UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The chemistry of thujone : the synthesis of rose oil components and germacrane analogues Gunning, Philip James 1991

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

Item Metadata

Download

Media
831-UBC_1991_A1 G86.pdf [ 13.41MB ]
Metadata
JSON: 831-1.0059759.json
JSON-LD: 831-1.0059759-ld.json
RDF/XML (Pretty): 831-1.0059759-rdf.xml
RDF/JSON: 831-1.0059759-rdf.json
Turtle: 831-1.0059759-turtle.txt
N-Triples: 831-1.0059759-rdf-ntriples.txt
Original Record: 831-1.0059759-source.json
Full Text
831-1.0059759-fulltext.txt
Citation
831-1.0059759.ris

Full Text

THE CHEMISTRY OF THUJONE: THE SYNTHESIS OF ROSE OIL COMPONENTS AND GERMACRANE ANALOGUES  by PHILIP JAMES GUNNING B.A. (Hons), St. John's College, Oxford University, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA May 1991 i  ©Philip James Gunning, 1991  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or  by  his or  her  representatives.  It  is  understood that  copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  DE-6 (2/88)  Abstract. This thesis is concerned with the synthesis of natural products from thujone (1), a readily available starting material obtained from Western red cedar. The first part of this study investigates the synthesis of the commercially important fragrances, (3-damascone (8) and pVdamascenone (24), which are components of rose oil. Thujone can be efficiently converted to the dimethylated thujone derivative 59 in a two step alkylation process. After the formation of the trimethylsilyl cyanohydrins, 77 and 78, the cyclopropane and isopropyl functionalities were cleaved to give the ketone 103. Further elaboration of 103 gave the key intermediates, 127 and 128. The unsaturated nitriles 128 and 127 can be converted to 8 and 24, respectively, by a reduction to the corresponding aldehydes followed by a Grignard reaction to attach the side-chain and subsequent oxidation. As a model study for the synthesis of 8 and 24 from a cyclohexanone derivative, the nitriles 127 and 128 were efficiently synthesised, in 81% overall yield, from 2,2,6trimethylcyclohexanone. Formation of the cyanohydrins 149 and 150 was followed by consecutive 'trans' and 'cis' eliminations to give the nitrile 128. The nitrile 127 was produced from 128 by allylic bromination, followed by hydrolysis and dehydration. A conversion of thujone into the ketone 179, using bromine to effect cyclopropane ringopening, was also studied. The second part of this study investigates the synthesis of ten-membered rings via a photo-induced oxidative cleavage of the alcohols 245 and 265. Treatment of 245 with lead tetraacetate under ultraviolet irradiation afforded, as the main isolated product, the tenmembered carbocycle 246. Treatment of 265 with iodobenzene diacetate under ultraviolet irradiation afforded, as the main isolated product, the bicyclic alcohol 297.  246  297  iv  Table of Contents Abstract  ii  List of Figures  ix  List of Tables  x  List of Abbreviations Acknowledgements Chapter 1. General Introduction.  xi xiv 1  Chapter 2. The synthesis of damascones 2.1. Introduction  4  2.1.1. The perception of smell and stereochemical dependence.  4  2.1.2. The damascones.  9  2.2. Results and discussion  19  2.2.1. The methylation of thujone.  21  2.2.2. The synthesis of P-damascone and {3-damascenone from thujone.  35  2.2.3. The synthesis of the key intermediates, p-cyclogeranonitrile and safronitrile, from 2,6-dimethylcyclohexanone. 2.2.4. The synthesis of compounds related to P-damascone.  66 79  2.2.5. The conversion of thujone to 3-(l-methylethyl)-2,6,6-trimethylcyclohex-2-en-1 -one. 2.3. Future developments.  84 89  2.4. Experimental. 2.4.1.  General Experimental.  92  2.4.2.  Monomethylated thujone 58 and dimethylated thujone 59.  95  2.4.3.  Dimethylated thujones 59 and 63 and trimethylated thujone 61. Method A .  97  Method B.  100  2.4.4.  Equilibration of 59 and 63.  100  2.4.5.  Silyl enol ether 67.  101  2.4.6.  Enol carbonate 71.  102  2.4.7.  Enol carbonate 73.  104  2.4.8.  Enamines 74 and 75.  105  2.4.9.  Trimethylsilyl cyanohydrins 77 and 78.  106  2.4.10. Alkenes 87/88, ketones 89/90 and alcohols 91/92.  107  2.4.11. Chlorides 97/98.  110  2.4.12. Chloride 95.  Ill  2.4.13. Alkenes 101 and 102.  112  2.4.14. Alcohols 105, 106, 107 and 108.  113  2.4.15. Chlorides 109 and 110.  114  2.4.16. Alkenes 111 and 112.  115  2.4.17. Ketones 103 and 104. Method A .  117  Method B.  118  2.4.18. Diols 121 and 122.  120  2.4.19. Aldehyde 123 and nitrile 115. Method A .  122  Method B.  123  2.4.20. Enone 124 and enol 125.  124  2.4.21. Alcohol 126. Method A .  125  Method B.  126  Method C.  127  2.4.22. Bromide 129.  127  2.4.23. 2,2,6-Trimethylcyclohexanone 142.  128  2.4.24. Trimethylsilyl cyanohydrins 153 and 154.  130  2.4.25. Cyanohydrins 149 and 150. Method A .  131  Method B.  132  Method C.  132  2.4.26. Acetates 155 and 156. Method A .  133  Method B.  134  Method C.  135  2.4.27. Nitriles 128 and 151.  vi  Method A.  135  Method B.  137  Method C.  137  Method D.  138  Method E.  138  Method F.  139  2.4.28. Ketone 152. 2.4.29. P-Cyclocitral (35).  140 141  2.4.30. Safronitrile 127. Method A.  142  Method B.  142  2.4.31. Safronal 134.  143  2.4.32. Alcohol 135.  144  2.4.33. Alcohols 139 and 140.  145  2.4.34. P-Damascenone (24). Method A.  147  Method B.  148  2.4.35. Aldehyde 160.  149  2.4.36. Alcohols 161,162 and 164.  150  2.4.37. Diketone 158 and keto alcohol 165.  152  2.4.38. Alcohols 166 and 167.  153  2.4.39. Silyl ether 168.  155  2.4.40. Aldehyde 169.  156  2.4.41. Bromide 175, dienone 176 and dibromide 177.  157  2.4.42. Bromo thujone derivative 180.  159  2.4.43. Dienone 178.  159  2.4.44. Enones 179 and 181.  160  Chapter 3. The synthesis of germacranes. 3.1. Introduction.  162  3.2. Results and discussion.  179  3.2.1. The synthesis of alcohol 265.  179  3.2.2. The synthesis of tertiary alcohol 245.  188  3.2.3. The ring-opening of 245 to a ten-membered ring.  193  vii 3.2.4. The photo-induced cleavage of the cyclopropane group in 265. 204 3.2.5. The synthesis of epoxy ketone 302 with known stereochemistry. 215 3.3. Experimental 3.3.1. 3.3.2.  Diketone268. Ketals 266 and 269.  220 221  3.3.3. Epoxide 273. 3.3.4. Alcohol 265. Method A. Method B.  224  3.3.5. Dienone 286. 3.3.6. Epoxide 290. 3.3.7. Alcohol 245. 3.3.8. Acetate 246. Method A. Method B. 3.3.9. Diene 296. 3.3.10. Alkenes 294 and 295. 3.3.11. Acetate 297. Method A. Method B. 3.3.12. Diene 297. Method A. Method B. 3.3.13. Diene300. 3.3.14. Diol 298. 3.3.15. Epoxy ketone 302. Method A. Method B. 3.3.16. Epoxy ketone 303. 3.3.17. Allylic alcohol 306. Method A. Method B. 3.3.18. Allylic acetate 309.  229 230 231  225 227  233 235 236 237 238 . 239 240 241 241 242 244 245 245 246 248 249  viii  3.3.19. Allylic acetate 310.  250  3.3.20. Epoxy alcohol 308.  251  References  252  Appendices X-ray structure report on 156. X-ray structure report on 245. X-ray structure report on 298. X-ray structure report on 308.  261 268 278 289  List of Figures  4  Figure 1. Drawing of the bisected nasal cavity. Figure 2.  Structures of the seven known damascones.  10  Figure 3.  *H nmr spectrum of dimethylated thujone 59.  25  Figure 4.  *H nmr spectrum of P-damascenone (24).  63  Figure 5.  Single crystal X-ray structure of acetate 156.  77  Figure 6.  Examples of naturally occurring germacranes.  162  Figure 7.  Single crystal X-ray structure of alcohol 245.  191  Figure 8.  !H nmr spectrum of acetate 246 .  197  Figure 9.  H nmr spectrum of acetate 297 (CDC1 ).  l  3  206  Figure 10. *H nmr spectrum of acetate 297 (C7D8).  208  Figure 11. !H nmr spectrum of diol 298.  210  Figure 12. Single crystal X-ray structure of diol 298.  211  Figure 13. Single crystal X-ray structure of epoxide 308.  219  Figure 14. Single crystal X-ray structure of acetate 156 (stereo view).  267  Figure 15. Single crystal X-ray structure of alcohol 245 (stereo view).  270  Figure 16. Single crystal X-ray structure of diol 298.  280  Figure 17. Single crystal X-ray structure of epoxide 308.  292  List of Tables Table 1.  Methylation of thujone with potassium r-butoxide and iodomethane.  23  Table 2.  Spinning band distillation of a mixture of 58, 59, 63 and 61.  29  Table 3.  Formation of 149 and 150 from 2,2,6-trimethylcyclohexanone.  71  Table 4.  The effect of reaction time in the reaction of 245 with lead tetraacetate. 202  Table 5.  The effect of stoichiometry in the reaction of 245.  202  Table 6.  The effect of acetic acid in the reaction of 245.  203  Table 7.  The effect of buffers and light in the reaction of 245.  204  Table 8.  The effect of varying stoichiometry on the isolated yield of 297.  213  Table 9.  Final atomic coordinates and B  Table 10.  Bond lengths [Compound 156].  264  Table 11.  Bond angles [Compound 156].  264  Table 12.  Torsion or conformation angles [Compound 156].  265  Table 13.  Final atomic coordinates and B  Table 14.  Bond lengths [Compound 245].  272  Table 15.  Bond angles [Compound 245].  273  Table 16.  Torsion or conformation angles [Compound 245].  274  Table 17.  Final atomic coordinates and B  Table 18.  Bond lengths [Compound 298].  282  Table 19.  Bond angles [Compound 298].  283  Table 20.  Torsion or conformation angles [Compound 298].  284  Table 21.  Final atomic coordinates and B  Table 22.  Bond lengths [Compound 308].  292  Table 23.  Bond angles [Compound 308].  293  Table 24.  Torsion or conformation angles [Compound 308].  294  e q  e q  e q  e q  [Compound 156].  [Compound 245].  [Compound 298].  [Compound 308].  263  271  281  291  xi  List of Abbreviations [oc] D  specific optical rotation recorded at 22°C using sodium D-line  Ac  acetyl  AD3N  2,2-azobis(2-methylpropionitrile)  anal.  elemental analysis  APT  attached proton test  ax  axial  br  broad  Bu  butyl, - ( C H ) 2 C H  r-Bu  f-butyl, - C ( C H ) C H  c  concentration (g/100 ml)  22  2  C  3  3  2  3  Celsius  cm"  wavenumber  8  chemical shift  d  doublet  DABCO  1,4-diazabicyclo[2.2.0]octane  DBU  1,8-diazabicyclo[5.4.0]undec-7-ene  DCC  1,3-dicyclohexylcarbodiimide  dd  doublet of doublets  ddd  doublet of doublet of doublets  dddd  doublet of doublet of doublet of doublets  DEAD  diethyl azodicarboxylate  DIBAL-H  diisobutylaluminium hydride  DMAP  4-dimethylaminopyridine  DME  dimethoxyethane  DMF  N, N-dimethylformamide  DMSO  dimethylsulphoxide  dq  doublet of quartets  dt  doublet of triplets  e  extinction coefficient  ee  enantiomeric excess  eq  equatorial  1  xii Et g GC GCMS h hv  ethyl gram gas-liquid chromatography gas-liquid chromatography-mass spectrometry hour energy  HMPA Hz I  hexamethylphosphoramide hertz impurity  ir /  infrared coupling constant  X  wavelength  1  litre  LDA LTA m M M max MCPBA Me mg MHz min |il  lithium diisopropylamide lead tetraacetate multiplet molar molecular ion maximum mera-chloroperoxybenzoic acid methyl milligrams megahertz minute microlitre  ml  millilitre  mmol  millimole  mol mp Ms m/z v  mole melting point methanesulphonyl mass to charge ratio frequency  NBS NIS  N-bromosuccinimide iV-iodosuccinimide  +  xiii  nm  nanometre  nmr  nuclear magnetic resonance  nOe  nuclear Overhauser effect  Ph  phenyl  ppm  parts per million  j-Pr  isopropyl, -CH(CH3)2  q  quartet  qd  quartet of doublets  Red-Al  sodium bis(2-methoxyethoxy)aluminium hydride  s  singlet  S  solvent  t  triplet  TBDMS  r-butyldimethylsilyl  TBHP  f-butyl hydroperoxide  td  triplet of doublets  tt  triplet of triplets  THF  tetrahydrofuran  TIPS  triisopropylsilyl  TMS  trimethylsilyl (or tetramethylsilane)  Ts  para-toluenesulphonyl  uv  ultraviolet  W  water  wh/2  width at half height  xiv  Acknowledgements I would like to thank my research director, Dr. Jim Kutney, for the unending help, support and enthusiasm which he has displayed over the last four years. I would also like to thank Dr. Krystyna Piotrowska and the other members, past and present, of the group who I have known during my time here. I would like to thank Dr. R. Clewley and Mr. J. Somerville for the teamwork displayed on the damascone project. I would like to thank all the support staff in the department for all the help they have given me. I am very grateful to Drs. L. Weiler, C. Stone and D. Williams and Mr. N . Pendleton for the help in the preparation of this thesis. Especial thanks go to my wife, Jennifer, for her unending support and to my son, Richard, for his help in rearranging my thesis.  XV  For Maud, Fred, Flo and Bert  "That's the trouble with being a chemist. Sometimes you can't actually think. You just look at a piece of blank paper waiting for it to speak to you." Gary Grant, Monkey Business (1953).  "A scientist who wants to do something original and important must experience a shock" Sir Peter Medawar.  Chapter 1. General Introduction.  Economic and environmental pressure on the forest industry of British Columbia has stimulated the search for more efficient use of the available resources. One of the largest waste products of any logging operation is the left-over bark, branches and leaves, generally termed "slash". This slash is often burnt under controlled conditions to provide nutrients for reforestation and to reduce the amount of combustible material that constitutes a fire hazard in dry spells. If the slash from Western red cedar (Thuja plicata Donn) is shredded and then steam distilled an essential oil is produced that contains as much as 1  88% thujone ( 1 ) . The steam distillation is done "on site" immediately after a logging operation, thus serving the dual purpose of removing the left-over slash and of providing  an inexpensive source of thujone. The essential oil can be sold for use in the perfumery industry or for the isolation, and subsequent use, of thujone as a chiral building block in the synthesis of biologically important, optically active molecules.  As environmental  pressure on the oil industry increases, the use of thujone (a renewable resource) as a general starting material in the synthesis of a host of organic compounds, with or without chirality, will surely increase.  Currently this oil is produced by Intrinsic Research and Development Inc., Vancouver, B.C. to which we are indebted for a gift of this oil. 1  2  A large amount of synthetic work using the monoterpene thujone ( 1 ) has been performed in our laboratory over the past two decades. The work has shown that thujone 2  is a viable chiral starting material for the synthesis of juvenile hormone analogues (2), pyrethroid insecticides (3), aryl terpenoids (4), sesquiterpenes (5), steroids (6) and insect antifeedants (7). Simple distillation of Western red cedar leaf oil yields an oil containing up to 95% thujone. The thujone is actually a mixture of the two isomers, (-)-3-isothujone (2) and (+)-3-thujone (3) (2:3 = 10:1). The separation of the two isomers has been reported (8), but for the purposes of this thesis separation was not required.  2  3  Originally there was confusion about the C4-methyl stereochemistry which gave rise to some inconsistency in nomenclature which has persisted until quite recently. Throughout this thesis the nomenclature proposed by Acharya (9) has been used in which the prefix iso is used when the C4-methyl is cis to the isopropyl group. For the mixture obtained by distillation of the essential oil the term thujone, implying no specific stereochemistry at C4, will be used. The thesis is divided into two parts: part one deals with the synthesis of the naturally occurring damascones from both thujone and dimethylcyclohexanone. The damascones are  For a review of thujone chemistry prior to 1971 see reference 1.  3  extremely important in the perfumery industry; part two deals with the ring cleavage of decalin-type compounds, derived from thujone, to yield germacranes. The germacranes are a series of sesquiterpenes possessing a ten-membered carbon ring.  Chapter  2.  The Synthesis of  Damascones.  2 . 1 . I n t r o d u c t i o n . This first part of the thesis is concerned with the synthesis of damascones, important perfumery constituents, and the potential synthesis of analogues thereof. Our laboratory has been involved with the synthesis of a variety of fragrances in the last few years. In this introduction a brief overview (10) of fragrance chemistry will be given followed by a discussion of the damascones and the various synthetic routes to this class of fragrances.  2.1.1. The Perception of Smell and Stereochemical Dependence. Man detects odour by inhalation of volatile compounds through the nose. The sensory cells that detect volatile, low-molecular weight organic (and a few inorganic) molecules are  Cribriform plate  naris  Figure 1. Drawing of the bisected nasal cavity. (Modified from Fragrance Chemistry, The Science of the Sense of Smell, edited by E. T. Theimer, Academic Press 1982).  olfactory receptor neurons, located in a relatively small area of the nasal cavity known as the olfactory sensory epithelium. Air bearing the volatile molecules enters the nasal cavity through the nose on inhalation passing over the epithelium on the way to the lungs. The main function of the receptor neurons is to detect, encode and transmit information about the intensity and 'quality' of the odorant to the higher cortical centres. It seems that the receptor cells are all identical, i.e. there are no separate cells to detect different odours. The ability of olfactory receptor neurons to discriminate among odours (man's ability to distinguish perceptually among numerous odours) and studies of specific olfactory deficiencies, strongly support the hypothesis that molecular receptor molecules (proteins) are integral components of the chemoreceptive membrane. There are many theories on the method of the actual binding of the odour molecules to the receptors, although at present there is no conclusive theory. The theory that the shape of a molecule is essentially responsible for its odour (a lock and key type interaction) has gained the most acceptance, especially since the discovery that the two enantiomers of a compound can have different odours (11). At present, one is unable to predict whether a compound would have an odour, nor what that odour would be.  Typically the stimuli are neutral organic compounds of  molecular weights up to about 400, with appreciable volatility at room temperature. Slight differences in structure can alter the odour. Consideration of this fact is important when it comes to the designing of analogues of the naturally occurring fragrances. For example, the eleven-membered straight chained ketones 4, 5 and 6 differ in smell, even though their only difference is the position of the carbonyl functionality (12).  6  It has been found that transposing a functional group can dramatically alter its odour characteristics. |3-Ionone (7) has a violet-type odour, but reversal of the functional group into P-damascone (8) leads to a rose odour (10).  7 "violet"  Isomerisation of double bonds can also change odoriferous properties. a-Gaidanium (Neogal®) (9) has an odour, whilst its double bond isomer pVgaidanium (10) is odourless (13). A reason put forward for this difference (14) is that the a,P-unsaturated ketone functionality adopts an S-trans coplanar arrangement for both molecules. This gives the  9  10  side chain a completely different spacial arrangement with respect to the six-membered ring in 10 than in 9, so that P-gaidanium is unable to elicit a response for the olfactory receptors.  Indeed, introduction of methyl groups at either the C2' or C6' position in  oc-gaidanium leads to a steric interference of the S-trans co-planar arrangement, resulting in loss of odour. Other examples of the addition of methyl groups altering the odour can be found in the musk fragrances. The synthetic musk 11 has a very strong odour. Versalide® (12), and  Tonalid® 13  14  8  Tonalid® (13), are synthetic musks of even greater potency (13). However, the addition of a further methyl group 14 leads to a complete loss of odour (14). Changes in double bond stereochemistry also have important consequences. For example, 15 has a violet-type odour whilst its isomer 16 has a strong sandalwood odour with tobacco undertones (13).  15  16  "violet"  "sandalwood"  There has also been shown to be differences in odour between diastereoisomers. c/s-Nerone® (17) has a green floral woody odour, whilst its diastereoisomer 18 is odourless (13). Some famous examples of enantiomeric differentiation are the enantiomers of carvone (19) and limonene (20). The isomer (S)-(+)-19 has the smell of caraway, whilst (R)-(-)-19 has the smell of spearmint. The isomer (S)-(-)-20 has a lemon smell, whilst (R)-(+)-20 smells strongly of oranges (15).  17 "floral-woody"  18  "odourless"  (S)-(+)-19  (R)-(-)-19  "caraway"  "spearmint"  H  2.1.2.  H  (S)-(-)-20  (R)-(+)-20  "lemons"  "oranges"  The Damascones.  Damascones have recently attained prominence as fragrances with the discovery of their occurrence in rose oil (16). The seven known damascones are shown in Figure 2. The damascones have been found in some types of tobacco (17), black teas (18) and in a host of other plants. One of the most widely used perfumes in Eastern Europe is Bulgarian rose oil, which contains up to 1% damascones. The pure compounds are said to have a 'green' and 'sweet' odour.  The damascone carbon skeleton consists of a trimethylated  cyclohexane moiety with a four carbon oc,p-unsaturated ketone side chain. Double bond isomerisation gives rise to the known members of the group.  Work performed  21 a-damascone  8 P-damascone  22 y-damascone  23  24  8-damascone  P-damascenone  25 y-damascenone  26 a-damascenone  Figure 2. The seven known damascones  at Firmenich  (19), a Swiss  perfumery company, has established  that  (S)-(-)-a-damascone (21) is by far a more valuable and powerful fragrance than the (R)-(+)- isomer.  (S)-(-)-21  (R)-(+)-21  11  The importance of the damascones in the perfumery and flavouring industry has prompted more than forty syntheses, with much of the work being reported in the patent literature. The syntheses can be divided into a few main routes as indicated in Scheme 1. From the many examples a representative of each route will be discussed.  Scheme 1 The main commercial synthesis at present proceeds via acid-catalysed cyclisation of geranyl or farnesyl derivatives. For example, cyclisation of geranic acid (27) to cyclogeranic acid (28) followed by esterification gave methyl cyclogeranate (29) (20)[Scheme 2].  12  27  28  29  a) BF .Et 0, benzene; b) C H N , Et 0 3  2  2  2  2  Scheme 2 Elaboration of the side-chain has been performed by direct Grignard addition of two molecules of the Grignard reagent, followed by a base-catalysed elimination of one of these new substituents (21), or by a modified Grignard reaction in which only mono addition occurs (22) [Scheme 3]. Subsequent double bond isomerisation gave the damascones. OH  32  31  a) C H = C H - C H - M g C l , T H F ; b) K H , H M P A ; c) C H = C H - C H - M g C l , 2  2  2  L D A , T H F ; d) / J - T S O H , T H F .  Scheme 3  2  13  By a similar cyclisation to that above, the acyclic "psuedodamascone" 33 could be cyclised to a-damascone (21) (20a, 23).  BF^EtjO benzene  33  30%  21  Methods have been developed for the cyclisation of (5-citral (34) into P-cyclocitral (35) (24) and subsequent synthesis of the damascones from 35 (16, 23) [Scheme 4].  .CHO  .CHO  a, b  35  34  8 a) PhNH2, E12O; b)  H2SO4;  36 c) 1-propenylmagnesium bromide, THF; d) C1O3.pyridine.  Scheme 4  14  A second route to the damascones involved the attachment of the side chain to 2,6,6trimethylcyclohexanone (or 2,6,6-trimethylcyclohexenone).  A simple synthesis via this  route was performed by Isoe (25) [Scheme 5]. p-Damascenone (24) was produced in two steps from 2,6,6-trimethylcyclohexenone (37) (no yields were reported).  39  24  a) HC=CCH(OH)CH , Li/NH , b) HC0 H, reflux. 3  3  2  Scheme 5  An elaborate synthesis has been reported involving a Diels-Alder reaction to produce the six-membered ring (26). The Meldrum's acid derivative 41, produced by the condensation of Meldrum's acid (40) with acetone, was used as the dienophile in a Diels-Alder reaction with 1,3-pentadiene to give the spiro compound 42 in 66% yield. Treatment of 42 with alkyllithium gave a mixture of keto acids 43 (a,P and P,y isomers), and 8-damascone (23). Decarboxylation of this crude mixture gave 8-damascone (23) in 66% yield from the cyclic adduct 42 [Scheme 6].  15  23  43  a) Acetone, pyridine, molecular sieves; b) 1,3-pentadiene, 130°; c) CH2=CH-CH2Li; d) NH4CI, THF.  Scheme 6  A fourth route that has been employed involves the isomerisation of the readily available ionones to damascones. There are many examples in the literature of this isomerisation. One of the latest syntheses of a-damascone employed such a route (27) [Scheme 7]. Reduction and acetylation of ct-ionone (44), followed by a reaction with trimethylsilyl cuprate gave the (E)-alkylsilane 46. Osmylation followed by Peterson elimination gave exclusively (E)-oc-damascol (48), which was oxidised to a-damascone (21) in 22% overall yield. A fifth route started with 3-methyl-2-cyclohexenone (49) (28) [Scheme 8]. Conjugate addition followed by a regiospecific acylation gave diketone 51. Grignard addition to 51, followed by dehydration, and elongation of the side chain by condensation with acetaldehyde gave y-damascone (22) in 27% yield.  a) NaBH , MeOH; b) A c 0 , Et N, DMAP; c) TMS CuLi.LiCN; d) Os0 , Me N0.3H 0; e) K H , THF; f) M n 0 , acetone. 4  2  3  2  4  3  2  Scheme 7  OH 53  22  52  a) MeMgl, Cul; b) AcCl,Et 0; c) MeMgl, Et 0; d) SOCl , pyridine; e) CH CHO, bromomagnesium N-methyl-anilide. 2  3  Scheme 8  2  2  2  17  A recent article by Fehr (19) reported the synthesis of the two enantiomers of a-damascone (21) by tandem Grignard reaction-enantioselective protonation [Scheme 9]. Enolate 56, obtained by a Grignard reaction either on the ester enolate 54 (30) or ketene 55 (31), was protonated with a chiral proton source. In their best case, using two equivalents of (-)-57(H) and one equivalent of the lithium conjugate base (-)-57(Li), they obtained an 84% ee of (S)-(-)-21. Use of the enantiomer (+)-57 gave an 84% ee of (R)(+)-21 (29). OLi  (S)-(-)-21  (R)-(+)-21  a) n-BuLi, THF, -78°C; b) -78°C -> 25°C; c) CH =CH-CH MgCl, THF; d) (-)-57(Li); 2  e) (-)-57(H); f) (+)-57(Li); g) (+)-57(H).  Scheme 9  2  18  In the primary scientific literature, there is no mention of the synthesis of synthetic analogues of the damascones. However, in the patent literature, there is much discussion on the synthesis of these analogues (20a, 32). A variety of analogues with extra methyl groups attached at the C2', C3' and C4 positions have been synthesised. However, the potency of the odours of these compounds is not well documented.  19  2.2. Results and  Discussion.  Our approach to the synthesis of damascones involved the use of thujone as a starting material.  Ring-opening of the cyclopropane ring present in thujone could lead to a  cyclohexane derivative that would be the basis for the backbone of damascones. We were not overly concerned with retaining optical purity although the synthesis of optically active damascones would be considered in the future. We did envisage a strategy which would permit the production of analogues, not previously studied, so as to establish a proper structure-activity relationship within this area. Methylation of thujone would give a trimethylated bicyclo[3.1.0]cyclohexanone system that could serve as a key intermediate.  Two routes from this intermediate to the  damascones are shown in Scheme 10. The first route involves attachment of the side chain present in the damascones either through a direct nucleophilic attack on the carbonyl group so as to attach the complete side chain, or by initial attack with a nucleophile, such as cyanide ion, followed by a chain extension. Cyclopropane ring-opening and isopropyl side chain cleavage would lead to the damascones. The second route involves cyclopropane ring-opening and isopropyl side chain cleavage directly after methylation of thujone to obtain a cyclohexenone system to which the desired side chain could be attached. The strategy employed should allow the synthesis of damascone analogues with substitution at C2', C3', C4', C5' and C6' positions of the 'cyclohexane' part of the damascones, as well as numerous analogues in which the side chain is altered. Even the cyclopropane and isopropyl functionalities present in thujone could be retained for the synthesis of interesting analogues. Before the synthesis of damascone analogues can be investigated, a direct synthesis of the natural products should first be completed. This section deals with the synthesis of the  20  (3-Damascone Scheme 10 natural products, (J-damascone (8) and (3-damascenone (24). Later in this chapter developments in the synthesis of damascone analogues will be discussed. Indeed, it is in the synthesis of damascone analogues that thujone, with its inherent multifunctionality and chirality, has benefit over the other starting materials that have been employed in the synthesis of damascones. O  .  