppm) were much more clearly resolved, allowing limited decoupling experiments to confirm some assignments. nOe Studies were also carried out, confirming the presence of a cis- and rranj-fused isomer, however the relative stereochemistry of the angular methyl and isopropyl groups could not be rigorously established. 130 Hydroboration was unsuccessful by comparison. Borane-dimethylsulphide proved too unselective, whereas9-BBN was found to be too hindered to effect reaction, even in THF at reflux. The epoxidation of a mixture of 118 and 119 with m-cpba was more useful. Two major epoxides resulted from column chromatography, each containing a ca. 1:1 mixture of isomers. The less polar mixture contained the products of epoxidation from one face for both 118 and 119, whereas the more polar mixture contained the products of epoxidation from the other face, as evidenced by lU nmr and mass spectrometry. No further separation of these mixtures was attempted but this experiment indicated epoxidation as the method of choice for characterising the IMDA cycloadducts from the enantiomerically pure triene (§ 4.4.8). 131 4 4.4 The Total Synthesis of a Vitamin D3 CD Ring Synthon. The results of the cyclisations of the model substrate, 94 showed that this demanding triene was reactive enough to cyclise to give in part, what we believed to be the desired rrans-fused product. The synthesis of an enantiomerically pure triene suitable for IMDA reaction to give the CD ring system of vitamin D3, including the cholestyl sidechain was carried out. The planned route is outlined in Scheme 66. After the required stereocentre had been introduced using an asymmetric Evans-enolate77*105 alkylation it was expected that chemistry developed for the model substrate would be applicable. 4 4 . I The optically pure natural product R-(+)-pulegone 122 is inexpensive (£0.15 g*1) and readily available in ca. 95% isomeric purity.102 Conversion to /?-(+)- dihydrocitronellic acid 123 which possesses the correct 3 R stereochemistry for the cholestyl sidechain was carried out according to a known procedure.103 Hydrochlorination of the unsaturated ketone proceeded smoothlyto give a mixture of diastereomeric chlorides 124 (Scheme 67). The fragmentation of this intermediate in aqueous potassium hydroxide gave /?-(+)-citroncllic acid 125 (61%, near-quantitative % ( yield based on recovered material) and a mixture of pulegone 122 and topulegone (38%) which were easily removed. R-(+)-citronellic acid may be purchased commercially, but its cost (£16.20 g*1) is prohibitive. Catalytic hydrogenation of the crude material proceeded in excellent yield to give fl-(+)-dihydrocitronellic acid, 123. a H 1. Oxidise 1. Evans alkylation Evans 2. /elimination Aldol TBDPSO XJU Scheme 66 1. Remove auxilliary auxilliary Remove 1. 2Reduce 3 3 steps O III!*' /?-(+)-pulegone 133 125 (üi) OH 123 Reagents and conditions: (i) HQ (g); (ii) KOH (aq); (iii) H 2, 10% Pd-C (cat.), MeOH. Scheme 67 Conversion of 123 to its acid chloride was accomplished using neat thionyl chloride. This was then coupled to the oxazolidinone auxilliary 126 derived from (15, )-norephedrine104 generating carboximide 127 (Scheme 68). O (0 86% (15,2/?)-norephedrine 126 oh ^71% A^Aa A \ / 123 127 / \ Reagents and conditions: (i) COCI2, PhMe, NaOH, H 2O; (ii) SOCI2; (iii) 123, n-BuLi, THF, -78°C, then add acid chloride. Scheme 68 134 4.4.2 Evans-enolate alkylation The Evans-enolate asymmetric alkylation105 is now a well established method in organic synthesis.77 It can be relied upon to provide reproducibly material of high enantiomeric excess in good yield. The anion derived from carboximide 127 suffers one disadvantage, viz. low chemical reactivity. It was hoped that by using the more reactive106 enolate formed with a sodium counterion, sufficient reactivity would be displayed to enable alkylation with electrophile 128 (Scheme 69). Unfortunately, very little desired material, 129 was isolated from this reaction indicating, as disclosed by Evans, that a more potent electrophile is required to achieve reaction with this enolate. Reaction of 127 with NaHMDS produced a stabilised anion which was allowed to react with allyl bromide at -50°C for 12 h giving 130 in excellent yield and diastereomeric excess. Reagents and conditions: (i) NaHMDS, THF, -78°C; (ii) 128, THF, -78°C to -50®C; (iü) CH2CHCH2Br, THF, -50°C, 12 h. Scheme 69 135 The more tightly coordinated and hence less reactive lithium enolate derived from the reaction of 127 with LDA gave a 34% yield of material in >95% d.e. as evidenced by 500 MHz ]H nmr. 4 .4 .3 This tactical change required the subsequent introduction of the primary hydroxyl functionality. Hydroboration with borane-dimethylsulphide complex and oxidative work-up gave uncharacterised sideproducts. 9-BBN Hydroboration proceeded smoothly on a 2.8 mmolar scale under carefully controlled conditions, giving alcohol 131. This alcohol could be lactonised to give 132 using catalytic potassium ferr-butoxide in THF with 20 equivalents of ferf-butanol (Scheme 71). These conditions were necessary to prevent epimerization at C-3 (Scheme 70). 132b Scheme 70 The lability of the centre a to the carbonyl was in agreement with the published observation107 that optically active 2-substituted S-valerolactones are very susceptible to 136 racémisation. (0 82% J — U -Â r ^ / ^ 130 (üi) (ii) 131 72% 94% 132 Reagents and conditions: 2.88 mmolar scale: (i) 9-BBN, THF; then NaOH, H2O2,0°C, 20 min; (ii) »-BuOK, i-BuOH, 20eq, THF, 0°C, 10-20 min; 46.5 mmolar scale: (iii) 9-BBN, THF; then NaOH, H 2O2,0°C, 40 min; then (ii). Scheme 71 Fortunately, the presence of a p-stereocentre produced a diastereomeric mixture if this épimérisation took place. This was immediately obvious from the appearance of new signals in the ÏH nmr spectrum, particularly that due to H-3 at 2.53 ppm. On a larger (46.5 mmolar) scale however, a slightly longer time was required for the complete conversion of the intermediate borate ester to alcohol 131 resulting in ca. 50% lactonisation. This mixture was then treated with catalytic base to give lactone 132 in excellent yield and of 95% d.e. as evidenced by !H nmr. 4 4.4 I twas envisaged that reduction of this lactone to lactol 133 would furnish material suitable for homologation by one of two approaches. Direct olefination of lactols using a Wittig or Wadswonh-Emmons process is known,108 although this often requires forcing conditions to achieve satisfactory conversion. Alternatively, selective protection of the open-chain aldehyde form of a related lactol has been carried out in good yield.109 137 (0 87-94% (ii) or (iii) or (iv) or(v) VY 134 X -I, Y= H; X ~ J ~ J 135 X = Br, Y ■ Br. HO/ Reagents and conditions: (i) DIBAL-H, PhMe, -78°C; H2O-THF (10% v/v), - 78°C; (ii) [ICH2PPh3]I, NaHMDS, THF, cool to -78°C then add 133; (iii) [ICH2PPh3lI. LiHMDS, THF, cool to -78°C then add 133; (iv) ICH2P(0 )(0 Et)2, n-BuLi or NaHMDS, -78°C to rt; (v) Zn dust (4 eq), PPh3 (4 eq), DCM. Scheme 72 The product aldehyde was then subsequently chain-extended using a Wittig rection. This sequence was never carried out on enantiomerically pure material, due to concern that the silylation conditions would cause extensive racemisation.109 Indeed, whichever method was used for the synthesis care would have to be exercised in its handling, since a- substituted aldehydes are known to be sensitive to epimerisation. Lactone 132 was successfully reduced to lactol 133 in excellent yield using DIBAL-H in toluene at low temperature.90 Attempted Wittig reaction with iodomethylenetriphenylphosphorane39 gave little or no identifiable product (Scheme 72). The only Wittig-like process to have produced any useful reaction was the modification due to Corey and Fuchs.91 Application of this method to lactol 133 gave tribromide 136 in moderate yield (Scheme 73). 138 Reagents and conditions: (i) CBu (6 eq),, PPh3 (3 eq), DCM. Scheme 73 In the second approach, attempted selective protection of the aldehydo-form of the lactol did not give any of the desired product. Instead, moderate yields of enol ether 137 were isolated (Scheme 74). Reagents and conditions: (i) TBSC1, Et3N, DMAP, DCM, 0°C. Scheme 74 4 .4 .5 It was clear that neither of these two routes would lead to the desired result, so a third, shorter route to alkynol 138 was proposed. If exocyclic enol ether 139 could be prepared, then it was envisaged that isomerisation to alkynol 138 might be possible via a base mediated ring-opening (Scheme 75). BASE'« 139 Scheme 75 139 The methylenation of lactone 132 was successfully carried out using the "Tebbe reagent",110 giving enol ether 139 (Scheme 76). 0) 86% JUJU. 132 A 1 139 "Tebbe reagent": Cp2Ti (ii) /''O 76% 139 140 Reagents and conditions: (i) "Tebbe reagent", 0.41 M in PhMe, PhMe-THF (3:2), pyridine (0.96 eq), -50°C to -20°C, 45 min; (ii) ¡ 2, Et3N, DCM, 0°C. Scheme 76 A number of attempts were made to effect the desired isomerisation. Reagents tried included potassium rerf-butoxide, n-butyllithium in the presence of boron trifluoride- etherate, and sodium amide. Other experiments in this series were invalidated due to the facility with which 139 rearranged to endocyclic enol ether 140. This product was also observed during the attempted iodination of exocyclic enol ether 139 (Scheme 76). 4 .4 .6 With the failure of the lactol and enol ether routes, there were two further options remaining to give access to alkynol 138. The first method involves reductive removal of the chiral auxilliary from 130 using standard procedures,105 followed by a lengthy sequence requiring differential protection. The second method would reduce the lactone to a diol and rely on the different environments of the two primary hydroxyl groups to effect a selective protection. 140 Despite literature precedent,105 in our hands the reductive removal of the oxazolidinone chiral auxilliary proved capricious. Use of a solution of lithium aluminium hydride in THF at low temperature led to moderate yields of material of variable diastereomeric excess (Scheme 77). Reduction of 130 using solid lithium aluminium hydride at 0°C in THF gave an improved yield of material, but the d.e. was now considerably worse: epimerization had obviously occurred during the reaction. 130 141 R ea g en tsand conditions: (i) UAIH4 (1.0M in THF), THF, -78°C (83% d.e); (ii) L1AIH4, THF, 0°C (60% d.e.). Scheme 77 The use of lithium borohydride,105 a milder reagent, resulted in incomplete reaction and was not investigated further. At this stage it was imperative to bring material forward for the synthesis of the required triene and it was decided to use the second method outlined above. The reduction of lactone 132 with lithium aluminium hydride in ether gave diol 142 in good yield (Scheme 78). Selective protection of 142 on a 25 mmolar scale gave monosilylated alcohol 143a in better than statistical yield. The other, undesired isomers 143b-c could be combined and recycled via TBAF deprotection. 141 Reagents and conditions: (i) L1AIH4, ether, 0°C; (ii) Et3N, DMAP, DCM, - 50°C; then add TBDPSC1,10 h; (iii) TBAF, THF, 12 h. Scheme 78 The next reaction, Swem oxidation30 of 143a to aldehyde 144 was performed with a minimum of triethylamine necessary to ensure high conversion (Scheme 79). This was to avoid epimerization of the sensitive a-stereocentre, of which none was evident by 500 MHz !H nmr. The next reaction was also potentially a source of epimerization, since unless reaction with the phosphorane was rapid at low temperature, the sensitive aldehyde would remain in basic medium for some time. Fortunately, reaction with iodomethylenetriphenylphosphorane39 produced Z-vinyl iodide 145 in stereospecific fashion without evidence of epimerization. The cis nature of the double bond was shown by the H-l/H-2 coupling constant (J = 7.4 Hz). None of the trans compound was visible by nmr. Elimination of vinyl iodide 145 using potassium rerr-butoxide and deprotection with TBAF27 gave alkynol 138 in good overall yield. 142 143a TBDPSO' Reagents and conditions: (i) (COCl)2. DMSO, DCM, -60°C, then add 143, then add Et 3N; (ii) [ICH2PPh3]I, NaHMDS, THF, n; then add 144, -78°C to rt; (iii) /-BuOK (2.2 eq), THF, -78°C to rt; (iv) TBAF, THF, rt Scheme 79 4 .4 .7 With this key intermediate 138 in hand, the rest of the synthesis of the desired triene closely paralleled the route developed for the model substrate. Zirconocene dichloride-mediated carbometallation87 with trimethylaluminium gave, after quench with iodine,93 an inseparable 5:1 mixture of vinyl iodide 147 and undesired terminal vinyl compound 148 (Scheme 80). The unfortunate introduction of ca. 15% of 148 arose from adventitious moisture in the iodine quench. This contaminant remained throughout the synthesis of the desired triene, thus all optical rotations of subsequnt compounds should be treated accordingly. Palladium(0)-catalysed cross-coupling with vinyl magnesium bromide94 gave E, E-diene 149 in excellent yield. Swem oxidation 30 smoothly converted this alcohol to aldehydes 150 and 151 in good yield (Scheme 80). 143 a a j (0 _ ~ • X — uk • 83% 138 1 3.5 : 1 Reagents and conditions: (i) ZrCp2Cl2, Mc3A1, C1(CH2)2C1, then add 138, 12 h; then add I 2, THF, -40°C; (ii) Pd(PPh3)4, PhMe, then add C^CHMgBr (l.OM THF); (iii) (C0 C1)2, DMSO, DCM, -60°C, then add 149, then add Et3N. Scheme 80 Reaction of this mixture with the lithio-anion of (phenylsulphonyl)methane gave 3- hydroxysulphones 152 admixed with a small amount of the vinyl compound. Mesylation and in situ elimination then furnished sulphonyl triene 153 together with vinyl sulphone 154 in a ratio of 3:1 (Scheme 81). Whilst this contaminant was unwelcome, it underwent an interesting side-reaction in the subsequent thermal IMDA process. This contamination would of course be avoidable by using scrupulously dry iodine in the carbometallation procedure; which could be achieved by drying the THF solution of iodine with activated molecular sieves. 144 (10% v/v), -78°C; (ii) MsCI, Et3N, DCM, -6°C. Scheme 81 4.4.8 IMDA Cyclisation of a Mixture of Enantiomerically Pure Triene and Diene. Thermolysis of a 3:1 mixture of (E, E, E)-sulphonyltriene 153 and E- sulphonyldiene 154 was carried out in ds-toluene in a sealed nmr tube at 180°C for 45 h. The crude mixture was shown by JH nmr analysis to contain a 1:1 ratio of cycloadducts 155 and 156 together with a third isomer 157 believed to be derived from intramolecular cne reaction of the diene 154 (Scheme 82). To our knowledge, this type of ene reaction is unknown for vinyl sulphones and would appear worthy of closer investigation. When the reaction was scaled-up in a Carius tube, the solution of a mixture of triene 153 and sulphonyldiene 154 was heated at 240°C for 48 h. Under these conditions, the product from the proposed ene reaction appeared to have undergone decomposition, since little starting 154 or product 157 was visible in the crude JH nmr of the reaction. The two [4+2] cycloaddition productsl55 and 156 were clearly visible 145 in ca. 1:1 ratio however. These products were isolated as an inseparable oil in 59% yield. Attepted HPLC separation of 155 and 156 proved fruitless, although it should be noted that the mixture was not subjected to argentation chromatography, which has proved useful on similar systems." Reagents and conditions: (i) 180°C, PhMe, 45 h, sealed nmr tube. Scheme 82 Scheme 83 146 The proposed transition states leading to the 1:1 mixture of isomers 155 and 156 are essentially the same as those depicted for the model substrate (§ 4.3.3). It is noteworthy how closely the synthesis and subsequent IMDA reaction of enantiomerically pure triene 153 paralleled that of the model substrate. 4 .4 .9 The final reaction performed on the IMDA cycloadducts was epoxidation with m-cpba (Scheme 84). Fortuitously, the resultant epoxidesl58 and 159 were separable by flash chromatography and sufficient material was isolated in a pure state for lH-1!! COSY and nOe studies to enable unambiguous assignment of their structures (Appendix 3). PhS02 155 156 Reagents and conditions: (i) m-cpba, DCM, rt; chromatography (SiC> 2 ). Scheme 84 147 4.5 Conclusions and Further Work This work has defined useful limits of reactivity for the IMDA reaction of sulphonyl-substituted trienes. The reaction of a sulphonyl-triene containing a 4- substituted 1,3-diene moiety is a demanding, but feasible reaction. It is clear that the concept of a dienophile-geometry controlled intramolecular Diels-Alder reaction has synthetic utility and a vitamin D3 CD-ring synthon has been synthesised via a novel IMDA disconnection. Work is currently in hand to develop the IMDA reaction of enantiomerically pure triene 153 into a formal total synthesis of vitamin D3. This could be achieved most rapidly by deoxygenation111 of epoxide 159 and hydrogenation of the resultant pure tetrahydroindene. This would lead to the same 8-sulphonyl indane fragment used by Nemoto et al. 74 In the synthesis of enantiomerically pure triene 153 the removal of the oxazolidinone chiral auxiliary (§ 4.4.6) without epimerization of the newly-formed stereocentre is a desirable goal and ought to be straightforward. This would allow access to quantities of material for further study and optimisation of the IMDA reaction leading to the trans-fused indene 157. Methodology suitable for the construction of enantiomerically pure (Z, £, £)- sulphonyl-substituted triene from aldehyde 150 is available (§ 2.2.3.6).56 Cyclisation of both this and the intermediate alkynylsulphone would certainly provide interesting results. An attractive route to vitamin D3 can be envisaged by cyclisation of a nitrile substituted triene, such as 160 (Scheme 85). Substitution by an additional electron withdrawing group would serve three purposes! (i) acceleration of the rate of reaction; (ii) improvement in the amount of desired trans-fused cycloadduct, since greater bond polarisation would enhance the asynchronous nature of the $ 148 Diels-Alder reaction; (iii) allow more direct entry to a hydroxysulphone sutable for reductive elimination (Scheme 85). Proposed transforations: (i) Heat; (ii) DIB AL-H, *78°C; (iii) couple with vinyl lithium A ring synthon; benzoylate; (iv) Na(Hg); (v) DeprotecL Scheme 85 One area of the cyclisation that has not been dealt with is the sulphone-aryl substituent. It is conceivable that replacement with other, more hindered groups such as rert-butyl would allow further optimisation of product selectivities. The intramolecular cne reaction of a vinylsulphone, described in § 4.4.8 is apparently a novel process and worthy of investigation, since these substrates are readily available from intermediates synthesised previously. 149 5.0 Experimental lH nmr spectra were recorded in CDC13 using either a Bruker AM-500, Bruker WM-250 or Jeol QX-270 nmr spectrometer. Infra-red spectra were recorded on a Perkin-Elmer 881 spectrophotometer. Mass spectra were obtained using VG-7070B, Jeol SX-102, VG 12-253 and VG ZAB-E instruments in the Imperial College Mass Spectrometry laboratory and the SERC Mass Spectrometry Service Centre, University College of Swansea. Elemental combustion analyses were performed in the Imperial College Chemistry Department microanalytical laboratory and the Rhone-Poulenc microanalytical laboratory, Dagenham. Melting points were measured on a Reichert hot stage apparatus. Optical rotations were measured using an Optical Activity AA-1000 polarimeter. Chromatography refers to column chromatography on Merck Kieselgel 60 (230-400 mesh) under pressure113 unless otherwise stated. HPLC was carried out on DYNAMAX 60A Si Columns. Tic refers to thin-layer chromatography performed on pre-coated Merck Kieselgel 60 F254 glass-backed plates and visualised using, where appropriate, UV irradiation, acidic ammonium molybdate(IV), acidic ethanolic vanillin, basic potassium permanganate and iodine vapour. Ether refers to diethyl ether and petrol to redistilled petroleum ether, bp 40-60°C. Where appropriate all solvents and reagents were purified before u se according to standard procedures.112 150 Preparation of 5-bromo-l-pentanoI (22a).26 Br 21a 22a 1,5-Pentanediol 6a (21 ml, 200 mmol) was dissolved in benzene (400 ml), aqueous hydrobromic acid (25 ml of a 48% solution, 0.75 eq) added and the stined mixture heated under reflux for 24 h whilst azeotropically removing the water with a Dean-Stark trap. When cool, the reaction mixture was poured into 6M aqueous sodium hydroxide (100 ml). The separated organic phase was washed with 2M aqueous hydrochloric acid (3 x 100 ml), dried (MgS04) and concentrated under reduced pressure to give 22a as a pale yellow oil (27.65 g) which was used crude in the next step. Preparation of 2-(5-bromopentyloxy)tetrahydro-2//-pyran (23a).26 22a 23a To a stirred solution of crude 5-bromo-l-pentanol 22a (27.65 g, 165.5 mmol) in dry dichloromethane (200 ml) under argon was added a trace of 10-camphorsulphonic acid (CSA). 3,4-Dihydro-2//-pyran (15.5 ml, 170 mmol, 1.03 eq) was then added dropwise via syringe to the solution at 0°C. After 30 min tic (40% ether-petrol) showed the reaction to be complete and the mixture was filtered through a plug of silica gel, rinsing thoroughly with more dichloromethane (2 x 300 ml). Removal of the solvent under reduced pressure gave 23a as a pale yellow oil (37.62 g) which was used crude in the next step. 151 Preparation of 2-[5-(phenyIsuIphonyl)pentyIoxy]tetrahydro-2//-pyran (24a). 4 1 T 41 Sodium phenylsulphinate (dried in vacuo at 120°C for 4 h; 30 g, 181 mmol) was added to a stirred solution of crude bromide 23a (37.62 g) in dry DMSO (170 ml) under argon at rt. An exothermic reaction ensued and after 30 min all the sulphinate salt had dissolved, to be replaced after a further 30 min by a white precipitate of sodium bromide. The reaction mixture was poured into water (250 ml) and extracted with ethyl acetate (3 x 120 ml). The combined organic layers were washed with water (3 x 100 ml), brine (3 x 100 ml), dried (MgS04) and concentrated to give a pale yellow oil. Purification by chromatography (30% - 60% ether-petrol) gave, in order of elution, 2-/5- (phenylsulphinyloxy)pentyloxy]tetrahydro-2U-pyran (2.10 g, 3% from 21a) as an oil; »max (film) 3594, 3060, 2944, 1445, 1353, 1323, 1201,1137, 1078, 1034 cm*i;8(270 MHz) 7.72-7.66 (2H, m, or/Ao-protons on Ph), 7.56-7.48 (3H, m, meta- and para- protons on Ph), 4.53 (1H, t, J 3Hz, H-2), 4.10-3.30 (6H, m, H-6, H -l’, H-5’), 1.90- 1.30 (12H, m, H-3, H-4, H-5, H-2’, H-3', H-4’); m/z (El) 312 (M+), 249, 218, 149, 115, 110, 101, 85, 69, followed by 2-[5-(phenylsuIphonyl)pentyloxy]tetrahydro-2U- pyran 24a (34.02 g, 52% from 21a) as an oil; vmax(film) 3066, 2941, 2868, 1586, 1120, 1147, 1305 cm'1; 8 (250 MHz) 7.94-7.85 (2H, m, ortho -protons oh Ph), 7.68- 7.51 (3H, m, meta- and para- protons on Ph), 4.55 (1H, t, J 3.4 Hz, H-2) 3.84-3.62 and 3.51-3.26 (4H, m, H-6, H-l'), 3.13-3.02 (2H, m, H-5'), 1.90-1.30 (12H, m, H-3, H- 4, H-5, H-2', H-3', H-4’); m/z (El) 312 (M+), 227 (M+ - C5H90), 211 (M+ - C5H90 2), 152 85 (C5H90 +) (Found: C, 61.84; H, 8.10. Ci6H240 4S requires C, 61.51; H, 7.74%). Preparation of 2-[(7E )-6-benzoyIoxy-5-(phenyIsulphonyl)-7- noneny!oxy]tetrahydro-2J/-pyran (25a). 4 24a 25a Sulphone 24a (32.04 g, 103.22 mmol) was dissolved in dry THF (300 ml) under argon and cooled to -78°C. The stirred solution was treated with n-butyllithium (44.2 ml of a 2.57M solution in hexanes, 1.1 eq) dropwise via syringe to give a lemon yellow solution of the anion. To this was added redistilled crotonaldehyde (9.4 ml, 113.54 mmol, 1.1 eq), causing the colour to fade slowly. After 30 min, redistilled benzoyl chloride (13.2 ml, 113.54 mmol, 1.1 eq) was added slowly via syringe and the reaction allowed to stir at -78°C for a further 2.5 h. The cooling bath was removed and after 3 h at rt the reaction was quenched with saturated aqueous sodium hydrogencarbonate (300 ml) and allowed to stir for 12 h. The organic phase was separated and the aqueous layer extracted with dichloromethane (3 x 100 ml). The combined organic layers were then washed with water (2 x 250 ml), dried (MgS04), and concentrated under reduced pressure to give the benzoyloxysulphones 25a (ca. 1:1 mixture of diastereomers by !H nmr; 51.83 g) as a pale yellow oil which was used crude in the following step. 153 Preparation of 2-[(5£, 7£)-5,7-nonadienyIoxy]tetrahydro-2//-pyran and 2- [(5Z, 7£)-5,7-nonadienyIoxy]tetrahydro-2/7-pyran (26a). OBz 25a 26a Crude benzoyloxysulphones 25a (51.83 g, ca. 103 mmol) were dissolved in dry THF (700 ml) under argon in a dry flask equipped with a large "rugby football"-shaped magnetic stirrer bar. Dry methanol (300 ml) and disodium hydrogenphosphate (60 g, 103 mmol, 1 eq) was added and the mixture was cooled to -20°C. 6% Sodium amalgam (160 g, 412 mmol Na, ca. two-fold excess) was added as two batches of finely-ground powder. Stirring became increasingly difficult until finally the mixture was allowed to stand overnight at -20°C. Water (700 ml) was added and the solution was decanted away from the mercury residues. The organic phase was separated and the aqueous layer extracted with petrol (3 x 250 ml). The combined organic layers were washed alternately with water (3 x 250 ml) and brine (3 x 250 ml), then dried (MgS04) and concentrated under reduced pressure to give a pale yellow oil (24 g). This was purified by chromatography (5% - 20% ether-petrol) to give the dienes 26a (6:1 ratio by JH nmr, 17.7 g, 77% from 24a) as a colourless oil; vmax (film) 3017, 2940, 2870,1035,987 cm- 1; 6 (270 MHz) (5E, 7£-isomer) 6.10-5.90 (2H, m, H-6', H-7'), 5.60-5.40 (2H, m, H- 5\ H-8'), 4.58 (1H, br. s, H-2), 3.95-3.66 and 3.55-3.30 (4H, m, H-6, H -l’) 2.09 (2H, q, J 6 Hz, H-4), 1.72 (3H, d, J 6 Hz, H-9), 1.70-1.35 (10H, m, H-3, H-4, H-5, H-2', H-3'); mil (El) 224 (M+), 140 (M+ - C5H80), 123 (M+ - C5H70 2) (Found: (M+), 224.1782. C14H240 2 requires (M+), 224.1778) 154 Preparation of (5 E, 7E)-5,7-nonadien-l-ol and (5Z, 7£)-5,7»nonadien*l-oI (27a). 9' 7' 5' r 8 * 6 ’ 26a 27a A solution of the THP ethers 27a (20.7 g, 92.2 mmol) in dry methanol (150 ml) containing a trace of CSA was stirred under argon at rt for 3 h, when tic (50% ether- petrol) showed the absence of starting material. The solution was concentrated under reduced pressure to one quarter of its original volume and then filtered through a pad of silica gel. The pad was washed exhaustively with ether (4 x 200 ml) and the combined filtrate and washings concentrated under reduced pressure to give an oil which was purified by chromatography (20% - 50% ether-petrol) to give the alcohols 27a (6:1 ratio by !H nmr; 12.2 g, 94%) as a colourless oil; umax (film) 3332 (br.), 3017, 2935, 1625, 1436, 1377, 1143, 1060, 986, 948, 926 cm-1; 6 (270 MHz) (5£, 7£-isomer) 6.05-5.93 (2H, m, H-6, H-7), 5.62-5.46 (2H, m, H-5, H-8), 3.60 (2H, t, J 6 Hz, H-l), 2.09 (2H, q, J 6 Hz, H-4), 1.88 (1H, br. s, OH), 1.72 (3H, d, J 6 Hz, H-9), 1.55-1.38 (4H, m, H-2, H-3); m/z (El) 140 (M+), 122, 107, 93, 81, 79, 68 (Found: C, 77.13; H, 11.62. C9H160 requires C, 77.09; H, 11.50%). 155 Preparation of (5E, 7E)-5,7-nonadienyl 3,5-dinitrobenzoate and (5Z, 7£)• 5,7-nonadienyl 3,5-dinitrobenzoate (28a). To a stirred solution of 3,5-dinitrobenzoyl chloride (8.73 g, 37.8 mmol, 1.1 eq) and DMAP (92 mg, 0.8 mmol, 0.02 eq) in dry dichloromethane (100 ml) at rt. under argon was added a solution of alcohols 27a (4.82 g, 34.4 mmol) in dichloromethane (50 ml). Triethylamine (5.3 ml, 37.8 mmol, 1.1 eq) was added and the solution stirred for 30 min at rt. The reaction was then poured into saturated aqueous sodium hydrogencarbonate (300 ml) and the aqueous phase extracted with dichloromethane (3 x 100 ml). The combined organic layers were washed with saturated aqueous sodium hydrogencarbonate (3 x 100 ml), water (3 x 100 ml), dried (MgS04) and concentrated under reduced pressure to give a dark orange oil. This was purified by chromatography (10% ether- petrol) to give a yellow solid which was crystallized to give the 3J-dinitrobenzoates 28a (12:1 ratio by !H nmr; 6.72 g, 58%) as yellow crystals, mp 42-43°C (ether-petrol); umax (film) 3104, 3018, 2928,1731,1630, 1599,1541, 1456,1346, 1277, 1167,1075, 991, 922, 825, 722 cm 1; 5 (500 MHz) (5E, 7£-isomer) 9.22 (1H, t, J 2 Hz, H-4), 9.14 (2H, d, J 2 Hz, H-2, H-6), 6.05-5.95 (2H, m, H-6', H-7'), 5.63-5.49 (2H, m, H-5’, H-8'), 4.45 (2H, t, J 6.5 Hz, H -l’), 2.15 (2H, q, J 7 Hz, H-4'), 1.84 (2H, m, H-2'), 1.72 (3H, d, J 6.5 Hz, H-9'), 1.58-1.51 (2H, m, H-3'); m/z (El) 334 (M+), 195 (C7H3N205), 149, 139 (M+ - C7H3N2O5), 121, 107, 94, 81, 75 (Found: (M+), 334.1165. Ci6H18N20 6 requires (M+), 334.1165). 