O  3'  8  24  21  2.2.1. The Methylation of Thujone. A key intermediate required for the synthesis of the damascones was the dimethylated thujone derivative 59. Preliminary studies on the methylation of thujone to obtain the desired dimethylated thujone were performed by J.W. Somerville using potassium 3  hydroxide in either methanol or ethanol to effect enolate formation, and iodomethane or dimethyl sulphate as the alkylating agent. In all cases, the only reaction observed was isomerisation of the C4-methyl group to give the equilibrium mixture of (-)-isothujone (2) and (+)-thujone (3) (2:3 = 33:67), equivalent to that reported by Hach (8). Somerville also attempted the alkylation of thujone using sodium hydride as the base, with iodomethane as the alkylating agent. Varying the quantity of reagents resulted, in the best case (three equivalents each of base and iodomethane), in a mixture of thujone (stereochemistry unknown) (13%), monomethylated product 58 (68%), and dimethylated product 59 (14%). When potassium hydride was used, a low yield of a compound thought to be the monomethylated compound 60 was obtained.  58  59  60  The next base Somerville used was potassium r-butoxide in tetrahydrofuran. Using 2.5 equivalents of both base and iodomethane a mixture of monomethylated thujone 58 (37%),  3  J.W. Somerville, Research report, October 1988,  U.B.C.  22  dimethylated thujone 59 (48%), and trimethylated thujone 61 (15%) was obtained. However, it was stated that separation of this mixture was very difficult.  1  58  59  61  Based on the above results it was decided that a two step process for alkylation was worthy of investigation to increase the yield of the ultimately desired dimethylated thujone 59 with concomitant reduction in the amount of products 58 and 61. Thus, the regiospecific monoalkyation of thujone at the C4 position was investigated, since monomethylation at the C2 position would give a product 60, that with subsequent methylation could give a mixture of dimethylated products 59 and 62. The strategy envisaged for the synthesis of the damascones from thujone made 62 an undesirable product.  1  60  59  62  Since Somerville had shown that an excess of potassium f-butoxide gave a high yield of polymethylated thujones, variation in the amount of base used was investigated. The  23  results are summarised in Table 1. Clearly, the reaction proceeds via methylation at C4 followed by methylation at C2, since none of the other dimethylated product 62 was obtained. The initial methylation at the more substituted C4 was expected, since the use of potassium f-butoxide is reported to give 'thermodynamic' control in the formation of enolates (35). The result of using 1.5 equivalents of potassium f-butoxide was the most encouraging. Since dimethylated thujone 59 was expected to be converted to 61 at a rate slower than alkylation of 58 to 59, a mixture of 58 and 59 could be tolerated for subsequent alkylation.  Table 1. Methylation of thujone with potassium r-butoxide and iodomethane.  A  r ——  KJ +  A  Eq of K O B u l  Solvent  EqofMel  A"  o  +  59  58  l  KJ  V 61  Percent Composition )  Yield  51  1  58  59  61  -  37  48  15  1  2.5b)  THF  2.5  2  2.2 )  THF  2.0  (?)  -  40  44  12  3  1.5*  THF/fBuOH  1.5  95  -  91  9  -  4  1.0>  THFABuOH  1.0  81  13  86  1  -  C  e  >100!  a) Determined by GC, calibrated with 1-menthol for entries 3 and 4; b) Performed by J. Somerville; c) Performed by R. Clewley; d) This work; e) Performed by U. Zoller.  24  It was found that cedar leaf oil could be substituted for pure thujone with no substantial reduction in the yield or increased difficulty in purification. In a typical reaction, 159 g of distilled cedar leaf oil (containing 86% thujone by G C calibration) gave 164 g of the alkylation reaction mixture as an oil. G C analysis (1-menthol as the internal standard) revealed that the mixture consisted of 58 (78.3%) and 59 (8.3%). Based on the purity of the starting material, the yield for 58 was 86%, and for 59 was 8%. Separation of the ketones 58 and 59 was effected by exhaustive flash chromatography. Dimethylated thujone 59, which was eluted first from the column, showed a molecular ion peak at m/z = 180 in the mass spectrum. The *H nmr spectrum [Figure 3] showed two 4  singlets at 8 0.99 and 1.09 ppm, a doublet at 8 0.98 ppm and two overlapping doublets at 8 1.03 ppm each integrating for three protons, the three cyclopropane protons at high field, and the characteristic septet of the isopropyl group. A quartet of doublets at 8 2.80 ppm, assigned to the C2 proton, was shown by decoupling to be coupled to one of the methyl doublets at 8 1.03 ppm and also to one of the cyclopropane protons by a W-coupling. Hence the C2 methyl had to be cis to the cyclopropane ring (endo form or explicitly the R configuration) to allow for a W-coupling of the C2 proton with a C6 proton. Also the 5  CS-Hexo proton at 8 0.54 ppm could be assigned through this W-coupling. The *H nmr spectrum of the monomethylated ketone 58 ( M at m/z = 166) clearly +  showed two methyl singlets at 8 1.12 and 1.03 ppm, and an A X system for the protons at  4  *H nmr spectral data discussed throughout this thesis were recorded in di-chloroform unless otherwise  indicated. 5  Previous workers (33) have used the term 'pseudo-axial' and 'pseudo-equatorial' for describing substituents  on the thujone system. However, since the five-membered ring is very close to planarity, and so can be distorted by substituents into either a 'boat' or 'chair'-like conformation, these terms do seem very ambiguous. For this work the terms exo and endo have been used to describe substituents on the bicyclic system. The term endo is used when the substituent is closer to the longer of the two unsubstituted bridges.  A 2  ppm  1  Figure 3. H nmr spectrum of dimethylated thujone 5 9 (400 MHz, l  J  1 CDCI3).  to  26  C2 (8 2.09 and 2.66 ppm). The signal at 8 2.66 ppm has long range coupling (/ = 2.5 Hz) corresponding to a W-coupling with C6-H  exo  (8 0.64 ppm), and so could be assigned as  the exo C2 proton. Given the high yield of monomethylated thujone derivative 58, a second methylation was required for the production of the desired dimethylated thujone 59. Since the use of excess potassium f-butoxide in the methylation of thujone gave rise to significant amounts of product 61, as well as the desired products, it was decided to use lithium diisopropylamide as the base. Using one equivalent of lithium diisopropylamide in tetrahydrofuran in the reaction with 58, and subsequent alkylation with iodomethane, gave a mixture that contained a predominance of two isomeric dimethylated compounds 59 (25%) and 63 (57%), with some starting material 58 (7%) and the fully methylated product 61 (6%).  The calibrated G C yield of the dimethylated compounds from this  reaction was 77%. It was found that after the methylation reaction had ceased, prolonging the reaction time resulted in an increase in the ratio of 59 to 63 with no change in the amount of 58 or 61. It was believed that the diisopropylamine present was effecting an  59  63  61  isomerisation of 63 into 59. Subsequently, the reaction mixture was treated with alcoholic potassium hydroxide to result in an almost complete isomerisation of 63 to 59. Since the methylation of the enolate generated from 58 is a kinetically controlled process,  27  compounds 63 and 59 are both formed in the methylation of 58.  Approach of  iodomethane can occur from either side of the enolate, with the exo face preferred (since the C6-endo proton hinders attack from the endo face) giving rise to a predominance of 63 over 59. However, ketone 59 is the more thermodynamically stable compound and thus becomes predominant on equilibration of the mixture. The structure of 63 was determined by its molecular ion peak at m/z = 180 in the mass spectrum, its isomerisation to 59, and the similarity of its H nmr spectrum to that of 59. l  Surprisingly, it crystallises to give a white solid (mp 53-54°C), since all the other methylated thujones are non-viscous oils. The *H nmr spectrum showed three signals corresponding to the cyclopropane protons at 8 -0.26, 0.72,  and 1.40 ppm, three sets of  methyl doublets, two methyl singlets, the isopropyl proton as a septet, and a quartet corresponding to the C2 proton at 8 2.31 ppm. The absence of additional coupling in this signal indicated the C2 proton was endo to the cyclopropane ring, since an exo proton would give rise to a W-coupling with the exo C6 proton. The structural determination of 61 was based on the molecular ion peak at m/z = 194 in the mass spectrum (indicating three methyl groups added to thujone) and four singlet methyl signals in the *H nmr spectrum at 8 1.01, 1.07, 1.19, and 1.22 ppm. The most likely explanation for the production of 61 in the alkylation of 58 with lithium diisopropylamide and iodomethane is shown in Scheme 11. The lithium diisopropylamide reacted completely with 58 to form the enolate 65. During the methylation step there was a mixture of 65 and 63 (or 59). The enolate 65 then abstracted a proton from 63 (or 59), to give enolate 66 and ketone 58. The enolate 66 was then methylated to give 61. Experimental observation seemed to indicate that this mechanism was occurring, since increasing the time lithium diisopropylamide was allowed to react with 58 gave no reduction in the amount of 58 that remained after alkylation.  28  61  66  58 Scheme 11  Since reaction of pure 58 gave rise to some fully methylated product 61, then in larger scale reactions, the crude reaction product from the initial methylation of thujone, which consisted of 58 (78.6%) and 59 (8.3%), was used. This resulted in somewhat poorer yields of dimethylated thujones, but eliminated the need for the first separation. A typical reaction on a 300 mmol scale gave the following reaction mixture: 58 (24%), 59 (34%), 63 (25%), and 61 (10%). After treatment with alcoholic potassium hydroxide the reaction mixture consisted of the following: 58 (24%), 59 (54%), 63 (1%), and 61 (11%). The separation of the crude reaction mixture obtained from the methylation reaction was laborious. Spinning band distillation with a very slow collection rate yielded fractions enriched in each compound, but completely pure compounds could not be obtained (see Table 2). Further purification using exhaustive column chromatography eventually yielded  29  Table 2. Spinning band distillation of a mixture of 58,59,63 and 61. Fraction  Temp of distillation ( ° Q  Composition %  Wtof fraction (g) Impurities from cedar leaf oil  Original  63  61  54  1  11  58  59  13  21  1  49a)  0.288  100  -  -  -  -  2  49-58  0.757  64  36  -  -  -  3  57-59  1.228  26  74  -  -  -  4  59  1.888  7  93  -  -  -  5  59-60  2.137  6  94  trace  6  60-61  2.035  3  83  13  1  -  7  60-61  1.927  3  53  42  2  -  8  64-66 >  0.778  -  65  30  2  -  9  66  0.056  4  62  33  1  -  10  66  0.783  3  63  31  2  trace  11  66  2.227  -  30  65  2  -  12  66  2.366  -  6  90  2  trace  13  66  2.502  -  4  89  2  2  14  66  4.040  -  2  87  2  6  15  66  4.642  -  2  85  2  9  16  66  3.858  -  1  86  2  8  17  76 )  1.350  -  1  94  1  1  18  76-78  3.856  -  -  88  2  7  19  78-80  1.154  -  -  26  1  69  20  80-81  1.784  -  -  3  trace  90  21  81  0.153  -  -  -  -  86  -  7.354  -  -  51  1  34  Still pot  b  c  trace  -  Take off ratio = 1:5 with 40 drops/min. Time for distillation = 12 h. a) Vacuum pressure = 5.2 Torr; b) vacuum drop to 6 Torr; c) vacuum drop to 8 Torr.  30  pure samples: 59 and 61 are essentially eluted at the same rate, as are 58 and 63. Using preparative H P L C in the separation of the crude reaction mixture, a sample containing only 59 and 61 can be separated in appreciable quantity (e.g. from 10 g of reaction mixture, a 2 g sample containing only 59 and 61 was obtained). Column chromatography of this sample gave a pure sample of 59, although this process was not very practical on a large scale (ca 50-100 g). One method for the separation of 59 from 61 that was investigated was the conversion of 59 to its silyl enol ether 67. Obviously, 61 would not be able to form an enol ether. A mixture of 59 and 61 (50% 59, 34% 61) was treated with a mixture of trimethylsilyl chloride, l,8-diazabicyclo[5.4.0]undec-7-ene (DBU), and silver nitrate in dichloromethane at reflux (38). After a one-step purification, silyl enol ether 67 was obtained in 46% yield (based on the purity of the starting material). The silyl enol ether was easily identified by the presence of the trimethylsilyl singlet (at 8 0.17 ppm), and a methyl singlet (at 8 1.56 ppm), indicative of a methyl attached to an olefin. Since acid hydrolysis can regenerate 59, this method was used to obtain 59 of high purity [Scheme 12]. The isomerisation of 63 to 59 was investigated. Treatment of either pure 63 or pure 59 with alcoholic potassium hydroxide, resulted in an identical mixture of 59:63 = 99:1. Treatment of a mixture of 58, 59, 63, and 61 (59:63 = 38:28) with thionyl chloride in methanol (a source of alcoholic hydrogen chloride) also resulted in isomerisation of 63 to 59 (59:63 = 65:3). Generation of 59 and 63 by acid hydrolysis of the trimethyl silyl enol ether 67 resulted in near quantitative yield of a mixture of 59 and 63 (99:1). Unlike the methylation of 58, where a 'kinetic' product ratio is formed, in the acid hydrolysis of 67 the products formed undergo equilibration to give the thermodynamic equilibrium mixture.  The isomerisation of 63 to 59 would be expected, since in 63 the near planarity of the cyclopentane ring forces an eclipsing relationship between the exo C2 methyl group and the isopropyl group. On isomerisation to 59, the C2 methyl group is cis to the cyclopropane ring, which is expected to be less sterically demanding than the isopropyl group.  67  59  (99:1)  63  32  Due to the difficulties encountered in the separation of the various methylated thujones, experiments were tried in which a blocking group could be added at the C2 position to prevent over-methylation. In the methylation of simple cyclohexanones, a common method is to first carboxylate the carbon a to the carbonyl group, followed by methylation, and then subsequent decarboxylation to remove the blocking group. For example, 2,2,6trimethylcyclohexanone (70) was obtained in 69% yield from carboethoxycyclohexanone (68) (34) (41% yield from cyclohexanone).  With thujone the blocking group needs  to be placed regioselectively at the C2 position. This could be accomplished by the use of the intermediate 'kinetic' enolate 64, which was expected to be obtained from thujone by the use of lithium diisopropylamide as base.  Although enolate 64 was obtained  regioselectively, subsequent alkylation occurred at the oxygen, exclusively giving enol carbonate 71. A similar result was obtained by the reaction of lithium diisopropylamide and ethyl chloroformate with 58 to give the enol carbonate 73. It had been hoped that ' C alkylation might occur to give the keto ester 72, which could be monomethylated and decarboxylated to give 59. The formation of the enol carbonates can be explained if one uses the hard/soft centre theory for alkylation (35).  Ethyl chloroformate is a hard'  alkylating agent, and so undergoes reactions with 'hard' nucleophiles, i.e. the oxygen anion, rather than the 'softer' carbon centre. By changing the leaving group from chloride to cyanide, for example, one would expect more ' C alkylation.  Investigation of the  33  reaction of thujone with base and ethyl cyanoformate, following the work of Mander (36), will be pursued in the future.  OCOOEt  1  64  71  73  A classical method for the introduction of a group into the less hindered position in an unsaturated ketone is via the use of enamine chemistry. Stork has shown (37) that in enamine formation the double bond is orientated towards the less substituted side of an unsymmetrical ketone. Alkylation can then occur such that the alkylating agent is delivered to the less substituted position. Stork stated that alkylation is a low yielding process, but that acylation, with ethyl chloroformate, is a very high yielding process for certain ketones.  34  The morpholine enamine of thujone was prepared in high yield (91%). The thujone used 6  consisted of a 11:1 mixture of 3-(-)-isothujone (2) and 3-(+)-thujone (3). However, in the formation of the enamine, two products, 74 and 75, were obtained in a 66:34 ratio . Since this was also the ratio of the equilibrium mixture of 2 and 3, it was suspected that  74  75  epimerisation of the C4 position had occurred prior to enamine formation. The *H nmr spectrum clearly showed two singlets at 8 4.56 and 4.57 ppm corresponding to the proton at C2 for the two isomers. No product was observed with the double bond orientated towards the C4 position. Employing the procedure used by Stork for acylation of the enamine with ethyl chloroformate resulted only in the recovery of thujone.  No acylated products were  obtained.  In summary, it has been shown that a mixture of methylated thujones, which contained 83% of the dimethylated thujone 59, could be obtained, in 77% overall yield, from thujone by a two step process of alkylation using initially potassium r-butoxide and subsequently lithium diisopropylamide as the base. Also thujone, in the form of distilled cedar leaf oil,  6  Stork stated that morpholine enamines give higher yields with acylating agents than the corresponding  pyrrolidine enamine.  35  can be converted in a 'two-pot' operation without intermediate purification to a mixture of methylated thujones, in which dimethylated thujone 59 was present as 54% of the mixture, in 84% overall yield. Separation of 59 from the methylation mixture can best be effected by spinning band distillation followed by column chromatography. It has been shown (see later, page 51) that a mixture of 59 and 61 can be used in the synthesis of the damascones since 59 showed selective reactivity towards subsequent reagents. Eventually it is hoped that the mixture of alkylated thujones without separation could be employed in the synthesis of the damascones.  2.2.2. The Synthesis of (3-Damascone and (3-Damascenone from  Thujone.  With the desired dimethylated thujone in hand, the elaboration of the side chain and ring-opening of the cyclopropyl group was considered. Work in our laboratory indicated 7  that direct addition of all four carbons of the damascone side chain to 59 may be difficult. Clewley prepared the dithiane 76 by reaction of 1,3-propanedithiol with crotonaldehyde (39). Treatment with base generates an anion that he had hoped would attack the carbonyl of the dimethylated thujone derivative 59, following the work of Zeigler (39). However, he reported that no reaction occurred.  Attention was turned to the addition of cyanide to generate a cyanohydrin. Since we hoped to use a newly developed method of cyclopropane ring-opening using ozone (7)  R . G . C l e w l e y , private communication.  36  followed by acid treatment (studied by others in our laboratory), the presence of double bonds was to be avoided. Clewley had looked at the direct addition of cyanide to 59 under a variety of conditions (various solvents, temperatures, crown ethers) but found no evidence of any cyanohydrin formation. Reports in the literature (40) indicate that ketones that do not form cyanohydrins can react with trimethylsilyl cyanide to give high yields of the corresponding silyl cyanohydrins. Application of this method to 59 (5 mmol) using 1.5 equivalents of trimethylsilyl cyanide with tetra-H-butylammonium cyanide (0.1 equivalent) as catalyst, in the absence of solvent at 70-80°C, did afford an essentially quantitative yield of the trimethylsilyl cyanides 77 and 78.  59  77  78  The use of 1.5 equivalents of trimethylsilyl cyanide was due to the fact that, on this scale, this amount (1 ml) was required to effect reasonable stirring. On a larger scale, the amount of trimethylsilyl cyanide was reduced to a 5% excess. The ratio of isomers 77:78 was 70:28. No attempt was made, either chemically or spectroscopically, to establish the configuration of each compound at this stage, though, following literature work on the stereochemical outcome of this reaction, some conclusions as to the stereochemistry can be inferred. Evans (41) established the stereochemical course of the trimethylsilyl cyanide addition to the relatively unhindered, but conformationally locked, 4-r-butylcyclohexanone 79. Kinetically the cyanide ion preferentially attacks the carbonyl from its axial face. A  37  thermodynamic equilibrium can be established in which the axial cyanide isomer also predominates [Scheme 13]. With the trimethylsilyl cyanohydrins of 59, the isomer with the cyano group endo to the cyclopropane ring is likely to predominate in a thermodynamic equilibrium, since the periplanar interaction of the C2 methyl and the nitrile would be less than the equivalent interaction of the C2 methyl and the oxysilyl group. Since the use of tetra-rt-butylammonium cyanide as a catalyst induces thermodynamic equilibrium (40b), then the major isomer was likely to be 77.  8  Separation of the isomers for characterisation  TMSCN, Z n l  2  OTMS  CN  9:1  80  81 'kinetic conditions'  OTMS  CN  81  80  'thermodynamic conditions Scheme 13  ° This argument assumes a near planarity of the five-membered ring in the products 77 and 78. It has been shown (42) that in 3-thujone (3) the five-membered ring is only slightly distorted from planarity.  38  proved difficult. As the trimethylsilyl group was intended to be eliminated at a later stage, there was no need to separate the isomers for the synthesis of the damascones. Along with the above mixture of isomeric silyl cyanohydrins we obtained the silyl enol ether 67 (present in 2% yield by GC). In larger scale reactions, 67 was separated from the silyl cyanohydrins. The generation of silyl enol ethers in the reaction of trimethylsilyl cyanide with ketones has been noted before (40a). The formation of 67 can be envisaged to occur by one of two pathways [Scheme 14]. Either the enol form of the ketone was (CN  77/78 Scheme 14  silylated by the reagent, or a portion of the silyl cyanohydrins that were formed in the reaction eliminate hydrogen cyanide. Since in subsequent elimination reactions of silyl  39  cyanohydrins, elimination of trimethylsilanol occurs in preference to hydrogen cyanide, it might be expected that the first pathway was the one that actually occurs. Having in hand a very high yielding process of introduction of a nitrile group, attention now turned to opening the cyclopropane functionality to generate the six-membered ring. Work simultaneously performed in our laboratory (7) [Scheme 15] had shown that ozonation of a thujone-derived tricyclic intermediate 82 involved oxidative attack exclusively at the tertiary carbon atom of the isopropyl group to give 83 and 84 in 71 % total yield. Ring-opening of the cyclopropane in 83 was then achieved using mineral acid 9  to give the five-membered ring bromide (or chloride) 85, which was rearranged to the sixmembered ring hydrocarbon 86 with tri-H-butyltin hydride.  86  85  Scheme 15  Hydroxylation of saturated hydrocarbons by 'dry' ozonation has been reported previously (43).  40  Application of this strategy of cyclopropane-opening/isopropyl cleavage to the present study was highly successful. Treatment of the mixture of silyl cyanohydrins 77/78 in ethyl acetate at 0 ° C , with a stream of ozone for eight hours, resulted in complete consumption of the starting material. The products, isolated as three separate mixtures of isomers, were the epimeric olefins 87/88 (1%), the epimeric ketones 89/90 (45%), and the epimeric alcohols 91/92 (21%). Attempts to separate the individual epimers from each of the mixtures by column chromatography for a detailed characterisation resulted only in  OH 87/88  89/90  91/92  the enrichment of each isomer in the mixture. Pure compounds were not obtained in this study. The mixture of alcohols 91 and 92, occurring as a white solid, consisted of a 66:34 mixture of the epimers similar to the ratio present in the starting material. The assumption is that the major product has the nitrile group endo to the cyclopropane ring. The infrared spectrum of this mixture clearly showed the O-H stretching frequency at 3600 c m . The -1  ratio of isomers was determined by G C , and by integration in the *H nmr spectrum. Since mixtures greatly enriched in either of the alcohols could be obtained by chromatography, identification of the resonances in the *H nmr spectrum was possible. The ketones 89 and 90, obtained in 4 5 % yield, occurred as an inseparable mixture (71:29). They were identified by a sharp carbonyl stretching frequency at 1690 c m in the -1  infrared spectrum, and by the presence of methyl singlets at 5 1.93 and 1.95 ppm,  41  corresponding to the acetyl protons in the *H nmr spectrum. Interestingly, the presence of a carbonyl group attached to the cyclopropane functionality results in a downfield shift of the cyclopropane  nmr resonances of between 0.6 and 0.8 ppm.  The third mixture isolated, which was obtained in 1% yield, was shown to be the alkenes 87 and 88, present as a 60:40 mixture determined by integration of the H nmr l  resonances at 8 4.78 and 4.85 ppm corresponding to the terminal methylene protons. Absence of the isopropyl group resonances, and the presence of methyl singlets at 8 1.70 ppm, characteristic of a methyl group attached to a double bond, are supportive of the assignment. The mechanism for the oxidation of tertiary carbons has been investigated by others (44). There are believed to be two mechanisms operating dependant on the temperature at which the reaction is run. Ozonation at low temperature results in the predominance of hydroxylated products resulting from an insertion of ozone into a C-H bond, with minor amounts of ketones formed by a direct insertion of ozone into a C-C bond (44a). This result has been corroborated in our laboratory (7) with the ozonation of thujone 1 at -78°C to afford a mixture of the alcohol 93 and the ketone 94 in a 3:1 ratio respectively, and an overall combined yield of 70% based on recovered starting material. Subjecting 93 to further ozonation resulted in a new unidentified product, rather than any of the ketone 10  94, which would indicate that 94 is not derived from 93 by an elimination/ozonolysis process. A free-radical mechanism has been postulated in the ozonolysis of hydrocarbons at temperatures of 10-40°C (44b). At these higher temperatures there is a predominance of ketone over hydroxyl products.  Y. Chen, private communication.  With the ozonation of 77/78 at 0 ° C there was a  42  OH 93  94  predominance of ketones 89/90 over the alcohols 91/92. However, monitoring of the reaction by G C indicated that initially 89 and 90 were produced at a rate equivalent to 91 and 92, but that with prolonged exposure to ozone an increase in the yield of 89 and 90, with a corresponding decrease in the yield of 91 and 92 occurred. This would indicate a third mechanism for the formation of the ketones was occurring in conjunction with the other two mechanisms, in which the alcohols undergo a dehydration to the alkenes 87/88, the latter then undergoing a conventional ozone insertion into the C=C bond. It should be noted that, with the use of ethyl acetate as solvent, an acidic medium is produced on prolonged ozonation. Initial investigation of the ring-opening of the cyclopropane ring began with the alcohols 91 and 92. Treatment of a mixture of the alcohols dissolved in dichloromethane with concentrated hydrochloric acid (45) gave a mixture of three isomeric chlorides in 52% overall yield, of which 95 was the predominant compound (83% of the mixture). Exhaustive purification by column chromatography resulted in a pure sample of 95. The 11  mass spectrum showed it to be a chloride of formula Ci6H28ClNOSi.  The *H nmr  spectrum showed the two methyl singlets attached to the double bond at 8 1.70 and 1.75 ppm. A n A M X system with two protons at 8 3.47 and 3.87 ppm, with a geminal  This purification was carried out by R. G. Clewley.  43  coupling of 11.5 Hz and the third proton considerably upfield (5 2.67 ppm), was consistent with the presence of the five-membered ring. If the structure of the chloride was a six-membered ring (i.e. 96), then one would expect a resonance integrating for one proton considerably downfield of the signals representing the other two protons in the  95  96  AMX system. The stereochemistry of the C3 side chain follows from the stereochemistry of the cyclopropane ring. The stereochemistry of the silyl cyanohydrin functionality was unknown. The two minor chlorides are probably isomers of 95 and/or 96. On monitoring the reaction of 91 and 92 with hydrochloric acid by G C , it was found that an initial product was formed, which then decreased as the percentage of the chlorides containing 95 increased. By performing the same reaction at 0°C, the initial product could be obtained after careful work-up. The product was a highly unstable 60:40 mixture of the chlorides 97 and 98, that rapidly decomposed to the alkenes 87 and 88. A *H nmr spectrum was obtained in both d6-benzene and di-chloroform. Decomposition in each of the solvents was detectable.  However, from the *H nmr spectrum recorded in  di-chloroform, the resonances due to the chlorides 97 and 98 could be detected. The cyclopropane ring was still intact, since the characteristic resonances for the C6 exo proton could still be observed at 8 0.63 and 0.75 ppm for the two isomers. The mass spectrum,  recorded immediately upon isolation of the crude products, showed the molecular ion peaks at m/z = 315/313 (1:3), indicating the presence of one chloride atom per molecule. The absence of resonances corresponding to protons attached to chlorine bearing carbons indicated that the chlorine atom was in the tertiary position. It was subsequently found that performing a sequence of reactions in which the chlorides were not purified led to significant increases in the yields. It was believed that the moderate yield of the chlorination step was due to the purification process. The five-membered ring chloride 95 was not an unexpected product, since previous studies in our laboratory had provided a similar ring-opening reaction (7) (see page 39). The stereochemical requirements that govern the direction of the ring-opening are currently being investigated in our laboratory. Fortuitously, a co-worker found that treatment of the mixture of chlorides (containing 12  95) with tri-n-butyltin hydride afforded the alkenes 101 and 102 (83:17) in 69% yield. The H nmr spectrum of 101 clearly showed that a rearrangement had occurred, since the l  methyl group doublet that would result from direct replacement of chlorine in 95 was absent.  Such a radical rearrangement has been noted previously (7).  Treatment of  chlorides with tri-«-butyltin hydride (46) leads to homolytic cleavage of the C-Cl bond to  This experiment was performed by R. G. Clewley.  45  generate a free radical. Usually, this radical abstracts a hydrogen atom from the tin hydride to propagate the free radical chain reaction. However, the radical can rearrange to a more stable radical before abstraction, as has apparently happened in this case [Scheme 16]. The primary radical 99 generated from 95 by tri-n-butyltin hydride may rearrange to the secondary radical 100.  This secondary radical is then trapped by a hydrogen atom.  101/102  100  Scheme 16  Davies (47) has investigated cyclopropyl radical rearrangements in which either the primary or secondary radicals were obtained on ring-opening, depending on the conditions and the initial stereochemistry of the cyclopropyl radicals. At this stage it was decided, since the alcohols 91/92 could be converted to the sixmembered ring alkenes 101/102, which on ozonolysis should be converted to the ketones  46  103/104,  13  that a study involving the conversion of the ketones 89/90 into the ketones  103/104 should be investigated.  101/102  103/104  As discussed previously, the ketones 89 and 90 (71:29) were produced in 45% yield in the prolonged ozonation of 77 and 78. Reduction with sodium borohydride gave, in 92% yield, a mixture of four isomeric alcohols 1 0 5 , 1 0 6 , 1 0 7 and 108 (38:35:14:13 as shown by integration of the C7 proton resonances in the *H nmr spectrum). The presence of the O-H functionality was clearly shown by an absorbance at 3475 cm" in the infrared 1  spectrum. Separation of this mixture was difficult, though during purification by column chromatography mixtures enriched in each of the isomers were obtained. From these mixtures, the partial assignment of the H nmr spectrum of each compound could be l  achieved. It was not considered necessary to obtain pure alcohols for the subsequent reactions, since it was expected that eventually a common product would be obtained from all of the isomers. Analysis of the product ratios in relation to the composition of the starting material suggested that, for each ketone, the reducing reagent attacked both the re or si face at equal rates.  1 3  A small scale ozonolysis of 101 performed by R. G. Clewley gave a crude product that was shown to  contain, by GCMS, a compound having a molecular ion peak equivalent to 103.  89/90  105/106  107/108  Reaction of this mixture of alcohols with hydrochloric acid resulted in the production of two chlorides 109 and 110 (63:37). Traces of the five-membered ring chlorides were detected. Compounds 109 and 110 were clearly identified by the presence in the H nmr l  105/106  107/108  109  110  spectrum of the signals corresponding to the cyclopropane protons, and the two signals at 8 3.75 and 4.11 ppm corresponding to the protons attached to the carbon atoms bearing chlorine. The simplification of the mixture from four isomers (for the alcohols) to two isomers (for the chlorides) was somewhat surprising, as was the retention of the cyclopropane functionality, since, with the alcohols 91 and 92, ring-opening to the fivemembered ring chloride 95 occurred under the same conditions. The formation of the chlorides 109 and 110 was of concern, since treatment with tri-/j-butyltin hydride might result in the eventual retention of the cyclopropane functionality. Luckily, on treatment with the tin reagent, a rearrangement of the initial  48  radical, formed by the homolytic cleavage of the C-Cl bond resulted in an efficient conversion of the chlorides into the alkenes 111 and 112 (111:112 = 72:28) in 95% yield, possessing the desired six-membered ring. The two alkenes were epimeric about the silyl cyanohydrin functionality. Support for the structures of 111 and 112 was provided by the presence of the two vinylic protons at 8 5.20 and 5.41 ppm, and the absence of resonances corresponding to the cyclopropyl protons in the *H nmr spectrum. The stereochemistry about the double bond could not be determined from the regular proton spectrum.  Since the stereochemistry was to be destroyed in the next step, further  experiments to determine this stereochemistry were not undertaken.  Ozonolysis of the mixture 111 and 112 gave, in high yield (93%), the six-membered ring ketone 103, contaminated with 5% of the epimer 104 (further purification gave samples of 103 of 98.5%  purity).  This result was highly surprising. Obviously an  epimerisation a to the carbonyl had occurred [Scheme 17]. The isomer 103 must consist of a pair of enantiomers to account for the high yield of the reaction after starting with a mixture of diasteroisomers of 72:28 ratio. Although the product had [a]D = -28°, it would be expected that 103 was not optically pure. Attempts at using a chiral shift reagent in the separation of the signals in the *H nmr spectrum were unsuccessful. The determination of the relative stereochemistry of 103 was based on *H nmr nOe difference experiments  49  performed on 103, reinforced by an *H nmr nOe difference experiment performed on a subsequent structure derived from 103 (see page 53 for a discussion of the stereochemistry of 103).  Scheme 17  The *H nmr spectrum of a pure sample of 103 showed signals corresponding to the trimethylsilyl protons at 8 0.23 ppm and the three methyl groups at 8 1.17,1.23, and 1.43 ppm with good separation of the resonances corresponding to the five other protons. The infrared spectrum showed the carbonyl absorption band at 1725 cm . -1  50  At this stage, having completed the conversion of the thujone into a trimethylated cyclohexanone with the nitrile placed for attachment of the side chain, repetition of the sequence from the cyanohydrins 77 and 78 through to the ketones 103 and 104 was performed on a large scale without intermediate purification. The products obtained by ozonation of a mixture of 77 and 78, the ketones 89 and 90, and the alcohols 91 and 92 were subjected without purification to reduction with sodium borohydride, to give a mixture of six identifiable alcohols: 105,106,107,108, 91, and 92. This mixture was subjected to treatment with concentrated hydrochloric acid, followed, after work-up, by treatment with tri-n-butyltin hydride to give a mixture of alkenes, predominantly containing 101,102, 111, and 112.  Without purification, ozonolysis of this mixture gave the  desired ketones 103 and 104 (103:104 = 95:5) in 35% overall yield for the five steps from 77 and 78. The five-membered ring ketones 113 and 114 (61:39) were isolated at this stage in 5% overall yield and characterised by the presence, for each compound, of two doublets each integrating for three protons in the *H nmr spectrum. Further purification afforded a pure sample of 114 which showed in the H nmr spectrum two methyl doublets l  at 8 1.05 and 1.35 ppm as well as two quartets each integrating for one proton at 8 2.28 and 2.48 ppm corresponding to the C2 and C5 protons. The corresponding resonances in 113 for the C2 and C5 protons were quartets of doublets due to W-coupling between these two protons, hence the stereochemistry shown.  113  114  51  Due to the difficulty in separating dimethylated thujone 59 from the trimethylated thujone derivative 61, an investigation of selective cyanohydrin formation was undertaken. Treatment of a 50:50 mixture of 59 and 61 with trimethylsilyl cyanide at 80°C for 1 h only, resulted in a mixture consisting of 37% 61, 4% 59, 47 % 77/78 and 7% of the cyanohydrins derived from 61 (as determined by GC). Obviously, a selective reaction was occurring.  In a subsequent repetition of the sequence 59 to 103, a sample of 59  contaminated with 61 (10%) was used. After a selective reaction of 59 with trimethylsilyl cyanide, the unreacted 61 was separated from 77/78 by chromatography. Having converted thujone into the desired trimethylated cyclohexanone moiety 103, elaboration of the side chain was attempted. A sequence of steps was envisaged in which the side chain would be derived, to be then followed by an elimination of the oxygens present in 1 0 3 to give p-damascenone [Scheme 18]. 14  Due to the experience of a co-worker, who found that reduction of the nitrile group by 15  diisobutylaluminium hydride (48) in 119 was hindered by the presence of a carbonyl functionality, such that only reduction of the carbonyl occurred to give 120, the carbonyl group in 103 was first subjected to reduction.  Reduction by sodium borohydride,  however, resulted in not only reduction of the carbonyl but an elimination of cyanide to CN  SBu  119  1 4  Only one of the enantiomers is drawn.  1 5  Z. Gao, MSc thesis, UBC (1989).  CN  120  SBu  OTMS  OTMS  OTMS  reduction  r  OH  CN  reduction  r  OH  115  CHO  116  Grignard reaction  Grignardreaction  O il  ""OTMS  118 elinnnation oxidation  24 Scheme 18 regenerate a carbonyl functionality, that was then reduced by the sodium borohydride to give the two diols 121 and 122 in 18% and 12% yield respectively. At the same time,  OH  OTMS '  C N  C  NaBH  4  X  o  OH  OH  103  121  122  .OH  53  reduction of 103 with diisobutylaluminium hydride was attempted. This resulted in the reduction of the carbonyl to give 115 in 41% overall yield, and the reduction of both the nitrile and the carbonyl to give the aldehyde 123 in 5% yield, the latter having lost the trimethylsilyl group. Repetition of the reaction with a longer reaction time resulted  103  115  123  in the exclusive production of the aldehyde 123 in 4 1 % yield. Clearly the carbonyl functionality was more reactive towards the reducing agent than was the nitrile. Alcohol 115 was characterised by its molecular ion peak at m/z = 255, indicating the addition of one mole of hydrogen to 103, and by the absorbances in the infrared spectrum at 3525 cm" , indicative of an O-H stretching frequency, and 2200 c m 1  -1  (weak), indicative  of a C=N stretching frequency (as noted before, cyanohydrins give weak absorbances in the C=N region), and by the appearance in the *H nmr spectrum of a broad singlet at 8 3.68 ppm, corresponding to the C3 proton. Alcohol 123 was characterised by strong absorbances in the infrared spectrum at 3500 cm  -1  (O-H), and at 1715 c m , corresponding to the carbonyl functionality. The *H nmr -1  spectrum clearly showed the aldehyde proton at 8 9.60 ppm, two exchangeable protons at 8 3.48 and 3.83 ppm, the proton a to the hydroxyl group at 8 3.83 ppm, and the absence of the trimethylsilyl protons normally resonating at 8 0.20 - 0.30 ppm. The relative stereochemistry of 103 and 123 was determined by *H nmr nOe difference experiments. Irradiation of the resonance corresponding to the C2 proton in 103 led to  54  enhancement of the resonances corresponding to the C2 methyl protons, one of the C4 methyl groups and one of the C6 protons, indicating an axial arrangement of the C2 proton. Irradiation of the resonance corresponding to the protons of the trimethylsilyl group in 103 led to enhancement of the resonances corresponding to the protons of the equatorial C4 methyl group and the C2 methyl group. The absence of an nOe of the C2 proton would support a diaxial arrangement of the C2 proton and the trimethylsilyloxy group. For compound 123 irradiation of the resonance corresponding to the C2 proton led to enhancement of the resonances corresponding to the C2 methyl protons, one of the C4 methyl groups and the aldehyde proton, indicating an axial arrangement of the C2 proton. Irradiation of the resonance corresponding to the aldehyde proton led to enhancement of the resonances corresponding to the protons of the two C4 methyl groups, the C2 methyl group and the C2 proton. These results indicated that the C2 proton and the aldehyde proton were in a cis relationship. The production of 115 and 123 in moderate yield limited pursuit of this route in the synthesis of the damascones, to be superceded by a more efficient route in which the oxygen functionalities in 103 were eliminated prior to manipulation of the nitrile group. Treatment of 103 with thionyl chloride and pyridine in dichloromethane returned only starting material. Treatment of 103 with thionyl chloride in methanol (a good source of alcoholic hydrogen chloride) generated the enone 124 very efficiently, with no elimination of the nitrile group. It was found that chromatography on silica gel resulted in the partial decomposition of 124 to an unknown compound. This unknown compound was found to be highly unstable, not reverting to 124 but undergoing polymerisation. The unknown compound showed a maximum at 306 nm in its ultraviolet spectrum, corresponding to a highly conjugated system, and absorbances corresponding to a  55  hydroxyl group (3500 cm" ), a nitrile (2195 cm" ), and a diene system (1645 and 1  1  1580 cm" ) in its infrared spectrum. The ! H nmr showed one olefinic proton at 8 4.88 1  ppm, as a triplet, coupled to a pair of geminal protons (7 = 6 Hz). The ketone 124 showed a maximum at 248 nm in the ultraviolet spectrum, which along with the absorbances in the infrared spectrum at 1690 and 1645 c m  - 1  indicated the  a,fi-unsaturated ketone functionality. Reduction of 124 with sodium borohydride in the presence of cerium III salts (49) gave 126  in essentially quantitative yield. To circumvent the production of the unknown  O  OH  124  126  compound on purification of the product 124 from the reaction of thionyl chloride in methanol with 103,  the two steps, elimination followed by reduction, were repeated  without intermediate purification. Alcohol 126 was obtained in 92% yield from 103.  The  shift of the absorption maximum in the ultraviolet spectrum from 248 nm (in 124) to 216 nm (in 126)  indicated the removal of the a,(3-unsaturated carbonyl chromophore. The  reduction to the hydroxyl group was shown by the presence of the absorbance at  56  3500 cm" and the absence of the absorbance at 1690 c m , corresponding to the ketone 1  -1  functionality in the infrared spectrum. Compound 126 was now envisaged to be a key intermediate in the production of fj-damascone and P-damascenone. Elimination of the hydroxyl group to generate the unsaturated  nitrile  127 would give access to the damascenones, whilst a  reductive-elimination would lead to the a,P-unsaturated nitrile 128, which would give access to the damascone series.  127  128  Compound 126 was converted to the bromide 129 in 92% yield by treatment with phosphorus tribromide in petroleum ether at 0°C, as evidenced by the disappearance of the absorbance at 3500 c m in the infrared spectrum, and by the molecular ion peaks at m/z = -1  229 and 227 in the mass spectrum. Reduction of the bromide, accomplished by sodium borohydride in dimethylsulphoxide (50), gave, in 95% yield, the oc,p-unsaturated nitrile 128 and the non-conjugated nitrile 130, as a 90:10 mixture.  57  The nitrile 128 has been reported previously in the literature (23a, 51). Ohloff (23a) has mentioned that an equilibrium mixture of 128 and 130 contained 88% 130 and 12% 128, although it was also mentioned that it was impossible to establish that equilibrium! A direct attack by an appropriate Grignard reagent would convert nitrile 128 into the desired P-damascone. Attempts at this strategy, using allylmagnesium bromide under the conditions reported by Cannonne (52) or Hall (53) failed to yield any of the desired ketone. Conversion of nitriles to carboxylic acids or esters has been well documented. Since the group at Firmenich have shown that methyl P-cyclogeranate (29) can be converted to P-damascone by either a one-step, or a two-step, process (see Introduction; page 12 Scheme 3), conversion of 128 to 29 would constitute a formal synthesis of P-damascone. To this end, reaction of 128 under a variety of acidic and basic conditions was attempted (54). Reaction with thionyl chloride in methanol, para-toluenesulphonic acid in ethanol and toluene, sodium hydroxide in aqueous ethanol, anhydrous hydrogen chloride in diethyl ether/methanol, concentrated hydrochloric acid in glacial acetic acid, perchloric acid in tetrahydrofuran/water and sulphuric acid resulted in the recovery of 128 (or in the case of concentrated sulphuric acid no recovery at all). The stability of this nitrile to these rather stringent conditions was very surprising. Attention was turned to the conversion of 128 into P-cyclocitral (35) by reduction with diisobutylaluminium hydride. At first this proved difficult. Reduction followed by standard work-up conditions, involving ammonium chloride solution, resulted in poor yields of the aldehyde. Air oxidation of 35 to the corresponding acid 131 was facile. Indeed, at one stage, direct conversion of the nitrile to the ester, by reduction followed by oxidation and esterification, was envisaged as the best route available to production of the ester 29.  Whilst this research was in progress, Picard (55) published work complementary to procedures we were employing. It was reported that the reduction of 128 by diisobutylaluminium hydride, with a work-up involving 5% sulphuric acid, gave the aldehyde, P-cyclocitral (35) in 70% yield. Repetition of this procedure with slight modifications (change in solvent and reaction time) afforded 35 in 90% yield after distillation. The presence of the unsaturated aldehyde functionality was shown by the absorbance at 1665 cm" in the infrared spectrum, and by the singlet at 8 10.11 ppm in the * H nmr spectrum. 1  As noted above, 35 had to be handled with care due to its facile oxidation. The aldehyde 35 had been previously converted into P-damascone (16, 23) (see Introduction; page 13, Scheme 4). Thus, the conversion of thujone into P-cyclocitral constitutes a formal synthesis of P-damascone. Having a synthesis of p-damascone, the production of P-damascenone  was  investigated. Treatment of 129 with l,8-diazabicyclo[5.4.0]undec-7-ene (DBU) resulted in the formation of the diene 127 in 27% yield. The conjugated diene was characterised by the absorbances at 1640 and 1575 cm" in the infrared spectrum and by the maximum at 1  285 nm in the ultraviolet spectrum corresponding to the chromophore of the unsaturated nitrile. The synthesis of 127, trivially named safronitrile, from the alcohol 126 by acidcatalysed dehydration has been reported (51), though no yields or procedure were presented. It was found that treatment of 126 with /wa-toluenesulphonic acid (0.18 equivalents) in toluene with azeotropic removal of water resulted in dehydration to give the diene 127 in 7 5 % yield.  59  p-TsOH, toluene  126  129  127  Reaction of safronitrile (127) with the Grignard reagent, allylmagnesium bromide, gave no isolated products. The same result has been reported by Buchi (22b) who also showed that the reaction of allyllithium with 127 gave only low yields of the products 132 and 133, arising from 1,4 and 1,6 addition, respectively.  Due to the difficulties experienced in hydrolysis of the nitrile 128, it was decided to pursue a route involving a reduction, followed by Grignard attack and oxidation similar to the synthesis of P-damascone from 128. Thus, the nitrile 127 was reduced to safronal (134) in 6 3 % yield and the latter was immediately subjected to reaction with allylmagnesium bromide to give the alcohol 135 in 9 1 % yield after distillation. Safronal (134) was characterised by the aldehyde proton at 8 10.15 ppm in the H nmr spectrum, l  the carbonyl stretching frequency at 1665 c m in the infrared spectrum and the maximum -1  60  127  134  135  at 309 nm in the ultraviolet spectrum. The alcohol 135 showed a maximum of 266 nm in the ultraviolet spectrum corresponding to the diene chromophore, an absorbance at 3425 cm  -1  in the infrared spectrum with no absorbance in the carbonyl region thereby indicating  that 1,2 addition of the Grignard reagent had occurred. Attempts at oxidation of 135 using fairly classical procedures (Collin's reagent (56), pyridinium chlorochromate (57), pyridinium dichromate (58), oxalyl chloride/dimethylsulphoxide (59)) resulted in decomposition of 135 with no formation of the ketone 136 or  135  136  24  p-damascenone (24). This was a surprise since workers at Firmenich (16, 23) had shown that, in the corresponding series in the production of P-damascone, the alcohol 137 resulting from reaction of allylmagnesium bromide with 35 could be oxidised by Collin's reagent efficiently to the corresponding ketone 138, which was then isomerised to p-damascone (8). However, with the present series, the oxidation of 135 must obviously  61  OH  35  137  O  138  be accompanied by rearrangements or polymerisations promoted by the presence of the diene functionality. Use of 'active' manganese dioxide (60) resulted in a near quantitative recovery of the alcohol. Obviously, the steric crowding around the endocyclic double bond must hinder the complexation of this double bond with the oxidising reagent. With no success at the oxidation of 135, safronal (134) was reacted with 1-propenylmagnesium bromide  16  to give a 1:1 mixture of allylic alcohols 139 and 140 in 89%  distilled yield. The two alcohols are formed since the commercial supply of 1-bromopropene, used in the generation of the Grignard reagent, exists as a 1:1 mixture of geometric isomers. It was found that the Z-isomer, 140, was unstable to chromatography, possibly undergoing an acid-catalysed rearrangement, whilst 139 was completely stable and could be isolated pure in 43% overall yield by chromatography of the mixture on silica gel. The assignment of the configuration in these alcohols could not be accomplished by  Prepared from a mixture of E- and Z- 1-bromopropene by the method described in reference 32a.  62  *H nmr spectroscopy. In 139, the four olefinic protons occur at the same chemical shift and in 140 the two side chain olefinic protons have resonances that were also coincident. Hence no information on the coupling constants of the C2 and C3 protons was available. However, assignments were made based on conversion to subsequent structures and relating back the configuration through to 139 and 140. Mild oxidation of 139 with 20 equivalents of 'active'  17  manganese dioxide (60) gave  quantitatively fJ-damascenone (24) that had spectral data consistent to the natural product [the *H nmr spectrum is given in Figure 4]. Oxidation of the mixture of alcohols 139 and 140 (49:51) under the same conditions afforded a mixture of the ketones p-damascenone (24) and its Z-isomer 141 (49:51) in quantitative yield. P-Damascenone (24) showed the two olefinic protons of the side chain at 8 6.20 and 6.85 ppm with a coupling constant of 16 Hz. Ketone 141 showed the equivalent protons as a multiplet resonating at 8 6.20 ppm. The coupling constant indicated the double bond to have the E configuration in P-damascenone (24). p-TsOH, THF  139/140  24  141  It was found that only manganese dioxide with an activity rating of I (60) would effect this reaction.  64  Separation of the mixture 24 and 141 was not required since it has been reported (21a) that acid-catalysed isomerisation of 141 to 24 occurs. Treatment of the mixture with paratoluenesulphonic acid in tetrahydrofuran resulted in the production of pure P-damascenone (24) exclusively, in 76% yield after purification.  In summary, the total synthesis of P-damascenone from thujone has been accomplished in 6% overall yield for 15 steps by the route as outlined in Scheme 19. It has been shown that only partial purification of the mixture obtained from the methylation of thujone is required. Purification of the intermediates in the synthesis of 103 non-essential.  has been shown to be  65  66  2.2.3. and  The Synthesis of the  Key Intermediates, fi-Cyclogeranonitrile  128  Safronitrile 127, from 2,6-DimethyIcyclohexanone.  The length of the sequence outlined in Chapter 2.2.2 from thujone to |3-damascenone would negate the commercial use of this route. A conversion of thujone into a trimethylcyclohexenone intermediate that has commercial applicability is oudined in Chapter 2.2.5. A synthesis of the nitriles 127 and 128 from commercially available 2,6-dimethylcyclohexanone was devised [Scheme 20] as a model study to investigate the feasibility of converting a cyclohexanone derivative of thujone into the damascones.  Scheme 20  2,2,6-Trimethylcyclohexanone (142) was synthesised in 85% yield from 2,6-dimethylcyclohexanone (144), by treatment with lithium diisopropylamide and iodomethane, by a 18  Commercially available, Aldrich Chemical Co.  67  procedure similar to that reported by Fitjer (61), although a slight excess of lithium diisopropylamide was employed to complete the reaction. The product, characterised by its molecular ion peak at m/z = 140 in the mass spectrum, contained a minor amount (3%) of 2,2,6,6-tetramethylcyclohexanone (145) (as shown by GC). This impurity was tolerated in preference to the impurity of cis/trans 2,6-dimethylcyclohexanone (144) in the synthesis of trimethylcyclohexanone reported by Fitjer (61).  144  142  1 4 5  The conversion to the a,|5-unsaturated nitrile was envisaged to proceed along similar lines to the procedure developed for the production of the unsaturated nitrile 128 from dimethylated thujone 59. Indeed, a similar sequence from 2,2-dimethylcyclohexanone to the nitrile 148 has been reported by Stille (62), via the intermediacy of the cyanohydrin 147. Thus, 2,2,6-trimethylcyclohexanone was reacted with sodium cyanide in aqueous  sodium bisulphite solution to give in quantitative yield a white solid that was shown, by G C and *H nmr spectroscopy, to contain a 23:77 mixture of cyanohydrins 149 and  68  150.  19  The mixture was characterised by the absorbances at 3580 and 3450 c m  -1  in the  infrared spectrum corresponding to free and hydrogen bonded O-H stretching frequencies  142  149  150  'trans'  'cis'  and by the weak nitrile stretching frequency at 2250 cm" . Dehydration of this mixture 1  using the same conditions employed by Stille, namely thionyl chloride and pyridine in benzene at reflux, which gave 148 from the cyanohydrin 147 in 86% yield, produced an interesting result. It was found that a mixture of nitriles 151 and 128 in a 74:26 ratio were  150/149  128  151  'cis'/'trans'  obtained in 94% overall yield. Compound 128 had been synthesised before from thujone [Section 2.2.2]. The nitrile 151 showed an absorbance corresponding to the nitrile functionality at 2260 c m  -1  and an absorbance corresponding to a double bond stretching  frequency at 1650 c m in the infrared spectrum. The presence of only two methyl signals, -1  a singlet at 8 1.79 ppm and a doublet at 8 1.10 ppm in the H nmr spectrum indicated that a l  Conventionally only one of each pair of enantiomers is drawn.  