156 Saponification of dinitrobenzoate ester (28a) 27a To a stirred solution of dinitrobenzoate ester 28a (6.72 g, 20.1 mmol) in THF (60 ml) was added 10% aqueous potassium hydroxide (120 ml, 201 mmol, ca. 10 eq). The colour of the mixture changed immediately from yellow to dark red and after 40 min tic (20% ether-petrol) showed the reaction to be complete. The solution was poured into water (200 ml), the organic phase separated and the aqueous layer extracted with ether (3 x 100 ml). The combined organic layers were washed with water (3 x 100 ml), brine (100 ml), dried (MgS04), and concentrated under reduced pressure. The product was purified by chromatography (15% - 20% ether-petrol) to give (5£, 7£>5,7-nonadien-l-ol 27a (contaminated with ca. 8% of the 5Z, 7E-isomer, 2.78 g, 98%) as a colourless oil. Preparation of (5E, 7E)-5,7-nonadienal and (5Z, 7E)-5,7*nonadienal (19a). 8 6 27a 19a To a stirred solution of oxalyl chloride (3.45 ml, 39.6 mmol, 2 eq) in dry dichloromethane (40 ml) under argon at -60°C was added a solution of DMSO (5.62 ml, 79.2 mmol, 4 eq) in dichloromethane (50 ml) dropwise via cannula. After 5 min a solution of alcohols 27a (2.78 g, 19.8 mmol) in dichloromethane (100 ml) was added and the reaction stirred for 15 min at -60°C. Triethylamine (13.80 ml, 99 mmol, 5 eq) 157 was added and the mixture allowed to warm to rt. The mixture was poured into water (200 ml) and extracted with ether (3 x 200 ml). The combined organic layers were then washed with saturated aqueous ammonium chloride (3 x 100 ml), water (3 x 100 ml), brine (2 x 100 ml), dried (MgS04) and concentrated under reduced pressure with ice cooling to give a pale yellow oil. This was purified by chromatography (2.5% - 10% ether-petrol) to give the aldehydes 19a (6:1 ratio by *H nmr; 2.72 g, 96%) as a colourless oil; umax (film) 3020,2930, 2855, 2720,1725,1450,990 cnr1; 6 (250 MHz) (5E, 7£-isomer) 9.76 (IH, t, J 2 Hz, H-l), 6.08-5.90 (2H, m, H-6, H-7), 5.80-5.33 (2H, m, H-5, H-8), 2.43 (2H, td, J 7, 2 Hz, H-2), 2.10 (2H, q, J 7 Hz, H-4), 1.73 (3H, d, J 6 Hz, H-9), 1.82-1.60 (2H, m, H-3); m/z (El) 138 (M+), 137 (M+ - H), 94 (M+ - C2H40), 81 (M+ - C3H5O), in agreement with data previously reported.11 Preparation of (6E, 8£)-l-(phenylsuIphonyl)-6,8*decadien*2-oI and (6Z, 8£)-l-(phenyl-suIphonyl)-6,8-decadien-2-ol (29a). 19a 29a To a stirred solution of (phenylsulphonyl)methane (dried in vacuo over P20 5; 729 mg, 4.66 mmol, 1.1 eq) in THF (20 ml) under argon at -78°C was added dropwise via syringe n-butyllithium (3.28 ml of a 2.57M solution in hexanes, 4.66 mmol, 1.1 eq) to give a colourless solution of the anion. After 10 min a solution of aldehydes 19a (freshly distilled, bpo.5 85°C; 586 mg, 4.24 mmol) in THF (5 ml) was added via cannula, rinsing with further THF (5 ml). After 15 min the reaction was quenched by the addition of a solution of acetic acid in THF (4.0 ml of a 1.75M solution, 1.3 eq). The mixture was allowed to warm to rt and poured into a 1:1 mixture of dichloromethane and saturated aqueous sodium hydrogencarbonate (100 ml). The layers were separated, and the aqueous layer was extracted with dichloromethane (2 x 50 ml) The combined organic 158 layers were washed with water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure to give a colourless oil. This was purified by chromatography (45% - 50% ether-petrol) to give the hydroxysulphones 29a (6:1 ratio by JH nmr; 1.12 g, 90%) as a viscous, colourless oil; omax (film) 3516 (br.), 3067, 3017, 2931, 1653, 1625, 1481, 1447, 1307, 1151, 1087, 1025, 991, 783, 746, 720 cm*»; 5 (250 MHz) (6E, 8 E- isomer) 7.98-7.88 (2H, m, ortho -protons on Ph), 7.65-7.52 (3H, m, meta- and para- protons on Ph), 6.02-5.86 (2H, m, H-7, H-8), 5.60-5.38 (2H, m, H-6, H-9), 4.20-4.10 (1H, m, H-2), 3.38 (1H, d, J 2.5 Hz, OH), 3.15-3.05 (2H, m, H-l), 2.00 (2H, q, J 7 Hz, H-5), 1.82 (3H, d, J 6.5 Hz, H-10), 1.60-1.30 (4H, m, H-3, H-4); m/z (El) 276 (M + - H20), 205, 199, 185, 156, 141, 135, 94, 77 (Found: C, 65.36; H, 7.79. Ci6H220 3S requires C, 65.27; H, 7.53%). Preparation of (IE, 6E, 8E)-l-(phenylsuIphonyI)-l,6,8*decatriene and (IE, 6Z, 8£>l-(pheny]sulphony!)-l,6,8-decatnene (17a). 7 5 To a stirred solution of hydroxysulphones 29a (1.26 g, 4.085 mmol) in dry dichloromethane (30 ml) under argon at -6°C was added, dropwise via syringe triethylamine (5.56 ml, 40.85 mmol, 10 eq) followed immediately by methanesulphonyl chloride (0.96 ml, 12.26 mmol, 3 eq). A light yellow precipitate developed and the reaction was allowed to warm to rt. After 90 min the mixture was poured into saturated aqueous ammonium chloride (100 ml) and extracted with dichloromethane (3 x 50 ml). The combined organic layers were washed with saturated aqueous ammonium chloride (2 x 100 ml), water (100 ml), dried (MgS04) and concentrated under reduced pressure. The crude product was purified by chromatography (20% ether-petrol) to give the trienes 17a 159 (6:1 ratio by *H nmr, 1.08 g, 91%) as a viscous, colourless oil; umax (film) 3020, 2925, 2855, 1625, 1445, 1305, 1150, 990 cnr1; S (500 MHz) (IE, 6E, 8E-isomer) 7.90 (2H, m, ortho -protons on Ph), 7.61 (1H, m, para -proton on Ph), 7.54 (2H, m, meta-protons on Ph), 6.99 (1H, dt, J 14.5, 6.5 Hz, H-2), 6.21 (1H, d, J 14.5 Hz, H-l), 6.03-5.81 (2H, m, H-7, H-8), 5.61-5.53 and 5.50-5.41 (2H, m, H-6, H-9), 2.24 (2H, q, J 6.5 Hz, H-3), 2.06 (2H, q, J 6.5 Hz, H-5), 1.72 (3H, d, J 6.5 Hz, H-10), 1.59-1.52 (2H, m, H-4); m/z (El) 276 (M+), 195 (M+ - C6H9), 135 (M+ - S02Ph), 95 (M+ - C3H4S02Ph) (Found: C, 69.60; H, 7.02. C16H20O2S requires C, 69.53; H, 7.29%). Preparation of 6-bromo-l-hexanol (22b).11 2lb 22b This was prepared analogously to 5-bromo-l-pentanol 21a on a 422 mmol scale to give a nearly colourless oil (97%) which was used crude in the following step. Preparation of 2-(6-bromohexyloxy)tetrahydro-2//-pyran (23b). 22b 23b This was prepared analogously to the lower homologue 23a on a 410 mmol scale to give a colourless oil (81 %) which was used crude in the following step. Preparation of 2-[6-(phenylsu!phinyIoxy)-hexyIoxy]tetrahydro-2J/-pyran and 2-[6-(phenylsu!phonyl)hexyloxy]tetrahydro-2Jy-pyran (24b). 160 THPO' 23b 4 24b R = S02Ph R - 0S(0)Ph Sodium phenylsulphinate (dried in vacuo at 120°C for 4 h; 73 g, 443 mmol, 1.1 eq) was dissolved in dry DMSO (400 ml) under argon with gentle heating. To the stirred solution was added a solution of crude bromide 23b (107 g, ca. 403 mmol) in DMSO (100 ml) via cannula, rinsing the flask with further DMSO (2 x 50 ml). After 2 h, tic (75% ether- petrol) showed no starting material to be present. The reaction mixture was poured into water (1000 ml) and extracted with ethyl acetate (3 x 300 ml). The combined organic layers were washed with water (3 x 300 ml), brine (3 x 250 ml), dried (MgS04) and concentrated to give a nearly colourless oil. Purification by column chromatography (30% - 60% ether-petrol) gave, in order of elution, 2 -[6-(phenylsulphinyloxy)- hexyloxyJtetrahydro-2U-pyran (6.9 g, 5% from 21b) as a colourless oil; umax (film) 2940, 1724, 1445, 1323, 1262, 1201, 1136, 1034, 975, 904, 756, 698 cnr1; 6 (270 MHz) 7.72-7.62 (2H, m, ortho- protons on Ph), 7.54-7.45 (3H, m, meta- and para- protons on Ph), 4.52 (IH, t, J 2.5 Hz, H-2), 4.05-3.28 ( 6H, m, H-6, H-l', H-6'), 1.85-1.25 (14H, m, H-3, H-4, H-5, H-2', H-3’, H-4', H-5'); mlz (El) 326 (M+), 307, 271, 243, 225, 209, 195, 185, 181, 170, 156, 143, 125, 101, 85 (Found: (M+ - C5H 9 O), 241.0904. C 17H 2404S requires (M+ - C 5H9 O), 241.0898), and 2-[6- (phenylsulphonyl)hexyloxyJtetrahydro-2H-pyran 24b (84.2 g, 61% from 21b) as a colourless oil; vmax (film) 2943, 2867, 1586, 1448, 1353, 1307, 1201, 1147, 1087, 1034 cm-1; 8 (250 MHz) 7.85 (2H, m, ortho- protons on Ph), 7.65-7.45 (3H, m, meta- and para- protons on Ph), 4.55 (IH, t, J 3 Hz, H-2), 3.85-3.20 (4H, m, H- 6, H-T), 3.05 (2H, m, H-6'), 1.80-1.20 (14H, m, H-3, H-4, H-5, H-2', H-3’, H-4’, H-5'); mlz (El) 326 (M+), 325 (M+ - H), 308 (M+ - H 20), 241 (M+ - C5H90), 225, 209, 197, 169, 155 (PhS02CH 2+), 143, 101, 85 (C 5H 9 0 +), 77 (Found: C, 62.26; H, 8.30. C i 7H 260 4S requires C, 62.54; H, 8.03%) (Found: (M + NH4+), 344.1896. 161 C17H2604S requires (M + NH4+), 344.1896). Preparation of 2-[(8£)-7-benzoyIoxy-6-(phenylsulphonyI)-8-decenyIoxy]- tetrahydro-2J/-pyran (25b). To a stirred solution of THP ether 24b (41.88 g, 128 mmol) in dry THF (220 ml) under argon at -78°C, was added /i-butyllithium (99.1 ml of a 1.42M solution in hexanes, 141 mmol, 1.1 eq) via cannula. After 10 min, freshly distilled crotonaldehyde (11.7 ml, 141 mmol, 1.1 eq) was added to the yellow solution. After a further 30 min benzoyl chloride (16.4 ml, 141 mmol, 1.1 eq) was added. The reaction mixture was stirred at -78°C for a further 3 h and allowed to warm to rt. Saturated aqueous sodium hydrogencarbonate (200 ml) was added and the mixture was stirred overnight. The layers were separated and the aqueous phase extracted with dichloromethane (3 x 75 ml). The combined organic layers were washed with water (3 x 100 ml), dried (MgS04) and concentrated under reduced pressure to give the benzoyloxysulphones 25b ( ca. 1:1 mixture of diastereomers by nmr; 75.9 g) as a pale yellow oil which was used crude in the following step. Preparation of 2-[(6£, 8£)-6,8-decadienyloxy]tetrahydro-2tf-pyran and 2- [(6Z, 8£)-6,8-decadienyloxy]tetrahydro-2//.pyran (26b). Crude benzoyloxysulphones 25b (72.9 g, ca. 146 mmol) were dissolved in THF (1000 162 ml) and methanol (400 ml) in a flask together with disodium hydrogenphosphate (60 g) and the mixture cooled to -20°C with overhead stirring. Freshly ground 6% sodium amalgam (224 g, 582 mmol of sodium, ca. two-fold excess) was added in two batches an hour apart. The thick suspension was stirred at -20°C for 12 h and then carefully decanted away from the mercury residues into water (1000 ml). The aqueous phase was extracted with petrol (3 x 500 ml). The combined extracts were washed with water (3 x 500 ml), brine (2 x 250 ml) and dried (MgS04). Concentration under reduced pressure and purification by chromatography (5% -10% ether-petrol) gave the dienes 26b (6:1 ratio by !H nmr, 24.3 g, 79%) as a colourless oil; vmtx (film) 3016, 2934, 2856, 1452, 1441, 1353, 1201, 1137, 1120, 1078, 1036, 986, 906, 870, 816 cm-*; 8 (250 MHz) (E, E-isomer) 6.10-5.90 (2H, m, H-7', H- 8 ’), 5.70-5.45 (2H, m, H-6’, H-9’), 4.57 (1H, br. s, H-2), 3.95-3.65 and 3.55-3.32 (4H, m, H- 6, H -l’), 2.08 (2H, q, J 5.5 Hz, H- 5’), 1.73 (3H, d, J 6.5 Hz, CH3), 1.90-1.20 (12H, m, H-3, H-4, H-5, H-2’, H-3’, H- 4’); mtz (El) 238 (M+), 220 (M+ - H 20), 154 (M+ - C5H8 0), 136 (M+ - C5HgO - H 20), 85 (C 5H9 0 +) (Found: C, 75.33; H, 11.21. C 15H260 2 requires C, 75.58; H, 11.00%). Preparation of (6E, 8E)-6,8-decadien-l-ol and (6Z, 8£)-6,8-decadien-l-oI (27b). 26b 27b To a stirred solution of THP ethers 26b (20.62 g, 86.5 mmol) in dry methanol (155 ml) was added a trace of CSA. After 3 h tic (50% ether-petrol) indicated the reaction to be complete and the solution was concentrated to one third of its original volume. The solution was filtered through a pad of silica gel which was rinsed exhaustively with ether. The solvent was removed under reduced pressure and the product purified by chromatography (20% - 25% ether-petrol) to give the alcohols 27b (12.2 g, 91%) as a 163 colourless oil which solidified on refrigeration at -5°C. The product gave spectral data identical with those reported previously .31 W Preparation of (6 E, 8E)-(6,8-decadien-l-yl) 3,5-dinitrobenzoate and (6Z, 8E)-(6,8-decadien-l-yl) 3,5-dinitrobenzoate (28b). This was carried out on a ca. 70 mmol scale in exactly analogous fashion to the lower homologues28a to give the 3^-dinitrobenzoates 28b (20:1 ratio by JH nmr; 24.26 g, 52%) as yellow crystals, mp 57-59°C (ether-petrol); umax (film) 2926,1728,1627,1462, 1377, 1346, 1290, 1170 cm*1; 8 (250 MHz) (£, E-isomer) 9.24 (1H, t, J 3 Hz, H-4), 9.15 (2H, J 3 Hz, H-2, H- 6), 6.08-5.80 (2H, m, H-7, H- 8 ’), 5.60-5.45 (2H, m, H-6', H-9 '), 4.45 (2H, t, J 7 Hz, H-l'), 2.15-2.05 (2H, m, H-5’), 1.90-1.80 (2H, m, H-2'), 1.72 (3H, t, J 7 Hz, CH3), 1.50-1.40 (4H, m, H-3', H-4'); m/z (El) 348 (M+), 195 (C 7H 3N 20 5+), 149, 135, 121, 107, 93, 81, 75, 68 (Found: (M+), 348.1321. C17H20N2O6 requires (M+), 348.1321). 164 Saponification of dinitrobenzoate ester (28b) 27b To a solution of dinitrobenzoate ester 28b (3.0 g, 8.61 mmol) in THF (75 ml) was added 10% aqueous potassium hydroxide (100 ml, 178 mmol, ca. 20 cq). The colour of the mixture changed immediately from yellow to dark red and after 40 min tic (20% ether- petrol) showed the reaction to be complete. The solution was poured into water (100 ml), the organic phase separated and the aqueous layer extracted with ether (3 x 50 ml). The combined organic layers were washed with water (3 x 50 ml), brine (50 ml), dried (MgS04), and concentrated under reduced pressure. The product was purified by chromatography (15% - 20% ether-petrol) to give ( 6E, 8 £)-6,8 -decadien-l-ol 27b (contaminated with ca. 5% of the 6Z, 8 £-isomer, 1.266 g, 95%) as a colourless oil. Preparation of ( 6E, 8£)-6,8-decadienal (19b). 27b I ’ b Swem oxidation was performed on an 8.2 mmol scale analogously to the lower homologue 19a to give the product (contaminated with ca. 5% of the 6Z, 8 £-isomer; 1 04 g, 84%) as a colourless oil, bp 0.2 80°C. Spectral data were identical with those reported previously.31^ Preparation of (7£, 9£)-l-(phenylsuIphonyl)-7,9-undecadien-2-ol (29b). 165 The procedure was carried out on a 3.8 mmol scale analogously to the lower homologue 29a to give the product (contaminated with ca. 5% of the 6Z, 8 E-isomer; 1.40 g) as a viscous, colourless oil which was used crude in the next step. Preparation of (IE, 7E, 9EM .(phenylsulphony1).l,7,9-undecatnene (17b). 8 Crude (7E, 9E)-l-(phenylsulphonyl)-7,9-undecadien-2-ol 29b (1.40 g) was treated analogously to the lower homologue 29a to give the product 17b (contaminated with ca. 5% of the IE, 6Z, 8 E-isomer; 762 mg, 69% from ( 6E, 8 E)-6,8 -decadienal) as a viscous, colourless oil; omax (film) 3015, 2928, 2857, 1655, 1625, 1572, 1540, 1447, 1319, 1307, 1148, 1087, 989, 819, 752, 716, 688 cm’1; 6 (270 MHz) 7.92-7.84 (2H, m, ortho -protons on Ph), 7.65-7.50 (3H, m, meta- and para-protons on Ph), 6.98 (1H, dt, J 16, 7 Hz, H-2), 6.31 (1H, dt, J 16. 1.5 Hz, H-l), 6.08-5.90 (2H, m, H- 8 , H-9), 5.68- 5.42 (2H, m, H-7, H-10), 2.23 (2H, m, H-3), 2.04 (2H, q, J 7 Hz, H- 6), 1.72 (3H, d, J 7 Hz, CH3), 1.55-1.32 (4H, m, 4-H, 5-H); mlz (El) 290 (M+), 195 (M+ - C 7Hn ), 81 (C6H9), 77 (Ph), 67 (C 5H7+) (Found: C, 70.43; H, 7.84. C 17H220 2S requires C, 70.31; H, 7.64%). 166 Preparation of bis(2,2,2-tnfluoroethyl) (phenylsulphonyl)methane phosphonate (30). PhS02'^*S'P(0Et)2 ______PhS02x<<*S' P(OCH2CF3)2 o o 30 A mixture of phosphorous pentachloride (1.18 g, 5.68 mmol, 3 eq) and dimethyl (phenylsulphonyl)methane phosphonate (500 mg, 1.89 mmol, 1 eq) under argon was heated at 70°C for 60 h. Excess phosphorous pentachloride and POCI 3 were removed by heating at 50°C/0.1 mm Hg for 1 h. The resulting oil was dissolved in dry benzene (1.5 ml) and treated with a solution of CF 3CH2OH {ex calcium sulphate/sodium hydrogencarbonate; 290 pi, 3.97 mmol, 2.1 eq) and diisopropylethylamine (757 pi, 4.35 mmol, 2.3 eq) in dry benzene (1 ml) under argon at 0°C. The mixture was stirred at 0°C for 1 h, after which time further CF 3CH2OH (138 pi, 1.89 mmol, 1 eq) and diisopropylethylamine (395 pi, 2.27 mmol, 1.2 eq) were added. After a further 30 min the reaction was quenched by the addition of saturated aqueous sodium hydrogencarbonate solution and the mixture extracted with ether (3 x 10 ml). The combined extracts were washed with brine, dried (MgSC^) and concentrated under reduced pressure to give a yellow oil. This was purified by chromatography (30% ethyl acetate-petrol) to give bis( 2,2,2-trifluoroethyl) (phenylsulphonyl)methanephos-phonate (330 mg, 44%) 30 as a white solid, mp 51-52.5°C (ether-petrol); umax (nujol) 2950, 2366, 1437, 1378, 1302, 1175, 1064, 962, 894, 854, 800, 735, 683 cm'»; 8 (250 MHz), 8.05-7.96 (2H, m, ortho- protons on Ph), 7.78-7.58 (3H, m, meta- and para- protons on Ph), 4.49 (4H, m, OC// 2CF3), 3.95 (2H, d, J 17.5 Hz, PhSO 2C //2); m/z (El) 400 (M+), 381, 336, 316, 296 (Found: C, 32.87; H, 2.61. Cn H n F 60 5PS requires C, 33.01; H, 2.77%). Preparation of bis(iV,Ar*dimethylamino) (phenylsulphonyl)methane phosphoramidate (33). ch3v P(NMe2)2 PhS02 II o O 34 33 To a stirred solution of bis(N//-dimethylamino) methanephosphoramidate 34 (248 mg, 1.65 mmol) in THF (5 ml) at -78°C under argon was added n-butyllithium (730 pi of a 2.50M solution in hexanes, 1.82 mmol, 1.1 eq). After 5 min, a solution of phenylsulphonyl fluoride (132 mg, 0.826 mmol, 0.5 eq) in THF (1 ml + 1 ml rinse) was added via cannula. The reaction was allowed to warm to rt whereupon tic (5% methanol- ethyl acetate) showed appearance of a single product, Rf 0.2 which was visualized using UV, I2 and KMnO^ The mixture was poured into saturated aqueous ammonium chloride (20 ml) and extracted with DCM (3 x 20 ml). The combined extracts were washed with water (3 x 20 ml), dried (MgSO^ and concentrated under reduced pressure to give a near-colourless oil. This was purified by chromatography (ethyl acetate - 50% methanol- ethyl acetate) to give bis(N.N-dimethylamino) (phenylsulphonyl)methane phosphonate 33 (222.5 mg, 93% based on phenylsulphonyl fluoride) as a colourless oil which crystallised slowly at -18°C to give a white solid mp 63-65.5°C (ethyl acetate); umax (nujol) 2923, 1465, 1378, 1311, 1207, 1153, 974, 785, 750, 686 cm-»; 5 (270 MHz) 7.88 (2H, dd, J 7, 2 Hz, orr/to-protons on Ph), 7.68-7.50 (3H, m, meta- and para- protons on Ph), 3.84 (2H, d, J 14 Hz, CH2), 2.67 (12H, d, J 10.2 Hz, N(C//3)2); m/z (El) 290 (M+), 247 (MH+ - NMe2), 246,226,203 (MH+ - (N(Me)2)2), 183, 168, 135, 92, 44 (N(Me) 2) (Found: C, 25.41; H, 6.54; N, 9.60. C 11H 19 N2O3PS requires C, 45.51; H, 6.60; N, 9.65%) (Found: (M+) 290.0854. C 11H19 N2O3PS requires (M+), 290.0854). 168 Preparation of tetramethyl [(2£)-l(phenylsulphonyl).2-octenyl]. phosphoramidate (35). To a stirred solution of sulphone 33 (37.5 mg, 0.129 mmol) in THF (0.7 ml) under argon at -78°C was added n-butyllithium (57 pi of a 2.50M solution in hexanes, 0.142 mmol, 1.1 eq). followed after 5 min by a solution of heptanal (redistilled; 20 pi, 0.142 mmol) in THF (0.5 ml + 0.3 ml rinse). The reaction was stirred at -78°C for 10 min then allowed to warm to rt during 20 min. The reaction was quenched after a further 15 min by the addition of saturated aqueous ammonium chloride (20 ml). The aqueous phase was extracted with ethyl acetate (3x10 ml) and the combined organics were washed with water (3 x 10ml), brine (10 ml), dried (MgSC> 4)and concentrated under reduced pressure to give a colourless oil. The product was purified by chromatography (ethyl acetate - 40% methanol-ethyl acetate) furnishing tetramethyl [(2E)-l(phenylsulphonyl)-2 - octenyljphosphoramidate 35 (6 mg, 11%) as a colourless oil; \>max (film) 2933, 1540, 1450, 1310, 1149, 1085, 990, 750, 688 cm**; 8 (250 MHz) 7.85 (2H, dd, J 8 , 2 Hz, ortho- protons on Ph), 7.65-7.45 (3H, m, meta- and para-protons on Ph), 5.49 (1H, ddd, J 17, 10.9, 6.4 Hz, H-3), 5.23 (1H, ddd, J 9.0, 7.1, 3.0 Hz, H-2), 4.28 (1H, dd, J 13.3, 9.7 Hz, H-l), 2.75 ( 6H, d, J 8.8 Hz (C//3)2), 2.70 (6H, d, J 8.8 Hz (C//3)2); m/z (El) 386 (M+), 342 (M+ - NMe2), 301, 252, 245 (M+ - PhS02), 200 (M+ - NMc 2 - PhS0 2), 154, 135, 92, 44 (N(Me)2) (Found: (M+) 386.1802. C18H31N20 3PS requires (M+), 386.1793). 169 Preparation of (lE, 7E, 9E)-1-(phenylsulphinyl)-1,7,9-undecatriene and (lZ, 7E, 9E)-1-(phenylsulphinyl)-1,7,9-undecatriene (39). 8 + 11 11 19b PhS(O) 39 Freshly distilled dimethyl methanephosphonate (1.10 g, 8.86 mmol, 2.1 eq) was dissolved in dry TI-IF (80 ml) under argon. The stirred solution was cooled to -78°C and treated with n-butyllithium (3.4 ml of a 2.57M solution in hexanes, 8.65 mmol, 2.05 eq) dropwise via syringe. To the colourless solution was added a solution of isopropyl phenylsulphinate (azeotropically dried with toluene (2 x 5 ml); 816 mg, 4.43 mmol, 1.05 eq) in dry TI-IF (10 ml) via cannula. After 5 min a solution of (6£, 8£)-6,8-decadienal 19b (freshly distilled; 642 mg, 4.22 mmol) in TI-IF (10 ml) was added via cannula and the reaction was then allowed to warm to 11 whereupon it was quenched by the addition of saturated aqueous ammonium chloride (75 ml). The mixture was extracted with dichloromethane (3 x 50 ml). The combined extracts were washed with water (3 x 50 ml), dried (MgS04), and concentrated under reduced pressure. The crude product was purified by chromatography (40% - 60% ether-petrol) to give, in order of elution, the £ and Z-vinylic sulphoxides (1:2 ratio by I H nmr; 1.005 g combined yield, 87%) as a colourless oil; (lE, 7E, 9E)-1-{phenylsulphinyl)-1,7,9-undecatriene: Umax (film) 3013, 2926,2855, 1617, 1580, 1442,1084, 1041,989,745,690 cm-I; 0 (250 MHz) 7.65- 7.40 (5H, m, Ph), 6.70 (1H, dt, J 15, 6 Hz, H-2), 6.22 (1H, dt, J 15, 1.5 Hz, H-l), 6.00-5.90 (2H, m, H-8, H-9), 5.65-5.40 (2H, m, H-7, H-IO), 2.30-2.15 (2H, m, H-3), 2.10-2.00 (2H, m, H-6), 1.85 (3H, d, J 7 Hz, CH3), 1.50-1.30 (4H, m, H-4, H-5); m/z (El) 274 (M+), 257, 149, 123,81,79,77 (Found: C, 74.29; H, 8.36. C17H 220S requires C, 74.40; H, 8.08%); (lZ, 7E, 9E)-1-(phenylsulphinyl)-1,7,9-undecatriene 39; umax (film) 2926, 1620, 1443, 1080, 1045,989,690 cm-I; 0 (250 MHz) 7.65-7.30 (5H, 170 m, Ph), 6.30 (4H, m, H-l, H-2, H-8 , H-9), 5.80-5.45 (2H, m, H-7, H-10), 2.70-2.50 (2H, m, H-3), 2.20-2.05 (2H, m, H-6), 1.75 (3H, d, J 7 Hz, CH3), 1.55-1.40 (4H, m, H-4, H-5); m!z (El) 274 (M+), 257, 177, 149, 123, 105, 81, 79, 77 (Found: C, 74.64; H, 8.37. C 17H22o s requires C, 74.40; H, 8.08%). Reaction of Z-sulphoxide (39) with neat S 0 2. 4 Z-Sulphoxide 39 (187 mg, 0.68 mmol) and hydroquinone (2 mg, 0.014 mmol, 2 mol%) were placed inside a teflon-lined steel autoclave together with a magnetic stirrer bar. Liquid sulphur dioxide (4 ml) was introduced via cannula and the system was sealed and heated to 70°C for 12 h. The resultant brown oil was extracted from the autoclave with dichloromethane, the solvent and excess sulphur dioxide removed under reduced pressure and the product purified by chromatography (40% -100% ether-petrol, then 75% -100% ethyl acetate-petrol) to give, in order of elution, recovered 39 (67 mg, 36%), and [2R*JS*]-2J-dihydro-2-methyl-5-[(5Z)-6-(phenylsulphinyl)-5-hexenyl]thiophene S,S- dioxide 40 (117 mg, 51%, 80% based on recovered 39) as a colourless oil; vmax (film) 2930, 1654, 1618, 1445, 1304, 1131, 1084, 1039 cm-»; 8 (270 MHz) 7.65-7.45 (5H, m, Ph), 6.25-6.15 (2H, m, H-5', H-6'), 5.94 (2H,m, H-3, H-4), 3.80 (1H, q, J 7 Hz, H-2), 3.68 (1H, m, H-5), 2.80-2.65 (1H, m, H-4’), 2.60-2.45 (1H, m, H-4’), 1.80- 1.50 (6H, m, H -l’, H-2’, H-3’), 1.40 (3H, d, J 7 Hz, CH3); m/z (El) 274 (M+ - S02), 149 , 123, 81, 79,77 (Found: (M+ - S02), 274.1396. Ci 7H2203S2 requires (M+ - S02), 274.1391. Found: C, 56.64; H, 5.99. Ci 7H220 3S2 requires C, 56.44; H, 5.92%). 171 Oxidation of Z-sulphoxide (40). 4 Z-Sulphoxide 40 (117 mg, 0.346 mmol) was dissolved in dichloromcthane (10 ml), anhydrous sodium acetate (31 mg, 0.381 mmol, 1.1 eq) added and the mixture cooled to 0°C. Peracetic acid (160 pi of a 32 wt.% solution in dilute acetic acid, 0.518 mmol, 1.5 eq) was added via syringe and the reaction was allowed to stir for 12 h, after which time tic (75% ethyl acetate-petrol) showed complete reaction. The mixture was poured into water (30 ml) and extracted with dichloromethane (3 x 20 ml). The combined organic layers were washed with 10% aqueous sodium thiosulphate (3 x 20 ml), saturated aqueous sodium hydrogencarbonate (3 x 20 ml), dried (MgS04), and concentrated under reduced pressure. The crude product was purified by chromatography (50% - 80% ethyl acetate-petrol) to give [2R*,5S*]-2J-dihydro-2-methyl-5-[(5Z)-6-(phenylsulphonyl)-5 - hexenyl]thiophene S,S-dioxide 41 (104 mg, 85%) as a colourless oil; vmax (film) 3064, 2929, 2865, 1625, 1448, 1305,1147 enr1; 6 (250 MHz) 7.90 (2H, m, orr/zo-protons on Ph), 7.70-7.50 (3H, m, meta- and para-protons on Ph), 6.35-6.20 (2H, m, H-5', H-6’), 5.92 (2H, br. s, H-3, H-4), 3.85-3.60 (2H, m, H-2, H-5), 2.75-2.60 (2H, m, H-4’), 1.90-1.50 (6H, m, H -l\ H-2\ H-3’), 1.40 (3H, d, J 7 Hz, CH3); m/z (El) 354 (M+), 336, 322, 293, 165, 149, 64, 79, 71, 64 (S02) (Found: C, 57.68; H, 6.41. C 17H 22O 4S2 requires C, 57.60; H 6.26%) (Found: (M + NH4+), 372.1303. C17H2204S2 requires (M + NH4+), 372.1303). 172 Cycloelimination of S0 2 from Z-sulphone (41). 8 Z-Sulphone 41 (61 mg, 0.172 mmol) was dissolved in toluene (10 ml) and the solution heated to 80°C for 1.5 h after which time tic (75% ethyl acetate-petrol) indicated complete reaction. The solution was concentrated under reduced pressure to one tenth of its original volume and the residue chromatographed (15% ether-petrol) to give (1Z, 7E, 9 E ) - 1 -(phenylsulphonyl)-l,7,9-undecatriene 18b (42 mg, 84%) as a colourless oil, umlx (film) 3015, 2928, .2859, 1625, 1587, 1555, 1447, 1306, 1148, 1086, 990, 779, 754 cm-i; g (250 MHz) 7.92 (2H, dd, J 8 ,1.5 Hz, ortho-proions on Ph), 7.66-7.50 (3H, m, meta- and para-protons on Ph), 6.33-6.18 (2H, m, H-l, H-2), 6.10-5.90 (2H, m, H- 8 , 5.66-5.42 (2H, m, H-7, H-10), 2.68 (2H, m, H-3), 2.04 (2H, m, H- 6), 1.73 (3H, d, J 6.5 Hz, CH3), 1.44-1.35 (4H, m, H-4, H-5); m/z (El) 290 (M+), 223 (M+ - C5H7), 209 (M+ - C 6H9), 195 (M+ - C 7Hn), 182, 165, 149 (M+ - S0 2Ph), 81 (Found: (M+), 290.1341. C 17H220 2S requires (M+), 290.1341). Reaction of THP ethers (26a) with S0 2-methanol. ‘OTHP ■£0 26a 42 A solution of THP ethers 26a (456 mg, 2.13 mmol) in methanol (5 ml) was heated to 84°C in a steel autoclave containing hydroquinone (5 mg, 0.041 mmol, 2 mol %) and liquid sulphur dioxide (1 ml, 20 mmol, 9.4 eq) for 5 h. When cool the dark oil was 173 dissolved in dichloromethane and the solvent and excess sulphur dioxide evaporated under reduced pressure. The residue was purified by chromatography (20% - 100% ethyl acetate-petrol) to give, in order of elution, E, E- and Z, E-alcohols 27a (3:1 ratio by lH nmr, 118 mg, 42%), identical with material prepared previously, and [2^*, 5'S*]-4- (2,5-dihydro- 5 -methylthiophen- 2 -yl)-l-butanol S,S-dioxide 42 (192 mg, 46%); umax (film) 3384 (br.), 2940,1653,1453,1300,1127,1078 cm*1; 6 (270 MHz) 5.93 (2H, m, H-3’, H-4’)» 3.78 (1H, m, H-5’), 3.73-3.64 (3H, m, H-l, H-2’), 2.00-1.90 (1H, m, H- 4), 1.70-1.50 (6H, m, H-2, H-3, H-4 [one proton], OH), 1.39 (3H, d, J 7 Hz, CH3); m/z (El) 205 (MH+), 186 (M+ - H 20), 174, 155, 140 (M+ - S02), 122, 94, 81, 79, 77 (Found: C, 53.20; H, 8.08. C ^ H j^ S requires C, 52.91; H, 7.90%). Swern Oxidation of alcohol (42) with in situ olefination. Oxalyl chloride (285 pi, 3.26 mmol, 2.2 eq) was added to THF (10 ml) under argon at - 78°C. After addition of dimethyl sulphoxide (463 pi, 6.53 mmol, 4.4 eq) the stirred solution was allowed to warm to -35°C for 3 min then re-cooled to -78°C. A solution of the alcohol 42 (303 mg, 1.48 mmol) in THF (5 m l) was added dropwise via cannula to give a cloudy white suspension. The reaction was allowed to warm to -35°C and stirred for 20 min. After re-cooling to -78°C, triethylamine (1.36 ml, 9.79 mmol, 6.6 eq) was added in one portion and the mixture warmed to rt. Analysis of the reaction mixture by tic (75% ethyl acetate-petrol) showed it to be complete after 20 min and the mixture was again cooled to -78°C. In a separate flask, dimethyl methanephosphonate (freshly distilled; 1.55 g, 12.46 mmol, 8.4 eq) was dissolved in THF (60 ml) under argon, cooled to -78°C and treated with n-butyllithium (4.82 ml of a 2.5M solution in hexanes, 12.16 174 mmol, 8.2 eq). After 5 min a solution of isopropyl phenylsulphinate (azeotropically dried with toluene (2 x 5 ml); 1.09 g, 5.93 mmol, 4.0 eq) was added dropwise via cannula. This pale yellow solution was then added via cannula at -78°C to the stirred Swem reaction mixture prepared above. The reaction mixture was allowed to warm to rt after 5 min, and stirred for a further hour. The mixture was poured into saturated aqueous ammonium chloride (200 ml) and extracted with ethyl acetate (3 x 100 ml). The combined organic layers were washed with saturated ammonium chloride (3 x 100 ml) and water (3 x 100 ml) alternately, followed by brine (100 ml), dried (MgS04) and concentrated under reduced pressure. The resulting crude oil was purified by chromatography (45% - 100% ethyl acetate-petrol) to give, in order of elution, [211*, y S * ] ^ .