69  rearrangement had occurred. The *H nmr spectrum also showed two resonances at 8 5.02 and 5.24 ppm corresponding to the terminal methylene protons. Further proof of the structure of 1 5 1 was provided by ozonolysis. The double bond was cleaved to give a product 1 5 2 which in the mass spectrum revealed a molecular ion consistent with a formula of C 9 H 1 3 N O . Since mass spectroscopy) of  1 5 1 had a molecular formula (determined by high resolution  C10H15N,  ozonolysis had resulted in the replacement of C H 2 by an  oxygen, a clear indication of a terminal double bond in 1 5 1 . The infrared spectrum  showed absorbances corresponding to a nitrile functionality (2260 cm ) and a ketone -1  (1725 cm" ). 1  The H nmr spectrum showed a singlet resonance at 8 2.41 ppm J  corresponding to the methyl ketone protons. The unusually low field resonance of the methyl ketone protons is possibly due to extra deshielding of the methyl protons by the nitrile group. It is well known that dehydration occurs preferentially via a fra«s-diaxial mechanism (63). Stille found only 1 4 8 after the reaction of 1 4 7 with thionyl chloride and pyridine, since the latter, being a conformationally mobile system, can provide a trans-diaxial arrangement of a proton and the hydroxyl group. However, in our case, with the 'els'  20  2 0  For this thesis the terms 'cis' and 'trans' have been adopted for compounds like 149 and 150 to indicate  the stereochemistry of a P proton relative to the hydroxyl group. A 'cis' compound would have the hydroxyl group and (5 proton on the same side of the ring plane. This convention has been used by other workers, i.e Picard (55) and Holm (64).  70  and 'trans' diastereomeric pair of cyanohydrins, only in the 'trans' product 149 could a rra/w-diaxial elimination occur to give the nitrile 128. In the 'cis' cyanohydrin 150, the C2-C3 bond and the C5-C6 bond are the only bonds that are orientated 180° to the leaving group in the conformation with an equatorial C6 methyl group. Migration of the C2-C3 bond occurs with a 'backside' displacement of the leaving group in a SN2 manner, resulting in the formation of the stabilised carbonium ion at what was the C2 carbon containing the geminal methyl groups. Loss of a proton from this carbonium ion to an olefin leads to formation of 151 [Scheme 21]. The stereochemistry of 151 was defined by need for a backside attack of the nucleophile.  151 Scheme 21  The assignment of 149 and 150 as the 'trans' and 'cis' cyanohydrins was at first based purely on the above results.  However, a single crystal X-ray determination of a  subsequent product confirmed the assignments (see later). It has been shown (65) that 'small' nucleophiles attack the carbonyl of cyclohexanone (and derivatives thereof) via an axial approach to give the equatorial alcohol. In the case of  71  2,2,6-trimethylcyclohexanone a 'kinetic' attack by the relatively small cyanide ion would afford 150 as the major product. However, 150 could also be the thermodynamically more stable product since a reversible reaction in the formation of 150 could be operating. With a view to increasing the proportion of 149 to 150 a systematic investigation of cyanohydrin production was undertaken [Table 3]. An increase in the temperature of the sodium cyanide reaction led to an increase in the production of the desired 'trans' isomer 149.  Using diethylaluminium cyanide (66), resulted in a slight increase in the  Table 3. Formation of cyanohydrins 149 and 150 from 2,2,6-trimethylcyclohexanone.  'cis' : 'trans'*)  Reagent  Temp/°C  Yield %  1  TMSCN  75-80  97  88 : 12  2  NaCN  0-5  100  80:20  3  tt  0-15  100  77 :23  4  it  20  100  80:20  5  II  60  97  70:30  6  Et2AlCN  -15-+5  97  65 : 35  b  a) Ratios determined by H nmr analysis; b) Obtained as the trimethylsilyl cyanohydrins. Conversion of cyanohydrins 153/154 to 149 and 150 confirmed stereochemical assignments. 1  ratio of 149 to 150. Use of trimethylsilyl cyanide (40), under the conditions employed in our studies on dimethylthujone 59, decreased the ratio of 149 to 150 (determined after deprotection of the hydroxyl group). It was found that the trimethylsilyl cyanohydrins 153 and 154 were converted, in 90% yield, to the corresponding cyanohydrins 149 and 150 by treatment with an alcoholic solution of hydrogen chloride. Reaction of the trimethylsilyl  72  cyanohydrins with phosphorus oxychloride and pyridine (40c) resulted in a moderate yield of the nitriles 128 and 151 (1:10, 41%).  OTMS  OTMS TMSCN nBuNCN"  CN  '>i.  'CN  4.  H 12:88  142  154 'cis'  HCl, MiOH/Etp  OH  P*CN H 150 'CIS'  After this series of reactions had been performed a complementary report, by Picard (55), on the synthesis of P-cyclocitral (35) from 2,2,6-trimethylcyclohexanone was published [Scheme 22]. Treatment of 2,2,6-trimethylcyclohexanone with trimethylsilyl cyanide under 'kinetic' conditions resulted in the production of the silyl cyanohydrins 21  154 and 153 as a 1:1 mixture. Picard found  22  that chromatography of the trimethylsilyl  cyanohydrins on silica gel resulted in the 'deprotection' of the 'trans' isomer to afford 149 whilst the 'cis' isomer was stable. In this manner, a separation of the 'trans' isomer for subsequent dehydration to the nitrile 128 could be achieved. This result is in contrast to  2 1  For a discussion on the thermodynamic vs kinetic control of trimethylsilyl cyanide addition to ketones  see section 2.2.2 page 36. 2 2  J.-P. Picard, private communication.  the present study since we found that flash chromatography on silica gel provided a quantitative recovery of the starting materials (153 and 154).  128  1 4 9  'trans' Scheme 22  154  'cis'  PICARD  Since the major product of all the various cyanohydrin formation reactions was the 'cis' isomer 150, investigations with respect to a 'cis' elimination were undertaken. It has been known for over 130 years that certain esters, when heated at 300-500°C, decompose into an olefin and a carboxylic acid. This pyrolytic elimination has been shown (67) to proceed via a 'cis' elimination, where the stereochemical requirement is the reverse of the transdiaxial elimination. The mixture of cyanohydrins, obtained from the sodium cyanide reaction, was first converted into the corresponding acetates. Two methods were adopted for this transformation.  The standard treatment with acetic anhydride under basic  conditions resulted in a poor yield of the acetates 155 and 156 (total yield 24%) with the major product being the original ketone 142 (51% yield). Better yields were obtained by  74  an acid-catalysed process using acetic anhydride with a catalytic amount of acetyl bromide (68).  This produced a 72:28 mixture of the 'cis' and 'trans' acetates, 156 and 155,  respectively, in 79% overall yield, that was isolated as a white solid and characterised by an absorbance in the infrared spectrum at 1750 c m  -1  and by the appearence in the *H nmr  spectrum of two resonances corresponding to the acetyl methyl protons at 8 2.10 and 2.11 ppm in a ratio of 72:28. Separation was not deemed necessary.  150  156  'cis'  'cis'  Two methods were used for pyrolysis of the acetates involving either a vacuum pyrolysis or a flash pyrolysis. The second method gave the more consistent results. Subjecting the mixture of acetates to vacuum pyrolysis with a furnace temperature of 450°C afforded the nitrile 128 in 72% yield. The starting materials 155 and 156 (155:156 = 91:9,28% yield) were also recovered, which supported the assignment of the minor acetate 155 as the 'trans' isomer that was unable to undergo the necessary 'cis' elimination. With such an efficient method for the conversion of the 'cis' isomer to the desired nitrile  75  128, and given that the two isomeric nitriles 151 and 128 were difficult to separate and that 149 and 150 cannot be separated by chromatography due to an incomplete reversion to the starting ketone, a re-investigation of the 'trans' elimination/rearrangement of 149 and 150 was undertaken. The reaction of thionyl chloride and pyridine with 149 and 150 was performed at lower temperatures in the hope that the production of 151 would be decreased. Indeed, reaction of a mixture of 149 and 150 (20:80) at 4 ° C in toluene for 17 h resulted in no production of 151. Attempted purification by column chromatography gave the nitrile 128 contaminated by the ketone 142 (due to decomposition of 150 on the column) and 150 also contaminated with 142. However, this result was encouraging since it was shown by H nmr spectroscopy that, prior to purification, 128 and the l  150  150  156  'cis'  'cis'  'cis'  'cis' cyanohydrin 150 were the exclusive products (128:150 = 20:80). Based on the preceding experiments involving 'cis' pyrolytic elimination the possibility of being able to utilise both the cyanohydrins in a 'one-pot' process to produce 128 became feasible.  76  Acetylation of the mixture of 128 and 150 gave 128 and the 'cis' acetate 156 in high yield.  The 'cis' acetate crystallised from diethyl ether and hexanes.  Proof of the  stereochemistry was obtained by a single crystal X-ray structure determination [Figure 5], that confirmed the cis relationship between the acetate functionality and the proton on C6. Based on the above results, an extremely efficient synthesis of the nitrile 128 from 2,2,6-trimethylcyclohexanone was achieved without intermediate purification [Scheme 23]. The steps followed were treatment of 142 with cyanide, treatment of the resultant mixture with thionyl chloride and pyridine at 4 ° C ('trans' elimination), followed by acetylation of the remaining 'cis' isomer and finally 'cis' pyrolytic elimination to give the desired nitrile 128. The overall yield of 128, from this four step conversion, was 81%. This procedure was performed on a 0.5 molar scale, but could easily be scaled-up further.  128  128 Scheme 23  OAc 'CN  H  Figure 5. Single crystal X-ray structure of acetate 156.  78  As previously mentioned, whilst this work was in progress, Picard published his sequence for the conversion of 142 into 128.  Since he was only able to utilise the 'trans'  silyl cyanohydrin that was produced in 50% yield, along with the use of extremely expensive reagents, it is felt that our sequence would have greater potential for commercial use. Being able to produce large quantities of the nitrile 128, the conversion of this nitrile to the nitrile 127, the precursor of P-damascenone, was investigated. This would be the reverse of the sequence from thujone, in which 126 was converted into 128. Allylic bromination has been reported previously on 128 (51) using bromine and photo-irradiation. Since N-Bromosuccinimide and benzoyl peroxide in refiuxing carbon tetrachloride (69) has been found to be a more convenient reagent system for allylic bromination, this was used to give a crude product that was shown by *H nmr spectroscopy to contain, as the major product, the bromide 129.  CN NBS. AIBN  This crude mixture was  NaCOjHp  CCI4  128 p-TsOH, toluene  127  79  immediately subjected to hydrolysis, employing a method used on a similar compound by Stille (62), to give, in 7 9 % yield, the alcohol 126, that had also been obtained from thujone. The enone 124 was obtained as a by-product. As no oxidation was thought to have occurred in the hydrolysis of the bromide 129, the 124 was assumed to arise from some by-product of the allylic bromination step, namely the dibrominated product 157. Since 124 is easily reduced to 126, then this by-product could be tolerated. The alcohol 126 has been converted to the nitrile 127 [Section 2.2.2].  In summary, 2,2,6-trimethylcyclohexanone has been shown to be an efficient precursor for the intermediates 127 and 128 used previously in the synthesis of P-damascone and P-damascenone. The nitrile 128 was produced in 81% overall yield and the nitrile 127 in 48% overall yield from 2,2,6-trimethylcyclohexanone. The efficiency of this sequence has indicated that a strategy in which thujone is first converted to a cyclohexanone derivative that is subsequently manipulated into the damascones and analogues thereof could be of commercial interest.  2.2.4.  The Synthesis of Compounds related to P-Damascone.  In the course of our studies on the synthesis of P-damascenone, the synthesis of related compounds with extra oxygen functionalities, namely the diketone 158 and the diol 159, were requested for evaluation by the Swedish Tobacco Company.  O 158  OH 159  80  The intermediate 126,  that was obtained both from thujone and 2,2,6-  trimethylcyclohexanone, seemed ideal for the synthesis of the above compounds. The nitrile group in 126 was reduced in good yield (80%) to the aldehyde 160 with diisobutylaluminium hydride. The aldehyde was characterised by the shift in the maximum from 216 nm (for 126) to 245 nm (for 160) in the ultraviolet spectrum, a new  OH  OH  i o  126  6  absorbance, in the infrared spectrum, at 1675 c m  -1  accompanied by the loss of the nitrile  absorbance at 2220 c m and by the appearance of a singlet at 5 10.13 ppm corresponding -1  to the aldehyde proton in the H nmr spectrum. Aldehyde 160 is a known compound, l  having been synthesised by Brook (70) and Sih (71) as an intermediate in the total synthesis of strigol from cc-ionone. Attachment of the side chain was accomplished by the Grignard reaction utilising 1-propenyl-magnesium bromide possible alcohols  24  23  with the aldehyde to give a mixture of all the four  161, 162, 163 and 164  (28:28:1:4 ) in 6 1 % overall yield.  Purification by column chromatography yielded pure samples of three of the isomers. The spectroscopic data on each isomer was very similar. The infrared spectra for each showed an O-H stretching frequency at 3400 cm" and an olefinic stretching frequency at 1  Produced as a 49:51 mixture of Z- and E- isomers from 1-propenyl bromide Conventionally only one of each pair of enantiomers is drawn.  81  1680 c m . The mass spectra showed the molecular ion peak at m/z = 210. The H nmr l  -1  spectra of each showed three methyl singlets (two in the 8 0.9-1.2 ppm region and one in the 8 1.8-2.0 ppm region), a methyl doublet (in the 8 1.7-1.8 ppm region), a triplet at  OH 163  OH 164  8 3.90-3.95 ppm for the C3' proton and a doublet at 8 4.9-5.2 ppm corresponding to the proton at CI, adjacent to the newly formed hydroxyl group. In the H nmr spectrum of l  161, the resonances for the C2 and C3 protons at 8 5.79 and 5.71 ppm, respectively, showed a coupling of 16 Hz. In 162, the C2 proton resonating at 8 5.82 ppm as a doublet of doublet of quartets and the C3 proton resonating at 8 5.64 ppm as a doublet of quartet of doublets showed a coupling of 10 Hz. The coupling constants indicated that 162 was the Z-isomer and 161 was the E-isomer. Since 161 and 162 were formed in a 1:1 ratio it can be assumed that 161 and 162 were isomers differing only in the configuration of the double bond. The stereochemistry of the hydroxy groups was not proven. There was some differentiation with respect to the face of the carbonyl in 160 to attack by the  82  Grignard reagent, since 161 and 162, accounting for 92% of the mixture, arose from attack on one side of the carbonyl, whilst 164 and 163, accounting for 8% of the mixture, arose from attack on the other face of the carbonyl.  Assuming 160 adopts a twist  conformation with the hydroxyl group orientated in a pseudo-equatorial position, molecular models indicate that the re face of the carbonyl group is less hindered giving rise to the stereochemistry depicted above for the major isomers. Oxidation of the mixture of alcohols with 'active' manganese dioxide led to the diketone 158 and the keto-alcohol 165. The alcohols 162 and 164, with the Z configuration in the  O  O  158  165  side chain, were oxidised at the C3' position but not at the CI position. Under more forcing conditions it might be possible to oxidise 165 to 158. Both hydroxyl groups in each alcohol, 161 and 163, were oxidised to carbonyl functionalities. Compound 158 was characterised by its mass spectrum that showed a molecular ion of m/z = 206 and by the absence of resonances in the *H nmr spectrum in the 5 3.0-5.0 ppm region corresponding to protons attached to hydroxyl bearing carbons. Resonances at 8 6.15 and 6.71 ppm, corresponding to the C2 and C3 olefinic protons respectively, indicated a coupling of 16 Hz between the two protons suggesting a E configuration to the double bond of the side chain. Compound 165 showed, in the  nmr spectrum, a resonance at  8 5.31 ppm corresponding to the CI proton that was coupled to an olefinic proton resonating at 8 5.74 ppm. The mass spectrum showed a molecular ion of m/z = 208  83  indicating that only one of the hydroxyl groups in the alcohols 162 and 164 had been oxidised.  The alcohols 161, 162 and 164 and the diketone 158 will be sent for  evaluation to the Swedish Tobacco Company. With the poor yields obtained in the sequence 160 to 158, investigation of the Grignard reaction with allylmagnesium bromide on 160 was tried. The two isomeric alcohols, 166 and 167 were obtained in moderate yields of 32% and 21% respectively. By the same argument as used for the stereochemistry of 161, it was assumed 166 had the  OH 166  OH 167  stereochemistry as shown. The *H nmr spectra of each alcohol showed the allylic resonances (as compared to 160) corresponding to the side chain. In 166, these signals consisted of a multiplet at 8 2.36 ppm and a doublet of doublet of doublet of triplets at 8 2.66 ppm corresponding to the C2 protons, a doublet of doublets at 8 4.34 ppm corresponding to the CI proton, adjacent to the newly formed hydroxyl group, a pair of terminal methylene protons at 8 5.16 and 5.19 ppm, and a multiplet at 8 5.91 ppm corresponding to the C3 olefinic proton. The poor yields dissuaded further investigation of this route until results from the evaluation by the Swedish Tobacco Company were obtained. A brief investigation of methods to block the active hydroxyl group in 160 was undertaken. Protection of the hydroxyl group of 126 with r-butyldimethylsilyl chloride led to the silyl ether 168 in 78%  84  yield. Compound 168 was characterised by the molecular ion peak at m/z = 279 that agreed with the addition of the silyl group. Reduction of 168 to the aldehyde 169 using diisobutylaluminium hydride was sluggish leading, after work-up, to an inseparable mixture of 168 and 169 (47:53) with near quantitative recovery. The *H nmr spectrum of this crude mixture showed the aldehyde proton at 5 10.13 ppm. The high yield indicated for this reaction should prompt further investigation of this reduction and subsequent manipulation of 169 to the diketone 158 in the future.  2.2.5.  The  Conversion of  Thujone to  3-(l-Methylethyl)-2,6,6-trimethyl-  cyclohex-2-en-l-one. From a commercial point of view, the synthesis of P-damascenone from thujone (as described in Section 2.2.2) is inefficient with the use of expensive reagents (namely trimethylsilyl cyanide and tri-n-butyltin hydride).  Investigation of a more efficient  conversion of thujone to the cyclohexanone-type system was thus undertaken. Clewley  25  in our laboratory had investigated the acid-catalysed cyclopropane ring-  opening in thujone and methylated thujones, following the precedent set by Eastman (33). Only low yields of a cyclohexenone system were obtained. Eastman has shown (72) that treatment of thujone with bromine resulted in the  R. G. Clewley, private communication.  85  formation of tribromide 174.  The mechanism of this reaction has recently been  investigated by Cocker (73) [Scheme 24]. It was thought that initial bromination occurred to give the a-bromoketone 170 which then via nucleophilic attack by bromide ion, with simultaneous ring-opening of the cyclopropyl-type ring system, provided the rearranged bromide 171. The double bond was then brominated to 172 with the latter undergoing dehydrobromination to 173. Allylic bromination of 173 finally affords the tribromide 174 as first isolated by Eastman.  Scheme 24  Treatment of 59 with bromine in petroleum ether at room temperature, using the same conditions as employed by Eastman on thujone, resulted in a complex product mixture.  Chromatography on alumina yielded three products that were assigned the structures 175, 176 and 177. The a,P-unsaturated carbonyl functionality in 175 was characterised by  the absorbances at 1670 and 1625 c m in the infrared spectrum and the maximum at 246 -1  nm in the ultraviolet spectrum. The *H nmr spectrum showed a doublet of doublets at 8 4.33 ppm integrating for one proton corresponding to the C5 proton and no signals corresponding to cyclopropyl protons. An analysis of the *H nmr spectrum of the crude reaction product indicated the desired product 175 to be present as approximately 25% of the mixture. The ultraviolet spectrum of 176 showed a maximum at 276 nm indicating a highly conjugated system, which along with the two singlets integrating for one proton each at 8 5.09 and 5.36 ppm in the *H nmr spectrum, led to the structural assignment for 176. The third major product isolated was tentatively assigned the structure 177 based on the appearance, in the *H nmr spectrum, of two singlets integrating for three protons each at 8 1.90 and 2.33 ppm. The conversion of 175 into the cyclohexenone 179 could be accomplished in low yield using tri-n-butyltin hydride, but since the use of this expensive reagent was to be avoided a classical two step elimination-reduction reaction was undertaken. Dehydrobromination using potassium hydroxide in ethanol afforded, in 93% yield, the diene 178, characterised  8 7  by its H nmr spectrum that showed an A B quartet centred at 5 6 . 1 2 ppm corresponding to 1  the two olefinic protons. A small scale hydrogenation reaction indicated that the  C 4 - C 5  double bond could be reduced selectively.  1  1 7 5  1 7 8  1 7 9  The bromination reaction on 59 was performed at 0 ° C in the hope of suppressing the formation of the five-membered ring product. At this temperature reaction occurred to give the cc-bromoketone 180 in almost quantitative yield. This compound was unstable to  5 9  1 8 0  purification, either by distillation (where conversion to 176 occurred) or chromatography (where decomposition occurred). Following the indicated mechanism for the reaction of thujone with bromine (73) [Scheme 2 4 ] in which an a-bromoketone is an intermediate that then undergoes nucleophilic attack by bromide ion to give a six membered ring, crude 180 was reacted with 4 8 % hydrobromic acid. The ratio of products in the crude mixture  88  differed little from that found in the room temperature bromination of 59.  Further  investigation on the nucleophilic ring-opening of 180 will be undertaken. The crude product mixture was dehydrobrominated and then hydrogenated in a "one-pot" operation to give two products 179 and 181, isolated in 20% and 36% yield respectively from 59. Enone 181 was characterised by the presence of three methyl doublets in the H nmr l  spectrum at 8 1.08, 1.19 and 1.22 ppm.  In summary, thujone can be converted into the trimethylated cyclohexenone 175, which can subsequently be used in the synthesis of damascones. Further studies are required on the nucleophilic cyclopropane ring-opening of 180 in order to increase the efficiency of the conversion.  89  2.3.  Future  Developments.  In conjunction with the future studies that have been outlined throughout this chapter further investigations on the use of thujone in the synthesis of damascones can be envisaged. The conversion of enone 175 to the damascones is expected to proceed using similar methods as used in the synthesis of P-damascenone from trimethylcyclohexanone [Scheme 25]. The isopropyl group originally present in thujone could be either cleaved (to give the damascones) or retained to produce a damascone analogue. Further analogues of the  O  175  CN  OH CN  182  184  P-damascenone analogues Scheme 25  183  124  p-damascenone.  90  damascones could be generated at the C6' or C2' positions by altering the alkylating reagent used in the initial alkylations of thujone. Substitution at the C5' position could be accomplished via an intermediate that retains the bromine present in 175. As mentioned in the introduction (S)-(-)-cc-damascone 21 is a more precious fragrance than (R)-(+)-21. A chiral synthesis of a-damascone can be envisaged employing the chirality inherent in thujone [Scheme 26].  Scheme 26  A n important series of fragrants, the irones (e.g. 192), are found in the roots of iris. They have a similar structure to the ionones but with an extra methyl group at the C5' position. This series has prompted much interest recently due to variety of fragrances present and due to the difficulty in their synthesis. A route to these important compounds could be envisaged from thujone [Scheme 27]. The ring expansion of thujone to the  91  carboxyhomothujone derivative 188 has been performed in our laboratory.  26  Methylation  of 188 followed by decarboxylation, cyanohydrin formation, ozonolysis and cyclopropyl ring-opening would lead to 190, that could be converted to the irones by a procedure similar to that employed in the synthesis of the damascones.  Scheme 27  Y. Chen, private communication.  92  2.4.  Experimental.  2.4.1. General Experimental. Melting points (with the recrystallisation solvents given in brackets) were determined using a Kofler block melting point apparatus and are uncorrected. Optical rotations were recorded on a Perkin-Elmer 141 automatic polarimeter in chloroform solution using a quartz cell of 10 cm pathlength with the concentration (in g/100 ml) given in brackets. The infrared spectra were recorded on Perkin-Elmer 710, 710B, and 1710 spectrometers in chloroform solution (using NaCl cells of 0.1 mm pathlength) or as thin film (using NaCl plates). The ultraviolet spectra were recorded on Cary 15 or Unicam SP800 spectrometers using quartz cells of 1 cm pathlength. The *H nmr spectra were recorded on Bruker WH-400, AE-200 or Varian XL-300 spectrometers (with solvent given in brackets) and the chemical shifts are reported on the delta (8) scale in ppm relative to tetramethylsilane. The structures presented after each title may have primed numbers for the substituents to facilitate the *H nmr assignments. A signal designated as a triplet (t) can be either a true triplet or an apparent triplet consisting of an overlapping doublet of doublets.  Assignments, where given, are based on a combination of chemical shift,  coupling constant, decoupling and nOe difference data. The  1 3  C nmr spectra were  recorded on Bruker AE-200 or Varian XL-300 spectrometers and the chemical shifts are reported on the delta (8) scale in ppm relative to tetramethylsilane. The mass spectra were recorded on AEI-MS-9 or KRATOS-MS-50 (using the electron impact ionisation method for low and high resolution analysis) or Delsi Nermag RIO-IOC (using the chemical ionisation method) spectrometers. High resolution mass measurements were obtained on samples that gave one peak on G C analysis, or in the case of mixtures of isomers no unidentified peaks on G C analysis. Gas chromatography-mass spectroscopy (GCMS)  93  analyses were performed using a Varian 6000 chromatograph coupled to a Delsi Nermag RIO-IOC mass spectrometer. Elemental analyses were determined using a combustion technique by Mr. P. Borda, Microanalytical Laboratory, University of British Columbia. Previously known compounds, some by-products or unstable intermediates may not have elemental analysis. Single crystal X-ray structure determinations were performed by Dr. S. Rettig with data obtained on Riguka AFC6S or Enraf-Nonius CAD4-F diffractometers (crystal structure data available in Appendices). Gas chromatography (GC) was performed on a Hewlett-Packard 5890A gas chromatograph, using a flame ionisation detector and a 25 m x 0.2 mm fused silica capillary column coated with cross-linked methyl silicone gum (HP1)  or a 14.5 m x 0.252 mm fused silica capillary  cyanopropylphenyl silicone gum (DP1701).  column coated with  Column chromatography was performed  unless otherwise stated by modified 'flash chromatography' (74) using columns of silica gel (230-400 mesh) with nitrogen gas pressure to obtain a suitable flow. Reactions were monitored by thin layer chromatography (TLC) analyses, which were carried out on commercial aluminium-backed silica gel plates (Merck Silica Gel 60 F254). Visualisation was accomplished with ultraviolet light and/or by spraying with 10% ammonium molybdate in 10% sulphuric acid, followed by heating. Bulb-to-bulb distillation was performed using a Kugelrohr distillation apparatus obtained from Aldrich Chemical Co. with the air bath temperature at which distillation occurred given in brackets. A l l reactions were performed under a positive pressure of dry nitrogen at room temperature unless otherwise stated. Reactions performed at -78°C were cooled using a dry ice/acetone bath. Photolysis reactions were performed in quartz vessels (for reactions requiring wavelengths < 300 nm) or Pyrex® vessels. Ultraviolet light for these reactions was generated by a high pressure Hanovia mercury lamp jacketed in a water cooled quartz immersion well.  94  A l l chromatography solvents were distilled prior to use. The term 'petroleum ether' refers to commercially available hydrocarbons boiling in the 35-60°C range. Anhydrous diethyl ether, tetrahydrofuran, benzene and toluene were prepared by distillation from a mixture containing sodium and benzophenone. Anhydrous dichloromethane was prepared by passage through a column of neutral alumina. Anhydrous methanol and ethanol were prepared by distillation from magnesium. Commercial reagents were purified, where necessary, by procedures contained in Perrin and Perrin (75). 'Active' manganese dioxide was prepared by the method of Attenburrow (68) and tested by the method of Fatiadi (60). The source of "thujone" was redistilled Western red cedar leaf oil which had a total thujone content ranging from 80-96%. Western red cedar leaf oil was obtained as a gift from Intrinsic Research and Development Incorporated.  2.4.2. (IS, 5S)  4,4-Dimethy]-l-(l-methy]ethyl)-bicyclo[3.1.0] hexan-3-one (58)  and  (IS, 2 R , 5S)  l-(l-methyIethyI)-2,4,4-trimethyIbicyclo[3.1.0] hexan-3-one (59).  58  59  The source of "thujone" was redistilled cedar leaf oil which had a total 3-(-)-isothujone/3-(+)-thujone content of 86%. Potassium r-butoxide (157.6 g, 1.40 mol) was dissolved in f-butanol (750 ml), diluted with anhydrous tetrahydrofuran (1500 ml) and the resulting solution heated to reflux. To this solution was added dropwise a solution of cedar leaf oil (158.6 g, 0.989 mol of thujone) in anhydrous tetrahydrofuran (750 ml) over 15 min.  The reaction was stirred at reflux for a further 3 h then cooled to room  temperature. Iodomethane (88 ml, 1.4 mol) was added to the mixture which was stirred for 85 min and then added to brine (2500 ml). The aqueous layer was extracted with petroleum ether (3 x 250 ml). The combined organic fractions were washed with brine (2 x 500 ml), dried (over MgSCU), filtered and evaporated in vacuo to give a yellow oil (164.2 g) that contained monomethylated thujone 58 (78.6%) and dimethylated thujone 59 (8.3%) by G C (1-menthol as internal standard). Based on the purity of the starting material the yield for 58 is 86.4% and for 59 is 8.4%. It is possible to separate 58 from  96  59 either by repeated flash chromatography using diethyl ether/petroleum ether (1:19) as eluent, or by spinning band distillation. Monomethylated thujone 58: bp 6 8 ° C / 8 Torr; [cc]  22 D  = -22.2°  Vmax (thin film): 2940, 2855, 1735 cm-*: *H nmr (400 M H z ,  J = 6, 4 Hz, C6-U ),  (CHC1 , 3  5: 0.01 (IH, dd,  CDCI3)  0.64 (IH, ddd, / = 8, 6, 2.5 Hz, C6-U ),  endo  c = 0.144); ir  0.96 (3H, d,  exo  / = 7 H z , C 7 - C H ) , 1.02 (3H, d, J = 7 Hz, C 7 - C H ) , 1.03 (3H, s, C 4 - C H ) , 3  3  3  1.12  (3H, s, C 4 - C H ) , 1.17 (IH, dd, / = 8, 4 Hz, C 5 - H ) , 1.39 (IH, septet, / = 7 Hz, C7-H), 3  2.09 (IH, d, / = 19 Hz, C2-U ), endo  2.66 (IH, dd, J = 19, 2.5 Hz, C2-H ) exo  ppm; mass  spectrum m/z: 166 ( M ) , 41 (100); high resolution mass measurement: calculated for +  CiiHigO:  166.1358; found:  166.1351; anal.: calculated for C n H i g O :  C 79.47;  H 10.91; found: C 79.53; H 10.90. ! H nmr decoupling experiment: irradiation of the doublet of doublets resonating at 5 2.66 ppm collapsed the resonance at 5 2.09 ppm to a singlet and simplified the doublet of doublet of doublets resonating at 8 0.64 ppm to a doublet of doublets (/ = 8, 6 Hz). Dimethylated thujone 59: bp 68°C / 8 Torr; [ a ]  2 2 D  = -49.2°  (thin film): 2950, 2855, 1740 cm-1; *H nmr (400 MHz, 4 Hz, C 6 - H  e n d o  CDC1 ) 3  ) , 0.54 (IH, ddd, / = 8, 6, 2 Hz, C6-H ), exo  3  m a x  8: -0.16 (IH, dd, 7 = 6, 0.98 (3H, d, / = 7 H z ,  C 7 - C H ) , 0.99 (3H, s, C 4 - C H ) , 1.03 (3H, d, / = 7 Hz, C 2 - C H 3  c = 3.282); ir v  (CHCI3,  3 e n  ^ ) , 1.03 (3H, d,  J = l H z , C 7 - C H ) , 1.09 (3H, s, C 4 - C H ) , 1.12 (IH, m, C 5 - H ) , 1.51 (IH, septet, 3  / = 7 Hz, C7-H), 2.80 (IH, qd, J = 1,2  3  Hz, C2-H ) exo  ppm; mass spectrum m/z:  180 ( M ) , 165, 137, 123, 109, 83 (100); high resolution mass measurement: calculated +  for  C12H20O:  180.1514; found: 180.1510; anal.: calculated for C i H o O : C 79.96; H  11.18; found: C 79.99; H 11.17.  2  2  97  *H nmr decoupling experiment: irradiation of the quartet of doublets resonating at 8 2.80 ppm collapsed the doublet resonating at 8 1.03 ppm to a singlet and simplified the signal resonating at 8 0.54 ppm to a doublet of doublets (7 = 8,6 Hz).  2.4.3. (IS, 2R, 5S) l-(l-MethylethyI)-2,4,4-trimethylbicyclo[3.1.0] hexan-3-one (59), (IS, 2S, 5S) l-(l-methylethyl)-2,4,4-trimethylbicycIo[3.1.0] hexan-3-one (63) and (IS, 5S)  l-(l-methylethyl)-2,2,4,4-tetramethyl-bicyclo[3.1.0] hexan-3-one (61).  59  63  61  Method A : To a solution of anhydrous diisopropylamine (0.93 ml, 6.6 mmol) in anhydrous tetrahydrofuran (50 ml) at -78°C was addedrc-butyllithium(1.55 M in hexanes, 4.25 ml, 6.58 mmol).  The reaction was stirred for 50 min, then monomethylated thujone 58  (1.103 g, 6.64 mmol) in anhydrous tetrahydrofuran (20 ml) was added. The reaction was stirred at -78°C for 100 min, warmed to 10°C and stirred for a further 60 min. Freshly purified iodomethane (1.23 ml, 19.8 mmol) was added and stirring continued for 40 min. The reaction was quenched by the addition of saturated ammonium chloride solution  98  (30 ml) and extracted with diethyl ether (3 x 30 ml). The combined organic fractions were diluted with petroleum ether (150 ml), washed with brine (50 ml), dried (over MgSCU), filtered and evaporated in vacuo to give a yellow oil (1.132 g) that contained (by G C , 1-menthol as an internal standard) monomethylated thujone 58 (7.1%), dimethylated thujone 59 (24.9%), dimethylated thujone 63 (56.6%) and trimethylated thujone 61 (5.7%). Yield based on the calibration: 58 (7.3%); 59 (23.6%); 63 (53.6%); 61 (5.0%). To this oil (1.126 g) dissolved in ethanol (25 ml) was added potassium hydroxide (ca 100 mg). The reaction was stirred for 7 h and the solvent evaporated in vacuo. The residue was diluted with diethyl ether (100 ml), washed with saturated ammonium chloride solution (3 x 50 ml), water (50 ml), dried (over MgSC<4), filtered and evaporated in vacuo to give a yellow oil (0.9706 g) that contained (by G C , 1-menthol as an internal standard) 58 (7.2%), 59 (83.2%), 63 (1.5%) and 61 (6.0%). In subsequent reactions a mixture of 58 and 59 was used. Thus, using the above procedure with 58:59 = 79:8 (54.405 g), 1.04 equivalents of lithium diisopropylamide and 3 equivalents of iodomethane an oil was obtained which was shown by G C (1-menthol as an internal standard) to contain 58 (24%), 59 (34%), 63 (25%) and 61 (10%). A solution of this oil and potassium hydroxide (4.20 g) in ethanol (400 ml) was stirred for 4.5 h. The reaction was quenched with saturated ammonium chloride solution (200 ml) and the organic solvent evaporated in vacuo. The residue was diluted with dichloromethane (200 ml) and the aqueous layer extracted with dichloromethane (2 x 200 ml). The combined organic layers were dried (over MgS04), filtered and evaporated in vacuo to afford an oil (56.34 g) that was shown by G C (1-menthol as an internal standard) to contain 58 (22%), 59 (54%), 63 (1%) and 61 (11%).  99  Spinning band distillation followed by column chromatography on silica using diethyl ether/petroleum ether (1:19), followed by chromatography using diethyl ether/petroleum ether (2:98) gave pure samples of each of the four compounds. Dimethylated thujone 63: mp 53-54°C; bp 67°C/8 Torr; [ a ] c = 0.63); ir v  m a x  2 2 D  2960, 1740 cnr*; l H nmr (400 MHz,  (CHCI3):  CDCI3)  (CHCI3,  5: -0.26 (IH,  dd, / = 6,  5 H z , C6-H ),  C6-K ),  0.98 (3H, d, / = 7 H z , C 7 - C H ) , 1.03 (3H, s, C4-CH ), 1.17 (3H, s,  ex0  endo  0.71 (3H, d, J = 7 H z ,  = +15.6°C  C 7 - C H ) , 0.72 3  3  C 4 - C H ) , 1.26 (3H, d, / = 8 Hz, C 2 - C H 3  (IH, m,  3  3 e X 0  ) , 1.40 (IH, dd, / = 9, 5 Hz, C5-H),  2.16 (IH, septet, / = 7 Hz, C7-H), 2.31 (IH, q, J = 8 Hz, C2-H ) endo  ppm; mass  spectrum m/z: 180 ( M ) , 165, 152, 137, 83 (100); high resolution mass measurement: +  calculated for  C12H20O:  180.1514; found: 180.1511; anal.: calculated for C i H o O : 2  2  C 79.96; H 11.18; found: C 79.84; H 11.17. *H nmr decoupling experiments:  irradiation of the signal resonating at 8 -0.26 ppm  affected the multiplet resonating at 8 0.72 ppm and simplified the doublet of doublets resonating at 8 1.40 ppm to a doublet (/ = 9 Hz); irradiation of the doublet resonating at 8 1.26 ppm collapsed the quartet resonating at 8 2.31 ppm to a singlet. Trimethylated thujone 61: bp 68°C/8 Torr; [ a ]  2 2 D  = +15.7°C (CHC1 c = 0.70); ir v x 3>  m a  (CHCI3): 2975, 1735 cm" ; *H nmr (400 MHz, CDCI3) 8: -0.24 (IH, dd, J = 6.5, 1  4 H z , C6-H ), endo  0.66 (IH, dd, J = 8.5, 4 Hz, C6-H ), exo  0.76 (3H, d, / = 7 H z ,  C 7 - C H ) , 1.00 (3H, d, J = l Hz, C7-CH ), 1.01 (3H, s), 1.07 (3H, s), 1.19 (3H, s), 3  3  1.22 (3H, s), 1.39 (IH, dd, J = 8.5, 6.5 Hz, C5-H), 2.20 (IH, septet, / = 7 H z , C7-H) ppm; mass spectrum m/z: 194 ( M ) , 179, 166, 123 (100); high resolution mass +  measurement: calculated for C 1 3 H 2 2 O : 194.1671; found: 194.1666; anal: calculated for C13H22O:  C 80.36; H 11.41; found: C 80.51; H 11.39.  100  H nmr decoupling experiments: irradiation of the doublet of doublets resonating at  l  8 -0.24 ppm simplified the doublet of doublets resonating at 8 0.66 ppm to a doublet (J = 8.5 Hz) and simplified the doublet of doublets resonating at 8 1.39 ppm to a doublet (/ = 8.5 Hz); irradiation of the septet resonating at 8 2.20 ppm collapsed the doublet resonating at 8 0.76 ppm to a singlet and collapsed the doublet resonating at 8 1.00 ppm to a singlet.  Method B: To a solution of silyl enol ether 67 (216.6 mg, 0.86 mmol) in methanol (10 ml) was added 2 M hydrochloric acid (1.5 ml). After stirring for 30 min, the solvent was evaporated in vacuo and diethyl ether (30 ml) added to the residue. The solution was washed with saturated sodium bicarbonate solution (30 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a colourless oil (150.2 mg, 97%) that was shown by G C (1-menthol as internal standard) to contain 59 (98.7%) and 63 (1.3%).  2.4.4.  Equilibration of 59 and  63  63.  59  A solution of dimethylthujone 63 (260.5 mg, 1.45 mmol) and potassium hydroxide (128 mg, 2.29 mmol) in ethanol (8 ml) was stirred for 4 h, then quenched with saturated  101  ammonium chloride solution (20 ml). The organic solvent was evaporated in vacuo and the residue extracted with diethyl ether (3 x 30 ml). The combined organic fractions were dried (over MgSC»4), filtered and evaporated in vacuo to leave a yellow oil (182.0 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:19) as eluent afforded a colourless oil (169.4 mg, 65% recovery) that was shown by GC (1-menthol as an internal standard) to contain 59 (98.6%) and 63 (1.4%). A solution of dimethylthujone 59 (149.0 mg, 0.83 mmol) and potassium hydroxide (60 mg, 1.1 mmol) in ethanol (4 ml) was stirred for 4 h. The solvent was evaporated in vacuo and diethyl ether (30 ml) was added. The solution was washed with saturated ammonium chloride solution (10 ml), dried (over MgSC<4), filtered and evaporated in vacuo to leave a yellow oil. Purification by flash chromatography using diethyl ether/petroleum ether (1:19) as eluent afforded a colourless oil (85.9 mg, 5 8 % recovery) that was shown by G C (1-menthol as an internal standard) to contain 59 (98.8%) and 63 (1.2%).  2.4.5. (IS, 5S) l-(l-MethylethyI)-3-trimethylsiloxy-2,4,4-trimethylbicycIo[3.1.0]hex-2-ene (67).  67  To a mixture of 59 and 61 (containing 59 50%, 61 35%, 5.998 g, 16.6 mmol in 59) in dichloromethane was added trimethylsilyl chloride (3.8 ml, 30 mmol), 1,8-diazabicyclo-  102  [5.4.0]undec-7-ene (DBU, 4.5 ml, 30 mmol) and silver nitrate (1.83 g, 10.8 mmol). After heating at reflux for 3 days, the reaction was cooled and the solvent evaporated in vacuo to leave a thick oil. Purification by flash chromatography using petroleum ether as eluent afforded 67 (1.950 g, 46% based on the purity of the starting material) as a colourless oil. Further elution using diethyl ether/petroleum ether (1:49) as eluent recovered a mixture of 59 and 61 (2.4886 g, 41%) as a colourless oil. Silyl enol ether 67:  [a]  2 2  D  = +46° (CHCI3, c = 0.15); ir v  m a  x (thin film): 2950,  1665 cm- ; U nmr (400 MHz, CDCI3) 5: 0.03 (IH, t, / = 5 Hz, C6-H ), 1  l  endo  s, Si-(CH )3), 0.58 (IH, dd, / = 8, 5 Hz, C6-H ), 3  exo  0.17 (9H,  0.73 (3H, d, / = 7 H z ,  C7-CH ), 0.85 (IH, dd, / = 8, 5 Hz, C5-H), 0.92 (3H, d, / = 7 Hz, C7-CH ), 0.96 3  3  (3H, s, C4-CH3), 1.08 (3H, s, C4-CH ), 1.56 (3H, s, C2-CH ), 1.82 (IH, septet, 3  3  / = 7 Hz, C7-H) ppm; mass spectrum m/z: 252 (M ), 237 (100); high resolution mass +  measurement: calculated for Cis^sSiO: 252.1910; found: 252.1910.  2.4.6. (IS, 4S, 5S)  3-EthyIcarbonyIdioxy-4-methyM-(l-methylethyl)bicyclo[3.1.0]hex-2-ene (71).  OCOOEt  To a solution of anhydrous diisopropylamine (0.46 ml, 3.3 mmol) in anhydrous tetrahydrofuran (25 ml) at -78°C was added «-butyllithium (1.55 M in hexanes, 2.1 ml, 3.3 mmol).  The reaction was stirred for 75 min, then thujone ((-)-isothujone/  103  (+)-thujone = 95:5, 482.4 mg, 3.17 mmol) in anhydrous tetrahydrofuran (5 ml) was added. The reaction was stirred at -78°C for 100 min, warmed to 0 ° C and stirred for a further 60 min. Freshly purified ethyl chloroformate (0.45 ml, 4.7 mmol) was added and stirring continued at room temperature for 60 min. The reaction was quenched by the addition of saturated ammonium chloride solution (30 ml) and extracted with diethyl ether (2 x 30 ml). The combined organic fractions were diluted with petroleum ether (50 ml), washed with brine (75 ml), dried (over MgS04), filtered and evaporated in vacuo to give a yellow oil (919 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:19) as eluent gave enol carbonate 71 (624.5 mg, 89%) as a colourless oil. (GC and *H nmr confirm isothujone enol carbonate/thujone enol carbonate ratio as 95:5):  irvmax (thin film): 2950, 1760, 1655, 1635 cm" ; l H nmr (400 MHz, CDC1 ) [signals 1  3  corresponding to the major isomer] 5: 0.28 (IH, t, / = 5 Hz, C6-H ^ ), 0.74 (IH, dd, en  / = 8, 5 Hz, C6-H ), exo  (3H, d, / = 7 H z ,  0  0.90 (3H, d, J = 7 Hz, C7-CH ), 0.94 (IH, m, C5-H), 0.98 3  C 7 - C H ) , 1.10 (3H, d, / = 7 H z , 3  C 4 - C H ) , 1.30 (3H, t, 3  J = 8 Hz, CH -CH3),1.36 (IH, septet, / = 7 Hz, C7-H), 2.62 (IH, br q , / = 7 Hz, 2  4.19 (2H, q, / = 8 Hz,  C4-H ), endo  ppm; mass spectrum m/z:  CH.2-CH3),  224 (M ), +  165,  5.66 (IH, br s, wh  152,  137  /2  = 4 Hz, C2-H)  (100); high resolution mass  measurement: calculated for Ci3H2o03: 224.1413; found: 224.1414; anal.: calculated for  Ci3H o0 : C 69.62; H 8.99; found: C 69.46; H 9.00. 2  3  104  2.4.7. (IS, 5S)  4,4-Dimethyl-3-ethylcarbonyldioxy-l-(l-methylethyl)bicyclo[3.1.0]hex-2-ene  (73).  OCOOEt  To a solution of anhydrous diisopropylamine (0.45 ml, 3.2 mmol) in anhydrous tetrahydrofuran (25 ml) at -78°C was added «-butyllithium (1.55 M in hexanes, 2.1 ml, 3.3 mmol).  The reaction was stirred for 30 min at -10°C, cooled to -78°C, then  monomethylated thujone 58 (537.0 mg, 3.23 mmol) in anhydrous tetrahydrofuran (10 ml) was added. The reaction was stirred at -78°C for 100 min, warmed to 10°C and stirred for a further 40 min. Freshly distilled ethyl chloroformate (0.922 ml, 9.64 mmol) was added and stirring continued at room temperature for 25 min. The reaction was quenched by the addition of saturated ammonium chloride solution (30 ml) and extracted with diethyl ether (2 x 30 ml). The combined organic fractions were diluted with petroleum ether (50 ml), washed with brine (75 ml), dried (over MgS04), filtered and evaporated in vacuo to give a yellow oil (1.180 g). Purification by flash chromatography using diethyl ether/petroleum ether (1:19) as eluent gave enol carbonate 7 3 (645.2 mg, 96%) as a colourless oil: ir v (thin film): 2945, 1760, 1630 cm- ; *H nmr (400 M H z , CDC1 ) 5: 1  / = 4 Hz, C6-U ), endo  3  0.61 (IH, dd, / = 8, 4 Hz, C6-U ), exo  m a x  0.30 (IH, t,  0.91 (3H, d, / = 7 Hz,  C 7 - C H ) , 0.99 (IH, m, C5-H), 0.99 (3H, d, / = 7 Hz, C7-CH ), 1.05 (3H, s), 1.18 3  (3H, s), 1.33 (3H, t, / = 8 Hz, C H 2 - C H 3 ) , 1.40 (IH,  3  septet, / = 7 Hz, C 7 - H ) , 4.20  (2H, q, / = 8 Hz, C H 2 - C H 3 ) , 5.66 (IH, s, C2-H) ppm; mass spectrum m/z: 238 (M+),  105  179, 165, 151 (100); high resolution mass measurement: calculated for  C14H22O3:  238.1569; found: 238.1567.  2.4.8. (IS, 4R, 5S)  4-MethyI-l-(l-methylethyl)-3-(4-morpholinyl)bicyclo[3.1.0]hex-2-ene (74)  and (IS, 4S,5S) 4-Methyl-l-(l-methylethyl)-3-(4-morpholinyl)bicyclo[3.1.0]hex-2-ene (75).  A solution of cedar leaf oil (total thujone content = 96%, 2:3 = 11:1, 5.000 g, 31.54 mmol of thujone), morpholine (5.7 ml, 65 mmol) and para-toluenesulphonic acid monohydrate (170 mg, 0.89 mmol) in toluene (20 ml) was heated at reflux with azeotropic removal of water for 40 h. The solvent was evaporated in vacuo to leave an oil (ca 10 g). Purification by bulb-to-bulb distillation (110-115°C, 10 Torr) afforded a 66:34 mixture of enamines 74 and 75 (6.375 g, 91%) as a yellow oil: ir v  m a x  (thin film): 2950, 2860,  1610 cm" ; H nmr (400 MHz, CDCI3) [partial signals corresponding to 74] 8: 0.00 (IH, 1  l  m), 0.62 (IH, dd), 3.65 (4H, br s, wh/ = 10 Hz), 4.56 (IH, s, C2-H) ppm, [partial 2  signals corresponding to 75] 8: -0.15 (IH, t, / = 4 Hz), 0.40 (IH, dd, 7 = 8 , 4 Hz), 3.65 (4H, br s, wh/ = 10 Hz), 4.57 (IH, s, C2-H) ppm; mass spectrum m/z: 221 M(+), 2  206 (100).  106  2.4.9. (IS, 2 R , 3S, 5S)  l-(l-MethyIethyl)-2,4,4-trimethyl-  3-trimethylsiloxybicyclo[3.1.0]hexane-3-carbonitrile and  (IS, 2 R , 3 R , 5S)  (77)  l-(l-methylethyl)-2,4,4-trimethyl-  3-trimethylsiloxybicyclo[3.1.0]hexane-3-carbonitrile  77  (78).  78  To a mixture of ketone 59 (906 mg, 5.03 mmol) and tetra-n-butylammonium cyanide (157 mg, 0.577 mmol) was added trimethylsilylcyanide (1.0 ml, 7.3 mmol). The solution was stirred at 7 5 ° C for 8 h, cooled and added to pentane (50 ml) (the flask being washed with dichloromethane (3 ml)). The solution was filtered, evaporated in vacuo and the residue dissolved in pentane (20 ml). Filtration through Celite followed by evaporation in vacuo gave a yellow oil (1.379 g, 98%) that was shown by G C to contain trimethylsilyl enol ether 67 (2%), trimethylsilyl cyanohydrin 77 (70%) and trimethylsilyl cyanohydrin 78 (28%). The material was used in subsequent reactions without further purification. Repeating the reaction on a larger scale followed by purification by flash chromatography using hexanes as eluent afforded a pure sample of trimethylsilyl enol ether 67. Mixtures o f 77 and 78 enriched in either 77 or 78 (up to 84%) were obtained. Silyl cyanohydrins 77/78: MHz,  1  CDCI3) [signals corresponding to 77] 5:  C6-H ), exo  (thin film): 2955, 2220 (weak) cm" ; *H nmr (400  irVmax  0.15 (IH, ddd, / = 8, 5, 1 Hz,  0.23 (9H, s, S i - ( C H ) ) , 0.85 (3H, d, / = 7 Hz, C 7 - C H ) , 0.87 (2H, m, 3  3  3  107  C5-H, C6-H ), endo  0.95 (3H, d, / = 7 Hz, C7-CH ), 1.06 (3H, s, C4-CH ), 1.07 (3H, 3  3  d, / = 7 Hz, C2-CH ), 1.24 (3H, s, C4-CH ), 1.50 (IH, septet, / = 7 Hz, C7-H), 3  3  2.72 (IH, qd, / = 7, 1 Hz, C2-H ) ppm; [signals corresponding to 78] 5: 0.24 (9H, exo  s, Si-(CH ) ), 0.25 (IH, m), 0.86 (3H, d, / = 7 Hz, C7-CH ), 0.88 (IH, m), 0.96 3  3  3  (3H, d, J = 7 Hz, C7-CH ), 1.00 (3H, s, C4-CH ), 1.22 (3H, d, / = 7 Hz, C2-CH ), 3  3  3  1.24 (3H, s, C4-CH ), 1.26 (IH, m), 1.44 (IH, septet, / = 7 Hz, C7-H), 2.40 (IH, qd, 3  / = 7, 1 Hz, C2-H ) ppm; mass spectrum m/z: 264 (M+-15), 252, 237 (100); CI mass exo  spectrum (NH ): 297 (M + NH4 ), 264; high resolution mass measurement: calculated +  3  for Ci5H26NSiO: 264.1783; found: 264.1782; anal.: calculated for Ci H 9NSiO: 6  2  C 68.76; H 10.46; N 5.01; found: C 68.82; H 10.35; N 5.04.  2.4.10. ( I S , 2 R , 5S)  l-(Methylethylene)-2,4,4-trimethyl-3-  trimethyIsiloxy-bicyclo[3.1.0]hexane-3-carbonitrile ( I S , 2 R , 5S)  (87)/(88),  l-acetyl-2,4,4-trimethyl-3-trimethylsiloxybicyclo[3.1.0] hexane-3-carbo-nitrile (89)/(90)  and (IS, 2 R , 5S)  l-(l-hydroxy-l-methylethyl)-2,4,4-trimethyl-3trimethy!siloxybicyclo[3.1.0]hexane-3-carbonitriIe (91)/(92).  87/88  89/90  91/92  108  Ozone was bubbled through a solution of trimethylsilyl cyanohydrins  27  (77:78 = 28:70, 1.899 g, 6.79 mmol) and sodium bicarbonate (2.25 g) in ethyl acetate (400 ml) at 0 ° C for 8 h. The reaction was flushed with nitrogen, filtered and evaporated in vacuo to give an oil (4.082 g). The residue was dissolved in dichloromethane (80 ml), washed with sodium bicarbonate solution (25 ml), dried (over MgS04), filtered and evaporated in vacuo to give an oil (2.355 g). Purification of a portion of this oil (1.330 g) by flash chromatography using ethyl acetate/petroleum ether (1:39) as eluent gave a mixture of epimeric alkenes 87:88 (60:40) (16.1 mg, 1%) as a colourless oil. Further elution with ethyl acetate/petroleum ether (1:19) gave a mixture of epimeric ketones 89:90 (71:29) (486 mg, 45%) as a colourless oil. Further elution with ethyl acetate/petroleum ether (1:9) gave a mixture of epimeric alcohols 91:92 (66:34) (241 mg, 21%) as a white solid. Alkenes 87/88: U nmr (400 MHz, CDC1 ) [signals corresponding to 87] 5: 0.23 (9H, l  3  s, S i ( C H ) ) , 0.58 (IH, ddd, / = 8, 4.5, 1.5 Hz, C6-H ), 3  3  exo  1.03 (3H, d, / = 7 H z ,  C2-CH ), 1.08 (3H, s, C4-CH ), 1.08 (IH, m), 1.27 (IH, m), 1.28 (3H, s, C4-CH ), 3  3  1.70 (3H, br s, wh  /2  3  = 3 Hz, C7-CH ), 2.88 (IH, qd, 7 = 7, 1.5 Hz, C2-H), 4.78 (IH, 3  br s, wh/2 = 4 Hz, C8-H), 4.85 (IH, br s, wh/ = 5 H z , C8-H) ppm; [signals 2  corresponding to 88] 5: 0.23 (9H, s, Si(CH ) ), 0.69 (IH, ddd, / = 8, 6.5, 1.5 Hz, 3  C6-H  eJC0  3  ), 1.05 (3H, s, C4-CH ), 1.08 (IH, m), 1.18 (3H, d, J = 7 Hz, C2-CH ), 1.25 3  3  (3H, s, C4-CH ), 1.27 (IH, m), 1.70 (3H, br s, wh/ = 3 Hz, C7-CH ), 2.56 (IH, qd, 3  2  3  J = l, 1.5 Hz, C2-H), 4.78 (IH, br s, wh/ = 4 Hz, C8-H), 4.85 (IH, br s, wh/ = 2  5 Hz, C8-H) ppm; mass spectrum m/z:  2  277(M ), 262, 250, 235, 107 (100); high +  resolution mass measurement: calculated for Ci6H27N0Si: 277.1862; found: 277.1863.  Contains 2% trimethylsilyl enol ether 67.  109  Ketones 89/90: ir v  (thin film): 2960, 1690 cm- ; *H nmr (400 MHz, CDCI3) [signals 1  m a x  corresponding to 89] 5: 0.24 (9H, s, Si(CH ) ), 1.12 (3H, s, C4-CH ), 1.13 (3H, d, 3  3  3  J = 7 Hz, C2-CH ), 1.24 (3H, s, C4-CH ), 1.25 (IH, m), 1.59 (IH, t, J = 6 Hz), 3  3  1.69 (IH, dd, / = 8, 6 Hz),  1.92 (3H, s, COCH3), 3.37 (IH, qd, J = 7, 1 H z ,  C2-H), [signals corresponding to 90] 5: 0.23 (9H, s, Si(CH ) ), 1.02 (3H, s, C4-CH ), 3  3  3  1.25 (IH, m), 1.29 (3H, d, / = 7 Hz, C2-CH ), 1.29 (3H, s, C4-CH ), 1.52 (IH, dd, 3  3  / = 8, 6 Hz), 1.71 (IH, dd, / = 9, 6 Hz), 1.94 (3H, s, C O C H ) , 2.95 (IH, qd, 3  J = 1,1 Hz, C2-H) ppm; mass spectrum m/z: 279(M ), 264, 252; high resolution mass +  measurement:  calculated for C15H25NO2SK  279.1654; found:  279.1658; anal.:  calculated for C i H 2 5 N 0 S i : C 64.47; H 9.02; N 5.01; found: C 64.25; H 9.20; N 5.00. 5  2  Alcohols 91/92: mp 35-37°C; ir v  m a x  (CHC1 ): 3601 (sharp), 2972 cm" ; *H nmr (400 1  3  M H z , CDC1 ) [signals corresponding to 91] 8: 0.23 (9H, s, Si(CH ) ), 0.45 (IH, ddd, 3  3  J = 8.5, 5, 1.5 Hz, C6-U ), exo  3  0.95 (IH, t, / = 5 Hz), 1.06 (3H, s, C4-CH ), 1.10 3  (IH, dd, J = 8.5, 5 Hz), 1.11 (IH, br s, wh/ = 8 Hz, exchangeable, OH), 1.15 (3H, d, 2  / = 7 H z , C 2 - C H ) , 1.16 (3H, s, C4-CH ), 1.24 (3H, s, C7-CH ), 1.26 (3H, s, 3  3  3  C7-CH ), 2.92 (IH, qd, / = 7, 1.5 Hz, C2-H) ppm; [signals corresponding to 92] 8: 3  0.23 (9H, s, S i ( C H ) ) , 0.58 (IH, ddd, / = 8.5, 6.5, 1.5 Hz, C6-U xo), 0.90 (IH, 3  3  e  dd, J = 6.5, 4 Hz), 1.02 (3H, s, C4-CH ), 1.11 (IH, br s, wh/ = 8 Hz, exchangeable, 3  2  OH), 1.13 (IH, m), 1.19 (3H, s, C4-CH ), 1.24 (3H, s, C7-CH ), 1.26 (3H, s, 3  3  C 7 - C H ) , 1.30 (3H, d, / = 7 Hz, C2-CH ), 2.58 (IH, qd, J = 1, 1.5 Hz, C2-H) 3  3  ppm; mass spectrum m/z: 295(M ), 280, 277, 262; high resolution mass measurement: +  calculated for Ci6H29N02Si:  295.1968; found:  295.1966; anal.: calculated for  C16H29NO2S1: C 65.04; H 9.89; N 4.74; found: C 64.81; H 9.93; N 4.87. ! H nmr decoupling experiments: irradiation of the quartet of doublets resonating at 8 2.58 ppm collapsed the doublet resonating at 8 1.30 ppm to a singlet and simplified the doublet  110  of doublet of doublets resonating at 5 0.58 ppm to a doublet of doublet (/ = 8.5, 6.5 Hz); irradiation of the quartet of doublets resonating at 5 2.92 ppm collapsed the doublet resonating at 8 1.15 ppm to a singlet and simplified the doublet of doublet of doublets resonating at 8 0.45 ppm to a doublet of doublet (/ = 8.5, 5 Hz).  2.4.11. (IS, 2 R , 5S)  l-(l-ChIoro-l-methylethyl)-2,4,4-trimethyl-  3-(trimethylsiloxy)bicyclo[3.1.0]hexane-3-carbonitriIe (97)/(98).  CI  To a solution of alcohols 91:92 (60:40, 59.0 mg, 0.227 mmol) in dichloromethane (2 ml) at 0 ° C was added cold concentrated hydrochloric acid (2 ml). After vigorous stirring for 20 min, the organic fraction was separated, diluted with cold dichloromethane (10 ml), washed with cold brine (10 ml), dried (over MgSC>4), filtered and evaporated in vacuo to give a red oil. G C analysis indicated a 60:40 mixture of the unstable chlorides 97/98. Partial *H nmr (400 MHz, C D C I 3 ) [approximately 30% decomposition to alkenes] [signals corresponding to 97] 8: 0.63 (IH, ddd, / = 8, 6, 1.5 H z , C6-U ), exo  3.11  (IH, qd, J = 1, 1.5 Hz, C2-H) ppm; [signals corresponding to 98] 8: 0.75 (IH, ddd, J = 9,1,  1.5 H z ,  C6-H  e x o  ) , 2.78 (IH, qd, 7 = 7, 1.5 H z ,  spectrum m/z: 315/313 (M+), 300/298, 277.  C2-H) ppm; mass  Ill  2.4.12. (2R, 4S)  4-Chloromethyl-3-(l-methylethylidene)-  2,5,5-trimethyl-l-(trimethylsiloxy)cyclopentane-l-carbonitriIe  (95).  To a solution of alcohols 91:92 (60:40, 192.0 mg, 0.650 mmol) in dichloromethane (10 ml) was added concentrated hydrochloric acid (10 ml). The reaction was stirred vigorously for 88 h. The aqueous layer was extracted with dichloromethane (2 x 50 ml). The combined organic fractions were washed with sodium bicarbonate (50 ml), dried (over MgSC»4), filtered and evaporated in vacuo to give a red oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded a mixture of these three isomeric chlorides (105.4 mg, 52%), of which 95 is predominant (83%): ir Vmax (thin film): 2974, 2232, 1673 cm" ; anal.: calculated for C i 6 H 8 N O S i C l : 1  2  C 61.22; H 8.99; N 4.46; found: C 61.50; H 9.10; N 4.63. Further repeated chromatography using diethyl ether/petroleum ether (3:97) as eluent gave a pure sample of chloride 95 as a colourless oil:  28  *H nmr (400 MHz,  CDCI3)  5: 0.28 (9H, s, S i ( C H ) ) , 1.10 (3H, s, C 5 - C H ) , 1.15 (3H, d, / = 7 Hz, C 2 - C H ) , 3  3  3  3  1.25 (3H, s, C 5 - C H ) , 1.70 (3H, br s, wh/ = 3 Hz, C=C-CH ), 1.75 (3H, br s, wh/ = 3  2  3  2  3 H z , C = C - C H ) , 2.67 (IH, dd, / = 8, 5.5 Hz, C4-H), 3.16 (IH, br q, J = 7 H z , 3  C2-H), 3.47 (IH, dd, / = 11.5, 5.5 Hz,  C4-CHC1),  This purification was carried out by R. G. Clewley.  3.87 (IH, dd, / = 11.5, 8 H z ,  112  C4-CHC1) ppm; mass spectrum m/z: 315/313 (M+), 300/298, 277, 262; high resolution mass measurement: calculated for Ci6H28NOSi Cl: 313.1629; found: 313.1631. 35  2.4.13. (2R)  3-(l-methylethyIidene)-2,6,6-trimethyll-(trimethylsiIoxy)-cyclohexane-l-carbonitriIe (101)/(102).  This experiment was performed by R. G. Clewley, but has been included here for 29  completeness.  A solution of the chlorides obtained in the preceeding section (314.0 mg, 1.00 mmol, containing 95 (83%)), 2,2 azobis(2-methylpropionitrile) (AIBN, 33.0 mg, 0.20 mmol) and tri-rt-butyltin hydride (1.08 ml, 4.00 mmol) in dry toluene (10 ml) was heated at reflux for 4 h. The solvent was evaporated in vacuo to leave an oil (> 1 g). Purification by flash chromatography using petroleum ether as eluent removed the tin compound present. Further elution using diethyl ether/petroleum ether (1:199) gave a mixture of alkenes 101:102 (83:17, 193.0 mg, 69%). Further purification by flash chromatography using diethyl ether/petroleum ether (1:199) afforded a pure sample of 101: *H nmr (400 MHz,  CDC1 ) 8: 0.29 (9H, s, Si(CH ) ), 1.03 (3H, s, C6-CH3), 1.07 (3H, d, / = 7 H z , 3  2 9  3  3  R. G. Clewley, research report, U.B.C. (1989).  