( 2 £-dihydro-5-methylthiophen- 2-yl)butanal S,S-dioxide 45 (28 mg, 9%) as a colourless oil; (film) 2934, 1720, 1451, 1304, 1129, 1078, 739, 641 cm-*; 8 (270 MHz) 9.77 (1H, t, J 1 Hz, H-l), 5.96-5.89 (2H, m, H-3', H-4'), 3.79 (1H, qt, J 7, 1 Hz, H-5'). 3.71-3.64 (1H, m, H-2’), 2.54 (2H, tt, J 7, 1 Hz, H-2), 2.00-1.60 (4H, m, H-3, H-4), 1.39 (3H, dd, J 7, 2 Hz, CH3); m/z (El) 203 (MH+), 185 (MH+ - H 20 ), 174, 167, 157, 153, 138 (M+ - S02), 123, 109, 94, 81 (Found: C, 53.27; H, 6.81. C9 H 140 3S requires C, 53.44; H 6.98%) (Found: (M + NH4+), 220.1007. C 9 Hi40 3S requires (M + NH4+), 220.1007), followed by [2R*. 5S*]-2,5-dihydro-2-methyl-5- [(4E)-5-(phenylsulphinyl)- 4-pentenyl]thiophene S,S-dioxide 44 (127 mg, 22.6%) as a colourless oil; umax (film) 2935, 1616, 1445, 1303,1131,1039, 750, 691, 640 cm*»; 8 (500 MHz) 7.64-7.60 (2H, m, ortho- protons on Ph), 7.55-7.45 (3H, m, meta- and para- protons on Ph), 6.59 (1H, ddt, J 15, 7, 3 Hz, H-4), 6.29 (1H, dd, J 15, 0.5 Hz, H-5), 6 00-5.88 (2H, m, H-3, H-4), 3.83-3.64 (2H, m, H-2, H-5), 2.36-2.28 (2H, m, H-3), 1.80-1.62 (4H, m, H -l\ H-2'), 1.41 (3H, d, J 7 Hz, CH3); m/z (El) 324 (M+), 260, 232, 179, 166, 149, 123, 91, 79, 64 (Found C, 59.53; H, 6.31%. C 16H20O3S2 requires C, 59.23; H, 6.21%) (Found: (M+ - S02), 260.1231. C 16H20O3S2 requires (M+ - S02), 260.1235), and finally [2R*. 5S*]-2J-dihydro-2-methyl-5-[(4Z)-5-(phenylsulphinyl)-4- 175 pentenyljthiophene S,S-dioxide 43 (217 mg, 38.6%) as a colourless oil; umax (film) 2936, 1623, 1558, 1444, 1303, 1131, 1086, 1037, 747, 691, 641 cnr’; 8 (500 MHz) 7.64-7.60 (2H, m, ortho -protons on Ph), 7.55-7.45 (3H, m, meta- and para -protons on Ph), 6.30-6.18 (2H, m, H-4’, H-5'), 5.94 (2H, m, H-3, H-4), 3.85-3.68 (2H, m, H-2, H-5), 2.75 (1H, m, H-3’), 2.63 (1H, m, H-3'), 2.05-1.68 (4H, m, H -l\ H-2'), 1.42 (3H, d, J 7 Hz, CH3); mlz (El) 260 (M+ - S02), 243, 232, 200, 179, 166, 149, 135, 123, 116, 109, 81, 64 (Found: C, 58.99; H, 6.25. C 16H2o0 3S2 requires C, 59.23; H, 6.21%) (Found: (M+ - S02), 260.1231. C 16H20O3S2 requires (M+ - S02), 260.1235). The presence of minor amounts of unidentified by-products in both product sulphoxides has not been taken into account when calculating the yields cited. Oxidation of Z-sulphoxide (43). To a stirred solution of Z-sulphoxide 43 (124 mg, 0.382 mmol) in dichloromethane (3.8 ml) containing anhydrous sodium acetate (47 mg, 0.573 mmol, 1.5 eq) at 0°C was added peracetic acid (117 pi of a 32 wt.% solution in dilute acetic acid, 0.573 mmol, 1.5 eq) via syringe. After 20 min the mixture was allowed to warm to rt and stirred for a further 8 h. The reaction was poured into water (50 ml) and extracted with dichloromethane (3 x 50 ml). The combined organic layers were then washed with 10% aqueous sodium thiosulphate (3 x 50 ml), saturated aqueous sodium hydrogencarbonate (3 x 50 ml), water (50 ml), dried (MgS04) and concentrated under reduced pressure. The crude product was purified by chromatography (50% - 75% ethyl acetate-petrol) to give /2R*, 5S*]- 2£-dihydro-2-methyl-5-l(Z)-5-(phenyl-sulphonyl)-4-pentenyl]thiophene S,S-dioxide 46 176 (86.4 mg, 66%) as a colourless oil; vmax (film) 3061, 2933, 1624, 1448, 1305, 1147, 1085, 783, 756, 689, 638 cm 1; 8 (270 MHz) 7.91 (2H, dd, J 8 , 2 Hz, or/Zio-protons on Ph), 7.66-7.52 (3H, m, meta- and para-protons on Ph), 6.36-6.20 (2H, m, H-4’, H- 5'), 5.92 (2H, m, H-3, H-4), 3.84-3.69 (2H, m, H-2, H-5), 2.76 (2H, m, H-3'), 1.96- 1.91 (1H, m, H-D , 1.79-1.55 (3H, m, H -l\ H-2’), 1.40 (3H, d, J 8 Hz, CH3); mlz (El) 276 (M+ - S02), 233, 221, 209, 195, 182, 165, 149, 147, 135, 95, 81, 77, 64 (S02) (Found: (M+- S02), 276.1180. Ci 6H20O4S requires (M+ - S02), 276.1184). Cycloelimination of S 0 2 from Z-sulphone (46). 7 Z-Sulphone 46 (18.7 mg, 0.055 mmol) was dissolved in dry toluene (2 ml) and heated to 92°C for 2 h, after which time tic (50% ethyl acetate-petrol) indicated the reaction to be complete. Concentration of the solution under reduced pressure and chromatography of the residue (20% ether-petrol) gave (1Z, 6E, 8E)-l-(phenylsulphonyl)-l, 6,8-decatriene 18a (13.7 mg, 92%) as a colourless oil; vmax (film) 3014, 2920, 1624, 1448, 1306, 1146,1086, 990,753, 688 cm'1; 8 (500 MHz) 7.91 (2H, dd, J 8 ,1.5 Hz, orf/to-protons on Ph), 7.66-7.60 (1H, m, para- proton on Ph), 7.58-7.52 (2H, m, me/a-protons on Ph), 6.30 (1H, br. d, J 11 Hz, H-l), 6.25 (1H, dt, J 11, 7 Hz, H-2), 6.05-5.95 (2H, m, H- 7, H-8 ), 5.55-5.63 (1H, m, H-6), 5.48 (1H, m, H-9), 2.67 (2H, br. q, J 7 Hz, H-3), 2.07 (2H, br. q, J 7 Hz, H-5), 1.74 (3H, d, J 7 Hz, CH3), 1.52-1.48 (2H, m, H-4); mlz (El) 276 (M+), 221 (M+ - C4H7), 209, 195, 182, 165, 149, 147, 135, 95, 81, 79, 77 (Found: (M+), 276.1180. C 16H20O2S requires (M+), 276.1184). 177 Preparation of [2 R*, 3S*]-(7E, 9£)-2-(phenylsuIphonyl)-7,9-undecadien- 3-ol (62a) and [2R*, 3/?*]-(7E, 9£')>2-(phenylsulphonyl)»7,9>undecadien- 3-oI (63a). 17a (Phenyl sulphonyl)ethaneM (dried in vacuo over P 20 5; 935 mg, 4.49 mmol, 1.1 eq) was dissolved in dry THF (20 ml) under argon and cooled to -78°C. n-Butyllithium (3.9 ml of a 1.42M solution in hexanes, 5.49 mmol, 1.1 eq) was added dropwise via syringe. After 10 min a solution of (5 E, 7£)-5,7-nonadienal 19a (freshly distilled; 690 mg, 4.99 mmol) in dry THF (5 ml) was added via cannula, rinsing with further dry THF (5 ml). The yellow anion colour faded slowly over the course of an hour to give a nearly colourless solution to which was added acetic acid in THF (4.6 ml of a 1.75 M solution, 1.3 eq). The reaction mixture was allowed to warm to rt, poured into saturated aqueous sodium hydrogencarbonate (50 ml) and the aqueous layer extracted with dichloromethane (3 x 50 ml). The combined organic layers were washed with water (2 x 50 ml), dried (MgSC> 4) and concentrated under reduced pressure. The crude product was purified by chromatography (35% - 40% ether-petrol) to give the hydroxysulphones (1:1 ratio by JH nmr; 1.298 g combined yield, 84%) as a viscous, colourless oil; less polar [2P*, 35*]- isomer 62a: vmax (film) 3516, 2935, 1630, 1448, 1302, 1146, 1086, 990, 760, 734, 690 cm-1; 6 (250 MHz) 7.90 (2H, dd, J 8 , 1.5 Hz, ortho-protons on Ph), 7.73-7.54 (3H, m, meta- and para -protons on Ph), 6.05-5.86 (2H, m, H- 8 , H-9), 5.65-5.40 (2H, m, H-7, H-10), 4.30-4.18 (1H, m, H-3), 3.02 (1H, qd, J 7, 1 Hz, H-2), 2.90 (1H, d, J 2.5 Hz, OH), 2.02 (2H, br. q, J 7 Hz, H- 6), 1.73 (3H, d, J 7 Hz, H -ll), 1.60-1.40 (4H, m, H-4, H-5), 1.32 (3H, d, J 7 Hz, H-l); m/z (El) 308 (M+), 307 (M+ - H), 290 (M+ - H20), 225, 213, 199, 181, 149, 141, 125, 94, 79, 77 (Found: (M+), 308.1451. 178 C17H24O3S requires (M+), 308.1446); more polar [2/?*, 3/?*]-isomer 63a: umax (film) 3516, 2926, 2858, 1630, 1448, 1305, 1147, 1083, 990, 734, 690 cm-*; 5 (250 MHz) 7.90 (2H, dd, J 8 , 1.5 Hz, on/ 10-protons on Ph), 7.74-7.52 (3H, m, meta- and para- protons on Ph), 6.05-5.89 (2H, m, H- 8 , H-9), 5.70-5.40 (2H, m, H-7, H-10), 4.08- 3.95 (1H, m, H-3), 3.85 (1H, d, J 2.5 Hz, OH), 3.15 (1H, quintet, J 7 Hz, H-2), 2.10- 2.00 (2H, m, H-6), 1.72 (3H, d, J 6.5 Hz, H -ll), 1.60-1.35 (4H, m, H-4, H-5), 1.12 (3H, d, J 7 Hz, H-l); mfz (El) 308 (M+), 307 (M+ - H), 290 (M+ - H 20), 225, 213,199, 149, 143, 107, 94, 79 (Found: C, 66.29; H, 8.03. C 17H240 3S requires C, 66.20; H, 7.84%). Preparation of 3S*]*(7£, 9E)-2-(phenyIsuIphonyl)-7,9*undecadien- 3-oI ( 4-methyI-phenyIsuIphonate) (64a). 8 Hydroxysulphone 63a (azeotropically dried with toluene (2x5 ml); 473 mg, 1.54 mmol) was dissolved in THF (15 ml) together with 1,10-phenanthrolinc (2 crystals) and the solution cooled to -78°C under argon. n-Butyllithium (560 nl of a 2.5M solution in hexanes, 1.4 mmol, 0.91 eq) was added dropwise to the stirred solution until a rust brown colour just persisted. After 5 min a solution of tosyl chloride (381 mg, 2.0 mmol, 1.3 eq) in THF (5 ml) was added via cannula. The reaction mixture was allowed to warm to 0°C whereupon saturated aqueous ammonium chloride (10 ml) was added. The aqueous phase was extracted with dichloromethane (3 x 50 ml) and the combined organic layers were washed with 1M aqueous sodium hydroxide (2 x 50 ml), water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure. The residue was purified by 179 chromatography (5% - 50% ether-petrol) to give, in order of elution, (2E, 7E, 9E)-2- (phenylsulphonyl)~2,7,9-undecatriene 61a (46.3 mg, 11%), as a colourless oil; umax (film) 3019, 2927, 2855, 1652,1446,1304, 1146, 1076,1024, 990, 761, 728, 690 enr l. 5 (250 MHz) 7.86 (2H, dd, J 8 , 2.5 Hz, ortho- protons on Ph), 7.64-7.48 (3H, m, meta- and para-protons on Ph), 6.89 (IH, td, J 6.5, 2.5 Hz, H-3), 6.06-5.88 (2H, m, H-8 , H-9), 5.70-5.30 (2H, m, H-7, H-10), 2.18 (2H, br. q, J 6.5 Hz, H-4), 2.08 (2H, br. q, J 6.5 Hz, H- 6), 1.80 (3H, s, H-l), 1.72 (3H, d, J 6.5 Hz, H -ll), 1.58 (2H, m, H-5); mlz (El) 209 (M+ - C 6H9), 149 (M+ - S0 2Ph), 77 40%) as a colourless oil; umax (film) 2936, 2905, 1598,1446, 1360, 1306,1189,1175, 1148, 1085, 924, 765, 731, 690 cm-»; 5 (270 MHz) 7.95-7.50 (7H, m, Ph and ortho- protons on Tol), 7.32 (2H, meta-protons on Tol), 6.00-5.90 (2H, m, H- 8 , H-9), 5.64- 5.30 (2H, m, H-7, H-10), 5.05 (IH, td, J 7, 3.5 Hz, H-3), 3.30 (1H, qd, 7, 2.5 Hz, H- 2) 2.42 (3H, s, Tol para-methyl), 2.05-1.90 (2H, m, H- 6), 1.75 (3H, d, J 6.5 Hz, H- 11), 1.40-1.05 (4H, m, H-4, H-5), 1.27 (3H, d, J 7 Hz, H-l); mlz (El) 462 (M+), 290 (M+ . TsOH), 213, 149, 94, 91, 77 (Found: (M + NH4+), 480.1878. C 29 H3o0 5S2 requires (M + NH4+), 480.1878). Preparation of (2E, 7E, 9E)-2.(phenyIsulphonyl).2,7,9-undecatriene (61a). 8 6 180 (phenylsulphonyl)ethane (286 mg, 0.619 mmol 64a) was dissolved in THF (10 ml) under argon and treated with a solution of potassium r-butoxide (513 pi of a 1.0M solution in THF, 0.513 mmol, 0.83 eq) at rt. After 20 min the reaction mixture was poured into water (25 ml) and extracted with dichloromethane (3 x 20 ml). The combined organic layers were washed with water (3 x 20 ml), dried (MgS04) and concentrated under reduced pressure. The product was purified by chromatography (15% - 50% ether-petrol) to give the title compound 61a as a colourless oil (58.2 mg, 32%), identical with material previously prepared, followed by a mixture of 62a, 64a and (phenylsulphonyl)ethane (177.5 mg, 0.223 mmol 64a). This mixture was re-subjected to the elimination conditions to give further 61a (58 mg, 32%; total yield 64%). Preparation of [2 R*, 3/?*]-(7E, 9E)-2-(phenyIsulphonyl)-7,9.undecadien. 3-oI 3-(4-methylphenylsulphonate) (65a). 8 63a 65a Hydroxysulphone 63a (azeotropically dried with toluene (2 x 20 ml); 378 mg, 1.225 mmol) and 1,10-phenanthroline (2 crystals) was dissolved in dry THF (15 ml) under argon and cooled to -78°C. To the stirred solution was added n-butyllithium (480 pi of a 2 5M solution in hexanes, 1.23 mmol, ca. 1 eq) until a rust-brown colour just persisted. A solution of tosyl chloride (408 mg, 2.141 mmol, 1.5 eq) in THF (10 ml) was added to the anion solution, causing slow discharge of the colour. The reaction was allowed to warm to rt, poured into water (50 ml) and extracted with dichloromethane (3 x 50 ml). The combined organic layers were washed with water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure to give a colourless oil (687 mg). The product was 181 purified by chromatography (15% - 50% ether-petrol) to give, in order of elution, (2Z, 7E, 9E)-2-(phenylsulphonyI)-2,7,9-undecatriene 60a (10 mg, 3%) as a colourless oil; umax (film) 3064, 2928,2859,1445, 1367, 1303, 1148, 1023, 997,760, 723, 690 cm->; 8 (250 MHz) 7.90 (2H, dd, J 8 , 2.5 Hz, ortho- protons on Ph), 7.70-7.50 (3H, m, meta- and para-protons on Ph), 6.11-5.92 (3H, m, H-3, H- 8 , H-9), 5.70-5.45 (2H, m, H-7, H-10) 2.68 (2H, br. q, J 6.5 Hz. H-4), 2.09 (2H, br. q, J 6.5 Hz, H- 6), 1.99 (3H, br. s, H-l), 1.75 (3H, d, J 6.5 Hz, H -ll), 1.48 (2H, quintet, J 6.5 Hz, H-5); m h (El) 290 (M+), 209 (M+ - C 6H9), 196 (M+ - C 7H10), 179,149 (M+ - S0 2Ph), 95, 79,77 (Found: (M+), 290.1341. C 17H220 2S requires (M+), 290.1341), and /2R*. 3R*J-(7E, 9E)-2- (phenylsulphonyl)-7,9-undecadien-3-ol 3-(4-methylphenylsulphonate) 65a (398 mg, 70%) as a colourless oil which crystallized from ether at -18°C, mp 92-94°C (ether- petrol); vmax (film) 2928, 2868, 1716, 1596, 1446, 1367, 1307, 1254, 1189, 1176, 1150, 1095, 1080, 914, 879, 816, 707, 688 c m 1; 8 (500 MHz) 7.88 (2H, dd, J 8 , 2.5 Hz, ortho- protons on S0 2Ph), 7.75-7.60 (5H, m, meta- and para- protons on S02Ph and ortho- protons on Tol), 7.32 2H, d, J 8 Hz, mefa-protons on Tol), 5.97 (1H, ddd, J 15, 11.5, 1.5 Hz, H-9), 5.85 (1H, dd, J 15, 11 Hz, H- 8 ), 5.58 (1H, dq, J 15, 6.5 Hz, H- 10), 5.32 (1H, dt, J 15, 6.5 Hz, H-7), 4.81 (1H, dt, J 13, 2 Hz, H-3), 3.72 (1H, qd, J 7, 3 Hz, H-2), 2.45 (3H, s, Tol para-methyl), 1.87-1.68 (3H, m, H-4 [one proton] and H-6), 1.74 (3H, d, J 6.5 Hz, H -ll), 1.62 (1H, m, H-4), 1.30 (3H, d, J 6.5 Hz, H-l), 1.18 (1H, m, H-5), 0.90 (1H, m, H-5); m/z (El) 462 (M+), 307 (M+ - Ts), 290 (M+ - TsOH), 255, 239, 225, 209, 196, 149, 94, 81, 77 (Found: C, 62.15; H, 6.41. C24H30O5S2 requires C, 62.31; H, 6.54%). 182 Preparation of (2Z, IE, 9E)-2-(phenylsulphonyl)-2,7,9-undecatriene (60a). 8 6 Tosyloxysulphone 65a (232 mg, 0.502 mmol) was dissolved in dry THF (5 ml) under argon and cooled to 0°C. To the stirred solution was added potassium r-butoxide (500 pi of a 1.0M solution in THF, 0.5 mmol, 1 eq) dropwise via syringe. This caused the appearance of a yellow colour followed by formation of a white precipitate. The reaction mixture was immediately poured into water (25 ml) and the aqueous layer extracted with dichloromethane (3 x 25 ml). The combined organic layers were washed with water (3 x 30 ml), dried (MgS04) and concentrated under reduced pressure to give a colourless oil (140 mg). The product was purified by chromatography (15% - 40% ether-petrol) to give, in order of elution, the title compound 60a (84 mg, 58%), identical with previously prepared material, followed by a mixture of trienes 61a and 60a (31 mg), followed by recovered 65a (6.4 mg, 3%). The 61a/60a mixture was re-chromatographed to furnish further 60a (16 mg; overall yield 100 mg, 69%). 183 Preparation of [2/?*, 3S*]-(8E, 10E)-2.(phenylsulphonyl).8,10- dodecadien-3-oI (62b) and [2 R*, 3E*,]-(8E, 10E).2-(phenylsulphonyl)- 8,10-dodecadien-3-ol (63b). 8 9 (Phenylsulphonyl)-ethane ( dried in vacuo over P 20 5; 790 mg, 4.63 mmol, 1.1 eq) was dissolved in dry THF (50 ml) and the solution cooled to -78°C. n-Butyllithium (1.85 ml of a 2.5M solution in hexanes, 4.63 mmol, 1.1 eq.) was added dropwise via syringe with stirring, causing the solution to turn yellow. After 5 min a solution of aldehyde 19b (freshly distilled; 642 mg, 4.21 mmol) in THF (5 ml + 2 ml rinse) was added via cannula to the reaction mixture. The reaction mixture was stirred for 1 h at -78°C after which time it was quenched by the addition of acetic acid in THF (4.9 ml of a 1.75M solution, 2 eq.) and allowed to warm to rt. The mixture was poured into saturated aqueous sodium hydrogencarbonate (50 ml) and extracted with dichloromethane (3 x 50 ml). The combined organic layers were washed with water (3 x 50 ml), dried (MgSO^ and concentrated under reduced pressure to give a colourless oil. The crude product was purified by chromatography (35% - 40% ether-petrol) to give, in order of elution, [2 R * , 3S*]-(8E, lOE)-2-(phenylsulphonyl)-8,10-dodecadien-3-ol 62b (642 mg, 47%) as a colourless oil; vmax (film) 3517, 2930, 2896, 1683, 1605, 1446, 1299, 1143, 1088, 990, 760, 735 cm 1; 8 (500 MHz) 7.92 (2H, dd, J 8 , 2 Hz, o rth oprotons - on Ph), 7.70- 7 56 (3H, m, m eta - and para-protons on Ph), 6.03-5.92 (2H, m, H-9, H-10), 5.58 (1H, m) and 5.48 (1H, m, both comprising H- 8 , H -ll), 4.26-4.21 (1H, m, H-3), 3.02 (1H, qd J 7, 2 Hz, H-2), 2.88 (1H, d, J 2.5 Hz, OH), 2.01 (2H, br. q, J 7 Hz, H-7), 1.73 (3H dt j 7 Hz, H-12), 1.65-1.56 (2H, m, H-4), 1.46-1.20 (4H, m, H-5, H-6), 1.31 184 (3H, d, J 7 Hz, H-l); nUz (El) 322 (M+), 265, 254, 239, 225, 213, 199, 180, 170, 163, 107, 95, 81, 77 (Found: C, 67.04; H, 8.13. Ci 8H260 3S requires C, 67.04; H, 8.13%), followed by [2R*, 3R*)-(8E, 10E)-2-(phenylsulphonyl)-8,10-dodecadien-3-ol 63b (513 mg, 38%) as a colourless oil; vmax (film) 3507, 2935, 1688, 1448, 1304, 1147, 1083, 999, 734, 691 cm*1; 6 (500 MHz) 7.90 (2H, dd, J 8 , 2 Hz, ortho -protons on Ph), 7.70- 7.55 (3H, m, meta- and para-protons on Ph), 6.04-5.94 (2H, m, H-9, H-10), 5.61-5.47 (2H, m, H-8 , H-l 1), 4.04-3.98 (1H, m, H-3), 3.82 (1H, dd, J 3, 1 Hz, OH), 3.17 (1H, dq, J 8 , 7 Hz, H-2), 2.05 (2H, br. q, J 7 Hz, H-7), 1.72 (3H, br. d, J 7 Hz, H -l2), 1.65-1.55 (2H, m, H-4), 1.50-1.32 (4H, m, H-5, H-6), 1.14 (3H, d, J 7 Hz, H-l); m/z (El) 322 (M+), 225 (M+ - C7H13), 199 (G jHjjO ^ ) , 181, 143,141 (PhS02+), 107, 94, 81, 77 (Found: (M+ - C 7H 13), 225.0590. C 18 H 260 3S requires (M+ - C 7H 13), 225.0585). Preparation of [2 R*, 3S*J-(8E, 10£)*2*(phenylsulphonyl)-8,10- dodecadien-3-ol 3-(4*methylphenylsuIphonate) (64b). 8 Hydroxysulphone 63b (589 mg, 1.828 mmol) and 1,10-phenanthroline (2 crystals) were dissolved in dry THF (30 ml) under argon and cooled to -78°C. n-Butyllithium (720 pi of a 2.57M solution in hexanes, 1.85 mmol, 1.01 eq) was added to the stirred solution until a rust brown colour just persisted. After 5 min a solution of tosyl chloride (524 mg, 2.742 mmol, 1.5 eq) in dry THF (10 ml) was added via cannula, slowly discharging the colour. When allowed to warm to rt a near-colourless solution resulted. The reaction was poured into saturated aqueous ammonium chloride (30 ml) and the aqueous phase 185 was extracted with dichloromethane (3 x 50 ml). The combined organic layers were washed with water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure. The product was purified by chromatography (5% - 60% ether-petrol), to give, in order of elution: (2E, 8E, 10E)-2-(phenylsulphonyl)-2,8,10-dodecatriene 61b as a colourless oil (46.7 mg, 8 %); v>max (film) 2928,2857,1652,1624,1447,1304,1144,1072 cm-i; 8 (250 MHz) 7.90-7.80 (2H, m, ortho- protons on Ph), 7.64-7.48 (3H, m, meta- and para- protons on Ph), 6.88 (1H, td, J 8 , 1.5 Hz, H-3), 6.08-5.90 (2H, m, H-9, H-10), 5.65- 5.40 (2H, m, H-8 , H -ll), 2.18 (2H, br. q, J 7 Hz, H-4), 2.08 (2H, br. q, J 7 Hz, H-7), 1.82 (3H, d, J 1Hz, H-l), 1.74 (3H, d, J 7.5 Hz, H-12), 1.50-1.30 (4H, m, H-5, H-6); m/z 304 (M+), 291 (M+ - CH3), 267, 238, 176, 143, 135, 125, 121, 109, 95, 91, 81, 77, followed by the title compound 64b (1:1 mixture of 62b and 64b by JH nmr; 492 mg, 37% 64b) as a colourless oil; vmax (film) 3062, 2935, 1597, 1584, 1446, 1405, 1362,1306, 1189, 1175,1146, 1087, 997, 783, 731, 690 cm:1; 8 (500 MHz) 7.84 (2H, dd, J 8.5, 1.5 Hz, ortho- protons on Ph) 7.77 (2H, d, J 8.5 Hz, ortho- protons on Tol), 7.70-7.50 (3H, m, meta- and para- protons on Ph), 7.32 (2H, d, J 8.1 Hz, meta-protons on Tol), 6.02-5.90 (2H, m, H-9, H-10), 5.58 (1H, m) and 5.42 (1H, m, all comprising H-8 , H -ll), 5.03 (1H, td, J 6.5, 3.5 Hz, H-3), 3.29 (1H, qd, J 7, 3.5 Hz, H-2), 2.44 (3H, s, Tol para- methyl), 1.94 (2H, br. q, J 7 Hz, H-7), 1.78-1.70 (5H, m, H-4, H- 12), 1.35-1.20 (4H, m, H-5, H-6), 1.27 (3H, d, J 7 Hz, H-l); mlz (positive FAB) 499 (MNa+), 289, 193, 154,107. 186 Preparation of (2 E t 8 E, 10E)-2-(phenyIsulphonyl)-2,8,10-dodecatriene (61b). 8 64b A mixture of hydroxysulphone 62b, tosyloxysulphone 64b and (phenylsulphonyl)ethane (294 mg of 64b by nmr, 0.617 mmol) was dissolved in THF (10 ml) and cooled to *20°C under argon. Potassium r-butoxide (700 pi of a 1.0M solution in THF, 0.7 mmol, 1.13 eq) was added dropwise to the stirred solution. When the reaction appeared to be complete by tic (2 x 50% ether-petrol) water (10 ml) was added to the rapidly stirred solution. The mixture was extracted with dichloromethane (3 x 25 ml), washed with water (3 x 25 ml), dried (MgS04) and concentrated under reduced pressure to give a colourless oil (327 mg). Purification by chromatography (15% - 50% ether-petrol) gave, in order of elution, the title compound 61b (157 mg, 84%) identical to material previously prepared, and an inseparable mixture of 62b and (phenylsulphonyl)ethane (135 mg). Preparation of [2/?*, 3R*]-(8E, 10£')-2-(phenylsulphonyl)-8,10* dodecadien-3-ol 3-(4-methylphenylsulphonate) (65b). 9 63b 65b 187 Hydroxysulphone 63b (513 mg, 1.59 mmol) and 1,10-phenanthroIine (2 crystals) were azeotropically dried with toluene (2 x 20 ml), dissolved in THF (16 ml) under argon and the solution cooled to -78°C. n-Butyllithium (620 pi of a 2.57M solution in hexanes, 1.59 mmol, 1.0 eq) was added dropwise via syringe until a rust brown colour just persisted. After 5 min a solution of tosyl chloride (455 mg, 2.39 mmol, 1.5 eq) in THF (5 ml) was added via cannula which caused the colour to fade slowly. The reaction was allowed to warm to rt over 40 min whereupon it was quenched by the addition of saturated aqueous ammonium chloride (50 ml). The aqueous phase was extracted with dichloromethane (3 x 50 ml) and the combined organic layers were washed with water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure. Chromatography of the crude product (5% - 50% ether petrol) gave, in order of elution, the title compound (567 mg, 75%) as a colourless oil; umax (film) 3060, 2939, 2864, 1721, 1596, 1446, 1366, 1307, 1250, 1189, 1176, 1150, 1096, 901, 817, 735, 690, 662 cnr1; 8 (270 MHz) 7.88 (2H, dd, J 8 , 2 Hz, ortho -protons on Ph), 7.75-7.45 (4H, m, meta- and para -protons on Ph and ortho -protons on Tol), 7.32 (2H, d, J 8.5 Hz, mew-protons on Tol), 6.05- 5.85 (2H, m, H-9, H-10), 5.66-5.30 (2H, m, H- 8 , H -ll), 4.78 (1H, ddd, J 9.5, 3, 2 Hz, H-3), 3.76 (1H, qd, J 7, 3 Hz, H-2), 2.45 (3H, s, para-methyl on Tol), 1.90-1.80 (3H, m, H-4 [one proton], H-7), 1.75 (3H, d, J 7 Hz, H-12), 1.62 (1H, m, H-4), 1.32 (3H, d, J 7 Hz, H-l), 1.20-1.00 (3H, m) and 0.90-0.70 (1H, m, all comprising H-5, H- 6); m/z (El) 476 (M+), 304 (M+ - TsOH), 172,163, (M+ - TsOH - P h S 0 2), 107, 94, 91, 81, 77 (Found: C, 63.18; H, 6.69. C25H3205S2 requires C, 62.99; H, 6.77%), and recovered 63b (15.3 mg, 3%). 188 Preparation of (2Z, 8£, 10£)-(2-phenylsulphonyl).2,8,10-dodecatriene (60b). Tosylate 65b (572 mg, 1.20 mmol) was dissolved in THF (15 ml) and cooled to -20°C under argon. Potassium r-butoxide (1.2 ml of a 1.0M solution in THF, 1.2 mmol, 1.0 eq) was added to the stirred solution dropwise via syringe. The reaction was quenched by the rapid addition of water (10 ml) and allowed to warm to rt. The aqueous phase was extracted with dichloromethane (3 x 50 ml), and the combined organic layers were washed with water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure. Purification of the product by chromatography (15% - 50% ether-petrol) gave, in order of elution, the title compound (265 mg, 72%) as a colourless oil; umax (film) 3014, 2926, 2851, 1642, ¿447, 1305, 1145, 1083 cm 1; 8 (250 MHz) 7.92-7.85 (2H, m, ortho- protons on Ph), 7.65-7.48 (3H, m, meta- and para-protons on Ph), 6.05-5.95 (3H, m, H-3 H-9, H-10), 5.55 (2H, m, H- 8 , H -ll), 2.65 (2H, m, H-4), 2.05 (2H, m, H-7), 1.95 (3H, d, J 1 Hz, H-l), 1.74 (3H, d, J 7 Hz, H-12), 1.45-1.30 (4H, m, H-5, H- 6); m/z (El) 304 (M+), 209, 196, 179, 163 (M+ - S0 2Ph), 125, 109, 81, 79 (Found: C, 71.14; H, 8.05. C 18H240 2S requires C, 71.01; H, 7.95%), followed by (2£, 8 £, 10E)- (2-phenylsulphonyl)- 2,8 ,10-dodecatriene 61b (29 mg, 8 %) and finally recovered 65b (13 mg, 2%). 189 IMDA reaction of (1 £, 7E, 9£)-l-(phenyIsuIphonyl)-l,7,9-undecatriene (17b). 1 H 9 8 7 z H 6 PhS02 50 53 A solution of triene 17b (azeotropically dried with toluene (3 x 50 ml); 497 mg, 1.71 mmol) in dry toluene (30 ml) was rigorously degassed by alternate sonification for 5 min, followed by degassing with argon for 5 min, repeated three times. The solution was then transferred via cannula to a dry, argon-filled resealable pressure tube and the system sealed. The tube was heated (Woods' metal bath) to 175°C for 96 h. After cooling, the toluene was evaporated under reduced pressure to give a pale crystalline solid. nmr (250 MHz) analysis of the crude product indicated the presence of a 6:1 mixture of cycloadducts. Purification by chromatography (20% ether-petrol) gave a white solid (457 mg, 92%) which was recrystallized to give /JR*, 4S*, JS*, 10S*]-3-methyl-4- (phenylsulphonyl)bicyclo[4.4.0]-2-decene 50 (270 mg, 54%), mp 133-134°C (benzene- petrol); umax (film) 3014, 2927, 2856, 1558, 1446, 1307, 1291, 1143, 1085, 935, 784, 746, 718, 690, 646, 616 cm*1; 8 (500 MHz) 7.92 (2H, dd, J 8 , 2 Hz, ortho-protons on Ph), 7.70-7.50 (3H, m, meta- and para-protons on Ph), 5.52 (2H, br. s, H-l, H-2), 2.80 (1H, m, H-4), 2.77 (1H, m, H-10), 2.69 (1H, m, H-3), 2.49 (1H, m, H-5), 1.74- 1.42 (5H, m) and 1.28-1.17 (3H, m, all comprising H- 6, H-7, H-8 , H-9), 0.95 (3H, d, J 7.5 Hz, CH3); m/z (El) 290 (M+), 258,226,148 (M+ - HS0 2Ph) (Found: C, 70.11; H 7.57. Q 7H22O2S requires C, 70.31; H, 7.65%). Repeated recrystallization (benzene- petrol) of the residue from the mother liquor gave /JR*. 4R*. 5R*, 1 OS*]-3-methyl-4 - (phenylsulphonyl)bicyclo[4.4.0]-2-decene 53, mp 133-134.5°C (benzene-petrol); umax (film) 2923, 1651, 1446, 1300, 1144, 1086 cm*1; 8 (500 MHz) 7.88 (2H, dd, J 8 , 1.5 190 Hz, orr/i¿»-protons on Ph), 7.65-7.60 (1H, m, para -proton on Ph), 7.55 (2H, m, mew- protons on Ph), 5.49 (1H, ddd, J 10, 5, 2.5 Hz, H-2), 5.33 (1H, d, J 10 Hz, H-l), 3.44 (1H, dd, J 10, 4.5 Hz, H-4), 2.41 (1H, m, H-3), 2.33 (1H, dd, J 12.5, 2.5 Hz, H-6cq), 1.90 (1H, m, H-5), 1.86 (1H, m, H-10), 1.80-1.70 (3H, m, H-7gq, H-S^, H-9eq), 1.35 (1H, m, H-7ax), 1.25 (1H, m, H-8 ax) 1.21 (3H, d, J 7 Hz, CH3), 1.12 (1H, qd, J 12, 3.5 Hz, H-9ax), 1.02 (1H, qd, J 12, 3.5 Hz, H- 6ax); mlz (El) 290 (M+), 225, 199, 169, 165, 149 (M+ - S0 2Ph), 81,79,77 (Found: C, 70.41; H, 7.68. C 17H220 2S requires C, 70.31; H 7.65%). IMDA reaction of (1Z, 7£, 9£)-l-(phenylsulphonyl)-l,7,9-undecatnene (18b). A solution of .triene 18b (azeotropically dried with toluene (2 x 10 ml); 32.4 mg, 0.116 mmol) in dry toluene was degassed as above and transferred to a dry, argon-filled resealable pressure tube via cannula. The solution was heated at 160°C for 70 h, 165 C for 48 h, and then 175°C for 28 h. After cooling, the solvent was evaporated under reduced pressure. JH nmr analysis of the crude material indicated the presence of a 3:1 mixture of cyclization products. Purification by chromatography (15% ether-petrol) gave a solid (30 mg, 92%) which was recrystallized to give the major adduct /JR*, 4S*, JR*, JOS*]-3-methyl-4-(phenylsulphonyl)bicyclo[4.4.0J-2-decene 55 mp 118-119°C (benzene-petrol); omax (film) 3008, 2914, 2848, 1638, 1539, 1480, 1443, 1368, 1303, 1287, 1145, HO8» 1083, 74°. 726’ 688 cm’1; 5 <500 MHz^ 7,92 ^2H’ dd’ J 8’ 2 Hz* ortho -protons on Ph), 7.65-7.50 (3H, m, meta- and para-protons on Ph), 5.50 (1H, br. d J 10 Hz, H-l), 5.43 (1H, m, H-2), 3.06 (1H, d, J 7 Hz, H-4), 2.64 (1H, m, H-3), 191 2.39 (1H, m, H-10), 2.05 (1H, qd, J 13, 3.5 Hz, H- 6ax), 1.90-1.80 (2H, m, H-7cq, H- 9eq), 1.78-1.68 (3H, m, H-5, H- 6eq, H- 8 eq), 1.38 (1H, qt, J 13, 3.5 Hz, H- 8 ax), 1.21 (1H, qt, J 13, 3.5 Hz, H-7ax), 0.97 (3H, d, J 7 Hz, CH3), 0.95 (1H, m, H-9.x); m/z (El) 290 (M+), 258, 249, 225, 208, 165, 148 (M+ - HS0 2Ph), 133, 105, 91, 81, 77 (Found: C, 70.05; H, 7.80. Ci7H2202S requires C, 70.31; H, 7.64%) (Found: (M+), 290.1341. C 17H220 2S requires (M+), 290.1341). The minor adduct, [3R*, 4R*, 5S*, 10S*]-3-methyl-4-(phenylsulphonyl)bicyclo[4A.0]-2-decene 56 could not be isolated in a pure state from the mixture; 5 (500 MHz) ( inter alia) 7.92 (2H, dd, J 8 ,2 Hz, ortho- protons on Ph), 7.68-7.55 (3H, m, meta- and para- protons on Ph), 5.60 (1H, dt, J 10, 3.5 Hz, H-l), 5.36 (1H, d, J 10 Hz, H-2), 3.33 (1H, dd, J 6.5, 2.5 Hz, H-4), 2.76- 2.70 (1H, m, H-3), 2.28-2.20 (2H, m) and 2.18-2.14 (1H, m, all comprising H-5, H- 10, H-6eq), 1.62 (1H, m, H- 6ax), 1.44 (3H, d, J 7.5 Hz, CH3), 1.08 (1H, m, H-9W). IMDA reaction of (IE, 6E , 8 E)-l-(phenyIsulphonyl)-l, 6 ,8 -decatrIene (17a). PhSO 17a 57 58 A solution of triene 17a (azeotropically dried with toluene (2 x 10 ml); 10 mg, 0.036 mmol) in dry xylene (10 ml) was degassed as above and transferred to a dry, argon-filled resealable pressure tube via cannula. The solution was heated at 145°C for 48 h. !H nmr analysis of the crude product indicated the presence of a 1:1 mixture of diastereomers. The product was purified by chromatography (15% ether-petrol) to give a colourless solid (9.3 mg, 0.034 mmol, 93%) which was fractionally crystallized to give [3R*, 4S*, 5S*, 9S*]-3-methyl-4-(phenylsulphonyl)bicyclo[4.3.0]-2-nonene 57, mp 130-131.5°C 192 (benzene-petrol); vmax (film) 2962, 1640, 1443, 1282, 1243, 1140, 1083, 763, 738, 717, 690 cm-1; 6 (500 MHz) 7.92 (2H, dd, J 8 , 2 Hz, ortho-protons on Ph), 7.70-7.50 (3H, m, meta- and para- protons on Ph), 5.45 (2H, br. s, H-l, H-2), 3.17 (1H, t, J 2.0 Hz, H-4), 2.82-2.74 (1H, m, H-3), 2.63 (1H, m, H-5), 2.55 (1H, m, H-9), 1.89-1.80 (1H, m), 1.78-1.70 (1H, m), 1.63-1.55 (2H, m) and 1.51-1.45 (2H, m; all comprising H-6, H-7, H-8 ), 1.10 (3H, d, J 7 Hz, CH3); m/z (El) 276 (M+), 223, 205, 182, 165, 149, 143, 141, 135 (M+ - S0 2Ph), 119, 105, 93 (Found: (M + NH4+), 294.1528. C16H20O2S requires (M + NH4+), 294.1528), and [3R*. 4R *, 5R*, 9S*J-3-methyl-4 - (phenylsulphonyl)bicyclo[4.3.0]-2-nonene 58, mp 132-134°C (benzene-petrol); umax (film) 3021, 2966, 1641, 1443, 1285, 1243, 1108, 1082, 1024, 841, 796, 764, 738, 716, 690 cm-1; 8 (500 MHz) 7.92 (2H, dd, J 8 , 2 Hz, ortho- protons on Ph), 7.68-7.50 (3H, m, meta- and para- protons on Ph), 5.73 (1H, d, J 9.5 Hz, H-l), 5.50 (1H, ddd, J 9.5, 5, 2.5 Hz, H-2), 3.50 (1H, dd, J 10, 4 Hz, H-4), 2.77 (1H, m, H-3), 1.98-1.55 (6H, m), 1.18 (1H, qd, J 10.5, 8.5 Hz) and 0.93 (1H, qd, J 10.5, 9.5 Hz, all comprising H-5, H- 6, H-7, H-8 , H-9), 1.28 (3H, d, J 7 Hz, CH3); m/z (El) 276 (M+), 223, 149, 143, 135 (M+ - S0 2Ph), 119, 107, 93, 79, 77 (Found: C, 69.56; H, 7.47. C16H20O2S requires C, 69.53; H, 7.29%). IMDA reaction of (1Z, 6 E, 8 E)-l.(phenylsulphonyl)-l,6,8-decatriene (18a). i H 7 + I H 6 PhSO, 59 60 A solution of triene 18a (34.3 mg, 0.124 mmol) in toluene (6 ml) degassed as above was transferred to a dry, argon-filled resealable pressure tube via cannula. The solution was 193 heated to 165°C for 60 h, after which time nmr analysis of the crude material indicated the presence of a 7:1 mixture of diastereomeric products. Chromatography (15% ether- petrol) gave a crystalline solid (21.7 mg, 63%) which was recrystallized to give the major isomer, [3R*, 4S*, 5R*, 9S*]-3-methyl-4-(phenylsulphonyl)bicylo[4.3.0]-2-nonene 60 mp 114-115°C (benzene-petrol); \>mM (film) 2960,2873,1624,1593,1448,1295,1193, 1164, 1145, 1084, 1025, 952, 764, 690 cm-*; 6 (500 MHz) 7.92 (2H, dd, J 8, 2 Hz, ortho- protons on Ph), 7.68-7.7.60 (1H, m, para- proton on Ph), 7.60-7.52 (2H, m, mew-protons on Ph), 5.88 (1H, d, J 10 Hz, H-l), 5.43 (1H, dt, J 10, 3 Hz, H-2), 3.34 (1H, d, J 4 Hz, H-4), 2.74 (1H, m, H-3), 2.65-2.57 (1H, m, H-5), 2.20-2.10 (1H, m, H-9), 1.98-1.89 (1H, m, H-6), 1.86-1.65 (4H, m, H-6, H-7, H-8 [one proton]), 1.15- 1.08 (1H, m, H-8), 0.95 (3H, d, J 7.5 Hz, CH3); m/z (El) 276 (M+), 211, 183, 143, 141 (S02Ph+), 134 (M+ - HS02Ph) (Found: (M+), 276.1180. C16H20O2S requires (M+), 276.1184). The minor isomer, [3 R*, 4R*, 5S*, 9S*]-3-methyl-4- (phenylsulphonyl)bicylo[43.0]-2-nonene 59 gave inter alia the following !H nmr data: 8 (500 MHz) 5.61 (1H, ddd, J 10, 5, 2 Hz, H-l), 5.37 (1H, dt, J 10, 2 Hz, H-2), 3.53 (1H, dd, J 5, 4 Hz, H-4), 2.38 (1H, qd, J 7.5, 4 Hz, H-3), 1.43 (3H, d, J 7.5 Hz, CH3). IMDA reaction of (2£, 7E, 9£)-2-(phenylsulphonyl)-2,7,9-undecatriene (61a). A solution of triene 61a (142 mg, 0.487 mmol) in dry toluene (10 ml) was degassed as above and transferred to an argon-filled resealable pressure tube via cannula. The 194 solution was heated to 175°C for 120 h, allowed to cool and concentrated under reduced pressure. !H nmr analysis of the crude product indicated the presence of a 1:1.5 mixture of cycloadducts. Purification by chromatography (20% ether-petrol) gave a semi-solid (73 mg, 52%) which was re-chromatographed (15% ether-petrol) to give, in order of elution, [3R*, 4R*. 5R*, 9S*]-3,4-dimethyl-4-(phenylsulphonyl)bicyclo[4.3.0]-2- nonene 67 (29 mg, 20%) as a crystalline solid, mp 132-133°C (benzene-petrol); umax (film) 2950, 2900,1600,1300, 1150,1120,1080 cnr*; 8 (270 MHz) 7.95 (2H, dd, J 8, 2 Hz, ortho-protons on Ph), 7.65-7.45 (3H, m, meta- and para- protons on Ph), 5.68 (1H, br. d, J 10 Hz, H-l), 5.52 (1H, ddd, J 10, 5, 2 Hz, H-2), 2.40 (1H, m, H-3), 2.12 (1H, td, J 12, 6 Hz, H-9), 2.00-1.60 (6H, m) and 1.05 (1H, m, all comprising H- 5, H-6, H-7, H-8), 1.40 (3H, d, J 7 Hz, C-3 CH3), 1.27 (3H, s, C-4 CH3); m/z (El) 223, 205, 176, 167, 149 (M+ - S02Ph), 133, 121, 107, 93, 79, 77 (Found: C, 70.23; H, 7.82. C17H220 2S requires C, 70.30; H, 7.64%), followed by a mixture of the two isomers (10 mg, 7%), and finally [3R*. 4S*, 5S*, 9S*]-3,4-dimethyl-4- (phenylsulphonyl)bicyclo[4.3.0]-2-nonene 66 (21.3 mg, 15%) as a crystalline solid, mp 126-132°C (benzene-petrol); vmax (film) 2960, 1585, 1448, 1381, 1299, 1129, 1075, 760, 733, 693 cm'1; 8 (500 MHz) 7.92 (2H, dd, J 8,2 Hz, ortho- protons on Ph), 7.70- 7.48 (3H, m, meta and para- protons on Ph), 5.50 (2H, s, H-l, H-2), 2.91-2.82 (2H, m, H-5, H-9), 2.37 (1H, q, J 7.5 Hz, H-3), 1.91-1.83 (1H, m) and 1.70-1.40 (5H, m, all comprising H-6, H-7, H-8), 1.37 (3H, s, C-4 CH3), 1.15 (3H, d, J 7.5 Hz, C-3 CH3); m/z (El) 149 (M+ - S02Ph), 133,119,107,93, 86, 81,79,77,49 (Found: C, 70.11; H, 7.40. C17H220 2S requires C, 70.30; H, 7.64%). IMDA reaction of (2Z, 7£, 9E)-2-(phenylsulphonyl)-2,7,9-undecatriene (60a). 60a 69 70 A solution of triene 60a (84.3 mg, 0.289 mmol) in dry xylene (10 ml) was degassed as above and transferred via cannula to a dry, argon-filled resealable pressure tube. The solution was heated to 190°C for 60 h, allowed to cool, and then concentrated under reduced pressure. 1H nmr analysis of the crude product indicated the presence of a 8:1 mixture of two diastereomeric cycloadducts. Purification by chromatography (15% ether- petrol) gave, in order of elution, (3 R*. 4R*, JS*. 9S*J-3,4'dimethyl-4- (phenylsulphonyl)bicyclo[43.0]-2-nonene 69 (6.8 mg, 8%) as an oil which crystallized from ether to give a solid, mp 126-128°C (benzene-petrol); \>max (film) 2944, 1623, 1448, 1379, 1300, 1149, 1127, 1073, 761, 737, 693 cm-»; 6 (500 MHz) 7.89 (2H, br. d, J 8 Hz, ortho -protons on Ph), 7.65 (1H, br. t, J 8 Hz, para -proton on Ph), 7.57 (2H, br. t, J 8 Hz, meta-protons on Ph), 5.57 (1H, ddd, J 10, 5, 2.5 Hz, H-2), 5.33 (1H, br. d, J 10 Hz, H-l), 2.56-2.48 (2H, m, H-3, H-9), 2.32-2.24 (1H, m), 2.12-2.04 (1H, m), 1.90 (1H, m), 1.75-1.65 (2H, m) and 1.58-1.49 (2H, m, all comprising H-5, H-6, H-7, H-8); m!z (El) 290 (M+), 149 (M+ - S02Ph), 133, 107, 93, 77 (Found: 290.1341. C 17H22O2S requires (M+), 290.1341), followed by a mixture of the two isomers as a colourless oil (15.4 mg, 18%), and finally ¡3R*, 4S*. 5R*, 9S*]-3,4- dimethyl-4-(phenylsulphonyl)bicyclo[4.3.0]-2-nonene 70 (57.7 mg, 69%) as a crystalline solid, mp 147-148.5°C (benzene-petrol); umax (film) 2961,2874,1649,1447, 1299, 1139, 731 cm*1; 8 (250 MHz) 7.90 (2H, dd, J 8, 2 Hz, ortho -protons on Ph), 7.65-7.50 (3H, m, meta- and para-protons on Ph), 5.84 (1H, dt, J 10, 1.5 Hz, H-l), 196 5.50 (1H, ddd, J 10, 5, 1.5 Hz, H-2), 2.90-2.70 (2H, m, H-3, H-9), 2.15-1.90 (2H, m), 1.80-1.60 (4H, m) and 1.22-1.08 (1H, m, all comprising H-5, H-6, H-7, H-8), 1.35 (3H, s, C-4 CH3), 0.90 (3H, d, J 7 Hz, C-3 CH3); m/z (El) 290 (M+), 149 (M+ - S 02Ph), 133, 119, 105, 91, 79, 77; (Cl) 308 (M + NH4+), 149 (M+ - S 02Ph) (Found: (M + NH4+), 308.1684. C17H220 2S requires (M + NH4+), 308.1684). IMDA reaction of (2 E, 8 E, 10E)-2-(phenylsulphonyI)-2,8,10-dodecatriene (60b). — 1 H 9 1 H 9 xfc PhS02^ \ ” 6 P h S O /\^ 6 J 71 72 60b W h A solution of triene 60b (304 mg, 1.0 mmol) in xylene (12 ml) was degassed as above and transferred via cannula to an argon-filled resealable pressure tube. The solution was heated to 170°C for 144 h, after which time an aliquot (ca. 10 mg) was withdrawn under a positive pressure of argon. !H nmr analysis of this crude mixture indicated ca. 50% conversion of 60b, and the presence of the cis- and trans-fused cycloadducts in a 10:3 ratio. The reaction was heated to 178°C for a further 168 h, allowed to cool and the solvent evaporated under reduced pressure. JH nmr analysis of the resultant dark brown oil showed it to contain the same two cycloadducts together with what appeared to be a third unassignable product, also containing a methyl doublet, at ca. 1.23 ppm. Purification by chromatography (15% - 20% ether-petrol) gave [3R*, 4R*, 5R*. 10S*]- 3,4 ~dimethyl-4~(phenylsulphonyl)bicyclo[4.4.0]-2-decern 72 (30 mg, 10%), followed by [3R*. 4S*. 5S*, 10S*]-3,4-dimethyl-4-(phenylsulphonyl)bicyclo[4.4.0]-2-decene 71 (18 mg, 6%) together with the unidentified product containing a methyl doublet Both 71 and 72 were unambiguously identified by comparison of their JH nmr spectra with those 197 of the products of méthylation of 50 and 53, respectively (see below). IMDA reaction of (2Z, 8E, 10E)-2-(phenylsuIphonyl)-2,8,10-dodecatriene (61b). 1 H 9 8 7 73 74 A solution of triene 61b (236 mg, 0.776 mmol) in dry xylene (10 ml) was degassed as above and transferred via cannula to an argon-filled rcsealable pressure tube. The solution was heated to 190°C for 120 h and allowed to cool. !H nmr analysis of the crude product showed the presence of a 1:2 mixture of cis- and trans-fused cycloadducts. Purification by chromatography (15% - 20% ether-petrol) gave, in order of elution, [3R*, 4R*. 5S*. 1 OS*]-3,4-dimethyl-4-(phenylsulphonyl)-bicyclol4.4.0]-2-decene 73 (51 mg, 22%), followed by [3 R *, 4S *, 5R *, 70S* ]-3,4 ‘dimethyl-4 - (phenylsulphonyl)bicyclo[4.4.0]-2-decene 74, (18 mg, 8%). Both 73 and 74 were unambiguously identified by comparison of their nmr spectra with those of the products of methylation of 50 and 53 respectively (see below). Methylation of [37?*, 4S*, 5S*, 9S*].3-methyl-4.(phenylsulphonyl). bicyclo[4.3.0]-2-nonene (57). 57 66 67 198 Sulphone 57 (contaminated with ca. 25 mol % 58 (by nmr); 17.3 mg, 0.063 mmol) was dried in vacuo over P20 5, dissolved in THF (0.6 ml) under argon and the solution cooled to -78°C. n-Butyllithium (27 pi of a 2.57M solution in hexanes, 0.07 mmol, 1.1 eq) was added dropwise to the stirred solution, giving a lemon yellow solution of the anion. Iodomethane (12 pi, 0.189 mmol, 3 eq) was then added, causing fading of the colour. After 15 min the reaction was allowed to warm to rt during which time a yellow colour developed. The mixture was poured into water (15 ml) and the aqueous phase was extracted with dichloromethane (3 x 10 ml). The combined organic layers were washed with water (3 x 10 ml), dried (MgS04) and concentrated under reduced pressure. iH nmr analysis of the crude product showed the presence of four isomeric products, indicating complete méthylation of both 57 and 58. Purification by chromatography (15% ether-petrol) gave, in order of elution, a mixture of 67 and 68 (3.2 mg and 5.8 mg, respectively (determined by >H nmr); 43% 68 based on 57) followed by a mixture of 66 and 69 (7.0 mg and 1.0 mg, respectively (determined by *H nmr); 51% 66 based on 57). Compounds 66,67,68 and 69 were unambiguously identified by comparison of their !H nmr spectra with those of the products of the IMDA reactions of trienes 60a and 61a (see above). Méthylation of [3 J?*t 4 JÏ*, 5J?*, 9S*]-3-methyl-4.(phenylsulphonyl). bicyclo[4.3.0]-2-nonene (58). Sulphone 58 (dried in vacuo over P20 5; 33.3 mg, 0.121 mmol) was dissolved in THF (1.2 ml) under argon and the solution cooled to -78°C. n-Butyllithium (52 pi of a 2.56M 199 solution in hexanes, 0.133 mmol, 1.1 C(j) was &ddcd dropwisc to the stirred solution, giving a lemon yellow anion colour. Iodomethane (23 ul, 0.363 mmol, 3 eq) was then added, causing slow decolourization. After 10 min the reaction was allowed to warm to rt, causing the development during 30 min of a deep yellow colour. The reaction mixture was poured into water (20 ml) and the aqueous phase extracted with dichloromcthane (3 x 20 ml). The combined organic layers were washed with water (3 x 20 ml), dried (MgS04), and concentrated under reduced pressure. Purification by chromatography (15% ether-petrol) gave, in order of elution, [3/?*, AR*, SR*, 95*]-3,4-dimethyl-4- (phenylsuIphonyl)bicyclo[4.3.0]-2-nonene 67 (13.3 mg, 38%) as a crystalline solid, identical by nmr to material prepared previously, followed by a colourless oil (19.8 mg) containing (by nmr) starting material 58 (4.4 mg, 13%), further 67 (4 mg, 12%; combined yield 50%) and its C-4 epimer 69 (11 mg, 31%), and finally further 58 (1 mg, 3%; total unreacted 58 amounted to 16%). The oil crystallized at -18°C to give [3/?*, 45*, 5 R*, 95*]-3,4-dimethyl-4-(phenylsulphonyl)bicyclo[4.3.0]-2-nonene 69, identical by nmr to material prepared previously (see above). Méthylation of [3K *, 4S *, 5S*, 105 *]-3-methyl-4-(phenyl. sulphonyI)bicydo[4.4.0]-2-decene (50). H 1 H 9 1 H 9 1 PhS02 50 70 72 1 H 9 + 7 2 0 0 A solution of sulphone 50 (dried in vacuo over P20 5; 100 mg, 0.344 mmol) in THF (3.5 ml) was cooled to -78°C under argon. n-Butyllithium (266 pi of a 1.42M solution in hexanes, 0.378 mmol, 1.1 eq) was added dropwise to the stirred solution and after 5 min iodomethane (52 pi, 0.833 mmol, 2.2 eq) was added via syringe to the bright yellow anion solution. The colour discharged slowly and after 10 min the reaction was allowed to warm to room temperature over 30 min and poured into water (15 ml). The aqueous phase was extracted with dichloromethane (3 x 20 ml) and the combined organic layers were dried (MgS04), filtered and evaporated under reduced pressure to give a colourless oil (111 mg). Purification by chromatography (15% ether-petrol) gave, in order of elution, [3R*, 4S*, 5S*, 10S*]-3-methyl-4-(2-methylphenylsulphonyl)bicyclo-{4.4.0]- 2-decene 74 (26 mg, 25%) as a crystalline solid, mp 137-139°C (benzene-petrol); umax (film) 2926, 2856, 1597, 1506, 1449, 1312, 1291, 1148, 1130, 1060, 807, 751, 706, 690 cm*1; 5 (500 MHz) 8.00 (1H, dd, J 8,2 Hz, ortho- proton on Ar), 7.52 (1H, td, J 8, 2 Hz, para-proton on Ar), 7.39 (2H, m, mefa-protons on Ar), 5.54 (2H, s, H-l, H-2), 2.82 (2H, m, H-4, H-10), 2.69 (3H, s, Ar-CH3), 2.68-2.60 (1H, m, H-3), 2.45-2.40 (1H, m, H-5), 1.70-1.40 (5H, m) and 1.25-1.15 (3H, m, all comprising H-6, H-7, H-8, H-9), 0.94 (3H, d, J 7.5 Hz, C-3 CH3); m/z (El) 277, 201, 183, 148 (M+ - C7H80 2S), 133, 119, 105, 91, 81, 77; (Cl) 322 (M + NH4+), 174, 163, 149 (M+ - C7H70 2S) (Found: (M + NH4+), 322.1840. C18H240 2S requires (M + NH4+), 322.1840), followed by [3R*. 4R*, 5S*, 10S*]-3,4-dimethyl-4-(phenylsulphonyl)bicyclo[4.4.0]-2- decene 72 (14.9 mg, 14%) as an oily solid; vmax (film) 2925, 2660, 2567, 2448, 2379, 1290, 1151, 1082, 760, 730, 692 cm*»; 5 (500 MHz) 7.88 (2H, dd, J 8, 2 Hz, ortho- protons on Ph), 7.70-7.50 (3H, m, meta- and para- protons on Ph), 5.62 (1H, dt, J 10, 3.5 Hz, H-2), 5.40-5.30 (1H, m, H-l), 2.70 (1H, dd, J 13, 2 Hz, H-10), 2.55-2.45 (1H, m, H-5), 2.35-2.25 (1H, m, H-3), 1.80-1.70 (2H, m, H-6eq, H-9eq), 1.66 (1H, m, H-9ax), 1.60 (3H, d, J 7 Hz, C-3 CH3), 1.50-1.33 (3H, m, H-7«,, H-8.,, H-8„), 1.30- 1.18 (1H, m, H-6ax), 1.23 (3H, s, C-4 CH3), 1.10-1.05 (1H, m, H-7«); m/z (El) 222, 201 177, 163 (M+ - S02Ph), 148, 133, 199, 105, 95, 81, 77, (Cl) 322 (M + NH4+), 163 (M+ - S02Ph) (Found: (M + NH4+), 322.1840. Ci 8H240 2S requires (M + NH4+), 322.1840), and Finally /JR*. 4S*, 5S*. 1 OS*]-3,4-dimethyl-4-(phenylsulphonyl)- bicyclo[4.4.0]-2-decene 70 (57 mg, 54%) as a crystalline solid, mp 86-87°C (benzene- petrol); umax (film) 3306, 2923, 2855,1551, 1447,1379,1291, 1147, 1078, 798, 754, 729, 693 cm 1; 6 (500 MHz) 7.85 (2H, dd, J 8 ,1.5 Hz, ortho- protons on Ph), 7.63-7.52 (3H, m, meta- and para- protons on Ph), 5.65 (1H, dt, J 10, 3.2 Hz, H-2), 5.55-5.53 (1H, m, H-l), 3.23 (1H, m, H-10), 2.98-2.96 (1H, m, H-3), 2.01 (1H, dt, J 12.5, 4 Hz, H-5), 1.78 (1H, dt, J 13.5, 2.5 Hz, H-9eq), 1.67 (1H, m, H-7eq), 1.53 ( 1H, tt, J 13.5, 4.5 Hz, H-9ax), 1.45-1.41 (2H, m, H- 8 eq, H- 6cq), 1.30 (1H, qd, J 12, 3 Hz, H- 6ax), 1.24 (3H, s, C-4 CH3), 1.20 (1H, qt, 13, 3.2 Hz, H- 8 ax), 1.11 (1H, qt, J 13, 3.2 Hz, H-7ax), 1.01 (3H, d, J 7.5 Hz, C-3 CH3); mfz (El) 218, 199, 184, 163 (M+ - S 0 2Ph), 162, 147, 118, 105, 95, 91, 77 (Found: C, 71.18; H, 8.03. ClgH 240 2S requires C, 71.01; H, 7.95%). Methylation of [3R *, 4R *, SR*, 105*]-3-methyl-4-(phenyl- sulphonyI)bicyclo[4.4.0]-2-decene (53). 1 H 9 H 53 1 H 9 8 7 75 76 2 0 2 Sulphone 53 (azeotropically dried with toluene (2 x 10 ml); 14.4 mg, 0.05 mmol) was dissolved in THF (0.8 ml) under argon and the solution cooled to -78°C. n-Butyllithium (39 ul of a 1.41 M solution in hexanes, 0.055 mmol, 1.1 eq) was added dropwise to give a yellow solution of the anion, followed by iodomethane (8 jil, 0.125 mmol, 2.5 eq). The reaction was stirred at -78°C for 10 min, then allowed to warm to room temperature and stirred overnight. Water (10 ml) was added and the aqueous phase was extracted with dichloromethane (3 x 10 ml). The combined organic layers were dried (MgS04) and concentrated under reduced pressure to give a colourless oil. Purification by chromatography (15% ether-petrol) gave, in order of elution, an inseparable mixture of /JR*, 4R*. JR*, 10S*]-3-methyl-4-(2-methylphenyl-sulphonyl)bicyclo[4.4.0)-2-decene 75 and [ 3 R *, 4R*, 5R *, 10S* ]-3,4-dimethyl-4-(2 -met hylphenyl- sulphonyl)bicyclo[4.4.0]-2-decene 76 (1 mg, 7%) as a semi-solid; umax (film) (mixture of 75 and 76) 2922, 2890, 1444, 1302, 1147, 1120, 791, 730, 690 cm'1; m/z (Cl) (36) 322 (M + NH4+), 163 (M+ - S02Ph), 141 (S02Ph), 108,91; (37) 336 (M + NH4+), 177 (M + - S0 2Ph), followed by /JR*, 4R*, JR *, 10S*]-3,4-dimethyl-4- (phenylsulphonyl)bicycIo[4.4.0]-2-decene 71 (0.5 mg, 3%); vmax (film) 2922, 2830, 1443, 1374, 1298, 1149, 1124, 1067,768, 731, 690 cm-*; 8 (250 MHz) 7.97 (2H, dd, J 8 , 2 Hz, ortho- protons on Ph), 7.65-7.46 (3H, m, meta- and para- protons on Ph), 5.57- 5.44 (1H, m, H-l), 5.28 (1H, d, J 9 Hz, H-2), 2.70 (1H, d, J 6.5 Hz, H-3), 2.40-2.30 (1H, m, H-10), 2.20-1.70 (5H, m) and 1.50-1.10 (4H, m, all comprising H-5, H- 6, H- 7, H-8 , H-9), 1.46 (3H, s, C-4 CH3), 1.32 (3H, d, J 7 Hz, C-3 CH3); m/z (Cl) 322 (M + NH4+), 180, 163 (M+ - S0 2Ph) (Found: (M + NH4+), 322.1840. C 18H240 2S requires (M + NH4+), 322.1841), and finally /JR*, 4S*, JR*, 10S*]-3,4-dimethyl-4- (phenylsulphonyl)bicyclo[4.4.0J-2-decene 73 (10.2 mg, 67%) as a crystalline solid, mp 128-129°C (benzene-petrol); vmax (film) 2922, 2850, 1590, 1444, 1379,1356, 1301, 1146, 1125, 1078, 1040, 850, 763, 724, 693 cm*1; 8 (500 MHz) 7.85 (2H, dd, J 8 , 1.5 Hz, ortho-protons on Ph), 7.62 (1H, m, para- proton on Ph), 7.50 (2H, m, meta-protons 203 on Ph), 5.49 (1H, ddd, J 10, 4.5, 2.5 Hz, H-2), 5.42 (1H, d, J 10 Hz, H-l), 2.57 (1H, m, H-3), 2.48 (1H, t with additional fine structure, J 9 Hz, H-10), 1.98 (1H, m, H ^ ) , 1.94-1.80 (3H, m, H- 6ax, H ^ , H-9eq), 1.74 (1H, m, H- 8 eq), 1.50 (1H, td, J 10, 3 Hz, H-5), 1.38 (1H, m, H- 8 ax), 1.34 (3H, s, C-4 CH3), 1.17 (1H, m, H-7ax), 0.95 (1H, qd, J 13, 3 Hz, H-9«), 0.89 (3H, d, J 7 Hz, C-3 CH3); m/z (Cl) 322 (M + NH4+), 180, 163 (Found: C, 71.11; H, 7.95. CjgH^C^S requires C, 71.01; H, 7.74%). Preparation of [1/?*, 2R*] (7E, 9E)-2.hydroxy-l-(methyl-sulphenyl)-l- (phenylsulphonyI)-7,9-undecadiene and [1/?*, 2S*]*(7£;, 9E)-2-hydroxy- l-(methyIsuIphenyI)-l-(phenylsulphonyl)-7,9-undeca-diene (79). To a stirred solution of methylsulphenyl(phenylsulphone )69 (dried in vacuo over P 20 5; 775 mg, 3.83 mmol, 1.1 eq) in THF (40 ml) under argon at -78°C was added n- butyllithium (2.64 ml of a 1.45M solution in hexanes, 3.83 mmol, 1.1 eq). To this yellow anion was added a solution of dienal 19b (530.2 mg, 3.48 mmol) in THF (5 ml). The yellow colour faded slowly and the reaction was allowed to warm to 0°C during 45 min. Saturated aqueous ammonium chloride (30 ml) was added and the aqueous phase extracted with DCM (3 x 50 ml). The combined extracts were washed with water (3 x 30 ml), dried (MgS04) and concentrated under reduced pressure to give a semi-solid. Purification by chromatography (20% - 50% ether-petrol) gave the title compounds 79 (1.254 g, 92%) as a colourless oil; umax (film) 3492 (br.), 3013, 2923, 2854, 1446, 1303, 1087, 989, 752, 717, 687 cm’1; 8 (270 MHz) 7.98 (2H, dd, J 8 , 2 Hz, ortho- protons on Ph), 7.75-7.50 (3H, m, meta- and para-protons on Ph), 6.08-5.90 (2H, m, H-8 , H-9), 5.62-5.44 (2H, m, H-7, H-10), 4.46-4.38 (1H, m, H-2 one diastereoisomer), 4.20-4.08 (1H, m, H-2 other diastereoisomer), 3.95 (1H, s, H-l one 204 diastereoisomer), 3.91 (1H, d, J 6.0 Hz, H-l other diastereoisomer), 3.68 (1H, s, OH one diastereoisomer), 3.39 (1H, d, J 6.0 Hz, OH other diastereoisomer), 2.10 (3H, s, SC//3), 1.78 (3H, d, J 6.5 Hz, H -ll one diastereoisomer), 1.75 (3H, d, J 6.5 Hz, H -ll other diastereoisomer), 1.60*1.20 ( 6H, m, H-3, H-4, H-5) used without further separation. Preparation of (l/?, 7 E, 9£)-l-(methylsulphenyl)-I-(phenyIsulphonyl). 1,7,9-undecatriene (77). 79 77 To a stirred solution of hydroxysulphones 79 (1.202 g, 3.39 mmol) in pyridine (35 ml) under argon at 0°C was added methanesulphonyl chloride (242 ml, 3.73 mmol, 1.1 eq). The reaction was allowed to warm to rt and after 2 h tic (50 % ether-petrol) indicated complete elimination. The reaction was quenched at 0°C by the addition of saturated aqueous sodium hydrogencarbonate (50 ml). The aqueous phase was extracted with DCM (3 x 50 ml) and the combined organic layers were washed with saturated aqueous sodium hydrogencarbonate (3 x 50 ml), water (50 ml), 2M aqueous hydrochloric acid (4 x 50 ml), water (50 ml), dried (MgSO,*) and concentrated under reduced pressure to give a near-colourless oil. Purification by chromatography (15% - 25% ether-petrol) gave (7E, 7E, 9E)-l-(methylsulphenyl)-l-(phenylsulphonyl)-l,7,9-undecatriene 77 (910.4 mg, 79%) as a pale yellow oil; umax (film) 3061, 3012, 2923, 2853, 1808, 1595, 1475, 1444, 1374, 1317, 1305, 1149, 1085, 1070, 1023, 988, 925, 874, 755, 722, 687, 633 cm’1; 5 (270 MHz) 7.96 (2H, dd, J 8 , 2 Hz, ortho -protons on Ph), 7.64-7.48 (4H, m, meta- and para -protons on Ph and H-2), 6.05-5.90 (2H, m, H- 8 , H-9), 5.70-5.42 (2H, m, H-7, H-10), 2.58-2.42 (2H, m, H-3), 2.21 (3H, s, SC// 3), 2.18-1.98 (2H, m, H- 6), 205 1.75 (3H, d, J 6.5 Hz, H -ll), 1.60-1.30 (4H, m, H-4, H-5); mlz (El) 336 (M+), 321, 223, 195, 179, 147, 120, 105, 87, 81 (Found: C, 64.02; H.7.20. C 18 H24 0 2S2 requires C, 64.24; H, 7.19%). Preparation of [1/?*, 25*, 3/?*, 4/?*, 55*, 10/?*]-3-methyl-l,2-epoxy-4- (phenylsulphonyl)bicyclo[4.4.0]decane (82) and [1/?*, 25*, 35*, 45*, SR*, 1 OS*]- 3 -m e th y l- 1 , 2 -epoxy *4-(phenyls ulphonyl)bicy clo [4.4.0]- decane (83). 50 82 83 To a stin-ed solution of alkene 50 (19 mg, 0.055 mmol) in DCM (1 ml) under argon was added m-CPBA (15.5 mg of 80% w/w reagent, 0.0715 mmol, 1.1 eq). After stirring for 20 h further m-CPBA (7.5mg of 80% w/w reagent, 0.0035 mmol, 0.5 eq) was added and tic (75% ether-petrol) showed complete reaction after a further 7 h. The reaction was diluted with DCM (20 ml) and washed with 10% aqueous sodium thiosulphate (2 x 20 ml), saturated aqueous sodium bicarbonate (2 x 20 ml), water (20 ml), dried (MgS04) and concentrated to give a colourless oil. Purification by chromatography (30% - 40% ether-petrol) gave two white solids: //R*. 2S*, 5R*, 4R*. 5S*, 1 OK*]-3-methyl-1,2- epoxy-4-(phenylsulphonyl)bicyclol4.4.0]decane 82 (3.1 mg, 16%) mp 93-94°C (benzene-petrol); umax (film) 2930, 2855, 1653, 1586, 1539, 1448,1289,1144, 1086, 1031, 935, 905, 855, 819, 778, 742, 717, 691, 616 cm*1; 8 (500 MHz) 7.90-7.50 (5H, m, Ph), 3.22 (1H, d, J 4.3 Hz, H-l), 3.13 (1H, t, J 4 Hz, H-2), 2.65 (1H, m, H-3), 2.61 (1H, s, H-4), 2.54 (1H, t, J 5 Hz, H-5), 2.35 (1H, m, H-10), 1.96 (1H, dd, J 13, 4Hz, H-9ax), 1-93 (1H, m, H-9eq), 1.72-1.69 (1H, m, H- 6eq), 1.66-1.53 (1H, m, H- 6ax)* 1.50-1.20 (4H, m, H-7), 1.12 (3H, d, J 7.3 Hz, C-3 Ctf3); mlz (El) 306 (weak; 206 M+), 165, 147, 135, 121, 105, 95, 81, 55, 41 (Found: C, 66.53; H, 7.43. C 17H22 0 3S requires C, 66.63; H, 7.24%); [1R*. 2S*, 3S*. 4S*, 5R *. 10S*]-3-methyl-l,2-epoxy- 4-(phenylsulphonyl)bicyclo-[4.4.0]decane 83 (9.8 mg, 49%); mp 145-146°C (benzene- petrol); vmax (film) 2930,2855, 1653, 1586, 1539, 1448,1289, 1144, 1086,1031,935, 905, 855, 819, 778, 742, 717, 691, 616 cn r1; 8 (500 MHz) 7.90-7.50 (5H, m, Ph), 2.99 (1H, t, 3.6 Hz, H-l), 2.97 (1H, dd, J 10.4, 6.0 Hz, H-4), 2.86 (1H, dd, J 4.1, 1.3 Hz, H-2), 2.71 (1H, quintet, J 7.0 Hz, H-3), 2.32 (1H, br. d, J 11 Hz, H-5), 2.17- 2.08 (2H, m, H-10, H-9 [one proton]), 1.78-1.70 (1H, m), 1.66-1.60 (1H, m), 1.52- 1.45 (1H, m), 1.42-1.20 (4H, m, all comprising H- 6, H-7, H-8 , H-9), 1.19 (3H, d, J 7.3 Hz, C-3 C//3); m/z (El) 306 (weak; M+), 213, 181, 165 (M+ - S0 2Ph), 147, 135, 121, 105, 95, 91. 81 (Found: C, 66.62; H, 7.21. Ci 7H220 3S requires C, 66.63; H, 7.24%). Preparation of [1**, 2S *, SR *, 10R •]-2-m«thyl-l-(phenyl- sulphonyl)bicyclo[4.4.0]decane (84). To a stirred suspension of 10% palladium-on-charcoal (15 mg, 0.0145 mmol Pd, 10 mol%) in ethyl acetate (1 ml) under argon was added a solution of alkene 50 (42.2 mg, 0.145 mmol) in ethyl acetate (2 ml). The mixture was purged with hydrogen and allowed to stir for 24 h under a positive pressure of hydrogen. Tic (3 x 20% ether-petrol) showed that the starting alkene had been replaced by a slightly less polar compound. The catalyst was removed by filtration through Celitc® and the filtrate was concentrated to give a white solid whose nmr (250 MHz) showed no olefin to be present. Purification by chromatography (15% ether-petrol) gave [1R*. 2S*. 5R*. 1 OR*)-2-methyl- 7- 207 (phenylsulphonyl)bicyclo[4.4.0]decane 84 (34.3 mg, 81%) as a white crystalline solid mp 155-157°C (ether-petrol); vmax (film) 2930,1448,1299,1274,1268,1146,740 cm* 1; s (500 MHz) 7.88 (2H, dd, J 8 , 2 Hz, ortho- protons on Ph), 7.62 (1H, m, para- proton on Ph), 7.55 (2H, m, mera-protons on Ph), 2.68 (1H, br. s, H-l), 2.42-2.36 (2H, m, H-2, H-10), 2.15 (1H, dt, J 13, 4 Hz), 2.03 (1H, tt, J 13, 5 Hz), 1.72 (1H, ddd, J 13, 4 Hz), 1.70-1.62 (1H, m) and 1.54-1.15 (9H, m, all comprising H-3, H-4, H-5, H-6, H-7, H-8 , H-9), 1.08 (3H, d, J 7 Hz, C-2 C// 3); m/z (El) 293 (M+), 151 (M+ - PhS02) (Found: C, 69.84; H, 8.27. C 17H2402S requires C, 69.82; H, 8.27%). Preparation of [IS*, 2S*, 5/?*, 9 S*]- 2 .methyl.l.(phenylsuIphonyl)- bicyclo[4.3.0]nonane (85). To a stirred suspension of 10% palladium-on-charcoal (2 mg, ca. 20 mol%) in ethyl acetate (1 ml), under argon was added a solution of alkene 58 (15.5 mg, 0.055 mmol) in ethyl acetate (1.5 ml). The mixture was purged with hydrogen and allowed to stir for 4 h under a positive pressure of hydrogen. Tic (3 x 20% ether-petrol) showed that the starting alkene had been replaced by a slightly less polar compound. The catalyst was removed by filtration through Celite® and the filtrate was concentrated to give a white solid. Chromatography (15% ether-petrol) gave [IS*, 2S*. 