113  C 2 - C H ) , 1.09 (3H, s, C6-CH3), 1.48 (2H, m, C 5 - H ) , 1.72 (6H, br s, wh/ = 4 Hz, 3  2  2  C = C - ( C H ) ) , 2.03 (IH, m, C4-H), 2.44 (IH, dtd, / = 14, 3, 1 Hz, C4-H), 3.28 (IH, 3  2  q, J = 7 Hz, C2-H) ppm.  2.4.14. (IS, 2R, 5S)  l-(l-Hydroxyethyl)-2,4,4-trimethyl-3-(trimethyl-  siloxy)bicyclo[3.1.0]hexane-3-carbonitrile (105), (106), (107) and (108).  To a solution of ketones 89/90 (504 mg, 1.80 mmol) in ethanol (20 ml) was added 30  sodium borohydride (130 mg, 3.42 mmol). After stirring for 1.5 h at room temperature, the solvent was removed in vacuo. Diethyl ether (50 ml) and saturated ammonium chloride solution (50 ml) was added to the residue. The organic layer was washed with brine (20 ml), dried (over MgSC<4), filtered and evaporated in vacuo to give an oil (554 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent afforded a mixture of four isomeric alcohols 105:106:107:108 (38:35:14:13) (411.6 mg, 92% based on purity of starting material) as a colourless oil: ir v 2960, 2210  m a x  (thin film): 3475 (br),  (weak) cm" ; *H nmr (400 MHz, CDC1 ) [partial assignments for 105] 5: 1  3  0.23 (9H, s, S i ( C H ) ) , 2.94 (IH, qd, J = 7, 1 Hz, C2-H), 3.60 (IH, q, / = 6.5 Hz, 3  3  C7-H) ppm; [partial assignments for 106] 5: 0.23 (9H, s, S i ( C H ) ) , 2.82 (IH, qd, 3  Sample contained alcohols 91/92 (60.3 mg).  3  114  J = 1,1 Hz, C2-H), 3.70 (IH, q, J = 6.5 Hz, C7-H) ppm; [partial assignments for  107] 5: 0.23 (9H, s, Si(CH ) ), 1.28 (3H, d, / = 7 Hz, C2-CH ), 2.60 (IH, qd, 3  3  3  / = 7, 1 Hz, C2-H), 3.56 (IH, q, J = 6.5 Hz, C7-H), [partial assignments for 108] 8: 0.23 (9H, s, Si(CH ) ), 0.45 (IH, ddd, / = 7, 6, 1 Hz, C6-H ), 3  3  exo  1.02 (3H, s,  C 4 - C H ) , 1.27 (3H, d, J = 1 Hz, C2-CH3), 2.50 (IH, qd, / = 7, 1 Hz, C2-H), 3.66 3  (IH, q, / = 6.5 Hz, C7-H) ppm; mass spectrum m/z: 281 ( M ) , 279, 266, 263, 248, +  239, 236; high resolution mass measurement: calculated for CisH27N02Si: 281.1811; found: 281.1807; anal: calculated for Ci5H 2 7N0 2 Si: C 64.00; H 9.67; N 4.98; found:  C 63.89, H 9.76, N 4.90.  2.4.15. (IS, 2R, 5S) l-(l-Chloroethyl)-2,4,4-trimethyl3-(trimethylsiloxy)bicyclo[3.1.0]hexane-3-carbonitrile  (109)/(110).  OSiMe-,  To a solution of alcohols 105:106:107:108 (38:35:14:13, 397.2 mg, 1.41 mmol) in dichloromethane (15 ml) was added concentrated hydrochloric acid (17 ml) and the reaction was stirred vigorously for 31 h. The aqueous layer was extracted with dichloromethane (10 ml). The combined organic layers were washed with water (25 ml), dried (over MgSC»4), filtered and evaporated in vacuo to give a red oil (419.6 mg), that was shown by G C to contain no starting alcohols. Purification by flash chromatography using diethyl  115  ether/petroleum ether (1:19) as eluent gave a mixture of chlorides 109:110 (63:37, 31  139.0 mg, 33%, 83% based on recovered starting material).  Further elution afforded  recovery of starting material (240.0 mg, 60%). Chlorides 109/110: i r v  m a  x (thin film): 2950, 2210, 1660 cm'- ; *H nmr (400 M H z , 1  CDCI3) [partial assignment of signals corresponding to 109] 8: 0.23 (9H, s, Si(CH3)3), 0.47 (IH, ddd, 7 = 9 , 7 , 1.5 Hz, C6-H ),  1.08 (3H, s, C4-CH ), 1.13 (3H, d, 3  exo  / = 7 H z , C2-CH ), 1.28 (3H, s, C4-CH ), 1.40 (3H, d, / = 7 Hz, C7-CH ), 3.03 3  3  3  (IH, qd, / = 7,1.5 Hz, C2-H), 4.11 (IH, q, / = 7 Hz, C7-H) ppm; [partial assignment of signals corresponding to 110] 8: 0.23 (9H, s, Si(CH3)3), 0.40 (IH, ddd, / = 9, 6, 1.5 H z ,  C6-H ), ex0  1.07 (3H, s, C4-CH ), 1.17 (3H, d, / = 7 H z , 3  C 2 - C H 3 ) , 1.29 (3H, s, C4-CH ), 1.55 (3H, d, / = 7 Hz, C7-CH ), 3.04 (IH, m, 3  3  C2-H), 3.75 (IH, q, / = 7 Hz, C7-H) ppm; mass spectrum m/z: 301/299 ( M ) , +  286/284, 263; high resolution mass measurement: calculated for CisH26NOSi Cl: 35  299.1472; found: 299.1471.  Contains minor amounts of five-membered ring compounds.  116  2.4.16. (2R) 3-(l-Ethylidene)-2,6,6-trimethyl-l-(trimethylsiloxy)cyclohexane-l-carbonitrile (111)/(112). OSiMe  5  3  ip-CN  6  ^ 7  To a solution of chlorides 109:110 (63:37, 152.3 mg, 0.508 mmol) and A I B N (ca 32  10 mg) in anhydrous toluene (4 ml) was added tri-n-butyltin hydride (179 (il, 0.632 mmol). The reaction was heated at reflux for 2 h, cooled and diluted with petroleum ether (50 ml). The reaction was washed with water (50 ml), dried (over MgSC>4), filtered and evaporated in vacuo to give an oil (664.0 mg). Purification by flash chromatography using petroleum ether as eluent removed the tin by-products. Further elution with diethyl ether/petroleum ether (1:49) gave a mixture of predominantly two alkenes 111:112 (72:28) as a colourless oil (127.6 mg, 95%): ir v  (thin film): 2960, 1660 cm" ; H nmr (400 1  m a x  l  M H z , CDCI3) [partial assignment of signals corresponding to 111] 8: 0.21 (9H, s, S i ( C H ) ) , 1.05 (3H, s, C 6 - C H ) , 1.20 (3H, d, / = 7 Hz, C 2 - C H ) , 1.24 (3H, s, 3  3  C6-CH ), 3  3  1.62 (3H, br d, J = 1 H z ,  3  C7-CH ), 3  1.95 (IH, m), 2.48 (IH, dt,  J = 14, 4 Hz, C4-H), 2.65 (IH, br q, / = 7 Hz, C2-H), 5.20 (IH, br q, J = 7 H z , C7-H) ppm; [partial assignment of signals corresponding to 112] 8: 0.29 (9H, s, S i ( C H ) ) , 1.04 (3H, s, C 6 - C H ) , 1.11 (3H, d, / = 7 Hz, C 2 - C H ) , 1.11 (3H, s, 3  3  C6-CH ), 3  3  3  1.65 (3H, dd, / = 7, 1 H z , C 7 - C H ) ,  See footnote previous page.  3  1.93 (IH, m), 3.27 (IH, q,  117  J = 7 Hz, C2-H), 5.41 (IH, qd, / = 7, 1.5 Hz, C7-H) ppm; mass spectrum m/z: 265 (M ), 250, 238, 223. +  2.4.17. (2R*, 3S*)  2,4,4-Trimethyl-3-trimethylsiloxycyclohexan-l-one-3-carbonitrile (103)  and (2R*, 3R*)  2,4,4-trimethyl-3-trimethylsiloxycyclohexan-l-one-3-carbonitrile (104).  o  o  103  104  Method A: Ozone was bubbled through a solution of alkenes 111:112 (72:28) (80.3 mg, 0.303 mmol) in ethyl acetate (10 ml) at -78°C for 30 min. Dimethyl sulphide (0.5 ml) was then added and the reaction allowed to warm to room temperature. The reaction was diluted with diethyl ether (25 ml), washed with sodium bicarbonate solution (25 ml), dried (over MgS04), filtered and evaporated in vacuo to give an oil (119.3 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent gave ketone 103 (70.9 mg, 93%), slightly contaminated with its isomer 104 (5%). Further purification gave enriched purity to ketone 103 (purity 98.5%), and a sample enriched in 104 (17%). Ketone 103: [ a ] nmr (400 M H z ,  2 2  D  = -28.0° (CHC1 , c = 0.3); ir v  CDCI3)  3  m a  x (thin film): 2950, 1725 cm" ; *H 1  5: 0.23 (9H, s, Si(CH ) ), 1.17 (3H, s, C4-CH ), 1.23 (3H, d, 3  3  3  118  / = 7 Hz,  C 2 - C H ) , 1.41 (3H, s, C4-CH ), 1.60 (IH, ddd, J = 14, 6.5, 2 H z , 3  3  C 5 - H ) , 2.01 (IH, td, J = 14, 5 H z , C 5 - H ) , 2.34 (IH, ddd, J = 14, 5, 2 H z , e q  ax  C 6 - H ) , 2.45 (IH, td, J = 14, 6.5 H z , C 6 - H ) , 2.89 (IH, q, J = 1 H z , C 2 - H ) eq  ax  ax  ppm; mass spectrum m/z: 253 (M ), 238 (100); high resolution mass measurement: +  calculated for C i 3 H 3 N 0 2 S i : 2  253.1498; found: 253.1496; anal.: calculated for  C13H23NO2S1: C 61.62; H 9.15; N 5.53; found: C 61.80; H 9.20; N 5.60. *H nmr decoupling experiment: irradiation of the quartet resonating at 8 2.89 ppm collapsed the doublet resonating at 8 1.23 ppm to a singlet. *H nmr nOe difference experiment: irradiation of the quartet resonating at 8 2.89 ppm led to enhancement of the signals resonating at 8 1.23, 1.41 and 2.45 ppm; irradiation of the singlet resonating at 8 0.23 ppm led to enhancement of the signals resonating at 8 1.17 and 1.23 ppm. Ketone  104: G C M S of second isomer m/z: 253(M+), 238; *H nmr (400 MHz,  [partial signals corresponding to 104]  8:  CDCI3)  0.26 (9H, s, Si(CH ) ), 1.74 (IH, ddd, 3  3  / = 14, 7, 3 Hz), 1.81 (IH, td, / = 14, 6 Hz), 2.74 (IH, q, / = 7 Hz) ppm.  Method B:  113  114  To a solution of cyanohydrins 77:78 (63:37) (5.6656 g, 20.27 mmol) in ethyl acetate (300 ml) was added solid sodium bicarbonate (2.0 g). The mixture was cooled to 0 ° C and  119  ozone bubbled through for 9.5 h. The reaction was flushed with oxygen, diluted with ethyl acetate (400 ml), washed with saturated sodium bicarbonate solution (2 x 100 ml), water (100 ml), dried (over MgS04), filtered and evaporated in vacuo. The residue was dissolved in ethanol (200 ml) and sodium borohydride (800 mg, 21.1 mmol) was added. After stirring for 30 min, the solvent was evaporated in vacuo. The residue was dissolved in diethyl ether (300 ml), washed with saturated ammonium chloride solution (2 x 100 ml), water (100 ml), dried (over MgSC<4), filtered and evaporated in vacuo to give a colourless oil (6.50 g).  This was dissolved in dichloromethane (300 ml) and concentrated  hydrochloric acid (300 ml). After stirring vigorously for 78.5 h, the aqueous layer was separated and extracted with dichloromethane (100 ml). The combined organic fractions were washed with water (2 x 200 ml), dried (over MgSCU), filtered and evaporated in vacuo to give a yellow oil (5.00 g). To a solution of this oil in anhydrous toluene (165 ml) was added A I B N (400 mg, 2.8 mmol) and tri-n-butyltin hydride (8.0 ml, 30 mmol). The reaction was heated at reflux for 4.5 h, cooled, washed with water (100 ml), dried (over MgSC»4), filtered and evaporated in vacuo to give an oil (12.11 g). Purification by flash chromatography using petroleum ether as eluent removed the tin compounds present. Further elution gave a mixture of alkenes (3.3427 g). This oil was dissolved in ethyl acetate (200 ml) and cooled to -78°C. Ozone was bubbled through the solution for 30 min until a blue colouration persists. Dimethyl sulphide (1 ml) was added and the reaction allowed to warm to room temperature. Most of the solvent was removed in vacuo and ethyl acetate (200 ml) was added. The solution was washed with saturated sodium bicarbonate solution (50 ml), water (50 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a colourless oil (3.21 g). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent gave a mixture of five-membered ring ketones  120  113/114 (61:39) (267.4 mg, 5% for five steps).  Further elution gave ketone 103  contaminated with 5% of ketone 104 (1.8153 g, 35% for five steps). Ketones 113 and 114:  !  H nmr (400 MHz, CDCI3) [Partial data corresponding to 113] 8:  0.24 (9H, s, Si(CH )3), 1.05 (3H, d, / = 7 Hz), 1.10 (3H, s, C 4 - C H ) , 1.15 (3H, d, 3  3  J = 7 Hz), 1.32 (3H, s, C 4 - C H ) , 2.18 (IH, qd, / = 7, 1 Hz), 2.91 (IH, qd, / = 7, 3  1 Hz) ppm. Further purification of ketones 113 and 114 by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded a pure sample of ketone 114: H l  nmr (400 MHz, CDC1 ) 8: 0.30 (9H, s, Si(CH ) ), 0.86 (3H, s, C 4 - C H ) , 1.05 (3H, d, 3  3  3  3  / = 7 Hz), 1.32 (3H, s, C 4 - C H ) , 1.35 (3H, d, J = 1 Hz), 2.28 (IH, q, / = 7 Hz), 3  2.48 (IH, q, / = 7 Hz) ppm; mass spectrum m/z: 253 (M+), 238, 211, 43 (100). *H nmr decoupling experiments:  irradiation of the quartet resonating at 8 2.28 ppm  collapsed the doublet resonating at 8 1.05 ppm to a singlet; irradiation of the quartet resonating at 8 1.26 ppm collapsed the doublet resonating at 8 1.35 ppm to a singlet.  2.4.18. ( I R * , 2 R * , 3S*)  2,4,4-Trimethylcyclohexan-l,3-diol  a n d (IR*, 2 R * , 3R*)  (121)  2,4,4-trimethylcyclohexan-l,3-diol  (122).  OH  A solution of ketone 103 (507.0 mg, 2.00 mmol) and sodium borohydride (114.0 mg, 3.00 mmol) in ethanol (20 ml) was stirred at room temperature for 1.5 h. The solvent was evaporated in vacuo and the residue dissolved in diethyl ether (50 ml) and water (30 ml). The organic layer was washed with saturated ammonium chloride solution (25 ml), brine  121  (25 ml), dried (over MgSC»4), filtered and evaporated in vacuo to leave an oil (410 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:4) as eluent gave diol 121  (57 mg, 18%) as a colourless oil.  Further elution using ethyl  acetate/petroleum ether (1:1) as eluent afforded diol 122 (37 mg, 12%) as a colourless oil. Diol 121: ir v m a x (thin film): 3300 (br), 2900 cnr*; *H nmr (400 MHz, CDC1 ) 5: 0.91 3  (3H, s, C4-CH ), 1.01 (3H, s, C4-CH ), 1.07 (IH, m), 1.18 (3H, d, J = 7 Hz, 3  3  C2-CH3), 1.77 (4H, m), 2.88 (2H, br s, wh = 6 Hz, exchangeable, O-H), 3.22 (IH, d, /2  J = 9 Hz), 3.82 (IH, br s, wh/ = 9 Hz) ppm; mass spectrum m/z: 140 (M+-H20), 125, 2  72 (100). Diol 122: ir v m a x (CH C1 ): 3625, 2950 cnr*; *H nmr (400 MHz, C D C I 3 ) 6: 0.90 (3H, 2  2  s, C4-CH ), 1.02 (3H, s, C4-CH ), 1.11 (3H, d, J = 7 Hz, C2-CH ), 1.25 (IH, m), 3  3  3  1.42 (2H, br s, wh/ = 8 Hz, exchangeable, O-H), 1.66 (4H, m), 3.36 (IH, d, 2  / = 11 Hz), 3.91 (IH, br s, wh/ = 9 Hz) ppm; mass spectrum m/z: 2  140 (M+-H 0), 2  125 (100); high resolution mass measurement: calculated for C Q H ^ O : 140.1201; found: 140.1198; anal.: calculated for C H 9  1 8  0 : C 68.31; H 11.46; found: C 68.09; H 11.35.  122  2.4.19. (IR*,  2 R * , 3R*)  2,6,6-TrimethyIcyclohexan1,3-diol-l-carboxaldehyde (123)  and (IR*, 2 R * , 3R*)  2,6,6-trimethyI-l-trimethylsiloxy-cyclohexan-3-ol-l-carbonitrile (115).  OH  OH  115  123  Method A: To a solution of ketone 103 (231.0 mg, 0.912 mmol) in anhydrous toluene (13 ml) at -78°C was added diisobutylaluminium hydride (DIBAL-H, 1.0M in toluene, 2.0 ml, 2.0 mmol). The reaction was stirred at -78°C for 1 h, then quenched with saturated ammonium chloride solution (10 ml) and extracted with diethyl ether (2 x 50 ml). The combined organic layers were washed with ammonium chloride solution (25 ml),water (50 ml), dried (over MgSC<4), filtered and evaporated in vacuo to leave an oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent gave 115 (96.0 mg, 41%) as a colourless oil. Further elution gave aldehyde 123 (8.0 mg, 5%) as an oil. Alcohol 115: i r v m  a x  3525, 2920, 2220 (weak) cm" ; *H nmr (400 MHz, 1  (CHCI3):  CDCI3) 8: 0.30 (9H, s, Si(CH ) ), 1.08 (3H, s, C6-CH ), 1.13 (3H, s, C6-CH3), 1.18 3  3  3  (IH, dt, / = 1 3 , 3 H z , C5-H), 1.31 (3H, d, / = 7 H z ,  C 2 - C H ) , 1.67 (IH, 3  tt,  / = 14, 3 Hz, C4-H ), 1.75-1.87 (2H, m), 2.00 (IH, qd, J = 7, 3 Hz, C2-H ), 3.05 ax  ax  123  (IH, br s, wh/2 = 20 Hz, exchangeable, O-H), 3.68 (IH, br s, wh/ = 16 Hz, C3-H) 2  ppm; mass spectrum m/z: 255 (M ), 240, 222, 213, 69 (100); high resolution mass +  measurement: calculated for Ci3H25N02Si: 255.1654; found: 255.1648. *H nmr decoupling experiments: irradiation of the signal resonating at 8 3.68 ppm simplified the quartet of doublets resonating at 8 2.00 ppm to a quartet (7 = 7 Hz) and simplified the triplet of triplets resonating at 8 1.67 ppm to a triplet of doublets (/ = 14, 3 Hz); irradiation of the signal resonating at 8 2.00 ppm collapsed the doublet resonating at 8 1.31 ppm to a singlet. Aldehyde 123: ir v  m a x  (CHC1 ): 3500, 2920, 1715 cm-1; *H nmr (400 MHz, CDC1 ) 5: 3  3  0.84 (3H, s, C6-CH3), 0.91 (3H, d, 7 = 7 Hz, C2-CH ), 1.06 (IH, m), 1.20 (3H, s, 3  C6-CH3),  1.84  (2H, m),  1.99  (IH,  7 = 7,3 Hz, C2-H), 3.48 (IH, br s, wh  td, / = 14, 4 H z , /2  C5-H),  2.34  (IH,  qd,  = 28 Hz, exchangeable, O-H), 3.83 (2H, br  s, wh = 5 Hz, 1 exchangeable, C3-H, O-H), 9.60 (IH, s, CHO) ppm. /2  *H nmr decoupling experiments: irradiation of the signal resonating at 8 3.83 ppm simplified the quartet of doublets resonating at 8 2.34 ppm to a quartet (7 = 7 Hz) and affected the multiplet resonating at 8 1.84 ppm; irradiation of the signal resonating at 8 2.34 ppm collapsed the doublet resonating at 8 0.91 ppm to a singlet. *H nmr nOe difference experiment: irradiation of the signal resonating at 8 2.34 ppm led to enhancement of the signals resonating at 8 0.91, 1.20 and 9.60 ppm; irradiation of the singlet resonating at 8 9.60 ppm led to enhancement of the signals resonating at 8 0.84, 0.91, 1.20 and 2.34 ppm.  Method B: To a solution of ketone 103 (199.0 mg, 0.786 mmol) in anhydrous toluene (13 ml) at -78°C was added diisobutylaluminium hydride (DIBAL-H, 1.5M in toluene,  124  1.0 ml, 1.5 mmol). The reaction was stirred at - 7 8 ° C for 1 h, then a further portion of D I B A L - H (1.5M in toluene, 1.0 ml, 1.5 mmol) was added. The reaction was allowed to warm to room temperature over 15 min, then stirred for 1 h at this temperature. The reaction was quenched with saturated ammonium chloride solution (100 ml) and diluted with diethyl ether (150 ml). The organic layer was washed with water (30 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave an oil (100 mg). Purification by flash chromatography on alumina using ethyl acetate/petroleum ether (1:4) as eluent gave aldehyde 123 (60.0 mg, 41%) as a colourless oil.  2.4.20. 2,4,4-Trimethylcyclohex-2-en-l-one-3-carbonitrile (124).  CN  O  Thionyl chloride (2.1 ml, 29 mmol) was added slowly to a solution of cyanohydrin 103 (1.0000 g, 3.953 mmol) in methanol (55 ml) and the reaction was stirred for 2 h. The solvent was evaporated in vacuo, and diethyl ether (75 ml) was added to the residue. The organic solution was washed with saturated sodium bicarbonate solution (30 ml), brine (30 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave an oil (820 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:9) as eluent gave an unknown compound as an unstable oil (337.0 mg, 52%). Further elution gave ketone 124 as a colourless oil (306.1 mg, 48%).  125  Unknown compound: ir v a x (thin film): 3500 (br), 2950, 2195, 1645, 1580 c n r ; uv 1  m  Amax (MeOH): 306 nm (e = 4400), 217 nm (e = 4550); l H nmr (400 MHz, CDC1 ) 5: 3  1.13 (6H, s, C6-(CH ) ), 2.05 (3H, s, C2-CH ), 2.22 (2H, d, / = 6 Hz, C5-H ), 4.88 3  2  3  2  (IH, t, / = 6 Hz, C4-H) ppm; mass spectrum m/z: 162 ( M M , 100). Ketone 124: ir v  m a  x (thin film): 2950, 2200, 1690, 1645 cm" ; uv Xmax  (MeOH): 248  1  nm (e = 12000); l H nmr (400 MHz, CDC1 ) 5: 1.22 (6H, s, C4-(CH ) ), 1.94 (2H, t, 3  3  2  / = 7 Hz, C5-H ), 2.06 (3H, s, C2-CH ), 2.57 (2H, t, / = 7 Hz, C6-H ) ppm; mass 2  3  2  spectrum m/z: 163 (M ), 148, 135, 121, 120 (100); high resolution mass measurement: +  calculated for C 1 0 H 1 3 N O :  163.0997; found: 163.0994.  2.4.21. 2,4,4-Trimethylcyclohex-2-en-l-ol-3-carbonitrile  (126).  OH  Method A: To a solution of enone 124 (163.0 mg, 1.00 mmol) in methanol (2.5 ml) was added cerium III chloride heptahydrate (372 mg, 1.00 mmol) followed by sodium borohydride (38.0 mg, 1.00 mmol). After stirring for 5 min, the reaction was quenched by addition of saturated ammonium chloride solution (30 ml), diluted with diethyl ether (30 ml) and the aqueous extracted with diethyl ether (10 ml). The combined organic extracts were washed with water (10 ml), dried (over MgS04), filtered and evaporated in vacuo to give allylic alcohol 126 (163.3 mg, 99%) as a viscous oil that crystallised on cooling: mp 25-26°C; uv>. ax (MeOH): 216 nm (e = 8100); i r v m  m a x  (thin film): 3400,2950,2220, 1645 cm- ; 1  126  *H nmr (400 MHz, CDC1 ) 8: 1.11 (3H, s, C4-CH ), 1.17 (3H, s, C4-CH ), 1.48 (IH, 3  3  3  ddd, J = 13.5, 9, 3 Hz), 1.68 (2H, m), 1.90 (IH, m), 2.08 (3H, s, C2-CH ), 2.58 (IH, 3  br s, wh/2 = 28 Hz, exchangeable, OH), 4.03 (IH, t, / = 6 Hz, Cl-H) ppm;  1 3  C nmr  (50 M H z , CDC1 ) 8: 19.8, 27.9, 28.0, 28.3, 33.0, 33.7, 68.3, 117.2, 119.9, 151.9 3  ppm; mass spectrum m/z:  165 (M ), +  150, 123, 109 (100); high resolution mass  measurement: calculated for C i o H ^ N O : 165.1154; found: 165.1155; anal.: calculated f o r C H N O : C 72.70; H 9.15; N 8.48; found: C 72.73; H 9:30; N 8.35. 1 0  1 5  Method B: Thionyl chloride (1.0 ml, 13 mmol) was added slowly to a solution of cyanohydrin 103 (1.659 g, 6.55 mmol) in methanol (65 ml) at -10°C. The reaction was warmed to room temperature, and stirred for 2.5 h. The solvent was evaporated in vacuo and diethyl ether (70 ml) was added to the residue. The organic solution was washed with saturated sodium bicarbonate solution (50 ml), water (50 ml), dried (over MgS04), filtered and evaporated in vacuo to leave an oil (1.0033 g). This oil was dissolved in methanol (15.5 ml) and cerium III chloride heptahydrate (2.291 g, 6.15 mmol) was added. After cooling to -78°C, sodium borohydride (245.0 mg, 6.48 mmol) was added in small portions. The reaction was allowed immediately to warm to room temperature with stirring proceeding at this temperature for 30 min. The reaction was diluted with diethyl ether (100 ml) and saturated ammonium chloride solution (50 ml). The aqueous layer was extracted with diethyl ether (2 x 50 ml). The combined organic layers were washed with water (20 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a yellow oil (1.028 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent gave alcohol 126 (990.5 mg, 92%) as a colourless oil that crystallised on cooling.  127  Method C: To a solution of unsaturated nitrile 128 (1.1254 g, 7.54 mmol) in carbon tetrachloride (25 ml) was added iV-bromosuccinimide (1.520 g, 8.54 mmol) and benzoyl peroxide (40 mg, 1.7 x 10  _1  mmol). The reaction was heated at reflux for 40 min, then cooled and  filtered to give a crude sample of bromide 129 (1.8548 g) as a red oil. Water (55 ml) and calcium carbonate (1.07 g, 10 mmol) was added and the reaction heated at reflux for 19 h. The reaction was cooled, diluted with diethyl ether (75 ml) and the aqueous layer extracted with diethyl ether (2 x 75 ml). The combined organic layers were washed with water (50 ml), dried (over MgSC^), filtered and evaporated in vacuo to give a yellow oil. Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent gave enone 124 (144.8 mg, 12%) as a colourless oil. Further elution afforded alcohol 126 (987.4 mg, 79%) as a colourless oil that crystallised on cooling.  2.4.22. 2 , 6 , 6 - T r i m e t h y l - 3 - b r o m o c y c l o h e x - l - e n e - l - c a r b o n i t r i l e  (129).  CN  Br  To a solution of alcohol 126 (355.2 mg, 2.15 mmol) in petroleum ether (5 ml) at 0 ° C was added phosphorus tribromide (225 |ll, 2.37 mmol). The reaction was stirred for lh at 0°C then diluted with petroleum ether (25 ml), washed with sodium bicarbonate solution (2 x 20 ml), dried (over MgSC^), filtered and evaporated in vacuo to give bromide 129 (449.0 mg, 92%) as a colourless oil: ir v (MeOH)  (thin film): 2965, 2210, 1626 cm" ; uv X 1  m a x  max  229 nm (e = 12300); *H nmr (400 MHz, CDC1 ) 5: 1.11 (3H, s, C6-CH3), 3  128  1.24 (3H, s, C6-CH3), 1.57 (IH, dt, J = 13, 3.5 Hz, C5-H), 1.96 (IH, ddd, J = 13, 10, 6 Hz, C5-H), 2.10 (2H, m, C4-H ), 2.13 (3H, s, C2-CH ), 4.63 (IH, t, / = 3 Hz, C3-H) 2  3  ppm; mass spectrum m/z: 229/227 (M ), 214/212, 148 (100); high resolution mass +  measurement:  calculated for C H 1 0  1 4  N  8 1  Br:  229.0291; found:  229.0288; anal.:  calculated for Q o H ^ N B r : C 52.64, H 6.18, N 6.16; found: C 52.85, H 6.19, N 6.30.  2.4.23. 2,2,6-TrimethyIcyclohexanone  (142).  0  To a solution of diisopropylamine (90 ml, 640 mmol) in anhydrous tetrahydrofuran (2 1) at -78°C was addedrc-butyllithium(1.6M in hexanes, 405 ml, 648 mmol) dropwise. After the addition was complete, the reaction was stirred for a further 45 min at -78°C and then 2,6-dimethylcyclohexanone (144) (75.00 g, 594.3 mmol) in anhydrous tetrahydrofuran 33  (50 ml) was added over 5 min. The reaction was stirred at -78°C for 1 h, warmed to 0°C, and stirred at this temperature for a further 1 h. Iodomethane (120 ml, 1.93 mol) was added over 15 min, and the reaction was stirred for a further 20 min at 0 ° C . After quenching with saturated ammonium chloride solution (900 ml), the reaction was extracted with petroleum ether (3 x 600 ml). The combined organic extracts were washed with water (500 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave a red oil. Purification by distillation at atmospheric pressure gave a fraction (70.85 g, 85%) that was  Obtained form Aldrich Chemical Co. as a 77:23 mixture of isomers.  129  shown by G C and *H nmr to contain 97% 2,2,6 trimethylcyclohexanone 142 and 3% 2,2,6,6, tetramethylcyclohexanone 145. Ketone 142: bp 178-180°C; ir v C D C 1 ) 8: 3  C2-CH3),  m a x  (thin film): 2925, 1700 cm" ; iJi nmr (400 MHz, 1  1.01 (3H, d, 7 = 6 . 5 Hz, C 6 - C H 3 ) , 1.05 (3H, s, C2-CH ), 1.18 (3H, s, 3  1.32 (IH,  qd, 7 = 13, 4 H z ,  C 5 - H ) , 1.55 (IH, ax  td, 7 = 1 3 , 4 H z ,  C 3 - H ) , 1.65 (IH, dquintets, 7 = 13, 4 Hz, C 4 - H ) , 1.78 (IH, dq, 7 = 1 3 , 4 H z , ax  eq  C 3 - H ) , 1.89 (IH, qt, 7 = 13, 4 Hz, C 4 - H ) , 2.07 (IH, ddq, 7 = 13, 6.5, 4 H z , e q  ax  C 5 - H ) , 2.66 (IH, dquintets, 7 = 13, 6.5 Hz, C 6 - H ) ppm; mass spectrum m/z: eq  ax  140 (M ), 125, 82 (100); high resolution mass measurement: calculated for C 9 H +  140.1201; found: 140.1205; anal.: calculated for C H 9  1 6  1 6  0:  0 : C 77.09; H 11.50; found:  C 77.30; H 11.68. l H nmr decoupling experiment: irradiation of the signal resonating at 8 2.66 ppm collapsed the doublet resonating at 8 1.01 ppm to a singlet, simplified the quartet of doublets resonating at 8 1.32 ppm to a triplet of doublets (7 = 13, 4 Hz) and affected the signal resonating at 8 2.07 ppm. The presence of a singlet in the *H nmr at 8 1.02 ppm corresponded to the presence of 2,2,6,6-tetramethylcyclohexanone (145) (3%).  130  2.4.24. (IR*, 6R*)  l-Trimethylsiloxy-2,2,6-trimethylcyclohexane-lcarbonitrile (153)  and (1R*,6S*)  l-trimethyIsiloxy-2,2,6trimethylcyclohexane-l-carbonitrile (154).  A mixture of 2,2,6 trimethylcyclohexanone (142) (476.9 mg, 3.40 mmol), trimethyl34  silyl cyanide (0.50 ml, 3.8 mmol) and tetra-n-butyl ammonium cyanide (25 mg, 0.09 mmol) was heated at 7 5 - 8 0 ° C for 1.5 h. After cooling, the product was chromatographed on flash silica using diethyl ether/petroleum ether (1:25 v/v) as eluent to give a mixture of silyl cyanohydrins 153 and 154 (12:88) as a colourless oil (789.5 mg, 97%): u*  (thin film): 2940 cm" ; H nmr (400 MHz, 1  1  CDCI3)  [partial signals corresponding to 153]  8: 0.25 (9H, s, S i ( C H ) ) , 1.04 (3H, d, / = 6 Hz, 3  3  Vmax  C6-CH3),  1.05 (3H, s, C 2 - C H ) , 3  1.14 (3H, s, C 2 - C H ) , 1.30-1.70 (6H, m), 2.00 (IH, m, C6-H) [partial signals 3  corresponding to 154] 8: 0.25 (9H, s, Si(CH ) ), 0.96 (3H, s, C 2 - C H ) , 1.05 (3H, d, 3  3  3  / = 7 H z , C 6 - C H ) , 1.14 (3H, s, C 2 - C H ) , 1.30-1.70 (6H, m), 1.85 (IH, m, C6-H) 3  3  ppm, ppm; mass spectrum m/z: 239 (M ), 224 (100); high resolution mass measurement: +  3 4  Contains 3% 2,2,6,6 tetramethylcyclohexanone. In this and subsequent reactions this impurity reacted  in a similar manner to 2,2,6-trimethylcyclohexanone (except in elimination reactions). However no compounds derived from this compound were isolated or even characterised in the various spectra. Crude yields reflect the presence of these derivatives however.  131  calculated for C H S i N O : 1 3  C H SiNO: 1 3  2 5  2 5  239.1705; found:  239.1702; anal.:  calculated for  C 65.20; H 10.52; N 5.85; found: C 64.97; H 10.37; N 6.00.  2.4.25. ( 1 R * , 6 R * )  l-Hydroxy-2,2,6-trimethylcyclohexane1-carbonitrile (149)  a n d ( I R * , 6S*)  l-hydroxy-2,2,6-trimethylcyclohexane-l-carbonitrile  149  (150).  150  Method A: To a mixture of 2,2,6 trimethylcyclohexanone (142) (12.6188 g, 89.94 mmol) and powdered sodium cyanide (9.166 g, 187.0 mmol) in water (23 ml) at 0 ° C ,  3 5  was  added a solution of sodium bisulphite (17.587 g) in water (40 ml). The reaction was stirred for 6 h at 0 - 1 5 ° C , then extracted with diethyl ether (4 x 100 ml). The combined organic layers were dried (over MgSC«4), filtered and evaporated in vacuo to give an oil (15.0255 g, 100%) that solidified on standing.  G C and iff nmr analysis indicated a  mixture of cyanohydrins 150:149 = 77:23. Purification by column chromatography using either  silica  (230-400 mesh)  2,2,6-trimethylcyclohexanone:  3 5  or Florisil®  resulted  mp 38-40°C; ir v ax m  in partial reversion  (CHCI3):  to  3580 (sharp), 3450 (br),  The reaction was repeated at various temperatures, 0°C, 20°C, 30°C, 80°C, with the ratio varying from  20:80 at 0°C up to 30:70 at 60°C.  132  2930, 2250 (weak) cm" ; *H nmr (400 MHz, CDC1 ) [signals corresponding to 149] 5: 1  3  1.13 (3H, d, / = 7 Hz, C6-CH3), 1.15 (3H, s, C2-CH ), 1.17 (3H, s, C2-CH ), 1.403  3  1.60 (6H, m), 2.06 (IH, m, C6-H), 2.35 (IH, br s, wh = 100 Hz, exchangeable, OH) /2  ppm, [signals corresponding to 150] 8: 1.01 (3H, s, C2-CH ), 1.13 (3H, d,J = 7 Hz, 3  C6-CH3), 1.20 (3H, s, C2-CH3), 1.33-1.62 (5H, m), 1.72 (IH, m), 1.91 (IH, m, C6-H), 2.35 (IH, br s, wh/ = 100 Hz, exchangeable, OH) ppm; 2  CDCI3) [signals corresponding to 149] 8:  1 3  C nmr (50 MHz,  16.1, 17.1, 23.2, 26.2, 27.7, 32.6, 35.1,  38.0, 78.2, 121.0 ppm; [signals corresponding to 150] 8: 16.4, 19.2, 20.6, 26.8, 32.2, 36.3, 37.6, 38.4, 81.4, 119.6 ppm; mass spectrum m/z: 167 (M ), 166, 152, 140, 82 +  (100); high resolution mass measurement: calculated for C ^ H ^ N O : 167.1310; found: 167.1306; anal.: calculated for C H N O : C 71.80; H 10.25; N 8.41; found: C 71.60; 1 0  1 7  H 10.24; N 8.31.  