5R*, 9S*J-2-methyl-1 * (phenylsulphonyl)-bicyclo[4.3.0]nonane 85 (14.9 mg, 97%) as a white solid which was recrystallised to give needles mp 113-115°C (benzene-petrol); umax (film) 2932, 2868, 1448, 1303, 1268, 749, 739 cm*1; 5 (270 MHz) 7.90 (2H, dd, J 7, 1 Hz, ortho- protons on Ph) 7.65-7.45 (3H, m, meta- and para-protons on Ph), 3.08 (1H, dd, J 10,5 Hz, H- 208 1), 2.35-2.20 (1H, m, H-9), 1.90-2.05 (1H, m, H-2), 1.85-1.20 (9H, m) and 1.00-0.80 (2H, m, all comprising H-3, H-4, H-5, H- 6, H-7, H-8 ), 1.15 (3H, d, J 7 Hz, C-2 C//3); m/z (El) 278 (M+), 223, 193, 185, 179, 167, 158, 137, 95, 81, 77 (Found: C, 69.20; H, 7.87. C 16H220 2S requires C, 69.02; H, 7.97). Preparation of l-(ter/-butyIdiphenyIsilyloxy)-( 6,8 )-decadiene (54). To a stirred solution of ( 6£)-6,8 -decadien-l-ol 27b (100 mg, 0.65 mmol) in dry dichloromethane (0;5 ml) under argon was added DMAP (5 mg, catalytic) followed by triethylamine (109 pi, 0.715 mmol, 1.1 eq), and fm-butylchlorodiphenylsilane (186 pi, 0.715 mmol, 1.1 eq). The mixture was stirred at rt after which time tic (25 % ether- petrol) indicated that the reaction was complete. The reaction was poured into saturated sodium hydrogencarbonate solution (10 ml) and extracted with DCM (3 x 10 ml). The combined organic extracts were washed with saturated sodium bicarbonate (10 ml), water (10 ml), dried (MgS04), and concentrated under reduced pressure. The product was purified by chromatography (eluant 1% - 5% ether-petrol to give a colourless oil (153 mg, 60 %); umax (film) 3071, 3015, 2931, 2858, 1625, 1473, 1429, 1112 cm-1; 8 (270 MHz) 7.70 (4 H, dd, J 10 Hz, 3 Hz, ortho -protons on Ph), 7.45-7.35 ( 6H, m, meta- and para- protons on Ph), 6.10-5.90 (2H, m, H- 8 , H-7), 5.65-5.50 (2H, m, H-9, H- 6), 3.65 (2H, t, J 6.0 Hz, H-l), 2.05 (2H, m, H-5), 1.85 (3H, d, J 7 Hz, H-10), 1.40-1.30 (4H, m, H-4, H-3), 1.05 (9H, s, rm-butyl); m b (El) 392 (M+), 335 (M+ - C 4H6), 199, 183, 149, 135, 105, 81, 77 (Found C, 79.40; H, 9.00. C 26H36OSi requires C, 79.53; H, 9.24 %). 209 Miscellaneous Experiments Reaction of THP ethers (26a) with neat S02. THP ethers 26a (1.15g, 5.13 mmol) were heated to 95°C with hydroquinone (11 mg, 0.1 mmol, 2 mol%) and liquid sulphur dioxide (5 ml, 103 mmol, 20 eq) in a steel autoclave (pressure reached ca. 4 atm. at 95°C) for 23 h. When cool the dark oil was dissolved in dichloromethane and the solvent and excess sulphur dioxide evaporated under reduced pressure. The resultant dark oil was purified by chromatography (20% - 100% ethyl acetate-petrol) to give, in order of elution, 5,7-nonadien-l-ols (27a) (3:1 5E,1E: 5ZJE ratio by nmr; 292 mg, 25%), as a colourless oil, identical with material prepared previously, followed by [2"R*,5"S*]-2-[4-(2J-dihydro-5-methylthiopken-2- yl)butoxy]tetra-hydro-2H-pyran S,S-dioxide as a colourless oil (397mg, 27%); umax (film) 3512, 2944, 1625, 1307, 1135, 1077, 1033 cm*1; 5 (270 MHz) 5.90 (2H, m, H- 3", H-4"), 4.55 (1H, t, J 3.5 Hz, H -l’), 3.90-3.60 (4H, m, H -6 [one proton], H -l’ [one proton], H-2", H-5"), 3.50-3.30 (2H, m, H- 6' [one proton], H-l' [one proton]), 2.00-1.90 (1H, m, H-4 [one proton]), 1.80-1.40 (11H, m, H-3, H-4, H-5, H-2’, H-3', H-4' [one proton]), 1.35 (3H, d, J 7 Hz, CH3); mlz (El) 288 (M+, weak), 224 (M+ - SO2), 205, 187, 149, 140, 123, 107, 101, 94, 85, 81, 79 (Found C, 58.35; H, 8.48. C 14H24O4S requires C, 58.30; H, 8.39%), and finally [2^*,5'S*]-4 -(2 ¿-dihydro-5- methylthio-phen-2-ylhl-butanol S,S-dioxide 42 as a colourless oil (160 mg, 15%); \>max (film) 3384 (br.), 2940,1653,1453,1300,1127,1078 cm*»; 6 (270 MHz) 5.93 (2H, m, H-3\ H-4'), 3-78 (1H, m, H-5'), 3.73-3.64 (3H, m, H-l, H-2'), 2.00-1.90 (1H, m, H- 210 4 [one proton]), 1.70-1.50 ( 6H, m, H-2, H-3, H-4 [one proton], OH), 1.39 (3H, d, J 7Hz, CH3); mil (El) 205 (MH+), 186 (M+ - H 20), 174, 155, 140 (M+ - S02), 122, 94, 81,79, 77 (Found: C, 53.20; H, 8.08. C 9 H160 3S requires C, 52.91; H, 7.90%). Acidic methanolysis of ^2,,/?*,5,,S*].2-[4.(2,5-dihydro.5-methylthio- phen- 2-yl)butoxy]tetrahydro- 2tf-pyran S,S-dioxide . 4' 4 OH 42a / 2"/?*, 5"5 *]- 2-[4-(2,5-dihydro- 5-methylthiophen- 2-yl)butoxy]tetrahydro- 2//-pyran ¿¿-dioxide (283 mg, 0.98 mmol) was stirred at rt in dry methanol (10 ml) with a trace of CSA until tic (75% ethyl acetate-petrol) indicated complete conversion to the alcohol. The solution was concentrated under reduced pressure and the residue was dissolved in ethyl acetate (20 ml) and filtered through a pad of silica gel, washing it exhaustively with ethyl acetate. The product was then purified by chromatography (20% - 100% ethyl acetate- petrol) to yield [2'R*,5'S*]-4-(2,5-dihydro-5-methylthiophen-2-yl)-l-butanol S,S- dioxide 42 as a colourless oil (170 mg, 85%), identical with previously prepared material. 211 Experimental Section for Part II 212 Preparation of 2-(3-chIoropropoxy)tetrahydro-2f/-pyran (99 ).80 4’ 5 ' 6' Cl OH Cl 99 To a stirred solution of 3-chloro-l-propanol (100 g, 1.06 mol) in dry DCM (200 ml) in a flame-dried flask under argon at 0°C was added CSA (250 mg) and 3,4-dihydro-2H- pyran (106 ml, 97.9 g, 1.16 mol, 1.1 cq), dropwise from a pressure-equalized dropping funnel. After stirring overnight at rt, the solution was filtered through a pad of silica gel, rinsed exhaustively with ether and concentrated under reduced pressure to give a near- colourless mobile oil (194 g). Distillation yielded 2-(3-chloropropoxy)tetrahydro-2//- pyran 99 (177.9 g, 94%) as a colourless oil (bp 0.6 68-71°C); \>max (film) 2941, 2871, 2735, 1448, 1440, 1363, 1352, 1322, 1299, 1274, 1201, 1185, 1172,-1130, 1120, 1082, 1035, 1021, 989, 987, 869, 653 cnr*; 6 (250 MHz) 5.56 (1H, m, J 3.6, 3.3 Hz, H-2'), 3.90-3.75 (2H, m, H-l, H- 6’1X), 3.62 (2H, t, J 6.0 Hz, H-3), 3.53-3.40 (2H, m, H -l, H-6'eq) 2.00 (2H, quintet, J 6.0 Hz, H-2), 1.80 (2H, m) and 1.58-1.40 (4H, m, all comprising H-3', H-4', H-5’); mtz (El) 177 (MH+), 123, 85 (C s^O *) (Found: C, 53.86; H, 8.38. CgH^ClC^ requires C, 53.78; H, 8.38%). Preparation of 2-(3-iodopropoxy)tetrahydro*2J/-pyran (98).80 99 98 2-(3-Chloropropoxy)tetrahydro-27/-pyran 99 (50 g, 0.28 mol) was dissolved in dry acetone (500 ml) together with sodium iodide(88.1 g, 0.59 mol, 2.1 eq) and heated under reflux in the dark for 48 h. The reaction mixture was allowed to cool, filtered and 213 the filtrate evaporated. The resultant yellow oil was dissolved in ether (300 ml) and more white ppt. removed by filtration. The filtrate was then washed with 5% aqueous sodium metabisulphite (3 x 150ml), water (2 x 150ml) and brine (150 ml), dried (MgS04) and concentrated under reduced pressure to give a pale orange oil (72 g). Fractional distillation gave 2-(3-iodopropoxy)tetrahydro-2//-pyran 98 (50.4 g, 67%) as a colourless oil (bpo.oi 60-64°C); vmax (film) 2960, 2850, 1460, 1440, 1280, 1260, 1235, 1201, 1184, 1132, 1117, 1076, 1032, 980, 924, 906, 869, 815, 754 cn r1; 8 (250 MHz) 4.58 (1H, m, H-2'), 3.90-3.74 (2H, m, H-l, H- 6’ax), 3.55-3.38 (2H, m, H-l, H- 6’eq), 3.28 (2H, t, J 7.0 Hz, H-l), 2.08 (2H, quintet, J 7.0 Hz, H-2), 1.90-1.65 (2H, m) and 1.60- 1.45 (4H, m, all comprising H-3’, H-4', H-5'); m/z (El) 270 (M+), 269 (M+ - H), 169, 143, 85 (C 5H9 0 +), 84, 43, 41 (Found: C, 35.78; H, 5.89. C 8 H 15I0 2 requires C, 35.57; H, 5.60%) (Found: (M+- H), 269.0047. C 8 H 14I 0 2 requires (M+ - H), 269.0039). Preparation of 2-(l-methylethyl)*5-(tetrahydro-2tf.pyran-2-yIoxy) pentanoic acid ( 1 0 0 ) 100 Isos aleric acid (0.5 ml, 4.90 mmol) was added dropwise to a stirred suspension of sodium hydride (215 mg, 5.39 mmol, 1.1 eq; pre-washed with dry hexane [3x5 ml]) in THF (50 ml) and diisopropylamine (3.8 ml of a 1.43M solution in THF, 5.43 mmol, 1.1 eq) at rt. The mixture was then heated under reflux for 15 min, cooled to 0°C and n- butyllithium (2.0 ml of a 2.50M solution in hexanes, 4.90 mmol, 1.0 cq) was added to the rapidly stirred solution. After heating briefly to 30°C to complete the metallation, the reaction was cooled to 0°C and a solution of 2-(3-iodopropoxy)tetrahydro-2H-pyran 99 214 (1.46 g, 5.39 mmol, l.leq) in THF (10 ml) was added via cannula. After 5 h the reaction was quenched by the addition of 0.1 M HC1 ( ca. 75 ml) to pH 6.0-7.0 (yellow endpoint with bromothymol blue indicator). The aqueous phase was extracted with petrol (3 x 50 ml) to remove any unconsumed electrophile, then neutralised (2 ml of 0.1M HC1) and sodium chloride added until saturated. The mixture was then extracted with ethyl acetate (3 x 75 ml), dried (MgSO^ and concentrated under reduced pressure to give a pale yellow oil (510 mg) which was used without purification in the next reaction. The product could be purified by column chromatography (10% - 30% ether-petrol) to give 2- (l-methylethyl)-5-(tetrahydro-2H-pyran-2-yloxy)pentanoic acid 100 as an oil; \>max (film) 3010 (br.), 2954, 1701, 1450, 1370, 1200, 1119, 1076, 1034,905, 868 , 814 cm* 1; 5 (500 MHz) 4.57 (1H, dd, J 4.2, 3.0 Hz, H-2'), 3.84 (1H, ddd, J 11.2, 7.9, 3.1 Hz, H-5), 3.72 (1H, m, H-6'ax), 3.52-3.45 (1H, m, H-5), 3.37 (1H, m, H-6’eq), 2.14 (1H, m, H-2), 1.88 (1H, septet, J 6.5 Hz, C//(CH3)2), 1.85-1.75 (2H, m, H-4), 1.70-1.46 (8 H, m, H-3, H-3", H-4", H-5"), 0.94 ( 6H, d, J 6.7 Hz, CH(C//3)2), no C 02H discernible; mil (El) no M+, 213, 199, 159 (M+ - C 5H9 0+), 143 (M+ - C5H9 0 2+), 101 (C5H9 0 2+), 85 (C 5H9 0+). Preparation of 2-[4-(l-methylethyI)-5-oxohexyIoxy)]tetrahydro-2W*pyran (102). 100 102 To a rapidly stirred solution of acid 100 (302 mg, ca. 1.2mmol) in THF (12 ml) at 0°C under argon was added, in one portion, methyllithium (3.5 ml of a 1.4 M solution in ether, 4.95 mmol, 4 eq). The reaction was allowed to stir at below rt for 2.5 h then 215 added via cannula to a rapidly stirred solution of saturated aqueous ammonium chloride and ice (100 ml). The aqueous phase was extracted with ether (3 x 50 ml) and the combined organic layers washed with water (3 x 50 ml), brine (50 ml), dried (MgS04) and concentrated under reduced pressure to give a nearly colourless oil. Purification by chromatography (20% - 30% ether-petrol) furnished 2-[4-(l-methylethyl)-5- oxohexyloxy)]tetrahydro-2U-pyran 102 (127 mg, 42%) as a colourless oil; vmax (film) 2943, 2873, 1713, 1507, 1456, 1354, 1324, 1261, 1202, 1165, 1121, 1078, 1035, 989, 905, 869, 814 cm-l; 8 (270 MHz) 4.50 (1H, t, J 3.5 Hz, H-2'), 3.85-3.75 (1H, m, H-D , 3.66 (1H, dt, J 9.8, 6.5 Hz, H- 6tx), 3.50-3.40 (1H, m, H -l’), 3.31 (1H, dt, J 9.8, 6.0 Hz, H- 6«,), 2.08 ( 3H, s, H-6’) 1.86-1.70 (1H, septet, J 6.9 Hz, C//(CH3)2), 1.65-1.35 (8 H, m, H-3’, H-3, H-4, H-5), 0.85 ( 6H, d, J 6.6 Hz, CH(C//3)2); m/z (El) 242 (M+), 157 (M+ - C5H 9 0+), 141, 98, 85 (C 5H 9 O+), 71, 67 (Found: (M+), 242.1882. C 14H260 3 requires (M+), 242.1882). Preparation of methyl 2-(l-methyIethyl)-5-(tetrahydro-2//.pyran-2- yloxy)pentanoate ( 101). A solution of acid 100 (93 mg, 0.381 mmol) in DCM (10 ml) in a large boiling tube was treated with diazomethane generated by adding aqueous potassium hydroxide ( 0.6 ml of a solution containing 1.08 g in 6 ml water) to an ethanolic solution of Diazald® (0.4 g, 1.87 mmol in 7 ml of ethanol) in an adjacent boiling tube. The resultant gaseous diazomethane was bubbled into the DCM solution via a glass tube with fire-polished ends using a gentle stream of argon. The reaction mixture turned yellow, then decolourised 216 after bubbling was continued for 30 min. The solvent was removed under reduced pressure and the product purified by chromatography ( 10% ether-petrol) to give methyl 2- (l-methylethyl)-5-(tetrahydro-2H-pyran-2-yloxy)pentanoate (61.7 mg, 63%) xx as a colourless oil; \>max (film) 2941, 1734, 1558, 1455, 1372, 1324, 1261, 1121, 1078, 1036, 992, 906, 870, 815, 771 cm**; 8 (270 MHz) 4.54 (1H, m, H-2’), 3.90-3.80 (1H, m, H-6'«), 3.75-3.65 (1H, m, H-5), 3.64 (3H, s, OC//3), 3.52-3.42 (1H, m, H-6’eq), 3.40-3.30 (1H, m, H-5), 2.13 (1H, br. q, J 5.7 Hz, H-2), 1.81 (1H, m, Ctf(CH3)2), 1.80-1.70 (2H, m, H-4), 1.70-1.40 ( 8 H, m, H-3, H-3’, H-4’, H-5'), 0.92 (3H, d, J 5.8 Hz, CH(C//3)(CH3)), 0.85 (3H, d, J 5.8 Hz, CH(CH 3)(C//3)); mlz (El) 173 (M+ - C5H9 0), 157 (M+ - C5H9 0 2), 143, 115, 101, 97, 85 (C 5H9 0 +), 73 (Found: (M+ - C5H9 0), 173.1178. C 9 H170 3 requires (M+ - C 5H9 0), 173.1178). Preparation of 2-(4-cyano-5-methylhexyIoxy)tetrahydro-2//-pyran (107). To a stirred solution of diisopropylamine (11.5 ml, 81.7 mmol, 1.1 eq) in THF (200 ml) under argon at -78°C was added n-butyllithium (33 ml of a 2.50M solution in hexanes, 81.7 mmol, 1.1 eq). After 20 min, wovaleronitrile ( 8.6 ml, 81.7 mmol, 1.1 eq) was added dropwise via syringe. The anion was added via cannula at -78°C to a stirred solution of 2-(3-iodopropoxy)tetrahydro-2//-pyran 98 (20 g, 74.3 mmol) in THF (200 ml) at -78°C over a period of 90 min. After a further 20 min at -78°C the reaction was allowed to warm to rt, by which time it had taken on a red colour. Water (300 ml) was added, discharging the colour and the organic phase was separated. The aqueous phase was extracted with ether (2 x 200 ml), the organic layers were combined and washed with water (3 x 200 ml) and brine (200 ml), dried (MgSO,*) and concentrated under reduced 217 pressure. The residue was purified by chromatography (10% - 30% ether-petrol gradient, 5% increments) to give 2-(4-cyano-5-methylhexyloxy)tetrahydro-2H-pyran (8.43 g, 50%) 107 as a colourless oil; vmax (film) 2943, 2875, 2237, 1559, 1470, 1373, 1353, 1324, 1261, 1201, 1122, 1078, 1035, 991, 905, 869, 815 cm-l ;8 (270 MHz) 4.56 (1H, t, J 3.2 Hz, H-2'), 3.90-3.70 (2H, m, H-l, H- 6'), 3.55-3.35 (2H, m, H-l, H-6'), 2.47 (1H, q, J 4.1 Hz, H-4), 1.95-1.60 (7H, m) and 1.60-1.45 (4H, m, all comprising H-2, H-3, H-5, H-3’, H-4’, H-5’), 1.05 ( 6H, d, J 6.9 Hz, H- 6, H-5"); m/z (El) 225 (M+), 224 (M+- H), 183, 142, 124, 85, 82 (Found: (M+- H), 224.1653. C13H23NO2 requires (M+ - H), 224.1657). Preparation of 4-cyano-5-methylhexanol (109). 107 109 2 -(4 -Cyano- 5 -methylhexyloxy)tetrahydro- 2 //-pyran, 105 (7.053 g, 31.3 mmol) was dissolved in dry methanol (150 ml) with a catalytic amount of CSA and stirred at rt for 18 h, after which time tic (40% ether-petrol) showed complete deprotection of the alcohol. The solution was concentrated to1/4 of the original volume, dissolved in ether(100 ml) and filtered through a pad of silica gel, rinsing with more ether (700 ml), to give, after removal of the solvents under reduced pressure, a near-colourless oil(4.81 g). This was used crude in the next, reprotection step. A small portion was purified by chromatography (20% -100% ether-petrol) to give4-cyano-S-methylhexanol 109 as a colourless oil; vmax (film) 3438 (br.), 2960, 2235, 1464, 1389, 1373, 1349, 1262, 1233, 1176, 1135, 1062, 1002, 967, 931 cnr1; 5 (270 MHz) 3.75-3.60 (2H, m, H-l), 2.45 (1H, dt, J 6.8 , 5.4 Hz, H-4), 1.85 (1H, m, H-5), 1.75-1.60 (4H, m, H-2, H-3), 1.06 (6H, d, 6.9 Hz, H- 6, H-5’); mlz (El) 140 (M+ - H), 126, (M+ - CH3), 122 (M+ - 218 H20), 111, 108 (M+ - H20 - CH3) 99, 96, 83, 81, 71, 69, 43; (Found C, 68.15; H, 10.74; N, 9.89. C 8 H 15NO requires C, 68.05; H, 10.71; N, 9.92%). Preparation of l-((l,l-diinethylethyl)diphenylsilyloxy)-4-cyano-5- methylhexane (110). To a solution of crude alcohol 109 (4.81 g, ca. 31.3 mmol) in DCM together with DMAP (38 mg, 0.31 mmol, 0.01 eq) under argon at 0°C was added triethylamine (5.3 ml, 37.6 mmol, 1.2 eq) followed by rerr-butylchlorodiphenylsilane (9.8 ml, 37.6 mmol, 1.2 eq). The reaction was allowed to stir for 30 min, then warmed to rt and stirred for a further 4 h after which time more triethylamine (436 pi, 3.13 mmol, 0.1 eq) and tert- butyldiphenylchlorosilane (814 pi, 3.13 mmol, 0.1 eq) were added. The reaction was quenched after another hour by the addition of saturated aqueous sodium hydrogencarbonate (100 ml) and the aqueous phase extracted with DCM (3 x 70 ml). The combined organic layers were washed with saturated aqueous ammonium chloride (3 x 100 ml), water (3 x 50 ml), dried (MgS04) and concentrated under reduced pressure. The yellowish residue was purified by chromatography (2% - 20% ether-petrol gradient elution; 5% increments) to give l-((l,l-dimethylethyl)-diphenylsilyloxy)-4-cyano-5- methylhexane 110 (9.44 g, 80% from 107) as a colourless oil; umtx (film) 3076,2963, 2862, 2237, 1540,1472, 1391,1362,1113,998, 824 cm*»; 8 (270 MHz) 7.66 (4H, dd J 8 , 2.2 Hz, ortho-protons on Ph), 7.45-7.35 ( 6H, m, meta- and para- protons on Ph), 3.71 (2H, t, J 5.4 Hz, H-l), 2.42 (1H, m, H-4), 1.84 (1H, m, J 6 Hz, H-5), 1.75-1.60 (4H, m, H-2, H-3), 1.08 (9H, s, rerf-butyl), 1.05 ( 6H, d, J 7.1 Hz, H-6, H-5); mtz (El) 379 (M+), 322 (M+ - C 4H9 ), 225, 199, 183, 148, 135, 105, 91, 77; (Found: (M+ - 219 C4H9 ), 322.1627. C2oH24NOSi requires (M+ - C 4H9 ), 322.1627). Preparation of l.((l,l-dimethylethyI)diphenyIsilyloxy).5-methyl-4. oxomethylhexane ( 111). To a stirred solution of nitrile 110 (12.46 g, 32.8 mmol) in toluene (330 ml) under argon at -78°C was added DIBAL-H (66 ml, of a 1.5 M solution in toluene, 98.5 mmol, 3 eq) drop wise via cannula. After 20 min the reaction was quenched by the addition of aqueous THF (59 ml of a 10% v/v solution, 328 mmol, 10 eq) (CAUTION! allow for the evolution of a large volume of gas). The reaction was allowed to stir to rt and at about 0°C it exothermed slightly and assumed a paste-like consistency. The mixture was poured onto solid sodium hydrogencarbonate and diluted with ethyl acetate (800 ml). After stirring vigorously for 30 minutes the now freely-flowing solid was removed by filtration through a pad of Celite® and the solution concentrated under reduced pressure. The residue was purified by chromatography (5% - 10% ether-petrol) to give 1-((1,7- dimethylethyl)diphenyl-silyloxy)-5-methyl-4-oxomethylhexane 111 (11.24 g, 90%) as a colourless oil; umax (film) 3072, 2959, 2859, 1726, 1591, 1539, 1473, 1429, 1390, 1362, 1114,999, 823, 742, 702 cm*1; 8 (250 MHz) 9.60 (1H, d, J 3.2 Hz, H-4'), 7.68 (4H, d, J 8 , 2 Hz, ortho- protons on Ph), 7.45-7.30 ( 6H, m, meta- and para-protons on Ph), 3.65 (2H, t, J 5.7 Hz, H-l), 2.10-1.90 (2H, m, H-4, H-5), 1.70-1.40 (4H, m, H- 2, H-3), 1.07 (9H, s, rerf-butyl), 0.97 (3H, d, J 6.4 Hz, CH(C// 3)(CH3), 0.96 (3H, d, J 6.4 Hz, CH(CH3)(C//3); m/z (El) 381 (M+ - H), 325 (M+ - C 4H9 ), 295, 269, 199, 183, 139, 109, 105, 91,77; (Found C, 75.48; H, 9.22. C 24H3402Si requires C, 75.48; H, 8.96%) (Found: (M+ - C 4H9 ), 325.1626. C20H25O2Si requires (M+ - C 4H9 ), 2 2 0 325.1624). Preparation of (lZ)-6-((l,l-dimethylethyI)diphenylsiIyloxy)-l-iodo-3-(l- methylethyl)-l*hexene ( 112). Ill 112 To a stirred suspension of iodomethyltriphenylphosphonium iodide (31.15 g, 58.8 mmol, 2.0 eq) in THF (250 ml) at rt under argon was added NaHMDS (58.8 ml of a 1.0M solution in THF, 58.8 mmol, 2.0 eq) via cannula. After 20 min the deep red solution of phosphorane was cooled to -78°C and a solution of aldehyde 111 (11.24 g, 29.4 mmol, 1.0 eq) in THF (50 ml) was added via cannula. After 15 min the reaction was allowed to warm to rt and quenched, after a further 30 min, with saturated aqueous ammonium chloride solution (300ml). The aqueous phase was extracted with ether (3 x 150 ml), discarding as much solid triphenylphosphine oxide as possible. The combined organic layers were washed with 5% aqueous sodium metabisulphite (3 x 200 ml), water (3 x 200 ml), brine (3 x 200 ml), dried (MgSC^) and concentrated to give a yellow oil. This oil was triturated with 5% ether-petrol and the white solid removed by filtration through a short pad of silica gel. After removal of the solvents under reduced pressure the product was purified by chromatography (3% -10% ether-petrol) to give (Z)-6-((l,l- dimethylethyl)diphenylsilyIoxy)-l-iodo-3-( 1 -methylethyl)-!-hexene 112 (11.96 g, 80%) as a colourless oil; vmax (film) 3070, 2958, 2860, 1624, 1570, 1559, 1541,1458,1429, 1387, 1301, 1266, 1112, 702 cm*1; 5 (270 MHz) 7.78 (4H, m, ortho- protons on Ph), 7.45-7.32 (6H, m, meta- and para - protons on Ph), 6.25 (1H, d, J 7.3 Hz, H-l), 5.89 (1H, dd, J 7.4, 9.7 Hz, H-2), 3.64 (2H, t, J 6.0 Hz, H- 6), 2.25 (1H, m, H-3), 1.70- 221 1.25 (5H, m, H-4, H-5, C//(CH3)2)), 1.05 (9H, s, /err-butyl), 0.91 (3H, d, 6.8 Hz, CH(C//3)(CH3)), 0.87 (3H, d, 6.8 Hz, CH(CH3)(C//3)); m/z (El) No M+, 449 (M+ - C4H9 ), 309, 199 (Ph 2SiOH+), 183, 123, 105, 81, 77 (Found: (M+ - C 4H9 ), 449.0798. C2iH36lOSi requires (M+ - C 4H9 ), 449.0798). Preparation of 6*((l,l-dimethylethyl)diphenylsiIyloxy)-4*(l>inethyIethyl)< 1-hexyne (113). 112 113 To a stirred solution of vinyl iodide 112 (8.55 g, 16.88 mmol) in THF (170 ml) under argon at -78°C was added potassium rm-butoxide (34 ml of a 2.0M solution in THF, 33.76 mmol, 2 eq) and then allowed to warm to rt over a period of 30 min. The reaction was quenched after 10 min by the addition of saturated aqueous ammonium chloride (200 ml). The aqueous phase was extracted with ether (3 x 150 ml) and the combined organic layers were then washed with water (3 x 100 ml), brine (2 x 100 ml), dried (MgSC> 4) and concentrated under reduced pressure to give a pale yellow oil. Purification by chromatography ( 1% - 10% ether-petrol) provided 6-((l ,1-dimethylethyl)- diphenylsilyloxy)-4-(l-methylethyl)-l-hexyne 113 as a colourless oil (5.569 g, 87%); vmax (film) 3305 (sharp), 3072, 2960, 2932, 2863, 1653, 1559, 1463, 1429, 1358, 1112, 823, 740,702 cm*1; 8 (270 MHz) 7.66 (4H, dd, J 8.0, 2.0 Hz, ortho -protons on Ph), 7.45-7.32 (6H, m, meta- and para- protons on Ph), 3.68 (2H, t, J 4.7 Hz, H- 6), 2.20 (1H, m, H-3), 2.02 (1H, d, J 2.4 Hz, H-l), 1.80 (1H, m, Ctf(CH3)2), 1.70-1.40 (4H, m, H-4, H-5), 1.05 (9H, s, rerr-butyl), 0.95 ( 6H, d, J 7.1 Hz, CH(C//3)2); m/z (El) no M+, 321 (M+ - C 4H9 ), 279, 265, 243, 217, 199, 183, 163, 105, 77 (Found: C, 2 2 2 79.27; H, 9.22. C 25H34OSi requires C, 79.31; H, 9.05%) (Found: (M+ - Q H 9 ), 321.1675. C21H25OSi (M+ - C4H9 ), requires 321.1674). Preparation of 4-(l-methyIethyl)-5-hexyn-l-oI (105). 113 105 To a stirred solution of alkyne 113 (6.405 g, 16.92 mmol) in THF (17 ml) under argon was added TBAF (34 ml of a 1.0M THF solution, 33.83 mmol, 2.0 eq) at rt. After 15 min, tic (30% ether-petrol) indicated complete desilylation and the reaction was quenched by the addition of saturated aqueous ammonium chloride (100 ml). The aqueous phase was extracted with ether (3 x 50 ml) and the combined organic layers washed with water (3 x 50 ml), brine (2 x 50 ml), dried (MgS04) and concentrated under reduced pressure to give a yellow oil. Purification by chromatography furnished 4-(l-methylethyl)-5- hexyn-l-ol 105 (2.36 g, 99%) as a colourless oil; vmax (film) 3609 (br.), 3308 (sharp), 2962, 2874, 1557, 1539, 1457, 1370, 1060 cm*l; 8 (270 MHz) 3.68 (2H, t, J 6.2 Hz, H-l), 2.24 (1H, m, H-4), 2.04 (1H, d, J 2.4 Hz, H-6), 1.80 (1H, m, C//(CH3)2), 1.70-1.40 (4H, m, H-2, H-3), 0.96 (3H, d, J 6.1 Hz, CH(Ctf 3)(CH3)), 0.94 (3H, d, J 6.1 Hz, CH(CH3)(C//3)); mlz (El) no M+, 123 (M+ - OH), 100, 81 (QHg) (Found: C, 77.38; H, 11.60. C9H160 requires: C, 77.09; H, 11.50%) (Found: (M+ - OH), 123.1174. C9 H15 requires (M+ - OH), 123.1174). Preparation of (£)-l*iodo-2-methyI-3-(l-methyIethyl).l-hexen»6-oI (xx). To a stirred solution of zirconocene dichloride (1.97 g, 6.75 mmol, 0.4 eq) in 1,2- dichloroethane (25 ml) under argon (bubbler) , was added trimethylaluminium (pyrophoric! 26 ml of a 2.0M solution in hexanes, 50.61 mmol, 3.0 eq). After 10 min the reaction was cooled to 0°C and treated with a solution of alcohol 105 (2.366 g, 16.87 mmol) in 1,2-dichloroethane (25 ml) via cannula. The yellow solution was allowed to warm to rt and stirred overnight, after which time it had assumed an orange colour. The solution was then cooled to -30°C and a solution of iodine (4.71 g, 18.56 mmol, 1.1 eq) in THF (20 ml + 2 x 5 ml rinse) was added via cannula. After stirring for a further 30 min the reaction was allowed to warm to 0°C whereupon it was quenched by the addition of saturated aqueous potassium carbonate (5 ml; NOTE: maintain rapid stirring and allow for the evolution of a large quantity of gas). The mixture was then poured onto solid sodium hydrogencarbonate and diluted with ethyl acetate (200 ml). After 30 min stirring the free-flowing white solid was removed by filtration through Cclite® and rinsed thoroughly with ethyl acetate. The filtrate was concentrated under reduced pressure to give a yellow oil. Purification by chromatography (25% - 75% ether-petrol) gave (E)-7- iodo-2-methyl-3-(l-methylethyl)-l’hexen-6-ol 115 (3.95 g, 83%) as a colourless oil; vm„ (film) 3334 (br.), 2957, 2870, 1613, 1469, 1386, 1270, 1150, 1063, 911, 771, 663 cm-1; 5 (270 MHz) 5.83 (1H, d, J 1.0 Hz, H-l), 3.60 (2H, dt, J 6.4, 1.2 Hz, H- 6), 1.87 (1H, dt, J 10.0, 4.0 Hz, H-3), 1.67 (3H, d, J 1 Hz, C-2 C//3), 1.64-1.20 ( 6H, m, H-4, H-5, C//(CH3)2, OH), 0.91 (3H, d, J 6.6 Hz, CH(C//3)(CH3)), 0.75 (3H, d, J 6.6 Hz, CH(CH3)(C//3)); m/z (El) 282 (M+), 239 (M+ - 'Pr), 221, 195, 181, 155 (M+ - I), 128, 94, 91, 71 (Found: (M+ - 'Pr), 238.9927. C 7H 12IO requires (M+ - 'Pr), 224 238.9933). During the attempted palladium (0) mediated coupling directly with vinyl bromide, proton quench resulted giving 4-(l-methylethyl)-5-methyl-5-hexen-lol 114 as a colourless oil; umax (film) 3313 (br.), 3073, 2936, 1645,1456,1376, 1166, 1063, 1001, 975, 888 cm- 1; 8 (270 MHz) 4.74 (1H, dt, J 3.9, 1.4 Hz, H-6), 4.64 (1H, dd, J 2.1, 0.5 Hz, H-6), 3.60 (2H, td, J 6.5, 2.2 Hz, H-l), 1.57 (3H, d, J 1.5 Hz, C-5 C//3), 1.65-1.15 (7H, m, H-2, H-3, H-4, C//(CH3)2, OH), 0.90 (3H, d, J 6.3 Hz, CH(Ctf3)(CH3)), 0.80 (3H, d, J 6.3 Hz, CH(CH3)(C//3)); m/z (El) 156 (M+), 138 (M+ - H20), 141 (M+ - CH3), 123, 112, 109, 97, 95, 81, 69, 67, 57, 55, 43 41 (Found: (M+), 156.1516. C10H20O requires (M+), 156.1514). Preparation of (5E, 7E)-5-methyl-4-(l-methylethyl)-5,7-octadienol (xx). 104 A solution of vinyl iodide 115 (3.27 g, 11.58 mmol) in toluene (100ml + 20 ml rinse) was added via cannula to a flask containing tetrakis(triphenylphosphine) palladium (0) (670 mg, 0.579 mmol, 5 mol%) under argon and stirred for 20 min. To this mixture was added vinyl magnesium bromide (34.7 ml of a 1M solution in THF, 34.7 mmol, 3 eq) at 0°C. After 20 min, tic (2 x 40% ether-petrol) indicated that the reaction was complete and it was quenched with saturated aqueous ammonium chloride (100 ml). The aqueous 225 phase was extracted with ether (3 x 100 ml) and the combined organic layers were washed with water (3 x 75 ml), brine (2 x 75 ml), dried (MgSC^) and concentrated under reduced pressure to give a red oil. Chromatography (20% - 70% ether-petrol gradient; 10% increments) gave a pale yellow oil which was decolourised by dissolving it in 1:1 ether-petrol (75 ml), treating with activated charcoal and filtering through a pad of Celite®. The product was then re-chromatographed (30% - 60% ether-petrol) to give (5E, 7E)-5-methyl~4-(l-methylethyl)-5,7-octadienol 104 (1.713 g, 81%) as a colourless oil; umax (film) 3382 (br.), 3084, 2956, 1647, 1559, 1508, 1456, 1385, 1166, 1063, 988, 898 cm-1; 5 (500 MHz) 6.59 (1H, ddd, J 16.8, 10.8, 10.3 Hz, H-7), 5.82 (1H, d, J 11 Hz, H-6, allylic coupling on one signal); 5.08 (1H, dd, J 16.8, 2.1 Hz, H- 8 ^ ) , 4.98 (1H, dd, J 10.1, 2.1 Hz, H- 8 cis), 3.60 (2H, m, J 6.2 Hz, H-l), 1.62 (3H, d, J 1.2 Hz, C-5 CH3), 1.65-1.50 (2H, m, H-4, CH(CH3)2), 1.45 and 1.33 (1H, m, H-3 one proton each), 1.35-1.25 (2H, m, H-2), 1.20 (1H, br. s, OH), 0.92 (3H, d, J 6.3 Hz, CH(C//3)(CH3)), 0.78 (3H, d, J 5.7 Hz, CH(CH 3)(C//3)); m/z (El) 182 (M+), 123 (M+ - C3H6), 121, 105 (C9 H 160 - H20), 95, 93, 79, 77, 55, 41 (C 3H5) (Found: (M+), 182.1671. C 12H22O requires (M+), 182.1671). i Preparation of (SE, 7E)-5-methyl-4-(l-methylethyl)*5,7-octadienal (xx). 