Method B: A  solution of trimethylsilyl cyanohydrins 1 5 3 / 1 5 4  (12:88, 83.3 mg,  0.348 mmol) in methanol/diethyl ether (1:3, 5 ml) was saturated with hydrogen chloride gas, stirred at 4°C for 44 h, then added to iced water (5 ml) and diethyl ether (20 ml). The aqueous layer was extracted with diethyl ether (20 ml). The combined organic layers were washed with water (30 ml), dried (over MgS04), filtered and evaporated in vacuo to give an oil (52.0 mg, 90%) that solidified on standing. G C and *H nmr analysis indicated a mixture of cyanohydrins 149:150 (12:88).  Method C: To a solution of 2,2,6-trimethylcyclohexanone (142) (292.7 mg, 2.09 mmol) in anhydrous toluene (4 ml) at -15°C was added a solution of diethylaluminium cyanide  133  (IM in toluene, 4.0 ml, 4 mmol). After stirring for 6 h at -15°C, a further portion of diethylaluminium cyanide (IM in toluene, 1.0 ml, 1 mmol) was added. The reaction was stirred at -15°C for 50 min, warmed to 0°-5°C and stirred for a further 1 h. The reaction was added to a cold mixture of methanol and 6M hydrochloric acid (7:10, 34 ml) and stirred for 30 min. The aqueous layer was extracted with diethyl ether (3 x 50 ml). The combined organic fractions were washed with water (100 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave a mixture of cyanohydrins 149 and 150 (35:65, 339.8 mg, 97%) as a white solid.  2.4.26. ( 1 R * , 6 R * ) l-Acetoxy-2,2,6-trimethylcyc!ohexanel-carbonitrile (155) and (1R*,6S*)  l-acetoxy-2,2,6-trimethyIcyclohexane-l-carbonitrile  155  (156).  156  Method A: A mixture of cyanohydrins 149 and 150 (30:70, 1.0332 g, 6.18 mmol), acetic anhydride (1.40 ml, 14.8 mmol) and acetyl bromide (100 jil, 1.35 mmol) was heated at 120°C for 6.5 h. The cooled reaction was poured into water (30 ml) and extracted with diethyl ether (100 ml). The organic layer was washed with water (50 ml), saturated sodium bicarbonate solution (2 x 50 ml), water (30 ml), dried (over MgSCH), filtered and evaporated in vacuo to leave a deep red oil (1.2004 g). Purification by flash chromato-  134  graphy using ethyl acetate/petroleum ether (1:9) as eluent afforded a mixture of cyanohydrin acetates 155 and 156 (28:72, 1.0200 g, 79%) as a white solid:  ir v  m a  x  (CHC1 ): 2950, 1750 cm- ; *H nmr (400 MHz, CDCI3) 5: 2.10 (0.72 x 3H, s), 2.12 1  3  (0.22 x 3H, s) ppm. Data on pure 156 is given below (section 2.4.26 Method B).  Method B: To a solution of cyanohydrins 149 and 150 (30:70, 328.5 mg, 1.96 mmol) and A/^V-dimethylaminopyridine (DMAP, 72 mg, 0.59 mmol) in dichloromethane (25 ml) was added triethylamine (400 p.1, 2.87 mmol) and acetic anhydride (300 |il, 3.18 mmol). After stirring for 78 h at room temperature, the reaction was added to iced water (100 ml). The aqueous layer was extracted with dichloromethane (50 ml). The combined organic layers were washed with water (30 ml), dried (over MgS04), filtered and evaporated in vacuo to give a brown oil (540.8 mg). Purification by flash chromatography using ethyl acetate/ petroleum ether (1:19) as eluent afforded in order of elution: 2,2,6-trimethylcyclohexanone (142) (135.1 mg, 51%), a pure sample of acetate 156 (56.3 mg, 14%) as a white solid, and a mixture of acetates 155 and 156 (14:86, 41.0 mg, 10%). Acetate 156: mp 67-68°C (ether/hexanes); ir v  m a x  (CHCI3): 2950, 1750 cm" ; *H nmr 1  (400 M H z , CDC1 ) 6: 1.06 (3H, s, C2-CH ), 1.07 (3H, d, / = 7 Hz, C6-CH3), 1.16 3  3  (3H, s, C2-CH3), 1.48 (4H, m), 1.74 (2H, m), 2.10 (3H, s, OCOCH3), 2.12 (IH, m, C6-H) ppm; mass spectrum m/z: 209 (M ), +  194, 167 (100); high resolution mass  measurement: calculated for Cj^H^NC^: 209.1416; found: 209.1414; anal.: calculated for C H N 0 2 : C 68.87; H 9.15; N 6.69; found: C 69.05; H 9.09; N 6.69. 1 2  1 9  Single crystal X-ray structure data is given in Appendix 1.  135  Method C: A mixture of nitrile 128 and cyanohydrin 150 (30:70 by G C , derived as indicated in section 2.4.27 Method C, 1.0077 g), acetic anhydride (1.0 ml, 10 mmol) and acetyl bromide (70 |il, 0.95 mmol) was heated at 120°C for 8 h. The cooled reaction was added to iced water (50 ml) and extracted with diethyl ether (3 x 70 ml). The combined organic layers were washed with water (50 ml), saturated sodium bicarbonate solution (50 ml), dried (over MgSC<4), filtered and evaporated in vacuo to leave a red oil (1.4386 g). Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded pure nitrile 128 (289.3 mg, 29%) as a colourless oil. Further elution afforded pure 'cis'-acetate 156 (807.6 mg, 63%) as a white solid.  2.4.27. 2,6,6-Trimethylcyclohex-l-ene-l-carbonitrile (128) and 2-methyl-l-(l-methylethylene)-cyclopentane-l-carbonitrile (151).  128  151  Method A: To a solution of cyanohydrins 149 and 150 (23:77, 567.6 mg, 3.39 mmol) in anhydrous benzene (2.5 ml) at 0 ° C was added pyridine (700 |il, 8.65 mmol) and thionyl chloride (500 (il, 6.36 mmol). The reaction was heated at reflux for 1 h. The cooled reaction was poured into water (10 ml) and extracted with diethyl ether (4 x 30 ml). The combined organic layers were washed with 2 M hydrochloric acid (20 ml), saturated  136  sodium bicarbonate solution (10 ml), water (20 ml), dried (over MgSC<4), filtered and evaporated in vacuo to leave an oil (664.6 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:24) as eluent afforded pure nitrile 151 (82.9 mg, 16%), a mixture of nitriles 151 and 128 (81:19, 335.1 mg, 66%) and pure nitrile 128 (63.2 mg, 12%) as colourless oils. Nitrile 151: i r v  m a  x (thin film):  3000, 2910, 2260, 1650 cm" ; *H nmr (400 MHz, 1  CDC1 ) 8: 1.10 (3H, d, 7 = 7 Hz, C2-CH ), 1.58 (IH, m), 1.75 (IH, m), 1.79 (3H, 3  3  br s, wh/2 = 3 Hz, C1'-CH ), 1.95 (3H, m), 2.13 (2H, m), 5.02 (IH, br s, wh 3  Cl'-H), 5.24 (IH, s, Cl'-H) ppm;  1 3  /2  = 4 Hz,  C nmr (50 MHz, CDC1 ) 8: 15.0, 18.5, 21.7, 32.1, 3  37.2, 41.2, 55.4, 114.1, 121.6, 140.6 ppm; mass spectrum m/z: 149 (M+), 148, 134, 121, 107, 94 (100); high resolution mass measurement: 149.1204; found: 149.1206; anal.: calculated for C  1 0  H  1 5  calculated for C i o H  1 5  N:  N : C 80.48; H 10.13; N 9.39;  found: C 80.30; H 10.00; N 9.60. *H nmr decoupling experiment: irradiation of the doublet resonating at 5 1.10 ppm affected the multiplet resonating at 8 2.13 ppm. Nitrile 128: i r v  m a  x (thin film): 2930, 2200, 1635 cm" ; uv ^ 1  m a x  (MeOH):  214 nm  (e = 11400); *H nmr (400 MHz, CDC1 ) 8: 1.16 (6H, s, C6-(CH ) ), 1.46 (2H, m, 3  3  2  C5-H ), 1.61 (2H, m, C4-H ), 1.96 (3H, s, C2-CH ), 2.07 (2H, t, J = 6 Hz, C3-H ) 2  ppm;  1 3  2  3  2  C nmr (50 MHz, CDC1 ) 8: 19, 24, 28, 32, 34, 37, 116, 118, 152 ppm; mass 3  spectrum m/z: 149 (M ), 134 (100); high resolution mass measurement: calculated for +  C  1 0  H  1 5  N:  149.1204; found: 149.1204; anal.: calculated for C  1 0  H  1 5  N : C 80.48;  H 10.13; N 9.39; found: C 80.18; H 9.99; N 9.48. *H nmr decoupling experiment: irradiation of the triplet resonating at 8 2.07 ppm affected the multiplet resonating at 8 1.61 ppm.  137  Method B: To a solution of trimethylsilyl cyanohydrins 153 and 154 (12:88, 241.0 mg, 1.01 mmol) in anhydrous benzene (1 ml) and pyridine (1.4 ml) was added phosphorus oxychloride (300 jil, 3.21 mmol). The reaction was heated at reflux for 7 days. The cooled reaction was added to iced water (10 ml) and extracted with diethyl ether (3 x 30 ml). The combined organic layers were washed with 2M hydrochloric acid (20 ml), saturated sodium bicarbonate solution (10 ml), water (20 ml), dried (over MgS04), filtered and evaporated in vacuo to give a yellow oil (200.2 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:24) as eluent afforded pure 'trans'trimethylsilyl cyanohydrin 153  (20.6 mg, 9%).  Further elution afforded a mixture of  nitriles 151 and 128 (10:1, 61.3 mg, 41%). Further elution using ethyl acetate/petroleum ether (1:20) afforded a mixture of cyanohydrin 150 and ketone 142 (20.7 mg). 'trans'-Trimethyl cyanohydrin 153: C D C 1 ) 5: 3  i r v a x (thin film): 2940 cm" ; H nmr (400 MHz, m  0.25 (9H, s, Si-(CH ) ), 1.04 (3H, d, J = 6 Hz, 3  3  1  l  C6-CH3),  1.05 (3H, s,  C2-CH ), 1.14 (3H, s, C2-CH ), 1.33-1.58 (6H, m), 2.00 (IH, m, C6-H) ppm. 3  3  Method C: To a solution of cyanohydrins 149 and 150 (20:80, 6.371 g, 38.08 mmol) in anhydrous toluene (30 ml) at 0 ° C was added pyridine (8.0 ml, 98 mmol) and thionyl chloride (5.6 ml, 71 mmol). The reaction was stirred at 4 ° C for 17 h. The reaction was poured into water (200 ml) and extracted with diethyl ether (4 x 200 ml). The combined organic layers were washed with 2M hydrochloric acid (100 ml), water (100 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave an oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded a mixture of  138  nitrile 128 and 2,2,6-trimethylcyclohexanone (142) (1.3677 g). Further elution gave a mixture of 'cis'-cyanohydrin 150 and 2,2,6-trimethylcyclohexanone (142) (4.696 g). Repetition of this experiment (using 149:150 = 30:70) without purification afforded a mixture of 128 and 150 (30:70, 97%).  Method D: A mixture of acetates 155 and 156 (23:77, 318.0 mg, 1.52 mmol) was vacuum pyrolysed (furnace temperature 450°C, 11 Torr) to obtain a colourless oil that was dissolved in diethyl ether (100 ml), washed with water (50 ml), dried (over MgSCU), filtered and evaporated in vacuo to give a colourless oil (306.7 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded pure nitrile 128 (163.3 mg, 72%). Further elution afforded a mixture of acetates 155 and 156 (91:9, 89.1 mg, 28% recovery).  Method E: To a solution of 2,2,6 trimethylcyclohexanone (142) (60.00 g, 428.0 mmol) and powdered sodium cyanide (30.00 g, 612.0 mmol) in water (108 ml) was added a solution of sodium bisulphite (84 g, 800 mmol) in water (192 ml). The reaction was stirred for 48 h at room temperature, then diluted with water (500 ml) and diethyl ether (500 ml). The aqueous layer was extracted with diethyl ether (2 x 500 ml). The combined organic layers were dried (over MgS04), filtered and evaporated in vacuo to give an oil (71.05 g, 99%) that solidified on standing. G C and *H nmr analysis indicated a mixture of cyanohydrins 149:150 = 20:80. This solid was dissolved in anhydrous toluene (300 ml), cooled to 0 ° C , and then pyridine (87 ml, 1.1 mol) and thionyl chloride (61 ml, 780 mmol) was added. The reaction was stirred at 4 ° C for 3 days. The reaction was poured into  139  water (1 1) and extracted with diethyl ether (3 x 500 ml). The combined organic layers were washed with 2M hydrochloric acid (300 ml), water (300 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a yellow oil (70.52 g). This oil was gently heated at reflux with acetic anhydride (75 ml, 790 mmol) and acetyl bromide (4.5 ml, 61 mmol) for 12 h. The cooled reaction was poured into water (11) and extracted with diethyl ether (3 x 500 ml). The combined organic layers were washed with water (300 ml), saturated sodium bicarbonate solution (300 ml), dried (over MgSC^), filtered and evaporated in vacuo to leave a deep red oil (110 ml). This oil was flash pyrolysed (furnace temperature 525-550°C) in 20-25 ml batches. The fractions collected were combined, diluted with diethyl ether (1.5 1), washed with water (300 ml), saturated sodium bicarbonate solution (300 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave a deep red oil (65.8.g). Purification by distillation afforded nitrile 128 (bp 99-102°C/ll Torr, 51.95 g, 81% for 4 steps).  Method F: A solution of bromide 129 (1.088 g, 4.77 mmol) and sodium iodide (980 mg, 6.54 mmol) in anhydrous dimethylsulphoxide (100 ml) was stirred at room temperature for 15 min, at which time sodium borohydride (630 mg, 16.5 mmol) and methanol (600 |il, 14.8 mmol) were added. After stirring for a further 1.5 h, the reaction was quenched with water (50 ml) and extracted with diethyl ether (2 x 100 ml). The combined organic layers were dried (over MgSCH), filtered and evaporated in vacuo to give a colourless oil (932.6 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:49) as eluent afforded a mixture of nitriles 128 and 130 (22:3, 501.9 mg, 71%). Further elution afforded a pure sample of nitrile 128 (99.8 mg, 14%) as a colourless oil.  140  Nitrile 130:  l  nmr (400 M H z , CDC1 ) [partial signals] 8:  H  wh/2 = 7 H z ,  3  C 2 - C H 3 ) , 2.75 (IH,  s, Cl-H), 5.65 (IH,  1.82 (3H, br s,  br s, w h  /2  = 8 Hz,  C3-H) ppm.  2.4.28. (IR*, 2R*)  l-Acetyl-2-methyl-cycIopentane-l-carbonitrile (152).  A stream of ozone was bubbled into a solution of nitrile 151 (68.9 mg, 0.46 mmol) in ethyl acetate (7 ml) at -78°C until a persistent blue coloration was present. Dimethyl sulphide (300 |ll) was added and stirring continued for a further 10 min. The reaction was diluted with ethyl acetate (30 ml), washed with saturated sodium bicarbonate solution (30 ml), dried (over MgSCU), filtered and evaporated in vacuo to give a colourless oil. Purification by flash chromatography using diethyl ether/petroleum ether (1:24) as eluent afforded ketone 152 (50.9 mg, 73%) as a colourless oil: ir v  r a a x  (thin film): 2980, 2260,  1725 cm" ; H nmr (400 MHz, CDCI3) 5: 1.19 (3H, d, / = 8 Hz, C2-CH ), 1.55 (IH, 1  l  3  m, C2-H), 1.80 (IH, m), 1.98 (2H, m), 2.25 (2H, m), 2.40 (IH, m), 2.41 (3H, s, -COCH3) ppm;  1 3  C nmr (50 MHz, CDCI3) 8: 17, 23, 29, 34, 38, 43, 61, 120, 201 ppm;  mass spectrum m/z:  151 (M+), 136, 109, 94, 82, 43 (100); high resolution mass  measurement: calculated for C9H13NO: 151.0997; found: 151.0994; anal.: calculated for C9H13NO: C 71.49; H 8.66; N 9.26; found: C 71.30; H 8.68; N 9.10. *H nmr decoupling experiment: irradiation of the doublet resonating at 8 1.19 ppm affected the multiplet resonating at 8 1.55 ppm.  141  2.4.29. 2,6,6-Trimethylcyclohex-l-ene-l-carbaIdehyde  (35).  To a solution of nitrile 128 (500 mg, 3.4 mmol) in anhydrous benzene (3 ml) at 0 ° C was added D I B A L - H (1.5M in toluene, 3.2 ml, 4.8 mmol). The reaction was stirred for 1.5 h at 0 ° C after which a further portion of D I B A L - H (1.5M, 1.0 ml, 1.5 mmol) was added. The reaction was stirred at 0 ° C for a further 30 min, then added to deoxygenated 5% sulphuric acid (25 ml) along with diethyl ether (20 ml). After stirring vigorously for 15 min, the aqueous layer was extracted with diethyl ether (3 x 50 ml).  The combined  organic layers were washed with water (40 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave a yellow oil (546 mg). Purification by bulb-to-bulb distillation (100°C, 15 Torr) afforded pure P-cyclocitral (35) (461.1 mg, 90%) as a colourless oil: ir Vmax (CHCI3): 2950, 1665, 1610 cm" ; *H nmr (400 M H z , CDCI3) 5: 1.20 (6H, s, 1  C6-CH3), 1.45 (2H, m), 1.64 (2H, m), 2.11 (3H, s, C 2 - C H ) , 2.21 (2H, t, J = 7 Hz, 3  C 3 - H ) , 10.11 (IH, s, -CHO) ppm; 2  1 3  C nmr (50 MHz, CDCI3) 5:  18.6, 19.3, 21.4,  27.8, 33.0, 35.7, 40.5, 128.2, 129.0, 192.2 ppm; mass spectrum m/z: 137 (100).  152 (M+),  142  2.4.30. 2,6,6-Trimethylcyclohexa-l,3-diene-l-carbonitriIe  (127).  Method A: A solution of alcohol 126 (24.6956 g, 149.5 mmol) and para-toluenesulphonic acid (5.00 g, 26.3 mmol) was heated at reflux with azeotropic removal of water for 30 h. The reaction was cooled, diluted with diethyl ether (11), washed with water (2 x 200 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a red oil (22.6472 g). Purification by distillation (100-102°C, 11 Torr) gave safronitrile (127) (16.4631 g, 75%) as a colourless liquid: ir Vmax (thin film): 2950, 2180, 1640, 1575 cm ; uv X.max -1  (MeOH): 285 nm (e = 7200); H nmr (400 MHz, CDCI3) 5: 1.17 (6H, s, C6-CH3), !  2.08 (3H, s, C2-CH3), 2.22 (2H, dd, / = 4, 2 Hz, C5-H ), 5.93 (IH, dt, J = 9, 2  2 Hz, C3-H), 6.07 (IH, dt, 7 = 9, 4 Hz, C4-H) ppm;  1 3  C nmr (50 MHz, CDC1 ) 3  5: 21.8, 27.7, 32.5, 37.7, 114.5, 118.2, 127.2, 131.9, 146.3 ppm; mass spectrum m/z: 147(M ), 132; high resolution mass measurement: calculated for C ^ H ^ N : 147.1048; +  found: 147.1046; anal.: calculated for C i o H N : 1 3  C 81.59; H 8.90; N 9.51; found:  C 81.76; H 8.83; N 9.49.  Method B: A solution of bromide 129 (151.6 mg, 0.664 mmol) and DBU (300 2.0 mmol) in benzene (10 ml) was heated at reflux for 4 h. After cooling, the reaction was diluted with diethyl ether (30 ml), washed with saturated ammonium chloride solution (10 ml), water (10 ml), dried (over MgS04), filtered and evaporated in vacuo to leave an oil  143  (153.2 mg). Purification by flash chromatography using diethyl ether/petroleum ether (1:9) as eluent gave 127 (26.5 mg, 27%) as a colourless oil.  2.4.31. 2,6,6-TrimethylcycIohexa-l,3-diene-l-carbaIdehyde  (134).  CHO  To a solution of safronitrile 127 (580.0 mg, 3.94 mmol) in anhydrous benzene (3.5 ml) at 0°C was added DIBAL-H (1.5M in toluene, 4.9 ml, 7.4 mmol). The reaction was stirred for 2.5 h at room temperature then added to deoxygenated 5% sulphuric acid (30 ml) along with diethyl ether (20 ml). After stirring vigorously for 15 min, the aqueous layer was extracted with diethyl ether (3 x 50 ml). The combined organic layers were washed with water (50 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a orange oil (543 mg). Purification by bulb-to-bulb distillation (100-105°C, 11 Torr) afforded pure safronal (134) (371.6 mg, 63%) as a colourless oil: ir v x (thin film): 2960, 2930, 2870, ma  1665, 1635 cm- ; uv ? i x (MeOH): 309 nm (e = 7500); *H nmr (400 MHz, CDC1 ) 1  3  m a  6: 1.20 (6H, s, C6-CH3), 2.16 (2H, dd, J = 4, 2 Hz, C5-H ), 2.18 (3H, s, C2-CH ), 2  3  5.93 (IH, dt, / = 10, 2 Hz, C3-H), 6.17 (IH, dt, / = 10, 4 Hz, C4-H), 10.15 (IH, s, CHO) ppm; mass spectrum m/z: 150 (M ), 135, 121, 107 (100); high resolution mass +  measurement: calculated for QrjH^O: 150.1045; found: 150.1045; anal.: calculated for C H O: 1 0  1 4  C 79.96; H 9.39; found: C 80.18; H 9.42.  144  2.4.32. l-(2,6,6-Trimethylcyclohexa-l,3-dienyl)-but-3-en-l-ol  (135).  OH  To a solution of safronal (134) (245.8 mg, 1.64 mmol) in anhydrous tetrahydrofuran (12 ml) at 0°C was added allylmagnesium bromide (IM in diethyl ether, 2 ml, 2 mmol). The reaction was stirred at 0°C for 2 h, quenched with saturated ammonium chloride solution (15 ml) and extracted with diethyl ether (2 x 50 ml). The combined organic layers were washed with water (30 ml), dried (over MgS04),filteredand evaporated in vacuo to leave a yellow oil (311.9 mg). Purification by bulb-to-bulb distillation (125-130°C, 11 Torr) afforded alcohol 135 (287.3 mg, 91%) as a colourless oil: ir v ax (thin film): 3425 m  (br), 2910, 1640 cm" ; uv X  (MeOH): 266 nm (e = 7200); *H nmr (400 MHz,  1  max  CDCI3) 8: 1.07 (3H, s, C6'-CH ), 1.11 (3H, s, C6'-CH ), 1.60 (IH, br s, wh = 8 Hz, 3  exchangeable,  3  /2  O-H), 1.91 (3H, s, C2'-CH ), 2.05 (2H, m, C5'-H ), 2.41 (IH, m, 3  2  C2-H), 2.63 (IH, dtt, / = 15, 9, 1 Hz, C2-H), 4.59 (IH, dd, J = 9, 4 Hz, Cl-H), 5.15 (IH, dq, / = 10, 1 Hz, C4-U ), cis  5.19 (IH, dq, / = 18, 1 Hz, C4-H rans), t  (2H, m, C3'-H, C4'-H), 5.90 (IH, dddd, J = 18, 10, 9, 6 Hz, C3-H) ppm;  1 3  5.73  C nmr (50  MHz, CDC1 ) 8: 19.2, 26.1, 27.0, 34.3, 40.6, 41.3, 70.3, 117.5, 124.8, 127.4, 130.3, 3  135.8, 139.7 ppm; mass spectrum m/z:  192 (M+), 177, 174, 159, 151 (100); high  resolution mass measurement: calculated for Ci H2oO: 192.1514; found: 192.1517. 3  *H nmr decoupling experiments: irradiation of the doublet of doublets resonating at 8 4.59 ppm simplified the signals resonating at 8 2.41 and 2.63 ppm; simultaneous irradiation of the signals resonating at 8 5.15 and 5.19 ppm simplified the signal resonating at 8 5.90;  145  irradiation of the signal resonating at 5 5.90 ppm affected the signals resonating at 8 5.15 and 5.19 ppm, simplified the signal resonating at 8 2.63 ppm to a doublet of doublet of triplets (/ = 15, 9, 1 Hz) and simplified the signal resonating at 8 2.41 ppm to a doublet of doublet of triplets (J = 15, 6, 1 Hz); irradiation of the multiplet resonating at 8 2.05 ppm simplified the signal resonating at 8 5.73 ppm to an AB quartet.  2.4.33. (E)-l-(2,6,6-TrimethyIcyclohexa-l,3-dienyl)-but-2-en-l-ol (139) and (Z)-l-(2,6,6-trimethylcyclohexa-l,3-dienyl)-but-2-en-l-ol (140).  139  140  To a solution of safronal (134) (256.7 mg, 1.71 mmol) in anhydrous tetrahydrofuran (12 ml) at 0°C was added 1-propenylmagnesium bromide (1.21M in tetrahydrofuran, 36  2.0 ml, 2.4 mmol). The reaction was stirred at 0°C for 10 min, quenched with saturated ammonium chloride solution (15 ml) and extracted with diethyl ether (2 x 30 ml). The combined organic layers were washed with water (50 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a yellow oil (344.2 mg). Purification by bulb-to-bulb distillation (115-125°C, 11 Torr) afforded a mixture of alcohols 139 and 140 (49:51, 293.9 mg, 89%) as a colourless oil: ir v  (thin film): 3400 (br), 2960, 1680 (br) cm" ; 1  m a x  *H nmr (400 MHz, CDCI3) [partial signals corresponding to 140] 8: 1.06 (3H, s,  Prepared as a mixture of isomers 49:51 by the method described in reference 32.  146  C6'-CH ), 1.11 (3H, s, C6'-CH ), 1.40 (IH, br s, wh/ = 8 Hz, exchangeable, O-H), 3  3  2  1.74 (3H, dd, / = 8, 1 Hz, C4-H ), 1.95 (3H, s, C2'-CH ), 2.07 (2H, t, / = 3 Hz, 3  3  C5'-H ), 5.40 (IH, d, J = 9 Hz, Cl-H), 5.58 (IH, dqd, J = 11, 8, 1 Hz, C3-H), 5.74 2  (2H, m, C3'-H, C4'-H), 5.83 (IH, ddq, J = 11, 9, 1 Hz, C2-H) ppm. Purification by flash chromatography of a portion of this oil using ethyl acetate/petroleum ether (1:19) as eluent afforded pure trans-alcohol 139 (79.2 mg, 43% overall) as a colourless oil: ir v (thin film): 3400 (br), 2960, 1680 (br) cnr*; uv ? i  m a x  m a x  (MeOH): 271 nm (e = 9040); *H  nmr (400 MHz, CDC1 ) 5: 1.04 (3H, s, C6-CH3), 1.10 (3H, s, C6-CH3), 1.53 (IH, br 3  s, wh/2 = 6 Hz, exchangeable, O-H), 1.72 (3H, dd, J = 7, 1 Hz, C4-H ), 1.88 (3H, s, 3  C2'-CH ), 2.06 (2H, t, J = 4 Hz, C5'-H ), 5.03 (IH, br d, / = 5 Hz, Cl-H), 5.73 3  2  (4H, m, C3-H, C4'-H, C2-H, C3-H) ppm; mass spectrum m/z: 192 (M+), 177, 174, 159, 121 (100); high resolution mass measurement: calculated for C H 2 Q 0 : 192.1514; 13  found: 192.1516. *H nmr decoupling experiments: irradiation of the doublet of doublets resonating at 8 1.74 ppm simplified the signal resonating at 8 5.58 ppm to a doublet of doublets (/ = 11, 1 Hz) and simplified the signal resonating at 8 5.83 ppm to a doublet of doublets (/ = 11,9 Hz); irradiation of the doublet resonating at 8 5.40 ppm simplified the signal resonating at 8 5.83 ppm to a doublet of quartets (/ = 11, 1 Hz) and affected the signal resonating at 8 5.58 ppm. Theris-alcohol140 decomposed on chromatography.  147  2.4.34.  (E)-l-(2,6,6-Trimethylcyclohexa-l,3-dienyl)but-2-en-l-one (P-Damascenone)  24  (24).  141  Method A: To a solution of rrarcs-alcohol 139 (18.9 mg, 0.098 mmol) in petroleum ether (3 ml) was added 'active' manganese dioxide (170 mg, 1.95 mmol). After stirring for 1 h the reaction was filtered and evaporated in vacuo to afford p-damascenone (24) (16.7 mg, 100%) as a colourless oil: ir v  m a x  (thin film): 2980, 2950, 1670, 1635, 1610 cm" ; uv 1  ^max (MeOH): 305 nm (e = 2220), 226 nm (e = 13800); *H nmr (400 MHz, CDC1 ) 5: 3  1.05 (6H, s, C6'-(CH )2), 1.64 (3H, s, C2'-CH ), 1.94 (3H, dd, J = l, 2 Hz, C4-H ), 3  3  3  2.12 (2H, dd, 7 = 5, 1 Hz, C5'-H ), 5.84 (2H, m, C3'-H, C4'-H), 6.20 (IH, dq, 2  / = 16, 2 Hz, C2-H), 6.85 (IH, dq, J = 16, 7 Hz, C3-H) ppm; mass spectrum m/z: 190 (M ), 175, 159, 121, 69 (100); high resolution mass measurement: calculated for +  C i H i 0 : 190.1357; found: 190.1355. 3  8  *H nmr decoupling experiments: irradiation of the doublet of quartets resonating at 8 6.20 ppm simplified the signals resonating at 8 1.94 ppm to a doublet (/ = 7 Hz) and simplified the signal resonating at 8 6.85 ppm; irradiation of the signal resonating at 8 6.85 ppm simplified the doublet of doublets resonating at 8 1.94 ppm to a doublet (/ = 2 Hz) and affected the signal resonating at 8 6.20 ppm; irradiation of the doublet of doublets resonating at 8 2.12 ppm simplified the signal resonating at 8 5.84 ppm to an A B quartet;  148  irradiation of the signal resonating at 5 1.94 ppm simplified the signal resonating at 8 6.20 ppm to a doublet (/ = 16 Hz) and simplified the signal resonating at 8 6.85 ppm to a doublet (/ = 16 Hz).  Method B: To a solution of alcohols 1 3 9 and 1 4 0 (49:51, 139.4 mg, 0.725 mmol) in petroleum ether (30 ml) was added 'active' manganese dioxide (700 mg, 8.05 mmol). After stirring for 2 h, a further portion of manganese dioxide (700 mg, 8.05 mmol) was added. After stirring for 1.5 h, the reaction was filtered and evaporated in vacuo to afford a mixture of P-damascenone  (24)  and enone  141 (24:141  = 49:51, 138.7 mg, 100%) as a  colourless oil. *H nmr (400 MHz, CDCI3) [signals corresponding to enone 1 4 1 ] 8: 1.10 (6H, s, C6'-(CH ) ), 1.74 (3H, s, C2'-CH ), 2.11 (2H, dd, / = 5, 1 Hz, C5'-H ), 3  2  3  2  2.16 (3H, d, J = l Hz, C4-H ), 5.84 (2H, m, C3'-H, C4'-H), 6.20 (2H, m, C2-H, 3  C3-H) ppm. The mixture of ketones and para-toluenesulphonic acid (20 mg, 0.1 mmol) in anhydrous tetrahydrofuran was heated at reflux for 20 h. The reaction was cooled and diluted with diethyl ether (150 ml). The organic layer was washed with water (30 ml), saturated sodium bicarbonate solution (50 ml), water (30 ml), dried (over MgSCU), filtered and and evaporated in vacuo to give a colourless oil (136.2 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded pure P-damascenone ( 2 4 ) (105.0mg, 76%) as a colourless oil.  149  2.4.35. 2,4,4-Trimethylcyclohex-2-en-l-oI-3-carboxaldehyde  (160).  CHO  OH  To a solution of nitrile 126 (562.8 mg, 3.41 mmol) in anhydrous toluene (30 ml) at -60°C was added DIBAL-H (1.5 M in toluene, 4 ml, 6 mmol). After stirring for 2 h, the reaction was quenched by the addition of saturated ammonium chloride solution (35 ml). After extraction with diethyl ether (5 x 100 ml), the combined organic layers were dried (over MgS04), filtered and evaporated in vacuo to give a yellow oil (493 mg). Purification by passage through a short column containing Florisil® using diethyl ether as eluent afforded aldehyde 160 (454.2 mg, 80%) as a slight yellow oil: ir v 3400 (br), 2945, 1675, 1605 cm" ; uv X  J  max  CDCI3) 8:  (thin film):  ax  (MeOH): 245 nm (e = 12200); H nmr (400  1  MHz,  m  1.15 (3H, s, C4-CH ), 1.21 (3H, s, C4-CH ), 1.40 (IH, ddd, 3  3  J= 14,9, 3 Hz, C5-H), 1.56 (IH, ddd, / = 14, 9.5, 3 Hz, C5-H), 1.70 (IH, dddd, / = 14, 9, 6, 3 Hz, C6-H), 1.92 (IH, dddd, / = 14, 9.5, 6, 3 Hz, C6-H), 2.07 (3H, s, C2-CH ), 4.08 (IH, t,/ = 6Hz, Cl-H), 10.13 (IH, s, CHO) ppm; mass spectrum m/z: 3  168 (M ), 153, 150, 139 (100); high resolution mass measurement: +  calculated for  C i o H i 0 : 168.1150; found: 168.1153. 6  2  *H nmr decoupling experiment:  Irradiation of the triplet resonating at 8 4.08 ppm  simplified the signal resonating at 8 1.92 ppm to a doublet of doublet of doublets (J = 14, 9.5, 3 Hz) and simplified the signal resonating at 8 1.70 ppm to a doublet of doublet of doublets (J = 14, 9, 3 Hz).  150  2.4.36.  (3-Hydroxy-2,6,6-trimethylcycIohex-l-enyl)but-2-en-l-ol (161), (162) and (164).  OH  OH  161  OH  162  164  To a solution of aldehyde 160 (114.0 mg, 0.678 mmol) in anhydrous tetrahydrofuran (5 ml) at 0 ° C was added 1-propenylmagnesium bromide (1.21M in tetrahydrofuran, 1.6 ml, 1.9 mmol). The reaction was stirred at 0 ° C for 5 min, then warmed to room temperature and stirred at this temperature for 30 min. The reaction was quenched with saturated ammonium chloride solution (20 ml) and extracted with diethyl ether (3 x 30 ml). The combined organic layers were washed with water (50 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a yellow oil (215.8 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent afforded in order of elution: diol 161 (33.0 mg, 23%) as a white solid; a mixture of diols 161 and 162 (9:11, 16.6 mg, 12%); diol 162 (29.3 mg, 21%) as a colourless oil; diol 164 as a colourless oil. Diol 161: i r v m C D C 1 ) 8: 3  a x  (thin film): 3400 (br), 2960, 1680 (br) cm" ; *H nmr (400 MHz, 1  1.00 (3H, s, C6'-CH ), 1.10 (3H, s, C6'-CH ), 1.38 (IH, ddd, J = 14, 6, 3  3  3 Hz), 1.50-1.70 (4H, m, 2 exchangeable H), 1.70 (3H, dd, J = 7 Hz, C4-H ), 1.83 3  (IH, tt, / = 14, 5 Hz, C4'-H), 1.94 (3H, s, C2'-CH ), 3.90 (IH, t, / = 4 Hz, C3'-H), 3  4.79 (IH, d, / = 5 H z , Cl-H), 5.71 (IH, m, C3-H), 5.79 (IH, dd, / = 16, 5 Hz,  151  C2-H) ppm; mass spectrum m/z: 210 (M ), 192, 177; high resolution mass measurement: +  calculated for C13H22O2: 210.1620; found: 210.1615. *H nmr decoupling experiment: irradiation of the doublet resonating at 8 4.79 ppm simplified the doublet of doublets resonating at 8 5.79 ppm to a doublet (/ = 16 Hz); irradiation of the triplet resonating at 8 3.90 ppm simplified the signal resonating at 8 1.83 ppm to a triplet of doublets (/ = 14, 5 Hz) and affected the multiplet resonating at 8 1.501.70 ppm. Diol 162: i r v  x (thin film): 3400 (br), 2960, 1680 (br) cm" ; H nmr (400 MHz, 1  m a  J  CDCI3) 8: 0.97 (3H, s, C6'-CH ), 1.14 (3H, s, C6'-CH ), 1.39 (IH, ddd, / = 14, 6, 3  3  3 Hz, C5'-H), 1.50 (2H, br s, wh/ = 56 Hz, exchangeable, OH), 1.63 (lH,m), 1.72 2  (IH, m), 1.80 (3H, dd, / = 7, 2 Hz, C4-H ), 1.87 (IH, m), 2.02 (3H, s, C2'-CH ), 3  3  3.92 (IH, t, / = 4 H z , C3'-H), 5.14 (IH, d, / = 8 Hz, Cl-H), 5.64 (IH, dqd, / = 10, 7, 1 Hz, C3-H), 5.82 (IH, ddq, J = 10, 8, 2 Hz, C2-H) ppm; mass spectrum m/z: 210 (M+), 192, 177. Diol 164: i r v  (thin film): 3400 (br), 2960, 1680 (br) cm" ; *H nmr (400 MHz, 1  m a x  CDC1 )*8: 0.92 (3H, s, C6'-CH ), 1.20 (3H, s, C6'-CH ), 1.39 (IH, m), 1.50-1.70 3  3  3  (4H, m, 2 exchangeable H), 1.79 (3H, dd, / = 7, 1 Hz, C4-H ), 1.85 (IH, m), 1.98 3  (3H, s, C2'-CH ), 3.96 (IH, t, / = 6 Hz, C3'-H), 5.13 (IH, d, / = 9 Hz, Cl-H), 3  5.63 (IH, dq, 7 = 9, 7 Hz, C3-H), 5.82 (IH, tq, / = 9, 1 Hz, C2-H) ppm; mass spectrum m/z: 210 (M+), 192, 177.  