104 To a stirred solution of oxalyl chloride (351 pi, 4.03 mmol, 2 eq) in DCM at -60°C under argon was added DMSO (572 pi, 8.05 mmol, 4 eq) as a solution in DCM (5 ml). After 5 min a solution of dienol 104 (367 mg, 2.013 mmol) in DCM (5 ml + 2 ml rinse) was added via cannula. After 20 minutes, triethylamine (1.4 ml, 10.06 mmol, 5 eq) was added via syringe; the reaction was stirred for 10 min at -60°C, then allowed to warm to rt 226 over a period of 45 min. The mixture was poured into ether - water (1:1, 100ml), organics separated and the aqueous phase extracted with ether (2 x 50 ml). The combined organic layers were then washed with saturated aqueous ammonium chloride (3 x 50 ml), water (3 x 50 ml), brine (50 ml), dried (MgSC^) and concentrated under reduced pressure to give a yellow oil. Chromatography (3% - 8 % - 15% - 20% ether-petrol) gave (5E, 7E)-5-methyl-4-(l-methylethyl)-5,7-octadienal 116 (307 mg, 85%) as a colourless o il; vmax (film) 3085, 2958, 1727, 1645, 1598, 1570, 1539, 1470, 1386, 989, 900 cm*l; 8 (500 MHz) 9.74 (1H, t, J 1.6 Hz, H-l), 6.58 (1H, dt, J 16.7, 10.8 Hz, H-7), 5.80 (1H, d, J 10.8 Hz, H- 6; allylic coupling on one signal), 5.11 (1H, dd, J 16.8, 2.0 Hz, H- 8 tr.ns). 5 00 (1H> dd’ J 10-2’ 2 0 Hz* H"8 ci.)- 2-30 <2H* H*2)’ L94 OH, m’ H’4)- 1.61 (3H, d, J 1.0 Hz, C-5 C//3),1.58-1.46 (3H, m, H-3, C//(CH3)2), 0.95 (3H, d, J 6.1 Hz, CH(0 / 3)(CH3)), 0.79 (3H, d, J 6.2 Hz, CH(CH 3)(C//3)); mlz (El) 180 (M+), 162 (M+ - H20), 121 (M+ - H20 - C3H5), 119, 109, 95, 93, 81, 79, 67, 55, 41 (Found: (M+), 180.1511. C 12H20O requires (M+), 180.1514). Preparation of (6E, 8E)-6-methyI-5-(l-methylethyl).l-(phenylsulphonyl). 6,8-nonadien-2-ol (117). OH 117 116 To a stirred solution of (phenylsulphonyl)methane (398 mg, 2.546 mmol, 1.1 eq) in THF (15 ml) under argon at -78°C was added dropwise n-buyllithium (1.11 ml of a 2.30M solution in hexanes, 2.546 mmol, 1.1 eq) to give a slightly yellow anion. After 10 min a solution of aldehyde 116 (422 mg, 2.315 mmol) in THF (10 ml + 5 ml rinse) was added via cannula. After 20 min tic (50% ether-petrol) showed complete consumption of the aldehyde and the reaction was quenched at -78°C by the addition of acetic acid (2.7 ml of 227 a 1.75M solution in THF (10% v/v), 4.63 mmol, 2.0 eq) and allowed to warm to rt. Saturated aqueous sodium hydrogencarbonate (30 ml) was added and the aqueous phase extracted with DCM (3 x 30 ml). The combined organic layers were washed with saturated aqueous ammonium chloride solution (3 x 50 ml), water (3 x 50 ml), brine (50 ml), dried (MgSC^) and concentrated under reduced pressure to give a yellow oil. Purification by chromatography (10% - 55% ether-petrol gradient, 5% increments) furnished (6E, 8E)-6-methyl-5-(l-methylethyl)-l-(phenyl-sulphonyl)-6,8-nonadien-2-ol xx (653 mg, 84%) as a colourless oil; vmax (film) 3526 (br.), 2953, 1646,1448, 1385, 1305, 1149, 1088, 901, 785, 747,719, 689 cm’1; 8 (500 MHz) 7.92 (2H, dd, J 7.3, 1.7 Hz, ortho- protons on Ph), 7.72-7.55 (3H, m, meta- and para- protons on Ph), 6.53 (1H, dt, J 16.5, 10.8 Hz, H- 8 ), 5.71, 5.73 (1H each, d, J 10.8 Hz H-7, two diastereoisomers), 5.07, 5.00 (1H each, dd, J 7.0, 1.9 Hz, H-9cis two diastereoisomers), 4.95 (1H, br. d, J 10.8 Hz, H-9 both diastereoisomers), 4.20- 4.06 (1H, m, H-2), 3.37-3.31 (1H, m, O-H both diastereoisomers), 3.18 (1H, dd, J 7.6, 3.7 Hz, H-l one diastereoisomer), 3.15 (1H, dd, J 7.6, 1.9 Hz, H-l other diastereoisomer), 1.80-1.10 ( 6H, m, H-3, H-4, H-5, C//(CH3)2), 1.56 (3H, s, C-6 C//3), 0.90, 0.88 (3H, d, J 6.5 Hz, CH(C// 3)(CH3) both diastereoisomers), 0.75, 0.73 (3H, d, J 6.5 Hz, CH(CH3)(C//3) both diastereoisomers); mlz (El) 336 (M+), 293, 275, 199, 195, 133, 123, 93, 81, 77 (Found: (M+), 336.1766. C 19 H 28 0 3S requires (M+),336.1759). Preparation of (1 £, 6E, 8E)-6-methyl-5-(l-methylethyl)-l.phenyl- su!phonyl-l,6,8-nonatriene (94). 117 94 228 To a rapidly stirred solution of hydroxysulphones 117 (653 mg, 1.941 mmol) in DCM (20 ml) at -6 °C under argon was added triethylamine (2.7 ml, 19.41 mmol, 10 eq) followed by methanesulphonyl chloride (378*0, 5.82 mmol, 3 eq). A white ppt developed and after 30 min tic (1:1 ether-petrol) showed no starting material and so the reaction was poured into saturated aqueous sodium hydrogencarbonate (50 ml). The aqueous phase was extracted with DCM (3 x 40 ml) and the combined organic layers were washed with water (3 x 20 ml), dried (MgS04) and concentrated under reduced pressure to a yellow oil. Purification by chromatography (15% - 30% ether-petrol) gave (]E, 6E, 8E)-6-methyl-5-(l -methylethyl)-] -phenylsulphonyl-1,6,8-nonatriene 94 (425 mg, 69%) as a colourless o il; vmax (film) 2958, 1641, 1448, 1385, 1320, 1148, 1087, 989, 901, 818, 754, 716, 689 cm*1; 8 (270 MHz) 7.88 (2H, dd, J 8.1, 1.6 Hz, meta- protons on Ph), 7.60-7.48 (3H, m, ortho- and para- protons on Ph), 7.95 (1H, ddd, J 14.5, 7.2, 6.0 Hz, H-2), 6.52 (1H, ddd, J 16.7, 10.8, 10.3 Hz, H- 8 ), 6.27 (1H, dt, J 15.0, 1.5 Hz, H-l), 5.68 (1H, br. d, J 10.9 Hz, H-7), 5.05-4.93 (2H, m, H-9), 2.14 (1H, m, H-5), 2.01 (1H, septet, J 6.7 Hz, C//(CH3)2), 1.80-1.30 (4H, m, H-3, H-4), 1.58 (3H, s, C -6 C//3), 0.89 (3H, d, J 6.1 Hz, CH(C// 3)(CH3)), 0.75 (3H, d, J 6.1 Hz, CH(CH3)(C //3)) m il (El) 318 (M+), 275, 133, 123, 95, 81 (Found: (M+), 318.1660. C] 9 H260 2S requires (M+), 318.1654). IMDA reaction of (IE, 6£, 8£)*6*methyl-5-(l-methylethyl).l.phenyl* su!phonyl-l, 6,8 -nonatriene (94). 94 118 119 A solution of triene 94 (azeotropically dried with toluene [2 x 10 ml]; 234 mg, 0.735 2 2 9 mmol) in dry toluene (15 ml) was degassed as previously and transferred to a dry, argon filled Carius tube via cannula. The tube was evacuated and cooled in liquid nitrogen, then sealed using a flame and allowed to warm to rt behind a safety screen. The tube was then heated in a Carius oven at 220°C for 48 h. After cooling, the tube was opened and the solvent removed under reduced pressure to give a slightly discoloured oil (229 mg). *H nmr analysis of the crude mixture showed the presence of two major products 118 and 119 in the ratio 1:1, together with a third uncharacterised cycloadduct (< 10%). Chromatography (15% - 20% ether-petrol) gave an inseparable 1:1 mixture of {4R*, 5R*, SR*, 9R*]-9-methyl-8-(l-methylethyl)-4-(phenylsulphonyl)bicyclo[4.3.0]nonene 1 1 9 and [ 4 R *, 5R *, SR *, 9S * J-9-methyl-8-( 1 -methylethyl)-4- (phenylsulphonyl)bicyclo[4.3.0]nonene 118 (185 mg, 79%) as a colourless oil; umax (film) 2961, 2874, 1652, 1586, 1506, 1472, 1405, 1370, 1310, 1180, 1100, 790, 750, 700 cm-1; 6 (500 MHz) both diastereoisomers: 7.88 (2H, m, ortho- protons on Ph), 7.64 (1H, m, para- protons on Ph), 7.56 (2H, m, meta-protons on Ph), 118 (trans) inter alia:: 5.80 (1H, ddd, J 10.0, 2.3, 1.0 Hz, H-l), 5.44 (1H, ddd, J 10.0, 5.6, 2.5 Hz, H-2), 0.97 (3H, s, C-9 CH 3), 119 (cis) inter alia: 5.73 (1H, dt, J 10.0, 1.8 Hz, H-l), 5.63 (1H, dt, J 10.0, 4.1 Hz, H-2), 3.13 (1H, m, H-4), 2.39 (1H, td, J 8.0, 3.8 Hz, H-3ax or H-5), 1.20 (3H, s, C-9 CH 3), both diastereoisomers: 0.96 (3H, d, J, 6.5 Hz, >Pr), 0.94 (3H, d, J, 6.5 Hz, ¡Pr), 0.88 (3H, d, J, 6.5 Hz, ¡Pr), 0.83 (3H, d, J, 6.5 Hz, ¡Pr); m h (El) 318 (M+), 237, 176 (M+ - S0 2Ph), 93; (Cl) 338 (MNH4+), 312, 300, 213, 251, 239, 179, 152, 123, 109, 95 (Found: (MNH4+), 338.2154. C 19 H32NO requires (MNH4+), 338.2154). 230 Preparation of [1/?*, SR*, 6R*, 9/?*]-5-methyl-6-(l-methylethyl)-l. (phenylsulphonyl)bicyclo[4.3.0]nonane (120) and [1/?*, SS*, 6R*, 9R*]- 5-methyl-6-(l-methylethyl)l-(phenyIsulphonyI)bicyclo[4.3.0]nonane (121) 118 119 120 121 To a stirred suspension of 10% palladium-on-charcoal (13 mg, 0.0121 mmol, 0.1 eq) in ethyl acetate (0.5 ml) under argon was added a solution of alkenes 118 and 119 (38.5 mg, 0.121 mmol) in ethyl acetate (0.5 ml + 0.5 ml rinse). The mixture was purged with hydrogen and allowed to stir for 8 h under a positive pressure of hydrogen. Analysis by tic (3 x 20% ether-petrol) showed no change in Rf. The catalyst was removed by filtration through Celite® and the filtrate was concentrated to give a colourless oil. Chromatography (15% ether-petrol) gave a 1:1 inseparable mixture of [1R*, 5R*, 6R*, 9R*]-5-methyl-6-(l-methylethyl)-l-(phenylsulphonyl)bicyclo[4.3.0]nonane 120 and [1R*, 5S*, 6R*, 9R*]-5-methyl-6-(l-methylethyl)l-(phenylsulphonyl)bicyclo[4.3.0]- nonane 121 (38.2 mg, 99%) as a colourless oil; umax (film) 3064, 2957, 1506, 1447, 1385, 1304, 1145, 1086, 731, 690 cm*l; 5 (500 MHz) 7.86 (2H, dd, J 8 .6, 2.0 Hz, o- protons on Ph, one diastereoisomer), 7.64 (2H, dd, J 8.5,2.0 Hz, or/Ao-protons on Ph, other diastereoisomer), 7.66-7.60 (1H, m, para-protons on Ph), 7.58-7.50 (2H, m, meta-protons on Ph), 3.20 (1H, m, H-l cis-fused isomer 121), 2.96 ppm (1H, ddd, J 11.0, 8 .8 , 4.0 Hz), 2.39 (1H, m, H-2ax or H-9), 2.08 (1H, m, H-9ax or H-2), 1.30 (3H, s, C-6 CH3, as-fused isomer 121), 0.89 (3H, s, C -6 CH3, trans -fused isomer 231 120), both diastereoisomers: 0.97 (3H, d, J 6.5 Hz, *Pr), 0.91 (3H, d, J 6.5 Hz, >Pr), 0.87 (3H, d, J 6.5 Hz, >Pr), 0.84 (3H, d, J 6.5 Hz, »Pr); mlz (El) 321 (MH+, weak), 294 (M+ - CH3), 251,239, 179 (M+ - S0 2Ph), 150, 135, 123, 109,97, 83 (Found: (M+ + NH4+) 338.2154. C 19 H2802S requires (M+ + NH 4+), 338.2154). Preparation of /J-(+)-citronelIic acid {(+)-[3/f]-(6E)-3,7-d!methyl-6- octenoic acid} (125). .O 8 6 3 1 122 125 Anhydrous hydrogen chloride was bubbled slowly through R-(+)-pulegone (257 ml, 1.58 mol, ex Aldrich) in a 3-necked 1 litre flask at - 6°C, with stirring, venting the waste gas to a water scrubber in a fume hood. The reaction exothermed mildly and after 2 h, tic (40% ether-petrol) of the dark brown mixture indicated less than 5% of the starting material remained. The system was allowed to warm to rt, sealed under an argon atmosphere and stirred overnight. The mixture was then poured cautiously into a rapidly stirred solution of potassium hydroxide (360 g, 6.32 mol, 4 eq) in water (3.5 1; ca. 10% w/v) with ice-cooling. The reaction was allowed to warm to rt over 30 min and stirred for a further 2.5 h, after which time tic (20% ether-petrol) indicated absence of any intermediate addition product. The alkaline aqueous phase was extracted with ether (3 x 700 ml) and then acidified to pH 3 - 2 with concentrated hydrochloric acid. An oil appeared which was separated and the aqueous phase was extracted with ether (4 x 500 ml). The combined acidic ethereal extracts and the oil were washed with brine (2 x 500 ml), dried (MgSC> 4) and concentrated under reduced pressure to give /?-(+)-citronellic acid 125 (165.1 g, 0.97 mol, 61%) as a yellow oil. The alkaline ethereal extracts were washed with brine (500 ml), dried (MgSC> 4) and concentrated under reduced pressure to give a mixture of pulegone and wo-pulegone (93.3g, 38%) as a pale yellow oil. The citronellic acid was split into two portions. One portion (117.2 g, 0.69 mol) was used in the next, hydrogenation reaction, without further purification and the other (48 g) was distilled to give /?-(+)-citronellic acid 125 (42 g) as an oil (bp 0.05 93°C) with a distinctive 233 odour; [a ]D20 +8.40° (neat); vm„ (film) 3200 (br.), 2923, 1713, 1538, 1412, 1381, 1302, 940 cm-1; 5 (270 MHz) 11.00-10.00 (1H, br. s, C02H), 5.09 (IH, t, J 7.5 Hz, H-6), 2.36 (IH, dd, J 14.9, 5.9 Hz, H-2), 2.14 (IH, dd, J 14.9, 8.3 Hz, H-2), 2.05- 1.90 (3H, m, H-5, H-3), 1.68 (3-H, s, H -8 or H- 8 '), 1.60 (3H, s, H-8 or H- 8 '), 1.38 (IH, m, H-4), 1.25 (IH, m, H-4), 0.98 (3H, d, J 6.6 Hz, H-3'); m/z (El) 170 (M+),152 (M+ - H20), 137, 127,115,109 (M+ - C 2H4O2), 95, 69, 55, 41, in agreement with data previously reported.10^ Preparation of (+)-[3/? I-dihydrocitronellic acid { (+)-[3/? ]-3,7- dimethyloctanoic acid} (123). OH 125 123 To a solution of /?-(+)-citronellic acid 125 (117.2 g, 0.69 mol) in methanol (21) under an atmosphere of argon was carefully added a slurry of 10% palladium-on-charcoal ( 7.3 g, 6.9 mmol, 1 mol%; CAUTION! Flammable) in methanol (100 ml). The flask was cooled in an ice-salt bath and evacuated, then purged with hydrogen (x 3). The reaction was then allowed to take up hydrogen at a rate of between 30-40 ml min*1, absorbing a total of 161 of hydrogen. The flask was evacuated, then filled with argon; repeated twice and the mixture filtered through a pad of Celite®. The solvent was removed under reduced pressure and the resultant oil distilled under reduced pressure to give (+)-( 3#)- dihydrocitronellic acid 123 (111.97 g, 94%) as a colourless oil (bp 0 05 95-98°C); [a ]D20 +5.83° (neat); umax (film) 3500-2900 (br.), 2960, 1710, 1465, 1412, 1384, 1294, 936 cm*1; S (500 MHz) 2.35 (IH, dd, J 15, 5.9 Hz, H-2), 2.14 (IH, dd, J 15, 8.2 Hz, H-2), 1.96 (1H, m, H-3), 1.52 (1H, septet, J 6.7 Hz, H-7), 1.35-1.25 and 1.25-1.10 ( 6H, m, H-4, H-5, H-6), 0.96 (3H, d, J 6.7 Hz, C-3 C//3), 0.86 ( 6H, d, J 6.7 Hz, H-8 , C-7 C //3); m/z (El) 172 (M+),157 (M+ - CH3), 129,113 (M+ - C 2H40 2), 87,60 (C 2H40 2), 234 57, in agreement with data previously reported .103 Preparation of (+)-[4/f, 5S]-4-methyI-5-phenyl-2-oxazolidinone (126). O 126 To a rapidly stirred two-phase solution of (IS, 2/?)-(+)-norephedrine (40 g, 265 mmol; ex Aldrich) in toluene and aqueous potassium hydroxide (670 ml of a 12.5% w/v solution, 1.53 mol, 5.8 eq) was added a solution of phosgene in toluene (425 ml of a 1.93M solution, 0.82 mol, 3.1 eq), maintaining the temperature at <5°C. After 40 min the reaction was allowed to warm slowly to rt and the organic phase was separated. The aqueous phase was neutralised with 2M aqueous hydrochloric acid to pH 6 and extracted with ethyl acetate (3 x 250 ml). The combined organics were washed with 1M aqueous sodium hydroxide (2 x 250 ml), water (3 x 250 ml), brine (250 ml), dried (MgS04) and concentrated under reduced pressure to give an off-white solid. This was dissolved in DCM (250 ml), treated with activated charcoal, filtered through Celite® and the solvent removed under reduced pressure. The residue, now free of phosgene was recrystallised from DCM-ether-petrol to give (+)-[4 R, 5S]-4-methyl-5-phenyl-2-oxazolidinone 126 as colourless crystals: 1st crop (23.91 g, 51%) mp 121.5-123.5°C; [a]D20 +169.5° (c 2.02, CHCI3); 2nd crop (12.92 g, 27.5%) mp 118.5-121°C; [ 3rd crop 3.72 g, 8 %; total yield 40.55 g, 86 %; umax (film) 3454 (sharp), 3022, 1758, 1457, 1397, 1345,1230,1124,1002, 960,777, 761, 743, 701,668 cm*1; 8 (500 MHz) 7.42-7.25 (5H, m, Ph), 5.72 (1H, d, J 7.5 Hz, H-5), 5.65 (1H, br. s, H-3), 4.21 (1H, dq, J 6.7, 5 Hz, H-4), 0.80 (3H, d, J 6.7 Hz, C-4' Ctf3); m/z (El) 177 (M+), 132 (M+ - C 0 2H), 117, 107 (PhCH20+), 91, 89, 79, 77, in agreement with data previously reported .104 235 Preparation of (+)-[3'/?, 4/?, 5S]-4-methyl-3-(l-oxo-3,7-dimethyl- octane)-5-phenyl-2-oxazolidinone (127). 123 (+)-[3/?]-Dihydrocitronellic acid 123 (25 ml, 22.4 g, 130 mmol, 1 eq) was heated at reflux with thionyl chloride (14.2 ml, 23.2 g, 195 mmol, 1.5 eq) under argon for 45 min. Excess thionyl chloride was removed by distillation at atmospheric pressure and the resultant oil distilled at reduced pressure to give dihydrocitronellyl chloride as a pungent citrus flavoured oil (bp0.i 45-46°C), vmax (film) 1808 cn r1. To a stirred solution of oxazolidinone 126 (23 g, 130 mmol) in dry THF (430 ml) at -78°C was added n- butyllithium (56.4 ml of a 2.3 M solution in hexanes, 130 mmol 1.0 eq) drop wise via syringe to give a blood-red anion. To this mixture was added a solution of the freshly prepared dihydrocitronellyl chloride in THF (50 ml) and the reaction was allowed to warm slowly to rt. The pale yellow reaction was quenched by the addition of saturated aqueous ammonium chloride (100 ml). The aqueous phase was extracted with ether (3 x 200 ml) and the combined organic layers washed with 1M aqueous sodium hydroxide (2 x 100ml), water (3 x 200 ml), brine (200 ml), dried (MgS04) and concentrated under reduced pressure to give a pale solid. This residue was dissolved in benzene (100 ml) and filtered through a 2" pad of silica gel, eluting with 20% ether-petrol until no more material could be detected in the filtrate by tic (40% ether-petrol). The solvents were removed to give a white solid which was recrystallised to give (+ H 3'R, 4R, 5S]-4- methyl-3-(1-oxo-3,7-dimethyloctane)-5-phenyl-2-oxazolidinone 127 (30.51 g, 71%) as large plates mp 90-93°C, (absolute ethanol); [a ]D20 +37.7° (c 1.03, CHCI3); umax (film) 3038, 2959, 1781, 1701, 1457, 1348, 1226, 1200, 792, 775, 765, 744 enr*; 6 (500 MHz) 7.44-7.35 (5H, m, Ph), 5.66 (1H, d, J 7.3 Hz, H-5), 4.78 (1H, quintet, J 6.8 Hz, H-4), 2.99 (1H, dd, J 16, 5.5 Hz, H-2'), 2.70 (1H, dd, J 16, 8.3 Hz, H-2'), 2.08 236 (1H, m, H-3'), 1.53 (1H, septet, J 6.7 Hz, H-7’), 1.36-1.10 (6H, m, H-4', H-5', H- 6’), 0.97 (3H, d, J 6.6 Hz, C-3’ C//3), 0.89 (3H, d, J 6.6 Hz, C//3), 0.87 (3H, d, J 6.6 Hz, C//3), 0.86 (3H, d, J 6.6 Hz, Ctf3); mtz (El) 332 (MH+), 316, 288, 272, 268, 246, 219, 202, 178, 160, 134, 118, 107, 71, 57, 43 (Found: C, 72.2; H, 8.9; N, 4.1. C20h 29n o 3 recluires C’72,47; H‘ 8,82; N’ 4*23%)* Preparation of (+)-[2'/f, 37?, 4/?, 5S]-4-methyl-3-[l-oxo*2-(propenyl)- 3,7-dimethyloctyl]-5-phenyI-2-oxazo!idinone (130). H 127 To a stirred solution of carboximide 127 (20.04 g, 60.5 mmol) in THF (200 ml) at -78°C under argon was added NaHMDS (66.5 ml of a 1M solution in THF, 66.5 mmol, 1.1 eq). After 30 min freshly distilled 2-propenyl bromide (21 ml, 242 mmol, 4.0 eq) was added via syringe. The reaction was allowed to warm to -50°C during 5 h and then stirred at this temperature for 12 h. After this period of time tic (20% ether-petrol) indicated complete consumption of the starting material, so the reaction was allowed to warm to 0°C during a further 3 h, whereupon it was quenched by the addition of saturated aqueous ammonium chloride (150 ml). The organic layers were separated and the aqueous phase extracted with ether (3 x 100 ml). The combined organic extracts were washed with 0.5M hydrochloric acid (3 x 100 ml), water (3 x 100 ml), brine (100 ml), dried (MgS 04) and concentrated under reduced pressure. The residue was purified by chromatography (5% - 8 % -15% - 25% ether-petrol) to give (+)-[2’R, 5 1 1 ,4K, 5SJ-4- methyl-3-[l-oxo-2-(2-propenyl)-3,7-dimethyloctyl]-5-phenyl-2-oxazolidinone 130 (19.62 g, 87%) as a colourless syrup containing <5% of the syn diastereoisomer by *H nmr, [a]D20 +12.5° (c 1.99, CHC13); vmax (film) 3068, 2958, 1789, 1697, 1643, 1457, 237 1345, 1193, 1148, 1120, 1090, 1068, 1032, 993, 916, 887, 768, 722, 700, 638; d (500 MHz) 7.43-7.28 (5H, m, Ph), 5.78 (1H, m, H-2"), 5.61 (1H, d, J 7.2 Hz, H-5), 5.03 (1H, ddd, J 17, 1.6, 1.5 Hz, H-l"trans), 4.95 (1H, dd with fine structure, J 10.3, 1.7 Hz, H -l”cis), 4.79 (1H, quintet, J 6.7 Hz, H-4), 3.94 (1H, ddd, J 10, 7.2, 4.4 Hz, H- 2'), 2.25 (2H, m, H-3”), 1.82 (1H, m, H-3'), 1.53 (1H, septet, J 6.6 Hz, H-7’), 1.48 (1H, m, H-4'), 1.38 ( 1H, m, H-4’), 1.26-1.10 (4H, m, H-5', H-6'), 0.97 (3H, d, J 6.8 Hz, C-3' C//3), 0.88 (3H, d, J 6.6 Hz) and 0.87 (3H, d, J 6.6 Hz, both comprising H- 8 ’ and C-7' C//3), 0.85 (3H, d, J 6.6 Hz, C-4 C//3); m/z (El) 371 (M+), 356 (M+ - CH3), 330, 286, 259, 194 (M+ - C 9 H 10NO2), 178, 134, 118, 109, 81, 69, 55, 41 (Found: C, 74.0; H 9.0; N 3.74. C 23H33N 0 3 requires C, 74.36; H, 8.95; N, 3.77%) (Found: (M+), 371.2460. C 23H33N 03, (M+) requires 371.2460). lH nmr data for syn diastereoisomer (+)-[2'S, 4K, 5S]-4-methyI-3-[ 1 -oxo-2-(2 - propenyl)-3,7-dimethyloctyl]-5-phenyl-2-oxazolidinone, inter alia 6 (500 MHz) 7.25- 7.15 (5H, m, Ph), 5.08 (1H, ddd, J 17, 1.6, 1.5 Hz, H - l " ^ ) , 4.98 (1H, m, H- l"cis), 3.87 ( 1H, ddd, J 10, 5.8, 3.6 Hz, H-2'), 2.25 (2H, m, H-3"), 1.90 (1H, m, H- 3'), 0.90 (3H, d, J 6.8 Hz, C//3). Preparation of (+)-[27f, 37?, 4/?, 5S]-4-methyl-3*[l*oxo-2-(3-propanoI)- 3,7-dimethyIoctyIl- 5-phenyl- 2-oxazolidinone (131). < ^ To a stirred solution of alkene 130 (17.28 g, 46.51 mmol) in THF (230 ml) under argon at rt was added, a solution of 9-BBN (186 ml of a 0.5M THF solution, 93.03 mmol, 2.0 238 eq). The reaction was allowed to stir for 5 hours at rt, after which time tic (75% ether- petrol) indicated complete consumption of starting material. The solution was cooled to 0°C and quenched cautiously with aqueous sodium hydroxide (93 ml of a 1M solution, 93.0 mmol, 2.0 eq), maintaining the internal temperature below 10°C. Hydrogen peroxide (38 ml, 335 mmol, 7.2 eq of a 30% w/v solution) was added DROPWISE to the cloudy solution, maintaining the temperature below 20°C. The reaction was quenched after 45 min by the addition of 0.5M aqueous hydrochloric acid ( ca. 250 ml) and the aqueous phase was extracted with ether (3 x 150 ml). The combined organic layers were washed with water (2 x 200 ml), saturated aqueous ammonium chloride (3 x 200 ml), water (2 x 100ml), brine (200 ml), dried (MgSC> 4) and concentrated under reduced pressure with ice-cooling to give a colourless oil. The product was redissolved in 60% ether-petrol (100 ml) and filtered through a 2" pad of silica gel, rinsing further with 60% ether-petrol (750 ml). The solvents were removed under reduced pressure to give a ca. 1:1 mixture (by *H nmr) of (+)-[2‘R, iTt, 4R, 5SJ-4-methyl-3-[1-oxo-2-(3-propanol)- 3,7-dimethyloctyl]-5-phenyl-2-oxazolidinone 131 and (+)-[3R, l'RJ-3-(6-methyl- 2 heptyl)- 2-oxotetrahydropyran 132 (16 g) as a colourless oil which was used without further purification in the next reaction. When carried out on a 2.77 mmolar scale, this reaction yielded after chromatography ( 20% - 80% ether-petrol gradient; 10% increments) (+)-[2'R, JR , 4R, 5S)~4-methyl-3-[l-oxo-2-(3-propanol)-3,7-dimethyloctyl]-5-phenyl- 2-oxazolidinone 131 (886 mg, 82%) as a colourless oil; [o ]D20 +21.7° (c 0.96, CHC13); vm.x (film) 3480’ 2955’ 1784’ 1698’ 1457* 1342’ 1230’ 1195’ 1121, 1068’ 1031’ 990‘ 768, 700 cm*1; 6 (500 MHz) 7.45-7.35 (3H, m, ortho- and para-protons on Ph), 7.30 (2H, dd, J 7.0, 1.4 Hz, mera-protons on Ph), 5.63 (1H, d, J 7.1 Hz, H-5), 4.79 (1H, quintet, J 7.2 Hz, H-4), 3.82 (1H, ddd, J 10.0, 6.5, 3.9 Hz, H-2'), 3.62 (2H, br. s, H- 1"), 1.79 (2H, m, H-3"), 1.65 (1H, m, H-3'), 1.54 (1H. m, J 6.5 Hz, H-7’), 1.50- 1.30 (5H, m, H-2", H-4', OH), 1.20-1.10 (4H, m, H-5', H-6’), 0.98 (3H, d, J 6.8 Hz, C-3' C //3), 0.90 (3H, d, J 6.6 Hz, C-4 C//3), 0.88 (3H, d, J 6.6 Hz) and 0.87 239 (3H, d, J 6.6 Hz, both C-7’ C //3, H- 8 '); m/z (El) 389 (M+), 277, 213 (M+ - C9 H 10NO2), 197, 178 (C 9 H 10NO2), 134, 127, 107, 100, 79 (Found: (M+), 389.2566. C23H35NO4 requires (M+), 389.2566). Preparation of (+M3/?,27?]-3-(6-methyl-2-heptyI)tetrahydro-2tf-pyran-6- one (132). r 3' r k 3 * J * 131 132 To a stirred solution of a 1:1 mixture of alcohol 131 and lactone 132 (16g, ca. 46.51 mmol) in THF (450 ml), was added redistilled rm-butanol (88 ml, 930 mmol, 20 eq) via cannula. The mixture was cooled to 0°C and treated with potassium tm-butoxide (930 jd of a 1.0M solution in THF, 0.930 mmol, 2 mol%). After 20 min, tic (1:1 ether-petrol; visualized using KMn 04 spray) indicated complete lactonisation and the reaction was immediately quenched by the addition of 0.5M aqueous hydrochloric acid (10 ml). Water (250 ml) was added and the organic layers were separated. The aqueous phase was extracted with ether (3 x 150 ml) and the combined organic layers were washed with saturated aqueous ammonium chloride (3 x 200 ml), water (3 x 200 mL), brine (2 x 200 ml), dried (MgSC> 4) and concentrated under reduced pressure with ice-cooling to give a near-colourless oil. Purification by chromatography (40-60% ether-petrol) furnished (+)- [3R, 2'R]-3-(6-methyl-2hepryl)tetrahydro-2H-pyran-6-one 132 (7.108 g, 72% from alkene 130) as a mobile colourless oil; [a ]D19 +48.0° (c 1.11, CHCI 3), 95% d.e. by 500 MHz jH nmr. Material >98% d.e. by *H nmr had [a ]D19 +52.8° (c 1.15, CHCI 3); (film) 2956, 1737, 1464, 1383, 1271, 1153, 1085, 953 cm-»; 8 (500 MHz) 4.33 (1H, ddt, J 11.1, 4.8, 1.6 Hz, H- 6eq), 4.23 (1H, ddd, J 11.1, 8.9, 4.3 Hz, H- 6ax), 2.45 (1H, 240 ddd, J 11.2, 7.1, 3.8 Hz, H-3), 2.18 (1H, m, H-2’), 1.95 (1H, ddd, J 12.6, 5.2, 1.7 Hz, H-4eq), 1.92-1.80 (2H, m, H-4ax, H-5ax), 1.66-1.58 (1H, m, H-5eq), 1.52 (1H, septet, J 6.6 Hz, H-6’), 1.38-1.30 and 1.25-1.08 ( 6H, m, all comprising H-3\ H-4', H- 5’), 0.96 (3H, d, J 7.0 Hz, C-2* CH3), 0.86 (3H, d, J 6.6 Hz, C-6’, C//3), 0.85 (3H, d, J 6.6 Hz, H-7'); m/z (El) 213 (MH+), 211 (M+ - H), 197 (M - CH3+), 141, 127, 100, (C5H 8 0 2+), 84, 69, 55 (Found: (MH+), 213.1855. Q 3H25O2, requires (MH+) 213.1855). Preparation of (+)-[5J?, 61?S, 27? ]-6-hydroxy-5*(6-methyl-2- heptyI)tetrahydro-2tf-pyran (133). 132 133 To a stirred solution of lactone 132 (717 mg, 3.38 mmol) in toluene (30 ml) at -78°C under argon was added DIBAL-H (2.7 ml of a 1.5M solution in toluene, 4.05 mmol, 1.2 eq). The mixture was allowed to react for 20 min before it was quenched with aqueous THF (6 ml of a 10% v/v solution of water in THF, 34 mmol, 10 eq). The reaction was warmed to rt, then the pasty mixture was diluted with ethyl acetate (50 ml) and solid sodium hydrogencarbonate added. After stirring for 30 min the free-flowing solid was removed by filtration through a pad of Celite® and the filtrate was concentrated under reduced pressure to give a colourless oil. This residue was redissolved in 1:1 ether-petrol (25 ml) and filtered through a pad of silica gel, eluting with 1:1 ether-petrol (150 ml) to give, after removal of the solvents, lactol 133 (633 mg, 87%) as a colourless oil ;[a ]D20 +8.5° (c 1.07, CHCI 3); vmax (film) 3397, 2953, 1725 (weak), 1540, 1466, 1367, 1073, 981, 910 cm*1; 8 (500 MHz) 9.66 (1H, s, H -6 open-chain aldehyde), 5.23 (1H, t, J 2.6 Hz, H-6 ax-anomer), 4.63 (1H, t, J 6.4 Hz, H -6 eq-anomcr), 4.00-3.90 (2H, m, H-2 241 both anomers), 3.58-3.53 (1H, m, H -6 one anomer), 3.48-3.42 (1H, m, H -6 other anomer), 3.10 (1H, d, J 6.1 Hz, O-H ax-anomer), 2.62 (1H, dd, J 3.2,0.9 Hz, O-H eq- anomer), 1.80-1.00 (13H, m, H-2’, H-3’, H-4’, H-5’, H- 6’, H-3, H-4, H-5), 0.92 (3H, d, J 6.1 Hz, H-l' one anomer), 0.91 (3H, d, J 6.9 Hz, H -l' other anomer), 0.86 (6H, d, J 6.6 Hz, H-7', C-6 C //3), 0.85 ( 6H, d, J 6.6 Hz, H-7', C-6' C//3); m/z (El) 214 (M+, weak), 196 (M+ - H 20), 181 (M+ - H20 - CH3), 140, 111, 84, 70, 55, 43, 41; (Cl) 215 (MH+), 197 (MH+ - H 20), 140, 129, 111, 84, 69 (Found: (MH+), 215.2011. Ci3H270 2, requires (MH+), 215.2011). Preparation of [3/?, 27?]-6-methyIene*5-(6-methyl*2-heptyl)tetrahydro- 2//-pyran (139). l' i" 1 4 2 3 132 139 To a stirred solution of lactone 132 (110.1 mg, 0.519 mmol) in THF (630 pi) and I toluene ( 1.11 ml) under argon at -50 °C was added pyridine (40 pi, 0.498 mmol, 0.96 eq) followed by a solution of "Tebbe reagent" in toluene (1.4 ml of a 0.41M solution, 0.570 mmol, 1.1 eq, prepared according to standard procedure 110 (*>)• (c)) added dropwise over a period of 5 min. The reaction was allowed to stir at between -50 and -45°C for 20 min then warmed to -20°C during 45 min. Tic (10% ether-petrol) at this point showed complete conversion of the lactone to a non-polar product (Rf 0.6 in 10% ether-petrol). Saturated aqueous sodium carbonate (3 ml) was added and the mixture stirred rapidly to rt. When effervesence had ceased, ether (50 ml) was added, followed by anhydrous sodium sulphate. After 10 min the orange solution was filtered through a pad of Celite®, rinsing with ether (70 ml) to give, after concentration under reduced pressure, an orange oil. When petrol (25 ml) was added a ppt formed immediately which was broken up by 242 sonication for 5 min, then removed by filtration through neutral activity III alumina (activity I + 3% w/w water). A slight colour persisted in the filtrate, so it was re-filtered through fresh neutral avivity III alumina to give, after concentration under reduced pressure, [3R, 2'R]-6-methylene-5-(6-methyl-2-heptyl)tetrahydro-2H-pyran 139 (94.1 mg, 86 %) as a pale yellow oil; umax (film) 2957, 2868, 1650, 1468, 1380, 1277, 1181, 1074, 854 cm-1; 8 (500 MHz) 4.43 (1H, s, H-l"), 4.09 (1H, s, H-l"), 3.93 (1H, dt, J 10.8, 4.8 Hz, H-2eq), 3.79 (1H. ddd, J 10.8, 8 .8 , 3.6 Hz, H-2ax), 1.95 (1H, dt, J 7.7, 5.4 Hz, H-5), 1.83 (1H, m, H-2’), 1.80-1.70 (2H, m), 1.58-1.48 (3H, m), 1.40 (1H, m), 1.30-1.10 (4H, m), 1.02 (1H, m, all comprising H-3, H-4, H-3’, H-4', H-5', H- 6'), 0.91 (3H, d, J 6.7 Hz, H-l'), 0.87 ( 6H, d, J 6.6 Hz, H-T, C-6' C//3). Preparation of [27f]-5-(6-methyl-2-heptyI)-3,4.dihydro-2if-pyran (137). r 137 To a stirred solution of triethylamine (26 pi, 0.189 mmol, 0.7 eq) and DMAP (10 mg, 0.081 mmol, 0.3 eq) in DCM (0.5 ml) at 0°C under argon was added tert- butylchlorodiphenylsilane (77 ml, 0.30 mmol, 1.1 eq). To this stirred mixture was added lactol 133 (57.8 mg, 0.27 mmol) in DCM (0.5 ml + 0.5 ml rinse) via syringe. After 6 h at rt, the reaction was quenched by the addition of saturated aqueous ammonium chloride (10 ml). The aqueous phase was extracted with DCM (3 x 10 ml), organics combined and washed alternately with water (3x10 ml) and saturated ammonium chloride (3x10 ml), dried (MgS 04 ) and concentrated under reduced pressure to give a pale yellow oil. Purification by chromatography (1% - 30% ether-petrol) gave / 2'R]-5-(6-methyl-2 - heptyl)-3,4-dihydro-2 H-pyran 137 (21.3 mg, 40%) as a colourless oil; vmax (film) 2955, 243 2928, 2867, 2662, 1466, 1368, 1235, 1160, 1129, 924, 858 cm-1; 6 (250 MHz) 6.22 (1H, br. s, H-6), 3.89 (2H, t, J 5 Hz, H-2), 2.00-1.80 (5H, m, H-3, H-4, H-2’), 1.51 (1H, m, J 6.6 Hz, H-6'), 1.35-1.05 (6H, m, H-3', H-4', H-5’), 0.97 (3H, d, J 6.9 Hz, H-l'), 0.85 ( 6H, d, J 6.6 Hz, H-7', C-6' C//3); m/z (El) 196 (M+), 181 (M+ - CH3), 111 (M+ - C5H9 O), 97, 93, 55, 41 (Found: (MH+), 197.1925. C i 3H2sO, requires (MH+), 197.1925). Preparation of [27?]-6-methyl-5-(6-methyl-2-heptyl)-3,4-dihydro-2J/. pyran (140). r i" JUJU,u 139 140 To a stirred solution of enol ether 139 (13.2 mg, 0.0682 mmol) in DCM (0.4 ml) under argon at 0°C was added triethylamine (26 pi, 0.188 mmol, 3 eq) followed by a solution of iodine (18 mg, 0.069 mmol, 1.1 eq) in DCM (0.3 ml + 0.2 ml rinse). The purple colour of the iodine solution was immediately discharged, and after stirring for 30 min a cream ppt had formed. The starting material spot on tic (1% ether-petrol Si 02 and alumina) had been replaced by material with the same Rf (0.35) but now exhibiting UV activity. The reaction was quenched by the addition of saturated aqueous sodium hydrogencarbonate (10 ml) and the aqueous phase extracted with petrol (3x10 ml). The combined organic layers were washed with saturated aqueous ammonium chloride (3 x 10 ml), water (3 x 10 ml), brine (2 x 1 0 ml), dried (MgS 04 ), filtered through a pad of activity III neutral alumina and concentrated under reduced pressure to give [2'K]-6-methyl-S-(6-methyl-2- heptyl)-3,4-dihydro-2H-pyran 140 (10 mg, 76%) as a colourless oil which now no longer showed UV activity on tic; umax (film) 2928, 2868, 1672, 1560, 1539, 1466, 244 1385, 1254, 1179, 1107 cm-*; 5 (500 MHz) 3.92-3.82 (2H, m, H-2), 2.40 (1H, m, H- 2'), 1.90-1.84 (2H, m, H-4), 1.82-1.75 (2H, m, H-3), 1.74 (3H, t, J 1.4 Hz, H-l"), 1.51 (1H, m, J 6.5 Hz, H-6'), 1.40-1.10 (6H, m, H-3', H-4’, H-5’), 0.93 (3H, d, J 6.6 Hz, H-l'), 0.87 ( 6H, d, J 6.6 Hz, H-7’, C-6' CH2 finely separated); m/z (El) 210 (M+), 195 (M+ - C//3), 125, 105, 71, 69, 57, 55, 43 (Found: (M+), 210.1984. C 14H260 requires, (M+), 210.1984). Preparation of [4/?, 5R]-l-bromo-4-(l,l-dibromoethen-2-yl)-5,9- dimethyldecane (136). B iv 1* Br A—kX 3 136 Triphenylphosphine (182 mg, 0.694 mmol, 6 eq) and tetrabromomethane (115 mg, 0.374 mmol, 3 eq) were dissolved in DCM (1ml) under argon and stirred at 0°C for 20 min. A solution of lactol 133 (24.8 mg, 0.116 mmol) in DCM (0.5 ml + 0.5 ml rinse) was then added to the orange solution. After stirring for 16 h, tic (40% ether petrol) evidenced >80% conversion, so petrol (5 ml) was added and the resultant sticky oil transferred onto a pad of silica gel and washed with 5% ether-petrol (100 ml). Removal of the solvents under reduced pressure gave [4K, 5R]-l-bromo-4-(l,l-dibromoethen-2- yl)-5,9-dimethyIdecane 136 (28 mg, 56%) as a colourless oil; umax (film) 2957, 2869, 1655, 1615, 1462, 1384, 1238, 1158, 777 cm*1; 6 (500 MHz) 6.20 (1H, d, J 10.3 Hz, H-2'), 3.42 (2H, t, J 6.7 Hz, H-l), 2.39 (1H, septet, J 4.7 Hz, H-4), 1.90-1.80 (2H, m), 1.68-1.60 (1H, m), 1.60-1.50 (2H, m), 1.50-1.40 (1H, m, all comprising H-3, H- 5, H-6, H-9), 1.35-1.10 ( 6H, m, H-2, H-7, H-8 ), 0.87 (9H, d, J 6.7 Hz, C-5 Ctf3, C-9 C//3, H-10 with 1.4 Hz separation between C-9/C-10 C//3s and C-5 CH2)\ m/z (El) 433 245 (M+ with Br3 isotope pattern), 353 (M+ - Br), 319, 217, 246, 196, 166, 113, 71, 57 (Found: (M+), 429.9506. C ^ H ^ ^ B ^ , requires (M+), 429.9506). Preparation of (+)-[4/?, 5/?]-5,9-dimethyl-4-hydroxymethyldecanoI (142). J— i j . 132 To a rapidly stirred solution of lactone 132 (6.96 g, 31.5 mmol) in dry ether (150 ml) under argon at 0°C was added solid lithium aluminium hydride (1.2 g, 31.5 mmol, 4 eq of "H*") portionwise over 20 min, following the reaction by tic (100% ether). After initial partial reduction to the lactol, complete reduction to the diol was evident after 40 min. Water (1.2 ml) was added to the reaction dropwise, followed by 3M aqueous sodium hydroxide (1.2 ml) and more water (3.6 ml). After stirring for 30 min, the white granular ppt was removed by filtration through a 5 cm pad of silica gel, rinsing exhaustively with ethyl acetate (1000 ml) to give, after removal of the solvents under reduced pressure, (+)-[4R, 5R]-5,9-dimethyl-4-hydroxymethyldecanol 142 (5.67 g, 83%) as a colourless oil; [a ]D20 +5.80° (c 1.88, CHCI 3); umax (film) 3304 (br.), 2948, 1467, 1383, 1168, 1038 cm’1; 8 (500 MHz) 3.67 (1H, dd, J 10.3, 4.7 Hz, H-l'), 3.65 (2H, td, J 7.8, 1.4 Hz, H-l), 3.51 (1H, dd, J 10.6, 6.4 Hz, H -l’), 1.95-1.77 (2H, br.s, OH), 1.64-1.54 (2H, m, H-4, H-5), 1.53 (1H, septet, J 6.6 Hz, H-9), 1.48-1.28 (7H, m) and 1.25-1.05 (3H, m, all comprising H-2, H-3, H- 6, H-7, H-8 ), 0.85 ( 6H, d, J 7.1 Hz, H-10, C-9 C //3), 0.84 (3H, d, J 7.1 Hz, C-5 C //3); m/z (El) 217 (MH+), 199 (MH+ - H20), 185 (MH+ - CH3), 180 (MH+ - 2 H 20), 168 (MH+ - 2 H20 - CH3), 152, 126, 111, 97, 83, 69, 55,43; (Cl) 234 (M+ + NH4), 217 (MH+), 199 (M+ - H 20), 181, 125, 109, 95, 82, 69 (Found: (MH+), 217.2168. C j 3H 2g 0 2, requires (MH+), 217.2168). 246 Preparation of (+)-[4/?, 5fl]-5,9-dimethyI-l-((l,l-dimethylethyl)- diphenylsilyloxy)-4-hydroxymethyldecane (143a), [4/?, 5R]-5,9-dimethyl- 4 -((l,l-dimethylethyl)diphenylsHyloxy)methyl)decanoI (143b) and [4R, 5/?]-5,9-dimethyI-l,r-bis((l,l-dimethylethyl)-dlphenyIsilyl-oxy)-4- methyldecane (143c). Rz « OH. 142 143b R1 ■ OH, R2 « OTBDPS. 143c R1« OTBDPS, R2 - OTBDPS. To a stirred solution of triethylamine (3.9 ml, 27.7 mmol, 1.1 eq) and DMAP (154 mg, 1.26 mmol, 5 mol%) in DCM (100 ml) under argon at -50 °C was added a solution of diol 142 (5.445 g, 25.3 mmol) in DCM (15 ml + 5 ml rinse). After 5 min a solution of rerr-butylchlorodiphenylsilane (7.2 ml, 27.7 mmol, 1.1 eq) in DCM (10 ml + 5 ml rinse) was added dropwise via cannula. The reaction was allowed to stir for 10 h at low temperature, after which time it was allowed to warm slowly to 0°C. Saturated aqueous ammonium chloride (150 ml) was added and the aqueous phase extracted with DCM (3 x 75 ml). The combined organic layers were washed with water (3 x 100 ml), dried (MgS 0 4 ) and concentrated under reduced pressure. The residue was purified by chromatography (5% - 75% ether-petrol gradient; 10% increments) to give, in order of elution: [4R, 5R}-5,9-dimethyl-l,l'-bis((l,l-dimethylethyl)diphenylsilyloxy)-4- methyldecane 143c (5.723 g, 34%) as a colourless oil; vmax (film) 2939, 2860, 1561, 1539, 1507, 1465, 1429, 1385, 1113, 824, 739,701 cm*1; 8 (500 MHz) inter alia 7.70- 7.60 (8 H, m, ortho- protons on Ph), 7.45-7.30 (12H, m, meta- and para-protons on Ph), 247 3.65-3.55 (3H, m, H-l, H -l’), 3.52 (1H, dd, J 10, 5 Hz, H-l'), 1.05 (18H, s, tert- butyl), 0.89 ( 6H, d, J 7.0 Hz, H-10, C-9 Ctf3), 0.82 (3H, d, J 6.8 Hz, C-5 C//3); m/z (El) 451, 435, 423, 381, 367, 333, 227, 199, 183,167, 139, 111, 97, 83, 69, 55 ; (+)■ [4 R, 5R]-5,9-dimethyl-l-((l ,l-dimethylethyldiphenyl)silyloxy)-4-hydroxymethyldecane 143a (3.90 g, 34%) as a colourless oil; [a ]D20 +3.39° (c 0.47, CHC13); umax (film) 3415 (br.), 3072, 2933, 2859, 1653, 1538, 1506, 1472, 1429, 1363, 1113, 822, 740, 702, 609 cm’1; 8 (500 MHz) 7.68 (4H, dd, J 7.9, 1.4 Hz, ortho- protons on Ph), 7.45-7.34 (6H, m, meta- and para- protons on Ph), 3.67 (2H, ddd, J 6.7, 3.2 Hz, H-l), 3.62 (1H, dd, J 10.5, 4 Hz, H-l'), 3.49 (1H, dd, J 10.5, 5.1 Hz, H-l'), 2.26 (1H, br. s, O-H), 1.62-1.55 (2H, m, H-4, H-5), 1.54 (1H, septet, J 6.7 Hz, H-9), 1.42-1.28 (4H, m), 1.20-1.10 (6H, m, all comprising H-2, H-3, H- 6, H-7, H-8 ), 1.07 (9H, s, rm-butyl), 0.88 (3H, d, J 6.7 Hz), 0.87 (3H, d, J 6.7 Hz, H-10 and C-9 Ctf3), 0.84 (3H, d, J 6.9 Hz, C-5 C//3); m/z (El) no M+, 377 (M+ - Ph), 335, 256 (TBDPSO+), 239 (TBDPS+), 227, 213, 199 (Ph 2SiOH), 181 (M+ - TBDPSO), 137,121,77; [4R, 5R]-5,9-dimethyl- 4.((l,l-dimethylethyl)diphenylsilyloxy)methyl)decanol 143b (1.055 g, 9.2%) as a colourless oil; vmax (film) 3328 (br.), 3073, 2932, 2860, 1690, 1440, 1420, 1100, 1080, 1060, 840 cm'1; 8 (500 MHz) 7.65 (4H, dd, J 6.7,1.1 Hz, orr/io-protons on Ph), 7.43-7.32 (6H, m, meta- and para-protons on Ph), 3.62-3.50 (4H, m, H-l, H -l’), 1.67- 1.58 (2H, m, H-4, H-5), 1.58 (1H, br. s, O-H), 1.36 (1H, m, J 6.7 Hz, H-9), 1.53- 1.43 (4H, m), 1.30-1.20 (2H, m) and 1.18-1.08 (4H, m, H-2, H-3, H- 6, H-7, H-8 ), 1.05 (9H, s, rcrf-butyl), 0.86 ( 6H, d, J 6.6 Hz, H-10, C-9 0 / 3), 0.81 (3H, d, J 7.0 Hz, C-5 C//3); m/z (El) no M+, 397 (M+ - C 4H9), 379 (M+ - C 4H9 - H20), 335, 319, 229, 199 (Ph 2SiOH), 181, 139, 125, 111, 97, 83, 69, 57 (Found: (M+ - C 4H9), 397.2563. C25H370 2Si requires (M+ - C 4H9), 397.2571) and finally (+)-[4R, 5R]-5,9-dimethyl-4- hydroxymethyldecanol 142 (1.089 g, 20%) as a colourless oil identical to material prepared previously. Hence the total mass balance was 90%. 248 Preparation of (-)-[4/?, 5/?]-5,9-dim ethyl-l.((l,l.dim ethylethyl). diphenyIsilyloxy)-4-oxomethyldecane (144). To a stirred solution of oxalyl chloride (1.50 ml, 17.15 mmol, 2 eq) in DCM (40 ml) at - 60°C under argon was added a solution of DMSO (2.43 ml, 34.3 mmol, 4 eq) in DCM (10 ml + 5 ml rinse) via cannula. After 5 min a solution of alcohol 143a (3.90 g, 8.58 mmol) in DCM (20 ml + 10 ml rinse) was added slowly via cannula to the reaction. The mixture was allowed to react for 25 min then triethylamine (5.02 ml, 36.02 mmol, 4.2 eq) was added in one portion, causing a colour change to yellow followed by a dense white ppt. After stirring for 10 min at -60°C, the mixture was allowed to warm to rt whereupon saturated aqueous ammonium chloride (100 ml) was added. The aqueous phase was extracted with ether (3 x 100 ml) and the combined organic layers were washed with 0.5M aqueous hydrochloric acid (3 x 75 ml), water (75 ml), saturated aqueous sodium hydrogencarbonate (2 x 75 ml), water (2 x 75 ml), brine (75 ml), dried (MgSC> 4) and concentrated under reduced pressure to give a yellow oil. Purification by chromatography ( 2% - 10% ether-petrol) gave a pale yellow oil which was redissolved in 5% ether-petrol, treated with activated charcoal, filtered through a pad of Celite® to yield. The filtrate was concentrated under reduced pressure to give aldehyde 143 (2.836 g, 73%) as a colourless oil; [a ]D20 -10.63° (c 1.60, CHCI3); vmix (film) 3068,2953, 2928, 2858, 1722, 1460, 1426, 1382, 1362, 1184, 1111, 987, 938, 832, 740, 702 cm **;8 (500 MHz) 9.63 (1H, d, J 3.2 Hz, H-T), 7.66 (4H, dd, J 7.8, 1.4 Hz, or/Aa-protons on Ph), 7.45-7.35 (6H, m, meta- and para-protons on Ph), 3.67 (2H, m, H-l), 2.16 (1H, m, H-4), 1.81 (1H, m, H-5), 1.54 (1H, septet, J 6.6 Hz, H-9), 1.60-1.45 (3H, m), 1.40-1.30 (2H, m), 1.25-1.10 (5H, m, all comprising H-2, H-3, H- 6, H-7, H-8 ), 1.05 249 (9H, s, rm-butyl), 0.94 (3H, d, J 6.9 Hz, C-5 C//3), 0.87 ( 6H, d, J 6.5 Hz, H-10, C-9 C //3); m/z (El) no M+, 395 (M+ - C 4H9), 381, 333, 287, 227, 199 (Ph 2SiOH) (Found: (M+ - C5H11), 381.2250. C 24H330 2Si requires (M+ -C 5Hn ), 381.2250). Preparation of (-)-[4/f, 5Jt]-(rZ)-5y9-dimethyl-l-((l,l-dimethyl- ethyl)diphenyIsilyloxy)-4-(l-iodoethen-2-yl)decane (145). To a stirred suspension of iodomethyltriphenylphosphonium iodide (6.41 g, 12.00 mmol, 2 eq), in THF (40 ml), under argon at rt was added NaHMDS (12.7 ml of a 1.0M THF solution, 12.70 mmol, 2.1 eq). After 15 min the deep red phosphorane solution was cooled to -78 °C and a solution of aldehyde 144 (2.738 g, 6.048 mmol) in THF (10 ml + 10 ml rinse) was added via cannula. The reaction was allowed to stir for 10 minutes i then warmed to rt. After 40 min saturated aqueous ammonium chloride (100 ml) was added. The aqueous phase was extracted with petrol (3 x 100 ml) and the combined organic layers were washed with 5% aqueous sodium metabisulphite (3 x 100 ml), water (3 x 100 ml), brine (2 x 100 ml), dried (MgS04) and concentrated under reduced pressure to give a pale yellow oil. Chromatography (1% ether-petrol) provided vinyl iodide 145 (3.082 g, 88 %) as a colourless oil [o ]D20 -4.5° (c 2.76, CHC13); vmax (film) 3071, 2959, 1590, 1558, 1471, 1429, 1384, 1364, 1299, 1265, 1111, 999, 939, 823, 739, 702, 613 cm '1; 6 (500 MHz) 7.68 (4H, dd, J 7.8, 1.4 Hz, orrito-protons on Ph), 7.45-7.35 (6H, m, meta- and para-protons on Ph), 6.23 (1H, d, J 7.4 Hz, H-l'), 5.92 (1H, dd, J 9.7, 7.4 Hz, H-2’), 3.70-3.60 (2H, m, H-l), 2.43-2.37 (1H, m, H-4), 1.60- 1.50 (5H, m, H-9, H-5, H-3, H- 6), 1.40 (1H, m, H-6), 1.35-1.25 (3H, m), 1.18-1.08 250 (3H, m, all comprising H-2, H-7, H- 8 ), 1.06 (9H, s, ferr-butyl), 0.88 ( 6H, d, J 6.8 Hz, H-10, C-9 CH3), 0.86 (3H, d, J 7.0 Hz, C-5 C//3); m/z (El) 575 (M+ - H), 561 (M+ - CH3), 533, 519 (M+ - C 4H9), 393 (M+ - C 4H9 - 1), 309, 199, 183, 163, 135, 109, 95, 81, 69, 57, 43 (Found: (M+ - C 4H9), 519.1580. (^gH^IOSi requires (M+ - C 4H9), 519.1580). Preparation of (+)-[4rt, 5/?]-5>9-dimethyI-1-((1,1-dimethylethyl)- diphenylsilyloxyM-ethynyldecane (146). To a stirred solution of vinyl iodide 145 (3.037 g, 5.27 mmol) in THF (50 ml) at -78°C under argon was added potassium rm-butoxide (10.5 ml of a 1.0M THF solution, 10.5 mmol, 2.0 eq), generating a deep orange colour. After 15 min, the reaction was allowed i to warm to rt and a cream coloured solid precipitated. Saturated aqueous ammonium chloride (100 ml) was added and the aqueous phase extracted with petrol (3 x 50 ml). The combined organic layers were washed alternately with water (3 x 75 ml) and brine (3 x 75 ml), then dried (MgS04) and concentrated under reduced pressure to give a yellow oil. This was redissolved in 0.5% ether-petrol and filtered through a pad of silica gel, to give, after removal of the solvents under reduced pressure, alkyne 146 (2.135 g, 90%) as a colourless o il; [afo *0 +1.14° (c 0.96, CHC13); umax (film) 3310, 3073,2931,2861, 1466, 1429, 1384, 1112, 998, 823, 740, 703, 614 cm-»; 6 (500 MHz) 7.69 (4H, dd, J 7.6, 1.4 Hz, ortho-protons on Ph), 7.45-7.35 ( 6H, m, meta- and para-protons on Ph), 3.70 (2H, ddt, J 10.3, 7.1, 6.2 Hz, H-l), 2.36 (1H, m, H-4), 2.01 (1H, d, J 2.6 Hz, H-l'), 1.81 (1H, m, H-5), 1.68-1.60 (1H, m), 1.58-1.45 (4H, m), 1.40 (1H, m), 1.35- 251 1.10 (5H, m, all comprising H-2, H-3, H- 6, H-7, H-8 , H-9), 1.07 (9H, s, rerr-butyl), 0.93 (3H, d. J 6.6 Hz. C-5 C//3), 0.89 ( 6H. d, J 6.6 Hz, H-10, C-9 C//3); m/z (El) no M+, 391 (M+ - C 4H9), 287, 265, 225, 199 (M+ - TBDPSO) (Found: (M+ - C 4H9), 391.2457. C 26H350 Si requires (M+ - C 4H9 ), 391.2448). Preparation of (-)-[4/?, 5/?]-5,9-diniethyI-4-ethynyM-decanol (138). To a stirred solution of alkyne 146 (2.103 g, 4.685 mmol) in THF (10 ml) under argon was added TBAF (9.4 ml of a 1.0M THF solution, 9.37 mmol, 2.0 eq) via syringe, to give a characteristic green colour. After 1.5 h, tic (50% ether-petrol) indicated that the reaction was complete and saturated aqueous ammonium chloride (100 ml) was added. The aqueous phase was extracted with ether (3 x 75 ml) and the combined organic layers were washed alternately with water (3 x 100 ml) and brine (3 x 100 ml). The solution was then dried (MgS04) and concentrated under reduced pressure to give a yellow oil. Chromatography (20% - 50% ether-petrol) provided (-)-[4R, 5K ]-5,9-dimethyl-4 - ethynyl-l-decanol 138 (918 mg, 93%) as a colourless oil; [a ]D20 -2.49° (c 2.89, CHC13); umax (film) 3311 (br.), 2961, 2872, 1506, 1462, 1383, 1262, 1058, 801, 627 cm*1; 6 (500 MHz) 3.69 (2H, t, J 6.3 Hz, H-l), 2.37 (1H, ddd, J 9.7, 4.6, 2.5 Hz, H- 4), 2.03 (1H, d, J 2.5 Hz, H-l'), 1.82 (1H, m, H-5), 1.70-1.60 (1H, m, H-3), 1.60- 1.46 (4H, m), 1.42-1.36 (1H, m), 1.35-1.20 (4H, m), 1.18-1.10 (2H, m, all comprising H-2, H-3 [one proton], H- 6, H-7, H-8 , H-9, O-H), 0.93 (3H, d, J 6.7 Hz, C-5 C //3), 0.86 ( 6H, d, J 6.6 Hz, H-10, C-9 OZ3); m/z (El) 211 (MH+), 181, 137, 123, 109, 95, 81, 77, 71, 57 (Found: (MH+), 211.2062. C] 4H2702 requires (MH+), 252 211.2062). Preparation of [4/f, 5J?]-(l,Z)-5,9-dimethyl-4-(l-iodopropen-2-yl)-l- decanol (147) and [4/?, 5/?]-5,9-dimethyl*4-(2-propen-2-yl)-l-decanol (148). 147 138 OH OH 3S2 148 I l ^ C To a stirTed solution of zirconocene dichloride (494 mg, 1.691 mmol, 0.4 eq) in 1,2- dichloroethane (6 ml) under argon (bubbler) was added trimethylaluminium (PYROPHORIC! 6.34 ml of a 2.0M solution in hexanes, 12.68 mmol, 3.0 eq) generating a yellow solution. The reaction was cooled to 0°C and a solution of alkyne 138 (889 mg, 4.23 mmol) in 1,2-dichloroethane (6 ml + 2 ml rinse) was added via cannula. After stirring for 16 h it was cooled to -30°C and a solution of iodine (1.18 g, 4.65 mmol, 1.1 eq) in THF (6 ml) was added to the orange reaction mixture. The reaction was stirred for 15 min at low temperature then allowed to warm to 0°C and quenched by dropwise addition of saturated aqueous potassium carbonate (3 ml), allowing for the evolution of a large quantity of gas. The glutinous mixture was diluted with ethyl acetate (100 ml) and stirred with solid sodium hydrogencarbonate for 20 min. The solid was removed by filtration through a silica gel pad, rinsing with ethyl acetate (750 ml) and the solution concentrated under reduced pressure to give a yellow oil. The 253 residue was purified by chromatography (20% - 40% ether-petrol gradient; 5% increments) to give an inseperable 5:1 mixture (]H nmr) of [4R, 5R]-(llL)-5,9-dimethyl- 4-( 1 -iodopropen-2-yl)-l-decanol 147 and [4R, 5R]-5,9-dimethyl-4-(2-propen-2-yl)-l- decanol 148 (1.1299 g, 83% total; i.e. 63% vinyl iodide 147, 20% terminal methylene 148) as a colourless oil; [a ]D20 +14.76° (c 1.66, CHCI3); vmax (film) 3324 (br.), 2950, 2866, 1608, 1458, 1379, 1266, 1151, 1058, 887, 771 cm'1; 8 (500 MHz) vinyl iodide 147 5.83 (1H, d, J 0.9 Hz, H -l’), 3.61 (2H, m, H-l), 1.98 (1H, ddd, J 10.9, 9.0, 3.9 Hz, H-4), 1.67 (3H, d, J 0.9 Hz, C-2' C //3), 1.65-1.58 (1H, m, H-5), 1.52 (1H, septet, J 6.8 Hz, H-9), 1.46-1.24 (7H, m), 1.20-1.08 (3H, m), 1.05-0.80 (1H, m, all comprising H-2, H-3, H- 6, H-7, H-8 , O-H), 0.87 (3H, d, J 6.6 Hz) and 0.86 (3H. d, J 6.6 Hz, H-10, C-9 C// 3), 0.74 (3H, d, J 6.7, C-5 C//3); terminal methylene 148 inter alia 4.79 (1H, dd, J 2.6, 1.4 Hz, H-l'), 4.64 (1H, dd, J 2.4, 0.7 Hz, H -l’), 0.80 (3H, d, 6.6 Hz, C-2’ 0 / 3); m/z (El) no M+, 239 (M+ - C 8 H17), 225 (M+ - 1), 221,194, 141, 96, 81 (Found: (M+ - C 8 H17), 238.9933. C 7H 12IO requires (M+ - C 8 H17) 238.9933). P rep aratio n of [41?, 5/?]-(3,E)-5,9-dimethyl-4«(l,3-pentadien»4« yl)decanoI (149). 147 149 + 148 A solution of vinyl iodide 147 contaminated with ca. 20% of 148 (1.046 g, 2.97 mmol) in toluene (20 ml + 5 ml rinse) was added to a stirred solution of tetrakis(triphenylphosphine) palladium (0) (172 mg, 0.15 mmol, 5 mol%) in toluene (5 ml). After 20 min, vinyl magnesium bromide (8.9 ml of a 1.0M solution in THF, 8.9 mmol, 3 eq) was added dropwise via syringe causing a mild exotherm. After stirring for 254 50 min, tic (2 x 20% ether-petrol) indicated complete reaction and saturated aqueous ammonium chloride (50 ml) was added to the dark solution. The aqueous phase was extracted with ether (3 x 50 ml) and the combined organic layers were washed with water (3 x 50 ml), brine (50 ml), dried (MgSO^ and concentrated under reduced pressure to give a red oil. This residue was purified by chromatography (25% - 40% ether-petrol) to give a yellow oil which was redissolved in 40% ether-petrol, treated with activated charcoal and filtered through Celite®. The filtrate was concentrated under reduced pressure to give an inseparable mixture of [4R, 5R]-(3 'E)-S,9-dimethyl-4-( 1 j-pentadien- 4-yl)decanol 149 and unchanged contaminant [4R, 5R]-5,9-dimethyl-4-(2-prop-en-2-yl)- J-decanol 148 (682.9 mg, 91%), 3.5:1 respectively (by JH nmr), as a colourless oil; [a ]D20 +8.43° (c 2.10, CHC13); vmax (film) 3328 (br.), 3079, 2951, 2867, 1639, 1595, 1462, 1379, 1212, 1168, 1059, 986, 896, 658 cm’1; S (500 MHz) dienol xx: 6.59 (1H, dt, J 16.8, 10.4 Hz, H-2'), 5.81 (1H, dd, J 10.8, 0.6 Hz, H-3’), 5.09 (1H, dd, J 16.8, 2.0 Hz, H -l'trans), 4.98 (1H, dd, J 10.3, 2.3 Hz, H -l’cis), 3.62 (2H, m, H-l), 1.78- 1.68 (1H, m, H-5), 1.62 (3H, d, J 1.1 Hz, C-4' C //3), 1.52 (1H, septet, J 6.3 Hz, H- 9 ), 1.50-1.00 (12H, m, H-6, H-7, H-8 , H-9, H-2, H-3, OH), 0.87 (3H, d, J 6.7 Hz) and 0.86 (3H, d, J 6.7 Hz, H-10 and C-9 C// 3), 0.76 (3H, d, J 6.8 Hz, H-6 C//3); m/z (El) 252 (M+), 226 (M+ - C2H2), 184, 167, 139, (M+ - CgH17), 114, 96, 81 (Found: (M+), 252.2453. C17H320 , requires (M+) 252.2453). Nmr data for terminal methylene 148 8 (500 MHz) inter alia 4.78 (1H, dd, J 2.6, 1.4 Hz, H -l’^ns), 4.64 (1H, dd, J 2.4, 0.6 Hz, H -l'cis), 1.59 (3H, dd, J 1.3, 0.7 Hz, C-2’ C//3), 0.80 (3H, d, J 6.7 Hz, C-5 C// 3). 255 Preparation of [4/?, 5/?]-(3'E)-5,9-dimethyI-4-(l,3-pentadien-4-yl)decanal (ISO) and [4/?, 5/f]-5,9-dimethyI-4-(2-propen-2-yl)-l-decanal (151) 149 + 148 150 + 151 To a stirred solution of oxalyl chloride (484 pi, 5.55 mmol, 2.2 eq) in DCM (10 ml) under argon at -60°C was added a solution of DMSO (787 pi, 11.10 mmol, 4.4 cq) in DCM (5 ml + 5 ml rinse). After 10 minutes a solution of a 3.5 :1 mixture of alcohols 149 and 148 (636.6 mg, 2.52 mmol) in DCM (5 ml + 5 ml rinse) was added via cannula. The reaction was stirred at -60°C for 20 min then triethylamine (1.93 ml, 13.87 mmol, 5.5 eq) was added via syringe. After a further 10 min it was allowed to warm to rt and poured into a 1:1 mixture of ether and water (100 ml). The layers were separated and the aqueous phase extracted with ether (2 x 50 ml). The combined organic layers were washed with saturated aqueous ammonium chloride (3 x 50 ml), water (3 x 50 ml), brine (2 x 50 ml), dried (MgSC^) and concentrated under reduced pressure. The yellow residue was purified by chromatography ( 2% - 20% ether-petrol) to give, after treatment with activated charcoal, a 4:1 mixture (]H nmr) of dienal 150 and enal 151 (511 mg, 81%) as a colourless oil with a characteristic odour, [a ]D20 +3.86° (c 1.32, CHCI 3); vmax (film) 3080, 2953, 2928, 1723, 1638, 1595, 1459, 1409, 1380, 1169, 987, 900 cm*»; 8 (500 MHz) 9.74 (1H, t, J 1.5 Hz, H-l), 6.56 (1H, dt, J 17.0, 10.3 Hz, H-2'), 5.80 (1H, d, J 10.8 Hz, H-3'), 5.10 (1H, dd, J 16.8, 1.9 Hz, H -l’trans), 5.00 (1H, dd, J 10.2, 1.9 Hz, H -rcis), 2.35-2.25 (2H, m, H-2), 1.94 (1H, m, H-4), 1.72 (1H, m, H- 5), 1.60 (3H, d, J 1.0 Hz, C-4’ C//3), 1.52 (1H, m, J 6.0 Hz, H-9), 1.50-1.30 (4H, m) and 1.25-1.00 (4H, m, H-3, H-6, H-7, H-8 ), 0.87 (3H, d, J 6.6 Hz) and 0.86 (3H, d, J 6.6 Hz, H-10 and C-9 C//3), 0.76 (3H, d, J 6.7 Hz, C -6 C//3); m/z (El) 250 (M+), 232, 256 209, 191,165 (M+ - C 5H7), 137 (M+ - C8 H17), 122,93,71, 57 Preparation of [2R S , SR , 6RM3'£)-6,10-dimethyl-2-hydroxy-5-(l,3- pentadien-4-yl)undecane (152) and [2R S, SR, 6R]-6,10-dimethyl-2- hydroxy-5-(2-propen-2-yl)undecane (152b). 150 + 151 To a stirred solution of (phenylsulphonyl)methane (337 mg, 2.16 mmol, 1.1 eq) in THF (15 ml) under argon at -78°C was added n-butyllithium (863 pi of a 2.50M solution in hexanes, 2.16 mmol, 1.1 eq), dropwise via syringe to generate a faintly yellow anion. After 10 min, a solution of a ca. 4:1 mixture of aldehydes 150 and 151(491.4 mg, 1.96 mmol) in THF (5 ml + 2 ml rinse) was added via cannula. The reaction was stirred at - 78°C for 1 h and then quenched by the addition of acetic acid in THF (2.3 ml of a 1.75M solution (10% v/v), 3.92 mmol, 2.0 eq) and allowed to warm to rt. Saturated aqueous sodium bicarbonate (50 ml) was added and the aqueous phase was extracted with DCM (3 x 20 ml). The organics were combined and washed with water (3 x 20 ml), dried (MgS04) and concentrated under reduced pressure to give a pale yellow oil. Purification by chromatography (20% - 40% ether-petrol) gave hydroxysulphones 152 contaminated with ca. 20% of /2RS, 5R, 6R]-6,10-dimethyl-2-hydroxy-5-(2-propen-2-yl)undecane 152b (736.2 mg, 84%) as a colourless oil; [a ]D20 0.00° (c 1.61, CHCI3); vmax (film) 3517 (br.), 2951, 2867, 1446, 1379, 1304, 1148, 1085, 1024, 997, 901, 841, 782, 7 4 6 , 719 , 687 cm*1; 8 (500 MHz) 7.92 (2H, dd, J 8.5, 2.8 Hz, ortho -protons on Ph), 7.70-7.55 (3H, m, meta- and para -protons on Ph), 6.52 (1H, dt, J 16.8, 10.6 Hz, H- 2'), 5.71 (1H, t, J 10.6 Hz, H-3'), 5.04 (1H, ddd, J 16.5, 14.5, 2.0 Hz, H -l’tr#ns), 257 4.96 (1H, m, H -lcis), 4.18-4.08 (1H, m, OH), 3.32 (1H, m, H-2), 3.20-3.18 (2H, m, H-l), 1.72-1.55 (2H, m, H-5, H-6), 1.56 (3H, t, J 0.9 Hz, C-4’ C//3), 1.50-1.25 (5H, m), 1.22-1.05 (5H, m) and 1.00-0.90 (1H, m, all comprising H-3, H-4, H-7, H- 8 , H-9, H-10), 0.86 (3H, d, J 6.6 Hz) and 0.85 (3H, d, J 6.6 Hz, H -ll, C-10 C//3); m h (El) 406 (M+), 363, 337, 321 (M+ - C5H7), 293, 247, 211, 141, 133, 109, 95 (Found: (M+), 406.2542. C24H380 3S (M+), requires 406.2542). Preparation of [5/?, 6/?]-(lE, 3'£)*6,10-dimethyl-5-(l,3-pentadien-4.y])- l-phenylsulphonylundecene (153) and [SRt 6/f]-(lE)-6,10-dimethyl-5-(2- propen- 2-yl)-l-phenylsulphonylundecene (154). To a stirred solution of hydroxysulphones 152 and 152b ( ca. 4:1) (718.2 mg, 1.77 mmol) in DCM (17 ml) under argon at - 6°C was added triethylamine (2.5 ml, 17.7 mmol, 10 eq) followed by methanesulphonyl chloride (344 ml, 5.3 mmol, 3 eq). Tic (40% ether-petrol) indicated complete elimination after 30 min and saturated aqueous sodium hydrogencarbonate (30 ml) was added. The aqueous phase was extracted with DCM (3 x 20 ml) and the combined organic layers were washed with saturated aqueous ammonium chloride (3 x 10 ml), water (3 x 10 ml), dried (MgS04) and concentrated under reduced pressure to give a yellow oil. Purification by chromatography gave an inseparable mixture of [5R; 6R]-(1E, 3E)-6,10-dimethyl-5-(l,3-pentadien-4-yl)-l - phenylsulphonylundecene 153 and [5R, 6R]-(lE)-6,10-dimethyl-5-(2-propen-2-yl)-l. phenylsulphonylundecene 154 (454.3 mg, 66%) as a colourless oil; [a ]D20 +1.17° (c 258 2.14, CHC13); umax 2951, 2866, 1637, 1445, 1379, 1318, 1147, 1086, 987, 901, 816, 752, 715, 687 cm*1; 8 (500 MHz) 7.87 (2H, dd, J 7.1, 1.5 Hz, ortho -protons on Ph), 7.65-7.50 (3H, m, meta- and para-protons on Ph), 6.96 (1H, ddd, J 13.8, 7.5, 6.3 Hz, H-2), 6.53 (1H, dt, J 16.9, 10.4 Hz, H-2'), 6.28 (1H, tt, J 15.0, 1.5 Hz, H-l), 5.68 (1H, d, J 10.8 Hz, H-3'), 5.01 (1H, dd, J 16.8, 2.0 Hz, H - l '^ ) , 4.97 (1H, dd, J 10.2, 2.0 Hz, H-l*cis), 2.22-2.10 (1H, m, J, H-3), 2.02 (1H, m, H-5), 1.74-1.62 (2H, m, H-3, H-6), 1.52 (1H, m, J 6.7 Hz, H-10), 1.48-1.28 (4H, m), 1.20-1.08 (3H, m) and 1.05-0.95 (1H, m, all comprising H-4, H-7, H- 8 , H-9), 0.87 (3H, d, J 6.7 Hz) and 0.86 (3H, d, J 6.7 Hz, H -ll, C-10 C//3), 0.73 (3H, d, J 6.7 Hz, C -6 C//3); m/z (El) 388 (M+), 373 (M+ - CH3), 359, 345, 331, 321 (M+ - C 5H7), 303, 276, 247, 218, 182, 135, 95 (Found: (M+), 388.2436. C 24H360 2S (M+), requires 388.2436). nmr Data inter alia for /5R , 6R]-(lE)-6,10-dimethyl’5-(2-propen-2-yl)’l - phenylsulphonylundecene 154 5 (500 MHz) 4.78 (1H, dd, J 2.5, 2.0 Hz, C//2), 4.57 (1H, d, J 2.0 Hz, CH2). 259 Thermolysis of [SR, 6R]-(1E, 3'£)-6,10-dimethyl-5-(l,3-pentadien-4-yI)- l-phenylsulphonylundecene (153) and [SR, 6/?]-(lZ?)-6,10-dimethyI-5-(2- propen-2-yl)-l-pheny!suIphonyIundecene (154). A solution of a 3:1 mixture of triene 153 and diene 154 (azeotropically dried with toluene [2 x 10 ml]; 162.2 mg, 0.417 mmol) in dry toluene (10 ml) was degassed as previously and transferred to a dry, argon filled Carius tube via cannula. The tube was evacuated and cooled in liquid nitrogen, then sealed using a flame and allowed to warm to rt behind a safety screen. The tube was then heated in a Carius oven at 240°C for 48 h. After cooling to k, the tube was opened and the solvent removed under reduced pressure to give a slightly discoloured oil. nmr analysis of the crude mixture showed the presence of two major products 155 and 156 in the ratio 1:1, together with a third compound, tentatively assigned as [IRS, 4R, 2'R]-l-[(phenylsulphonyl)methyl}’3- methylene-(6-methylhept-2-yl)cyclohexene 157. Chromatography (15% - 20% ether- petrol) gave a 1:1 mixture of [4R, 5R, 8R, 9S, 2 R]-9-methyl-8-(6-methylhept-2-yi)-4 - phenylsulphonylbicyclo[4.3.0]nonene 155 and [4R, 5R, SR, 9R, 2'R]-9-methyl-8-(6 - methylhept-2-yl)-4-phenylsulphonylbicyclo[4.3.0]nonene 156 (96.2 mg, 59%) as a 260 colourless oil; vmax (film) 3043, 2955, 1539, 1508, 1466, 1446, 1385, 1306, 1144, 1086,690 cm'1; 6 (500 MHz) both diastereoisomers: 7.88 (2H, m, orf/io-protons on Ph), 7.63 (1H, m, para-protons on Ph), 7.55 (2H, m, meta-protons on Ph); 155 inier alia : 5.80 (1H, ddd, J 10.1, 2.5, 1.4 Hz), 5.43 (1H, ddd, J 10.0, 5.5, 2.9, H-2), 3.12-3.03 (1H, m, H-4), 0.96 (3H, s, CH 3 C-9); 156 inter alia: 5.71 (1H, dt, J 10.2,1.8 Hz, H- 1), 5.63 (1H, dt, J 10.3, 4.3 Hz, H-2), 3.16-3.00 (1H, m, H-4), 1.20 (3H, s, C-9 CH3), both diastereoisomers, not assigned: 0.94 (3H, d, J 6.4 Hz, H-l'), 0.92 (3H, d, J 6.3 Hz, H-D, 0.86 (3H, d, J 6.6 Hz, H-7’ or C- 6’ CH3). 0.84 (3H, d, J 6.5 Hz, H-7’ or C- 6' CH3), 0.84 ( 6H, d, J 6.5 Hz, H-7' or C- 6' CH3 twice); m h (El) 388 (M+), 362, 246, 220, 205,135,107,93,77 (Found: (M+ + NH 4+) 406.2780. C24H360 2S requires (M+ + NH4+), 406.2780. nmr Data for 157 5 (500 MHz) inter alia : 4.67 (1H, m, CHi), 4.59 (1H, m, C// 2), 3.02 (2H, dd, J 6.3,3.0 Hz,.Ctf2S0 2Ph). 261 Preparation of [1R, 2S, 4R, 5R, 8 R, 9R, 2'R]-2-epoxy-9-methyl-8-(6- methyIhept-2-yI)-4-phenylsuIphonylbicycIo[4.3.0]nonane (159) and [1R, 2S, 4R, 5R, 8 R, 9S, 2'R]-2-epoxy-9-methyI-8-(6-methyIhept-2*yl)-4- phenylsulphonylbicyclo[4.3.0]nonane (158). To a stirred solution of a 1:1 mixture of alkenes 155 and 156 (40.2 mg, 0.1035 mmol) in DCM (0.5 ml) was added m-CPBA (54 mg of 80% w/w reagent, 0.1552 mmol, 1.5 eq) and the reaction was allowed to stir for 8 h. Analysis of the reaction mixture by tic (3 x 20% ether-petrol) showed the presence of two major and two minor epoxides. The reaction was diluted with DCM (20 ml) and washed with 10% aqueous sodium thiosulphate (2 x 20 ml), 2M aqueous sodium hydroxide (3 x 20 ml), water (20 ml), dried (MgS 04) and concentrated to give a colourless oil. Purification by chromatography (30% - 40% ether-petrol) gave, in order of elution epoxide 158 (8.9 mg, 21%) as a colourless oil; 8 (500 MHz) 7.86 (2H, dd. J 8.2, 1.4 Hz, meia-protons on Ph), 7.65 (1H, t, J 7.4 Hz, para -proton on Ph), 7.56 (1H, t, J 7.9 Hz, orr/ro-protons on Ph), 3.25 (1H, ddd, J 12.7, 9.3, 5.0 Hz, H-4), 3.22 (1H, t, J 4.4 Hz, H-2), 3.00 (1H, d, J 4.5 Hz, H-l), 2.25 (1H, dt, J 14.3, 4.6 Hz, H-3a), 2.13 (1H, td, J 9.6, 3.6 Hz, H-5), 2.06 (1H, dd, J 14.2, 12.7 Hz, H-3p), 1.74 (1H, m, H-2’), 1.66-1.60 (2H, m), 1.60-1.30 (6H, m) and 1.18-1.04 (3H, m, all comprising H- 6, H-7, H-3\ H-4’, H-5’, H-6’), 1.18 262 (3H, s, C-9 CH 3), 1.08-1.02 (1H, m, H- 8 ), 1.00 (3H, d, J 6.6 Hz, C-l* CH3), 0.85 (6H, d, J 6.7 Hz, C-6' CH3, H-7'); m/z (El) 405 (MH+), 404 (M+), 389, 333, 320, 291, 279, 263, 245, 149,133,109 (Found: (M+) 404.2385. C 24H36O3S requires (M+), 404.2385) followed by epoxide 159 ( 8.6 mg, 20.5%) as a colourless oil; vmax (film) 2943, 1624, 1571, 1522, 1456, 1307, 1147, 1084, 690 cm-1; 8 (500 MHz) 7.85 (2H, dd, J 8.3,1.3 Hz, meta-protons on Ph), 7.65 (1H, t, J 7.4 Hz, para-proton on Ph), 7.56 (1H, t, J 8.0 Hz, arr/to-protons on Ph), 3.31 (1H, t, J 3.9 Hz, H-2), 3.19 (1H, ddd, J 12.6, 10.9, 4.3 Hz, H-4), 3.06 (1H, d, J 4.3 Hz, H-l), 2.17 (1H, dt, J 14.2, 3.9 Hz, H-3p), 2.01 (1H, m, H- 6a), 1.91 (1H, dd, J 14.0, 13.1 Hz, H-3a), 1.82 (1H, ddd, J 10.8, 8.0 Hz, H-5), 1.78-1.66 (2H, m, H-7 [one proton], other unassigned), 1.52 (1H, septet, J 6.6 Hz, H-6’), 1.55-1.45 (2H, m, H-2’, H-7 [one proton]), 1.42-1.32 (2H, m) and 1.18-1.05 (3H, m, all comprising [including 6 1.78-1.66] H-3’, H-4’, H-5’), 1.22 (1H, m, H-6p), 1.01 (3H, d, J 6.6 Hz, H -l1), 1.00 (3H, s, C-9 CH 3), 0.99 (1H, m, H- 8 ), 0.87 (3H, d, J 6.6 Hz, C-6’ CH3 or H-7'), 0.86 (3H, d, J 6.6 Hz, C-6’ CH3 or H- 7'); m/z (El) 405 (MH+), 404 (M+), 389 (M+-0), 333, 291, 279, 245, 149, 143, 109 (Found: (M+ + NH 4+) 422.2729. C 24H36O3S requires (M+ + NH 4+), 422.2729). Appendix 1 264 Figure 1 Figure 2 266 H = H PhS02 55 Figure 3 267 58 Figure 4 268 60 Figure 5 269 Figure 6 270 H Figure 7 271 H Figure 8 82 Figure 9 273 X-Ray crystal data115 All data were measured on a Nicolet R3m diffractometer, with Cu-Ka radiation (X => I . 54178 A, graphite monochromator) using to-scans, with 20 á 116°. The data were all corrected for Lorentz and polarization factors; no absorption corrections were applied. All structures were solved by direct methods. The non-hydrogen atoms were refined anisotropically. Unless stated otherwise, the positions of all hydrogen atoms were idealized, C-H = 0.96Á, assigned isotropic thermal parameters, 17(H) = 1.2t/eq(C), and allowed to ride on their parent carbon atoms. All methyl groups were refined as rigid bodies. All computations were carried out using the SHELXTL 113 programme system. Crystal data for (65a): C 24H30O5S2, M - 462.6, orthorhombic, a = 9.413(8), b = II . 967(8), c = 21.721(17) A, V = 2447 Á3, space group P2 12,21, Z - 4, Dc = 1.26 g cm*3, p(Cu-Ka) = 22 cm 1, F(000) = 984. 1897 Independent reflections were measured of which 1853 had IF0I > 3a(IF 0l), and were considered to be observed. The leading hydrogen atoms on the methyl groups on the sp 2 centres were located from an A F map. The absolute configuration of the molecule was determined by an F-factor test. Refinement was by block-cascade full-matrix least squares to R ■ 0.064, Rw * 0.064 [w* 1 _ ai(f) + O.OOOOOF2]. The maximum and minimum residual electron densities in the final AF map were 0.24 and -0.41 eA *3 respectively. The mean and maximum shift/error in the final refinement were 0.012 and 0.080 respectively. Crystal data for (50): C 17H 22 0 2S, M = 290.4, monoclinic, a ■ 8.414(4), b - 9.531(4), c = 19.254(9) A, p = 90.58(4)°, V = 1544 A3, space group P2j/c, Z ■ 4, Dc * 1 25 g cm*3, p(Cu-Ka) =18 cm*1, F(000) = 624. A crystal of dimensions 0.50 x 0.27 x 0 10 mm was used. 2080 Independent reflections were measured of which 1928 had IF0I > 3a(IF0l), and were considered to be observed. Refinement was by block-cascade full- 274 matrix least squares to R = 0.044, R w = 0.052 [w 1 = o 2(F) + 0.00051F2]. The maximum and minimum residual electron densities in the final AF map were 0.27 and - 0.24 eA"3 respectively. The mean and maximum shift/error in the final refinement were 0.001 and 0.005 respectively. Crystal data for (53): C17H 2 2 O 2 S, M = 290.4, monoclinic, a = 21.559(14), b = 8.203(4), c = 17.662(9) A, p = 105.77(5)°, V = 3006 A3, space group C2/c, Z = 8 , Dc = 1.28 g cm-3, p(Cu-Ka) = 19 cm-1, F(000) = 1248. 2025 Independent reflections were measured of which 1837 had IF0I > 3o(IF 0l), and were considered to be observed. Refinement was by block-cascade full-matrix least squares to R = 0.059, Rw = 0.073 [w- i = o 2(F) + 0.00080F2]. The maximum and minimum residual electron densities in the final AF map were 0.31 and -0.48 eA ’3 respectively. The mean and maximum shift/error in the final refinement were 0.013 and 0.073 respectively. Crystal data for (55): C 17H2202S, M = 290.4, orthorhombic, a = 5.849(1), b = 14.012(4), c = 19.042(5) A, V = 1561 A3, space group P2xcn, Z = 4, Dc = 1.24 g cm-3, p(Cu-Ka) = 18 cm-1, F(000) = 624. A crystal of dimensions 0.20 x 0.30 x 0.83 mm was used. 1177 Independent reflections were measured of which 1153 had IF0I > 3o(IF 0l), and were considered to be observed. Refinement was by block-cascade full- matrix least squares to F = 0.029, F w = 0.033 [w *1 = o 2(F) + 0.00088F2]. The maximum and minimum residual electron densities in the final AF map were 0.15 and - 0.24 eA*3 respectively. The mean and maximum shift/error in the final refinement were 0.017 and 0.070 respectively. Crystal data for (58): Ci 6H 2o0 2S, M = 276.4, monoclinic, a = 8.255(4), b = 10.020(7), c = 17.716(8) A, p = 100.29(4)°, V = 1442 A3, space group P2x/n, Z = 4, Dc = 1.27 g cm-3, p(Cu-Ka) = 19 cm-1, F(000) = 592. 1932 Independent reflections were 275 measured of which 1816 had IF0I > 3a(IF 0l), and were considered to be observed. Refinement was by block-cascade full-matrix least squares to R = 0.045, Rw = 0.055 [w- l s o 2(F) + 0.00090F2]. The maximum and minimum residual electron densities in the final AF map were 0.32 and -0.25 eA *3 respectively. The mean and maximum shift/error in the final refinement were 0.008 and 0.038 respectively. Crystal data for (60): C 16H2o0 2 S, M *= 276.4, monoclinic, a « 8.457(3), b * 15.739(6), c = 10.970(4) A, p = 96.28(3)°, V - 1451 A3, space group F 2j/c, 1 - 4, Dc = 1.26 g cm'3, p(Cu-Ka) = 19 cm-1, F(000) = 592. A crystal of dimensions 0.07 x 0.33 x 0.37 mm was used. 1954 Independent reflections were measured of which 1792 had IF0I > 3o(IF 0l), and were considered to be observed. Refinement was by block-cascade full-matrix least squares to R = 0.041, Rw = 0.049 [w -1 = o 2(F) + 0.00061F2]. The maximum and minimum residual electron densities in the final a F map were 0.37 and - 0.19 eA *3 respectively. The mean and maximum shift/error in the final refinement were 0.030 and 0.288 respectively. Crystal data for (67): CpI^C^S, M = 290.4, monoclinic, a - 12.952(4), b = 7.435(3), c - 15.907(6) A, p = 101.29(3)°, V = 1502 A3, space group F 2j/c, Z = 4, Dc = 1.28 g cm-3, p(Cu-Ka) = 19 cm-1, F(000) = 624. A crystal of dimensions 0.33 x 0.33 x 0.67 mm was used. 2022 Independent reflections were measured of which 1843 had IF0I > 3a(IF0l), and were considered to be observed. A AF map revealed the positions of all the hydrogen atoms. Refinement was by block-cascade full-matrix least squares to R s 0.048, F w = 0.056 [w -1 = a2(F) + 0.00174F2]. The maximum and minimum residual electron densities in the final AF map were 0.43 and -0.26 eA *3 respectively. The mean and maximum shift/error in the final refinement were 0.016 and 0.142 respectively. Crystal data for (73): CjgH^C^S, M = 304.5, monoclinic, a - 12.995(8), b = 276 9.026(6), c = 14.605(12) A, p = 107.36(6)°, V = 1635 A3, space group P ljc , Z = 4, £>c = 1.24 g cm3, p(Cu-Ka) = 17 cm-1, F(000) = 656. 2196 Independent reflections were measured of which 2053 had IF0I > 3a(IF0l), and were considered to be observed. The bridgehead hydrogen atoms were located from a AF map and refined isotropically. Refinement was by block-cascade full-matrix least squares to R = 0.075, Rw = 0.098 [w- l = a2(F) + 0.00238F2]. The maximum and minimum residual electron densities in the final AF map were 0.54 and -0.78 eA-3 respectively. The mean and maximum shift/error in the final refinement were 0.074 and 0.320 respectively. Appendix 2 278 Entry Triene T (°C) t (h)* Yield (%)b Products h ratio0 1 17b 175 96 92 50 Xp-'Xp53 PhSOj PhS02 2 18b 165-175 146 92 56 Xt>jdD 55 PhSOj PhSOj 3 17a 145 48 93 57 Xp"Xp58 PhSOj PhSOj 4 18a 165 60 H H 5 60a 175 120 6 61a 190 60 7 60b 170-178 312 8 61b 190 120 Notes: “total reaction time; isolated yield after chromatography of combined IMDA products; determined by *H nmr analysis of crude products. Table 1: IMDA Reactions of trienes 17, 18, 60 and 61 279 Entry Substrate Yield (%)a Products Notes: ‘yield of combined products; bbased on isolated yields after chromatography of methylated products Table 2: Méthylation reactions of IMDA products 50, 53, 57 and 58 Appendix 3 281 nOe Experiments All nOe experiments were conducted on a Bruker WM-500 spectrometer at 300K 158 Irradiated Signal Enhanced Signal Percentage nOe SI ppm Assignment SI ppm Assignment 2.01 H-6a 7.85 ortho-protons 0.7 on Ph 1.48 H-7 4.0 1.82 H-5 4.5 3.19 H-4 7.85 ortho-protons 2.1 on Ph 2.17 H-3p 2.1 3.31 H-2 3.06 H-l 8.8 2.17 H-3p 1.6 1.91 H-3a 2.1 1.91 H-3a 3.31 H-2 5.0 2.17 H-3p 14.9 7.85 ortho -protons 1.9 on Ph 3.06 H-l 3.31 H-2 9.5 1.00 C-9 CH3 2.5 1.01 h - t 2.0 282 0.85 H-7', C-6' CH3 1.49 H-6' 5.6 1.01 C-9 CH3, H -l\ 3.06 H-1 13.9 H-8 1.01 C-9 CH3, H -l\ 1.82 H-5 8.1 H-8 1.49 H-2 6.2 157 Irradiated Signal Enhanced Signal Percentage nOe SI ppm . Assignment SI ppm Assignment 1.81 C-9 CH3 1.00 H -r 2.1 1.42 - 5.4 1.74 H-2' 3.5 2.06 H-3p 3.6 2.13 H-5 12.3 3.00 H-1 6.6 1.74 H-2' 3.00 H-1 7.3 1.81 C-9 CH3 2.7 1.00 C-l' c h 3 3.2 1.00 C-1’ CH3 1.21 C-9 CH3 3.4 1.40 . 4.5 283 1.50 - 2.3 1.74 H-2' 6.0 3.00 H-l 6.0 2.06 H-3ß 2.25 H-30 27.0 1.21 C-9 CH3 4.0 3.22 H-2 4.1 3.00 H-l 1.5 3.00 H-l 3.22 H-2 5.4 1.74 H-2' 6.5 1.18 C-9 CH3 2.0 1.00 H -l' 1.7 3.22, 3.25 H-2, H-4 3.00 H-l 8.6 7.86 ortho -protons 2.3 on Ph 1.18 C-9 CHS 3.4 ■ a 04 2.25 Q 2.06 H-3p 22.2 3.22 H-2 6.2 3.25 H-4 5.0 7.86 ortho- protons 2.0 on Ph 284 PPM 285 PhS02 159 2D Spectrum _ l.CO . 2.00 _ 3.00 _ 4.00 _ 5.00 _ 6.00 _ 7.00 _ 8.00 PPM 286 PPM o co 287 PhSO. 2D Spectrum 00 00 00 00 00 00 00 00 B. 00 7.00 6.00 5.00 4.00 3.00 2.00 288 7.0 References 1. Maehr, H. J. Chem. Ed. 1985, 62, 114. 2 (a) . Brieger, G. J. Am. Chem. Soc. 1965, 85, 3763. (b) . Klemm, L. H. Tetrahedron Lett. 1963, 1243. (c) . House, H. O.; Cronin, T. H. J. Org. Chem. 1965, 30, 1061. (a) . Carruthers, W. Cycloaddition Reactions in Organic Synthesis ; Pergamon: Oxford, 1990. (b) . Taber, D. F. Intramolecular Diels-Alder Reactions and Alder Ene Reactions', Springer Verlag: New York, 1984. (a) . Craig, D. Chem. Soc. Rev. 1987, 16, 187. (b) . Edwards, M. P. Ph.D. thesis, University of London, 1982. (c) . Fallis, A.G. Can. J. Chem. 1984, 62, 183. (d) . Ciganek, E. Org. React. 1984,32,1. (e) . Brieger,G.; Bennett, J. N. Chem. Rev. 1980, 80, 63. (f) . Oppolzer,W. Synthesis 1978, 798. (g) . Oppolzer,W. Angew. Chem., Int.. Ed. Engl. 1978, 16, 798. 5. Boeckman, R. K. Jr.; Alessi, T.R. J. Am. Chem. Soc. 1982, 104, 3216. 6 . Roush, W. R.; Essenfeld, A.P.; Warmus, J. S. Tetrahedron Lett. 1987, 28, 2447 and references cited therein. 7. Roush, W. R.; Hall, S. E. /. Org. Chem. 1982, 47, 4825. 8. Lin, Y.-T.; Houk, K. N. Tetrahedron Lett. 1985, 26, 2269. 9. Roush, W. R., Ko, A. I.; Gillis, H. R. J. Org. Chem. 1980, 45, 4264. 10. Roush, W. R.; Gillis, H. R.; Ko, A. I. J. Am. Chem. Soc. 1 9 8 2 , 104, 2269. 11. Kurth, M. J.; O’Brien, M. J.; Hope, H.; Yanuck, M. J. Org. Chem. 1985, 50, 2626. 289 12. Brown, F. K.; Houk, K. N. Tetrahedron Lett. 1985,26, 2297 and references cited therein. 13. Knochel, P.; Retherford, C. Tetrahedron Lett. 1990, 32 ,441. 14. Fleming, I. Frontier Orbitals and Organic Chemical Reactions', Wiley: New York, 1976. 15. For example, in E-a,p-unsaturated sulphones, the p-proton resonates downfield and the y-protons upfield compared with the Z-isomer. See, for example: Craig, D.; Ley, S. V.; Simpkins, N. S.; Whitham, G. H.; Prior, M. J. J. Chem. Soc., Perkin Trans. 1 1985, 1949. 16 (a). Simpkins, N. S. Sulphones in Organic Synthesis', Pergamon: Oxford, manuscript in preparation. (b) . Simpkins, N. S. Tetrahedron 1990, 46, 6951 (c) . Trost, B. M. Bull. Chem. Soc. Jpn. 1988, 61, 107. (d) . Fuchs, P. L.; Braish, T. F. Chem. Rev. 1986, 86, 903. (e) . Magnus, P. D. Tetrahedron 1977, 33, 2019. (f) . The Chemistry of Sulphones and Sulphoxides; Patai, S.; Rappoport, Z.; Stirling, C. Eds.; Wiley: Chichester, 1988. 17. See references 3 (a), 15 (a), (b) and also: DeLucchi, O.; Pasquato, L. Tetrahedron 1988, 44, 6755. 18 (a). Weichert, A.; Hoffmann, H. M. R. J. Chem. Soc., Perkin Trans. 1 1990, 2154. (b). Chou, S.-S. P.; Wey, S.-J. J. Org. Chem. 1990, 55, 1270. 19 (a). Carr, R. V. C.; Paquette, L. A. J. Am. Chem. Soc. 1980, 102, 853. (b) Carr, R. V. C.; Williams, R. V.; Paquette, L. A. J. Org. Chem. 1983, 48, 4976. 20. Kametani, T.; Aizawa, M.; Nemoto, H. Tetrahedron 1981, 37 , 2547 21. Kametani, T.; Nemoto, H. Tetrahedron 1981, 37, 3. 290 22 (a). Shishido, K.; Hiroya, K.; Heno, Y.; Fukumoto, K.; Kametani, T.; Honda, T. J. Ckem. Soc., Perkin Trans. 1 1986, 829. (b). Shishido, K.; Hiroya, K.; Fukumoto, K.; Kametani, T. Ibid. 1986, 837. 23. Corey, E. J.; Da Silva Jardine, P.; Rohloff, J. C. J. Am. Chem. Soc. 1988, 110, 3672. 24. Strekowsi, S.; Kong, S.; Battiste, J. C. J. Org. Chem. 1988, 53, 901. 25. (a). Preliminary communication: Craig, D.; Fischer, D. A.; Kemal, Ö.; Plessner, T. Tetrahedron Lett. 1988, 29, 6369. (b). Full account: Craig, D.; Fischer, D. A.; Kemal, Ö.; Marsh, A.; Plessner, T.; Slawin, A. M. Z.; Williams, D. J. Tetrahedron 1991, 47, 3095. 26. Kang, S.-K.; Kim, W. S.; Moon, B.-H. Synthesis 1985, 1161. 27. Greene, T. W. Protective Groups in Organic Synthesis; Wiley: New York, 1981. 28 (a). Acetone as solvent: Schank, K. Justus Liebigs Ann. Chem. 1967, 702, 75. (b) . Idem.; Ibid.; 1968,774, 117. (c) . Aqueous DMF/ultrasonication: Biswas, G. K.; Jash, S. S.; Bhattacharyya, P. Ind. J. Chem. 1990, 29(B), 491. (d) . DBU/Acetonitrile: Biswas, G. K.; Mal, D. J. Chem. Res. (C) 1988, 308. 29 (a). Julia, M.; Paris, J.-M. Tetrahedron Lett. 1973, 4833. (b) . Kocienski, P. J.; Lythgoe, B.; Ruston, S. J. Chem. Soc., Perkin Trans. 1 1978, 829. (c) . Kocienski, P. J.; Lythgoe, B.; Waterhouse, I. Ibid. 1980,1045. 30. Mancuso, A. J.; Swem, D. Synthesis 1981, 165. 31 (a) For previous syntheses see: reference 11 and Larsen, S. D.; Grieco, P. A. J. Am. Chem. Soc. 1985, 107, 1768. (b) . Samain, D.; Descoins, G; Commeron, A. Synthesis 1978, 388. (c) . Fouget, G.; Schlosser, M. Angew. Chem., Int. Ed. Engl. 1974, 13, 82. 291 32. Roush, W. R.; Hall, S. E. J. Am. Chem. Soc. 1981, 103, 5200. 33. Nystrom, R. F.; Brown, W. G. J. Am. Chem. Soc. 1947, 69, 2548. 34. For a related procedure for the synthesis of vinylic sulphoximines, see: Bailey, P. L.; Clegg, W.; Jackson, R. F. W.; Meth-Cohn, O. J. Chem. Soc., Perkin Trans. 1 1991, 200. 35. For example, see: Ley, S. V.; Anthony, N. J.; Armstrong, A.; Brasca, M. G.; Clarke, T.; Culshaw, D. Greek, C.; Grice, P.; Jones, A. B.; Lygo, B.; Madin, A.; Sheppard, R. N.; Slawin, A. M. Z.; Williams, D. J. Tetrahedron 1989, 45, 7161. 36. (a) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405. (b) For a comparison of the effect of cation, temperature, and solvent on the £: Z ratio of olefinations using trimethylphosphonoacetate, see: Heathcock, C. H.; Thomson, S. K. J. Org. Chem. 1990,55, 3386. 37. See reference 15. 38 (a). Corey, E. J.; Kwiatkowski, G. T. J. Am. Chem. Soc. 1966, 88, 3652. (b) . Idem. Ibid. 1966,85, 5653. (c) . Idem. Ibid. 1968, 90, 6816. (d) . Corey, E. J.; Cane, D. E. J. Org. Chem. 1969, 34, 3053. 39. Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 2173. 40. Seyferth, D.; Vaughan, L. G J. Am. Chem. Soc. 1964, 86, 883. 41. (a). Synthesis of PhS02F: personal communication, Diorazio; L. (b) . Phenylsulphonate esters are also recognised as a source of ArSC>2+: Baarschers, W. H.; Can. J. Chem. 1976, 54, 3056. (c) . For the attempted reaction of a lithioalkyne with PhS020Ts, see: Shipman, M.; Ph.D. thesis, University of London, 1990. 42. Labadie, S. S. J. Org. Chem. 1989, 54, 2496. 43. (a). Labadie, J. W.; Stille, J. K.; J. Am. Chem. Soc. 1984, 106, 6417. 2 9 2 (b). Goure, W. F.; Wright, M. E.; Davis, F. P.; Labadie, S. S.; Stille, J. K. J. Am. Chem. Soc. 1984, 106, 6417. 44. Craig, D.; Daniels, K.; Marsh, A.; Rainford, D.; Smith, A. M.; Synlett, 1990, 531. 45. (a). Mikolajczyk, M.; Grzejszczak, S.; Zatorski, A. J. Org. Chem. 1975, 40, 1979. (b). Mikolajczyk, M.; Midura, W.; Grzejszczak, S.; Zatorski, A. Chefczynska, A. J. Org. Chem. 1978, 43, 473. 46. Cooper, G. D. J. Am. Chem. Soc. 1954, 76, 3713. 47. Trost, B. M.; Curran, D. P. Tetrahedron Lett. 1981,22, 1287. 48. Trost, B. M.; Braslau, R. J. Org. Chem. 1988,22, 1287. 49. Carlsen, P. H.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. J. Org. Chem. 1981, 46, 3936. 50. McKillop, A.; Tarbin, J. A. Tetrahedron Lett. 1983, 2 4 ,1505. 51. For an example of chemospecific oxidation in the presence of an isolated, disubstituted double bond using this reagent system, see: Ley, S. V.; Edwards, M. P.; Lister, S. G.; Palmer, B. D.; Williams, D. J. J. Org. Chem. 1984, 49, 3503. 52. Palmer, J. T.; Fuchs, P. L. Synth. Commun. 1988, 233. 53 (a). Backer, H.; Bottema, J. Reel. Trav. Chim. Pays Bas, 1932, 51, 294. (b) . For more recent work on thiophene-S, 5-dioxides in synthesis, .see: Guziec, F. S., Jr.; Sanfilippo, L. J. Tetrahedron 1988, 44, 6241. (c) . It has been reported that S, 5-thiophene dioxides undergo alkylation at C-2 leading to a convenient entry into this system. Several workers have found this to be a capricious reaction however, due to the instability of the anion: personal communication, Clasby, M. 293 54 (a) . Griffith, W. P.; Ley, S. V.; Whitcombe, G.P.; White, A. D. J. Chem. Soc., Chem. Commun. 1987, 1625. (b) . Griffith, W. P.; Ley, S. V. Aldrichimica Acta 1990, 23, 13. 55. For an example of a Wittig reaction with methoxycarbonylmethylene- triphenylphosphorane of a labile aldehyde generated in situ, see: Ireland, R. E.; Hàbich, D.; Norbeck, D. W. J. Am. Chem. Soc. 1985, 107, 3271. 56. Ryn, I.; Kusumoto, N.; Ogawa, A.; Kambe, N.; Sonoda, N. Organometaliics 1989, 8 , 2279. 57. Daeuble, J. F.; McGettigan, C.; Stryker, J. M. Tetrahedron Lett. 1990, 31, 2397 and references cited therein. 58. Lee, J. W.; Oh, D. Y. Synlett. 1990, 290. 59. Crystal grown by Dr D. Craig. 60. For example, see: Wattanasin, S.; Kathawala, F. G.; Boeckman, R. K., Jr.; J. Org. Chem. 1985 ,50, 3810. 61. For example, see reference 5 62. (7£>2-phenylsulphonyl-1,7,9-decatriene undergoes partial isomeriastion to (6£, 8£)-2-phenylsulphonyl-l,6,8-decatriene under thermal IMDA conditions (180°C, PhMe, 72 h): Clasby, M.; personal communication. 63. For isomerisation of a Z, Z, E-triene in a thermal IMDA, see: Borch, R. F.; Evans, A. J.; Wade, J. J. /. Am. Chem. Soc. 1975, 97, 6282 and Borch, R.F.; Evans, A. J.; Wade, J. JJbid. 1977, 99, 1612. 64. Prepared by reaction of sodium phenylsulphinate with bromoethane in DMSO (51 % after recrystallisation). 65. For a similar elimination procedure, see: Julia, M.; Launay, M.; Stacino, J.- P.; Verpeaux, J.-N. Tetrahedron Lett. 1982, 23, 2465. 66. These reactions carried out by Drs. D. Craig and Ô. Kemal. For example see:Eisch, J. J.; Galle, J. E. J. Org. Chem. 1979, 44, 3279. 294 67 Sulphones may also as a ketene-equivalent by virtue of the fact that they may be removed oxidatively: (a) . MoOPH oxidation of an a-sulphonyl anion: Little, R. D.; Myong, S. O. Tetrahedron Lett. 1980, 21, 3339. (b) . (TMS)20 oxidation of an a-sulphonyl anion: Hwu, J. R. J. Org. Chem. 1983, 48, 4432. 68. Torisawa, Y.; Satoh, A.; Ikegami, S.Tetrahedron Lett. 1988, 2 9 ,1729. 69. Ogura, K.; Yahata, N.; Hashizume, K.; Tsuyama, K.; Takahishi, K.; Iida, H. Chem. Lett. 1983, 767. 70. Craig, D. unpublished work, this laboratory. 71. Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents', In Best Synthetic Methods Series, Katritzky, A. R.; Meth-Cohn, O.; Rees, C. W.; Eds.; Academic: London, 1988. 72. Rylander, P. N. Hydrogenation Methods', In Best Synthetic Methods Series, Katritzky, A. R.; Meth-Cohn, O.; Rees, C. W.; Eds.Acadcmic: London, 1985. 73. Gais, H.-J.; Hellmann, G.; Günther, H.; Lopez, F.; Lindner, H. J.; Braun, S. Angew. Chem., Int. Ed. Engl. 1989, 28, 1027. For a review on the crystal structures of lithiosulphones see: Boche, G. Angew. Chem., Int. Ed. Engl. 1989, 28, 277. 74 (a). Nemoto, H.; Kurobe, H.; Fukumoto, K.; Kametani, T. Heterocycles 1985, 23, 567. (b). Kocienski, P. J.; Lythgoe, B.; Ruston, S. J. Chem. Soc., Perkin Trans. 1 1979, 1290. 75. Roush, W. R.; Peseckis, S. M. J. Am. Chem. Soc. 1981, 103, 6696. 76. Wadsworth, W. S. Org. React. 1977, 25, 73. 77. Evans, D. A. Aldrichimica Acta 1982, 15, 23. 295 78. Burke, S. D.; Powner, T. H.; Kageyama, M. Tetrahedron Lett. 1983, 00, 4529. 79. Herczegh, P.; Zsely, M.; Szilagy, L.; Dinya, Z.; Bognar, I. Tetrahedron 1989, 45, 5995. 80. For the synthesis of a related electrophile see reference 4 (b). 81. Finkelstein, H. Chem. Ber. 1910,43, 1528. 82. Creger, P. L. J. Am. Chem. Soc. 1970, 92, 1938. For a review, see: Petragnani, N.; Yonashiro, M. Synthesis, 1982, 521. 83. Seebach, D. Chem. Ber. 1985,27, 632. 84. Aldrichimica Acta 1983, 16, 3. 85. Bare, T. M.; House, H. O. Org. Syn. Coll.', Baumgarten, E. Ed.; Wiley: New York, 1973, vol. V, p 775. 86. Prepared in 93% yield by the reaction of sodium phenylsulphinate with allyl bromide in DMSO. 87 (a). For a comprehensive review, see: Negishi, E.; Takahashi, T. Aldrichimica Acta 1985, 18, 31. (b). Matsushita, H.; Negishi, E. J. Am. Chem. Soc. 1981, 103, 2882. 88. Negishi, E.; Okukado, N.; King, A. O.; Van Horn, D. E.; Spiegel, B. I. J. Am. Chem. Soc. 1978, 100, 2254. 89. House, H. O. Modern Synthetic Reactions, 2nd. Edn.; Benjamin: London, 1972, 90. Winterfeld, E. Synthesis 1975, 617. 91. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769. 92. Matsumo, M.; Kuroda, K. Tetrahedron Lett. 1999,4021. 93. Negishi, E.; Van Horn, D. E.; King, A. O.; Okukado, N. Synthesis 1979, 501. 94. Dang, H. P.; Linstrumelle, G. Tetrahedron Lett. 1978,1027. 2 9 6 95. Personal communication, Tilbrook, M. 96. The author thanks Dr N. S. Isaacs, SERC High Pressure Reactions Facility, Reading University for performing this experiment. 97. The vapour pressure for toluene in a sealed vessel at 178°C is ca. 5 bar; CRC Handbook of Chemistry and Physics, 66th Edn; Chemical Rubber Company: Boca Raton, Florida, 1976. 98. For an example of a pressure mediated IMDA reaction, see: Harwood, L. M.; Leeming, S. A.; Isaacs, N. S.; Jones, G. Pickard, J.; Thomas, R. M; Watkin, D. Tetrahedron Lett. 1988,29,5017. 99 (a). Parker, K.; Iqbal, T. J. Org. Chem. 1982, 47, 337. (b). Parker, K.; Iqbal, T. J. Org. Chem. 1987, 52, 4369. 100 Performed using Macromodel on a CAChE system. The author thanks Dr. P. Grice for help with this experiment. 101. Karplus-equation, for dihedral<(>, J in Hz: 0° angle£ $ :£ 90°, J * 8.5 cos^ 4» - 0.28; 90° £ <|> <. 180°, J = 9.5 cos 2 1959 ,30, 1. 102. Technical grade /?-(+)-pulegone purchased from Aldrich Chemical Co. was shown to be of high purity by optical rotation and 500 MHz JH nmr assay. 103. Plesek, J. Collect. Czech. Chem. Soc. 1957, 22, 644. 104. Newman, M. S.; Kutner, A. J. Am. Chem. Soc. 1951, 73, 4199; after Crowther, H. L.; McCombie, R. J. Chem. Soc. 1913, 103, 27. 105 (a). Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981, 103, 1737. (b). Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982, 104, 2127. For full experimental details, see: Evans, D. A.; Dow, R. L.; Shih, T. L.; Takacs, J. M. J. Am. Chem. Soc. 1990, 112, 5290. and references therein. 297 106. The sodium enolate has been found to be more reactive than the lithium counterpart: reference 105 (b). 107. Helmchen, G.; Nill, G.; Hocherzi, D.; Youssef, M. S. K. Angew. Chem., Int. Ed. Engl. 1979, 18, 163. 108 (a). Culshaw, D. Ph.D. thesis, University of London, 1983. (b). Hudlicky, T.; Luna, H.; Price, S. D. J. Org. Chem. 1990, 55, 4683. 109 See references 4 (b) and 51. 110. (a). Pine, S. H.; Zahler, R.; Evans, D. A.; Grubbs, R. H. J. Am. Chem. Soc. 1980, 102, 3270. (b) . Prepared according to: Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am. Chem. Soc. 1978,700, 3611. (c) . Thanks to Hartmuth Kolb for providing high-quality Tebbe reagent. 111. (a). Deoxygenation usingTMS-I: Caputo, R.; Mangoni, L.; Ncri, O.; Palumbo, G. Tetrahedron Lett. 1981 ,22, 3551; Denis, J. N.; Magnane, R.; Van Eenoo, M.; Krief, A. Nouv. J. Chem. 1979, 3, 705. (b). Using Sml2: Girard, P.; Namy, J. L.; Kagan, H. J. Am. Chem. Soc. 1980, 102, 2693. 112. Perrin, D.D.; Armarego, W. L. Purificatiuon of Laboratory Chemicals, 3rd Edn.; Pergamon: Oxford, 1988. 113. Still, W. C.; Khan, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923. 114. Atomic coordinates are available on request from the Director of the Cambridge Crystallographic Data Centre, University Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW. 115. Sheldrick, G. M. SHELXTL, An Integrated System for Solving, Refining and Displaying Crystal Structures from Diffraction Data, Revision 5.2; University of Gottingen: 1985. 298 8.0 Corrigenda 299