152  2.4.37. (E)-l-(3-Oxo-2,6,6-trimethylcyclohex-l-enyl)but-2-en-l-one (158) and  (Z)-l-(3-oxo-2,6,6-trimethyIcyclohex-l-enyl)-but-2-en-l-ol  (165).  158  To a solution of aldehyde 160 (282.0 mg, 1.68 mmol) in anhydrous tetrahydrofuran (12 ml) at 0 ° C was added 1-propenylmagnesium bromide (1.2M in tetrahydrofuran, 4.0 ml, 4.8 mmol). The reaction was stirred at 0 ° C for 5 min, then warmed to room temperature and stirred at this temperature for 60 min. The reaction was quenched with saturated ammonium chloride solution (30 ml) and extracted with diethyl ether (3 x 50 ml). The combined organic layers were washed with water (50 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a yellow oil (398.6 mg). A solution of this oil and 'active' manganese dioxide (2.92 g, 33.6 mmol) in petroleum ether/dichloromethane (1:1, 60 ml) was stirred for 48 h at room temperature. The reaction was filtered and evaporated in vacuo to leave a yellow oil (361.3 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:4) as eluent afforded diketone 158 (81.0 mg, 23%) as a colourless oil. Further elution afforded keto alcohol 165 (53.2 mg, 15%) as a colourless oil. Diketone 158: i r v  m a x  (thin film) 2950, 1670, 1635 cm" ; 1  nmr (400 MHz, CDC1 ) 5: 3  1.15 (6H, s, C6'-CH ), 1.58 (3H, s, C2'-CH ), 1.90 (2H, t, / = 6.5 Hz, C5'-H ), 3  3  2  153  1.94 (3H, dd, J = 1,2 Hz, C4-H ), 2.52 (2H, t, J = 6.5 Hz, C4'-H ), 6.15 (IH, dq, 3  2  / = 16, 2 Hz, C2-H), 6.71 (IH, dq, / = 16, 7 Hz, C3-H) ppm; mass spectrum m/z: 206 (M ), +  191; high resolution mass measurement: calculated for C i H i 8 0 : 206.1307; 3  2  found: 206.1304. Keto alcohol 165: CDC1 )8: 3  ir v  m a x  (thin film) 3400, 2950, 1670, 1635 cm" ; *H nmr (400 MHz, 1  1.11 (3H, s, C6'-CH ), 1.32 (3H, s, C6'-CH ), 1.81 (5H, m, C4-H , 3  C5'-H ), 1.90 (IH, br s, wh 2  /2  3  3  = 20 Hz), 1.95 (3H, s, C2'-CH ), 2.46 (2H, m, C4'-H ), 3  2  5.31 (IH, d, / = 8 H z , Cl-H), 5.74 (2H, m, C2-H, C3-H) ppm; mass spectrum m/z: 208 (M ), +  193, 179, 165; high resolution mass measurement: calculated for C i H o 0 : 3  2  2  208.1463; found: 208.1456. *H nmr decoupling experiment: irradiation of the doublet resonating at 5 5.31 ppm affected the multiplet resonating at 8 5.74 ppm.  2.4.38. (3-Hydroxy-2,6,6-trimethylcyclohex-l-enyl)but-3-en-l-ol (166) and (167).  OH 166  To a solution of aldehyde 160  OH 167  (78.5 mg, 0.467 mmol) in anhydrous diethyl ether  (5 ml) at -20°C was added allylmagnesium bromide (1.0M in diethyl ether, 1.0 ml, 1.0 mmol). The reaction was stirred at 0 ° C for 5 min, then quenched with water (20 ml)  154  and extracted with diethyl ether (2 x 50 ml). The combined organic layers were dried (over MgSC»4), filtered and evaporated in vacuo to leave a yellow oil (75.1 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent afforded diol 166 (31.0 mg, 32%) as a white solid. Further elution afforded diol 167 (21.0 mg, 21%) as a white solid. Diol 166: CDCI3) 5:  irv  m a x  (CHCI3): 3600, 3450 (br), 2925, 1640 cm" ; H nmr (400 MHz, 1  l  1.02 (3H, s, C6-CH3), 1.10 (3H, s, C6-CH3), 1.38 (IH, ddd, / = 14, 8,  4 Hz, C5'-H), 1.53-1.73 (4H, m, 2 exchangeable H), 1.86 (IH, m, C4'-H), 2.03 (3H, s, C 2 ' - C H ) , 2.36 (IH, m, C2-H), 2.66 (IH, dddt, 7 = 14, 11, 7, 1 Hz, C2-H), 3.92 3  (IH, t, / = 4 H z , C3'-H), 4.34 (IH,  dd, 7 = 11, 3 Hz, Cl-H), 5.16 (IH,  7 = 9 Hz, C4-H fr), 5.19 (IH, br d, 7 = 13 Hz, C4-H ), C  mass spectrum m/z:  trans  210 (M ),  br d,  5.91 (IH, m, C3-H) ppm;  192, 177, 169, 123 (100); high resolution mass  +  measurement: calculated for C13H22O2: 210.1620; found: 210.1616. ! H nmr decoupling experiment: irradiation of the doublet of doublets resonating at 8 4.34 ppm simplified the signal resonating at 8 2.36 ppm to a doublet of doublet of triplets (7 = 14, 6, 1 Hz) and simplified the signal resonating at 8 2.66 ppm to a doublet of doublet of triplets (/ = 14, 7, 1 Hz); irradiation of the triplet resonating at 8 3.92 ppm affected the signal resonating at 8 1.83 ppm and affected the multiplet resonating at 8 1.53-1.73 ppm. Diol 167:  mp 120-121°C; ir v ax (CHCI3): 3700, 3250 (br), 2940, 1645 cm" ; ! H nmr  (400 M H z , CDCI3) 8:  1  m  1.00 (3H, s, C6'-CH ), 1.18 (3H, s, C6-CH3), 1.40 (IH, ddd, 3  7 = 13.5, 8, 3 Hz, C5'-H), 1.50-1.73 (4H, m, 2 exchangeable H), 1.88 (IH,  m,  C4'-H), 2.00 (3H, s, C2'-CH ), 2.30 (IH, m, C2-H), 2.68 (IH, dddt, / = 14, 10, 7, 3  1 H z , C2-H), 3.94 (IH, t, 7 = 5 Hz, C3'-H), 4.37 (IH, dd, 7 = 11, 3 Hz, Cl-H), 5.16 (IH, dq, 7 = 9, 1 Hz, C4-H ), cis  5.20 (IH, dq, 7 = 1 6 , 1 Hz, C4-K ), trans  5.90  (IH, dddd, 7 = 16, 9, 7, 6 Hz, C3-H) ppm; mass spectrum m/z: 210 (M+), 192, 177,  155  169, 123 (100); high resolution mass measurement: calculated for  210.1620;  C13H22O2:  found: 210.1627.  2.4.39. l-/-Butyldimethylsiloxy-2,4,4-trimethylcyclohex-2-ene -3-carbonitrile (168).  OTBDMS  A solution of alcohol 126 (177.2 mg, 1.07 mmol), D M A P (55.0 mg, 0.45 mmol), r-butyldimethylsilyl chloride (185.0 mg, 1.23 mmol) and triethylamine (165 (il, 1.18 mmol) in anhydrous dichloromethane (5 ml) was heated at reflux for 12 h. After cooling, the reaction was diluted with dichloromethane (50 ml), washed with water (2 x 50 ml), dried (over MgS04), filtered and evaporated in vacuo to give an oil (354 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent afforded silyl ether 168 (234.6 mg, 78%) as a white solid: mp 35-36°C; ir Vmax (CHCI3): 2960, 2859, 2216 cm" ; uv X 1  5:  max  (MeOH) 216 nm (e = 1200); *H nmr (400 MHz, CDCI3)  0.10 (3H, s), 0.11 (3H, s), 0.91 (9H, s), 1.15 (3H, s, C4-CH ), 1.19 (3H, s, 3  C4-CH ), 1.47 (IH, m), 1.68 (2H, m), 1.83 (IH, m), 2.04 (3H, s, C2-CH ), 4.05 (IH, 3  3  t, / = 5 Hz, Cl-H) ppm; mass spectrum m/z: 279 (M+), 264, 22 (100); high resolution mass measurement: calculated for Ci6H29N0Si: 279.2019; found: 279.2020.  156  2.4.40.  l-/-ButyIdimethylsiloxy-2,4,4-trimethylcyclohex-2-ene -3-carboxaldehyde (169).  OTBDMS  To a solution of 168 (157.7 mg, 0.564 mmol) in anhydrous toluene (10 ml) at -60°C was added DIBAL-H (1.5M in toluene, 0.4 ml, 0.6 mmol). After stirring for 15 min, the reaction was quenched by the addition of saturated ammonium chloride solution (15 ml) and extracted with diethyl ether (4 x 30 ml). The combined organic layers were washed with water (30 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave an oil (160.3 mg) that was shown by G C and *H nmr analysis to contain 168 (47%) and 169 (53%): H nmr (400 MHz, CDCI3) [signals corresponding to 169] 8: 0.12 (3H, s), 0.13 l  (3H, s), 0.91 (9H, s), 1.18 (3H, s, C4-CH ), 1.24 (3H, s, C4-CH ), 1.40 (IH, m), 1.60 3  3  (IH, m), 1.83 (2H, m), 2.13 (3H, s, C2-CH ), 4.10 (IH, t, / = 5 Hz, Cl-H), 10.13 3  (IH, s, CHO) ppm.  157  2.4.41. 5-Bromo-3-(l-methylethyl)-2,6,6-trimethylcyclohex-2-en-l-one (175) and 4-methyIene-3-(l-methyIethyl)-2,5,5-trimethylcycIopent-2-en-l-one (176) and 5,6-dibromo-3-(l-methylethyl)-2,5,6-trimethylcycIohex-2-en-l-one (177).  175  176  177  To a solution of dimethylated thujone 59 (1.013 g, 5.62 mmol) in petroleum ether (6 ml) was added dropwise bromine (1.0 ml, 19 mmol). The reaction was stirred for 30 min, then diluted with saturated sodium thiosulphate solution (50 ml) and petroleum ether (50 ml). The organic layer was washed with water (50 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a red oil (1.51 g). A portion of this oil (619.3 mg) was purified by chromatography on neutral alumina using diethyl ether/petroleum ether (1:24) as eluent to give bromo enone 175 (79.0 mg, 13%) as a pale yellow oil containing a minor unidentified impurity. Further elution afforded a mixture of 175 and 176 (3:2, 70.0 mg) and a sample of 176 (108.0 mg, 11%) as a pale yellow oil. Further elution using ethyl acetate/petroleum ether (1:19) gave an oil (98 mg, 5%) tentatively assigned as 177.  158  Bromo enone 175:  irv  37  m a x  (thin film): 2975, 1670, 1625 cm- ; uv X 1  max  (MeOH):  246 nm (e = 7030); H nmr (400 MHz, CDC1 ) 5: 1.05 (3H, d, / = 7 Hz, C7-CH ), l  3  3  1.06 (3H, d, 7 = 7 Hz, C7-CH ), 1.15 (3H, s, C6-CH3), 1.26 (3H, s, C6-CH ), 1.80 3  3  (3H, t, / = 1 H z , C 2 - C H ) , 2.76 (IH, ddq, J = 18, 10, 1 Hz, C4-H), 2.88 (IH, 3  brdd, 7 = 1 8 , 5 Hz, C4-H), 3.00 (IH, septet, / = 7 Hz, C7-H), 4.33 (IH, dd, / = 10, 5 Hz, C5-H) ppm; mass spectrum (CI) m/z: 278/276 (M+NH4+), 261/259. *H nmr decoupling experiments: irradiation of the signal resonating at 5 4.33 ppm simplified the signal resonating at 8 2.76 ppm to a broad doublet (/ = 18 Hz) and simplified the signal resonating at 8 2.88 ppm to a broad doublet (/ = 18 Hz); irradiation of the signal resonating at 8 1.80 ppm simplified the signal resonating at 8 2.76 ppm to a doublet of doublets (/ = 18, 10 Hz) and simplified the signal resonating at 8 2.88 ppm to a doublet of doublets (/= 18, 5 Hz). Dienone 176: uv X  (MeOH): 276 nm (e = 4840); *H nmr (400 MHz, CDC1 ) 8: 3  max  1.11 (6H, s, C5-(CH ) ), 1.29 (6H, d, / = 7 Hz, C6-(CH ) ), 1.89 (3H, s, C2-CH ), 3  2  3  2  3  3.11 (IH, septet, / = 7 Hz, C6-H), 5.09 (IH, s, C9-H), 5.36 (IH, s, C9-H) ppm. *H nmr decoupling experiment: irradiation of the septet resonating at 8 3.11 ppm collapsed the doublet resonating at 8 1.29 ppm to a singlet. Dibromide 177: ir v  m a x  (thin film): 2975, 1685, 1630 cm- ; uv k 1  m a x  (MeOH): 264 nm  (e = 38,000); H nmr (400 MHz, CDC1 ) 8: 1.12 (6H, d, / = 7 Hz, C7-(CH ) ), 1.75 l  3  3  2  (3H, t, / = 1 Hz, C2-CH ), 1.90 (3H, s), 2.33 (3H, s), 3.00 (2H, br s, wh/ = 7 Hz, 3  C 4 - H ) , 3.05 (IH, septet, / = 7 Hz, C7-H) ppm; mass spectrum m/z: 2  (M -HBr), 170 (100). +  Minor unidentified impurity present.  2  258/256  159  2.4.42. (IS, 5S)  2-Bromo-l-(l-methylethyl)-2,4,4-trimethyl bicycIo[3.1.0]hexan-3-one (180).  To a solution of dimethylthujone 59 (599.3 mg, 3.32 mmol) in petroleum ether (15 ml) at 0 ° C was added dropwise bromine (200 |il, 4 mmol). The reaction was stirred for 30 min, then diluted with saturated sodium thiosulphate solution (10 ml) and diethyl ether (100 ml). The organic layer was washed with water (100 ml), dried (over MgSCU), filtered and evaporated in vacuo to give 180 (835.7 mg, 97%) as a pale red oil: H nmr l  (200  M H z , C D C 1 ) 8: 3  -0.20 (IH, dd, / = 8, 5 Hz), 0.76 (3H, d, / = 7 H z ,  C7-CH ), 0.87 (IH, m), 1.05 (3H, s, C4-CH ), 1.08 (3H, d, / = 7 Hz, C7-CH ), 1.51 3  3  3  (3H, s, C4-CH ), 1.54 (IH, dd, / = 8, 4 Hz), 1.78 (3H, s, C2-CH ), 2.68 (IH, septet, 3  3  / = 7 Hz, C7-H) ppm; mass spectrum m/z: 260/258 (M ), 179 (100). +  .  2.4.43. 3-(l-Methylethyl)-2,6,6-trimethylcycIohexa-2,4-dien-l-one (178).  160  A solution of bromo enone 175 (139.5 mg, 0.538 mmol) and potassium hydroxide (200 mg, 3.57 mmol) in ethanol (20 ml) was stirred at room temperature for 30 min. The solvent was evaporated in vacuo and the residue dissolved in diethyl ether (40 ml). The organic solution was washed with 2M hydrochloric acid (20 ml), water (20 ml), dried (over MgSC>4), filtered and evaporated in vacuo to give diene 178 (89.3 mg, 93%) as a colourless oil: U nmr (200 MHz, CDC1 ) 5: 1.04 (6H, d, / = 7 Hz, C7-(CH ) ), 1.11 l  3  3  2  (6H, s, C6-(CH ) ), 1.81 (3H, s, C2-CH ), 2.97 (IH, septet, / = 7 Hz, C7-H), 6.12 3  2  3  (2H, A B quartet, / = 9 Hz, C4-H, C5-H) ppm.  2.4.44. 3-(l-Methylethyl)-2,6,6-trimethylcyclohex-2-en-l-one  (179)  and 3-(l-methylethyl)-2,4,5,5-tetramethylcycIopent-2-en-l-one (181).  179  181  To a solution of dimethylated thujone 59 (111.5 mg, 0.618 mmol) in petroleum ether (3 ml) at 0 ° C was added bromine (42 |il, 0.81 mmol) dropwise. The reaction was stirred for 45 min, then 48% hydrobromic acid (3 ml) was added. After stirring for 10 h, the reaction was diluted with diethyl ether (25 ml). The organic layer was separated, washed with water (2 x 10 ml), dried (over MgS04), filtered and evaporated in vacuo to leave an oil (130.2 mg). This oil was dissolved along with potassium hydroxide (200 mg) in ethanol (20 ml) and stirred for 2 h. After addition of palladium on charcoal (10%, 50 mg), the  161  reaction was stirred under an atmosphere of hydrogen for 1 h. The reaction was filtered and the solvent removed in vacuo. The residue was dissolved in diethyl ether (50 ml), washed with 2 M hydrochloric acid until the washings were neutral, water (20 ml), dried (over MgSCU), filtered and evaporated in vacuo to leave an oil. Purification by flash chromatography using diethyl ether/petroleum ether (1:24) as eluent afforded enone 175 (21.9 mg, 20%) as a colourless oil. Further elution afforded enone 181 (39.6 mg, 36%) as a colourless oil. Enone 175:  irv ax m  (CHCI3):  2975, 1655, 1625 cm" ; uv X 1  (e = 17,700); H nmr (400 MHz, l  CDCI3)  max  (MeOH):  244 nm  5: 1.05 (6H, d, / = 7 Hz, C7-(CH ) ), 1.08 3  2  (6H, s, C6-(CH )2), 1.75 (2H, t, J = 6 Hz, C 5 - H ) , 1.79 (3H, t, / = 2 Hz, C 2 - C H ) , 3  2  3  2.28 (2H, tq, J = 6,2 Hz, C 4 - H ) , 2.98 (IH, septet, J = 7 Hz, C7-H) ppm; mass 2  spectrum m/z: 180 (M ), 124, 109 (100); high resolution mass measurement: calculated +  f o r C H O : 180.1514; found: 180.1513. 1 2  2 0  *H nmr decoupling experiment: irradiation of the signal resonating at 8 2.28 ppm collapsed the triplet resonating at 8 1.75 ppm to a singlet and collapsed the triplet resonating at 8 1.79 ppm to a singlet. Enone 181: i r v  m a  x (CHCI3):  2970, 1680, 1625 cm-1; * H nmr (400 M H z , CDCI3)  8: 0.97 (3H, s, C 5 - C H ) , 1.03 (3H, s, C 5 - C H ) , 1.08 (3H, d, J = 7 Hz, C 4 - C H ) , 3  3  3  1.19 (3H, d, J = 7 Hz, C6-CH3), 1.22 (3H, d, J = 7 Hz, C6-CH3), 1.73 (3H, d, / = 1 H z , C 2 - C H ) , 2.51 (IH, qq, / = 7, 1 Hz, C4-H), 2.90 (IH, septet, / = 7 H z , 3  C6-H) ppm; mass spectrum m/z:  180 ( M ) , 165, 137 (100); high resolution mass +  measurement: calculated for C H Q O : 180.1514; found: 180.1511. 1 2  2  *H nmr decoupling experiment: irradiation of the signal resonating at 8 2.51 ppm collapsed the doublet resonating at 8 1.08 ppm to a singlet and collapsed the doublet resonating at 8 1.73 ppm to a singlet.  162  3.1.  Introduction.  The substantial difficulties encountered in the synthesis of medium-ring carbocycles and heterocycles has led to much study on the production of both natural and unnatural compounds possessing a medium ring. The germacrane sesquiterpenes have figured prominently in this development, initially, because of their pivotal role as both biosynthetic and synthetic precursors to a variety of sesquiterpene families, and more recently, because of the wide range of biological activity associated with certain members of this class. The formidable problems associated with the setting of the stereochemistry in such chemically labile and conformationally mobile systems has prompted the interest of synthetic chemists. The first member of this class to have its structure identified was germacrone (193), a crystalline compound found as the principal component of Bulgarian zdravets oil (76).  Germacrone  Since then, a range of sesquiterpene hydrocarbons, ketones and lactones belonging to this class have been identified (77) [Figure 6]. Of the number of total syntheses of germacranes to date, five main routes have been followed. One of these routes is the Cope or oxy-Cope rearrangement of 1,5-dienes, pioneered by Still (78). An illustrative example is provided by the synthesis of the cytotoxic eucannabinolide (206) (78b)[Scheme 28]. (+)-Carvone (200) was transformed  163  Isabelin  Linderalactone  Figure 6. Examples of naturally occurring germacranes  in a multistep sequence into the 3-hydroxy-l,5-diene 202, which on treatment with potassium hexamethyldisilazide at 85°C afforded 203 in high yield. functional group manipulation gave 206 in an overall yield of 0.57%.  Subsequent  164  (a) LiAlH , Et 0,0°C; (b) MCPBA, CH C1 , 25°C; (c) PhCH OCH Cl, i-Pr NEt, 25°C; (d) PhSeK-LiBr, 4  2  2  2  2  2  2  THF, 25°C; 30% H 0 , NaHC0 , NaOAc, THF, 60°C; (e) Jones' reagent, 0°C; (f) 207, n-BuLi, THF, -70°C; 2  2  3  (g) KN(TMS) DME, 85°C; (h) K C 0 2>  2  3>  MeOH, 25°C; (i) (COOH) silica gel, CH C1 , 35°C; (j) H 0 , 2>  2  2  2  2  Ti(o-/-Pr) , i-Pr NEt, Et 0, -30°C; (k) NaBH , MeOH, 0°C; (1) K C 0 , MeOH, 25°C; (m) Cr0 -pyridine, 4  2  2  4  2  3  3  CH C1 ,25°C; (n) DBU, THF, 25°C; (o) NaBH , MeOH, 0°C; (p) H -20% Pd(OH)2/C, EtOH, 25°C; 2  2  4  2  (q) (trimethylsilyl)imidazole, C H N, CH C1 ,25°C; (r) LDA, THF; HCHO(g), -70°C; (s) MsCl, Et N, DMAP. 5  5  2  2  3  CH C1 ,25°C; (t) DBU, dioxane, 70°C; (u) Bu NF, THF, 25°C; (v) AcOH, DCC, 4-pyrrolidinopyridine; 2  2  4  (w) 208, DCC, 4-pyrrolidinopyridine; (x) C H NH OTs", MeOH, glycol. +  5  5  Scheme 28  W. C. STILL  165  A second route involves the direct cyclisation of a ten-membered carbon chain, as shown by the synthesis of hedycaryol (212) by Ito (79)[Scheme 29]. trans,trans-Famesy\ phenyl sulphide (209) was converted to the epoxide 210 in 54% yield. Cyclisation of 210 was performed usingrc-butyllithiumin the presence of l,4-diazabicyclo[2.2.0]octane to give the two isomeric hydroxy thioethers 211 and 213 in 35% and 25% yield, respectively. Desulphurisation of 211 gave hedycaryol (212).  (a) NBS; Na C0 ; (b) n-BuLi, DABCO; (c) Li/EtNH 2  3  Scheme 29  2  S. ITO  Marshall has devised a strategy to synthesise the ten-membered rings that involves the ring contraction of a thirteen-membered cyclic ether (80).  Geranyl acetate (215)  was  chain-elongated to the chloro alcohol 216 that was cyclised to the ether 217 [Scheme 30]. Treatment with n-butyllithium resulted in a facile [2,3] Wittig rearrangement giving the  166  fra/w-isopropenylcyclodecynol 218 in 92% yield. Subsequent alcohol inversion, reduction of the triple bond and carbonylation afforded aristolactone (219).  (a) Se02, TBHP, CH C1 ; NaBH EtOH ; (b) TBDMSC1, DMAP, Et N, CH C1 , 25°C; (c) K C 0 , 2  2  4>  3  2  2  2  3  MeOH, 5°C; (d) MsCl, LiCl, 2,6-lutidine, DMF, 0°C; (e) TIPSC=CCH MgBr, Cul, THF, -23°C; Bu NF, 2  4  THF; (f) MsCl, LiCl, 2,6-lutidine, DMF, 0°C; (g) n-BuLi, THF; CH 0; (h) EtMgBr, THF-HMPA, 50°C; 2  (i) n-BuLi, THF-pentane, -78°C; (j) Ph-3P> DEAD, PhC0 H, benzene; KOH, MeOH; (k) Red-Al, THF; 2  NIS, THF; (Ph P)4Pd, CO, «-Bu SnH, benzene. 3  3  Scheme 30  J . A . MARSHALL  167  A fourth route is illustrated by the synthesis of the germacranolide dilactone isabelin ( 2 2 5 ) (81).  The synthesis involved pyrolysis of the  tricyclo[4.4.0.0 ' ]decane 2 5  intermediate 224, which was derived from the photo-addition of 3-methylcyclohex2-en-l-one (221) with methyl cyclobut-l-ene-1-carboxylate (220) [Scheme 31]. The fifth strategy involves the cleavage of the central bridge in a decalin system (two fused six-membered rings).  It was this route that 'caught our eye'.  Following the  successful synthesis of P-cyperone, a sesquiterpene possessing the decalin system, in our laboratory (5a), efforts at the generation of ten-membered rings were undertaken using thujone as the chiral starting material. The ring cleavage of decalins to ten-membered ring germacranes has been studied extensively. This strategy can be broken into two general groupings, either involving an internal photochemical rearrangement or a Grob-type fragmentation.  The internal  rearrangement methodology is illustrated in the synthesis of dihydrocostunolide (232) by Corey (82) [Scheme 32]. The starting material was santonin (227) which was reduced in a multistep process to the allylic alcohol 229 which underwent a base-catalysed elimination to the diene 230. Transformation of the diene to a ten-membered carbocycle 231 was accomplished photochemically using the well-known photofission of 1,3-cyclohexadiene systems. Compound 231 was then catalytically hydrogenated to dihydrocostunolide (232). Similar approachs have been adopted by Yoshikoski (83) and Fujimoto (84).  168  H  (a) hv, CH C1 -78°C; (b) LDA, THF, -78°C; TMSC1; (c) Pd(OAc) , CH CN; (d) Cl , H 0; K C 0 , acetone; (e) Me CuLi, HMPA, Et 0, -25°C; CH =CHCH I, HMPA, -20°C; NH C1; (f) NaBH , MeOH, -10°C; (g) IM HCl; (h) 0 , MeOH/H 0, NaOAc, -78°C; Me S; (i) Ag C0 -celite, benzene, 80°C; (j) LDA, THF, -60°C; CH NMe I"; (k) Mel, THF/MeOH; Na C0 , CH C1 , H 0; (1) 200°C; toluene. 2  2>  2  2  2  3  2  3  2  2  2  4  2  2  2  2  4  3  +  2  2  2  Scheme 31  3  2  2  2  P. A. WENDER  3  169  (a) H , 2% Pd/SrC0 EtOAc; (b) Br , CHC1 ,0°C; (c) LiBr, LiC0 , DMF, 120°C; (d) Al(0-j-Pr) , 2  3)  2  3  3  3  i-PrOH, 80°C; (e) alumina/pyridine, 220°C; (f) hv, MeOH, -18°C; (g) H , Raney Nickel, -18°C. 2  Scheme 32  E . J. COREY  170  The second strategy involves the cleavage of the central bond by a Grob-type (85) fragmentation.  Marshall (86) has investigated the cleavage of these systems using  boronate fragmentation. For example, treatment of the enone tosylate 233 with diborane, followed by base and protection yielded 234 (87). Indeed, this reaction was the key step in the attempted synthesis of dihydronovanin (240) laboratory [Scheme 33]. The strategy  38  by L. Barnes, a co-worker in our  was to synthesise the enone acetate 238 by a  route similar to that previously published (5e). Treatment of the enone acetate 238 with OS0 CH 2  3  (a)BH, (b) NaOMe (c) ArCOCl  ArCOO  233  234  diborane and base would be expected to lead, after oxidation, to the germacranolide 239 that would then be converted to dihydronovanin (240). A ten-membered ring has also been produced by a similar reaction involving the cleavage of a y-hydroxy tosylate. Reduction and tosylation of the known ketone 241 followed by epoxidation yielded 242.  Acid treatment of 242 yielded the hydroxy tosylate  243, which then afforded stereoselectively the germacradienone 244 upon treatment with potassium f-butoxide (88)[Scheme 34].  L. Barnes and J. P. Kutney, private communication.  171  (a) 2,2-Dimethyl-l,3-propanediol, benzene,p-TsOH, 80°C; (b) KMn0 , KOH, f-BuOH-H 0; (c) AcOH, hexanes; (d) NBS, THF-H 0; (e) 2,2-dimethyl-l,3-propanediol, benzene, p-TsOH, 80°C. 4  2  Scheme 33  2  172  O  OTs  (a) NaBH , EtOH; (b) TsCI, pyridine; (c) MCPBA, CH CI ; (d) BF .Et 0, toluene, -30°C; (e) KOfBu, rBuOH, 42°C. 4  2  Scheme 34  A .  2  G.  3  2  GONZALEZ  The well documented synthesis of decalin and tricyclo[4.4.0.0 ' ]decane systems in our 2 4  laboratory prompted the investigation of germacrane synthesis. Since the cleavage of the decalin system by a Grob-type fragmentation was under investigation by L. Barnes, a strategy involving the direct cleavage of the two bridges in the tricyclo[4.4.0.0 ' ]decane 2 4  system was chosen. Since this system could be readily prepared from thujone (1), then ring-opening to the ten-membered ring could proceed as an early step in the overall synthesis of germacranolides. Multistep functionalisation of the ring-opened product would lead to a variety of germacranes.  173  A n initial study by P. Grice indicated that fragmentation of the alcohol 245 with lead 39  tetraacetate proceeded to give a compound tentatively assigned the structure 246. The overall objective of this project was the investigation of this ring-opening reaction.  In the sixties, Mihailovic developed the use of lead tetraacetate, primarily for the formation of steroid cyclic ethers (89). The reaction of lead tetraacetate with an alcohol leads, it is assumed, to the reversible formation (Equation 2.1) of a readily cleavable derivative 247. It is quite probable, depending on the relative concentration of the  n ROH  + Pb(OAc)  4  (RO) Pb(OAc) . n  4 n  + nCH COOH 3  247  reactants, the structure of the alcohol and the reaction conditions, that lead (IV) alkoxides with different numbers of alkoxy groups can be formed. Since the intermediate lead (IV) alkoxides are less stable than lead tetraacetate itself, the (R-O-)-Pb bond undergoes homolytic cleavage under thermal or photolytic decomposition to give the alkoxy radical 249 [Scheme 35]. The resultant alkoxy radical can undergo a variety of subsequent reactions. Primary or secondary alkoxy radicals can lose an adjacent a-hydrogen atom to  3 9  P. Grice, research report, U.B.C.  ^  O Pb(OAc)  OH  3  R  Pb(OAc)  4  R'  R  4  +  1  R' 4  3  248 5  /  R  R  3  AcOH  2  * Pb(OAc)  3  O' 8-  R  \  *  5  R'  R  4  'Pb(OAc)  0  iJ^R  R  4 /  1  3  *Pb(OAc)  3  3  R ^F 2  R  R  4'  3  O  a  249  251  250  __L_ Pb(OAc)  3  R =H - aH*  6H*  R ' 4  R 252 3  + AcO'  \  OAc  R  1  R  4  R 253  R 3  2  R' 4  R^ 3  2  R  R  4  255  254  3  R  oxidation  cyclisation  fragmentation  Pathway A  Pathway B  Pathway C Scheme 35  R 4  256  3  175  give a carbonyl compound [Pathway A]. An alkoxy radical with a 5-hydrogen can undergo a [l,5]-hydrogen transfer leading eventually to a cyclic ether [Pathway B]. Alternatively P-fragmentation can occur to give the ketone 251 and the radical 250, which subsequently gives the acetate 255 or the alkene 256 [Pathway C]. It appears that p-fragmentation and intramolecular ether formation are competing processes (90). Several factors control the course of the reaction and determine the ratio of products. Important factors in intramolecular hydrogen abstraction leading to cyclic ether formation are the relative position and orientation of the attacking hydroxyl oxygen with respect to the 8-(or e-)carbon hydrogen as well as the presence of steric hindrance. Although the rate of P-fragmentation is dependent on the stability of the initially formed carbon radical 252, it is also affected by other factors, such as the stability of the carbonylcontaining fragment or the relief of strain. Intramolecular cyclic ether formation decreases as the factors influencing P-fragmentation of the alcohol become more and more favourable. Indeed, when P-fragmentation of alcohols affords a stable allylic radical or benzyl radical the ring closure to a cyclic ether may be entirely suppressed (91). The direct oxidation of primary and secondary alcohols [Pathway A] to the corresponding aldehydes and ketones with lead tetraacetate in aprotic solvents usually affords low yields, rarely exceeding 15% in the most favourable of cases. The reaction of 5-hydroxy-steroids with lead tetraacetate leads via P-fragmentation to the formation of 5,10-seco-steroids.  Mihailovic  (92)  found that treatment  of  3P-acetoxycholestan-5a-ol (257) with one equivalent of lead tetraacetate in the presence of anhydrous calcium carbonate gave the two seco-steroids 258 (15%) and 259 (30%). Similarly 3P-acetoxycholestan-5p-ol (260) afforded the same seco-steroids in comparable yield.  176  260  A co-reagent often used in lead tetraacetate reactions is iodine. It is suspected that the initial alkoxy-lead-(IV)-acetate intermediate reacts with iodine to generate a hypoiodite that subsequently photolytically decomposes to an alkoxy radical (93).  The chemistry of  hypervalent organoiodine compounds has been an active area of research (94).  The  discovery of iodobenzene diacetate, a white stable solid, has led to much investigation of its use in cyclic ether formation and P-fragmentation of alcohols (95). Iodobenzene diacetate decomposes only at elevated temperatures and is not hydrolysed by atmospheric moisture. It is sensitive to daylight to a small extent but easily undergoes photochemical decomposition. The photo-induced reaction of iodobenzene diacetate with alcohols, in the presence of iodine, is very similar to the reaction of lead tetraacetate with alcohols. It has been suggested that the same alkoxy radicals are generated by the reaction of the alcohols with  177  the acetyl hypoiodite that is produced in situ (95a) (Equation 4). Acetyl hypoiodite has never been isolated, but has been postulated as an intermediate in numerous reactions.  PhI(OAc) AcOI  +  2  +  V I 2  2  ROH  Phi  +  2AcOI  ROI  +  AcOH  [4] AcOI  AcO*  +  I*  ROI  RO*  +  I*  The iodobenzene diacetate/iodine system has been found to be superior to the heavy metal derivatives that are used for intramolecular hydrogen abstraction (95b). Suarez has reported (95c) the f3-fragmentation of alkoxy radicals, generated in the presence of the iodobenzene diacetate/iodine system, in steroidal lactones. Ring expansion, for example, of the y-lactol 261 was accomplished by a photochemical reaction with iodobenzene diacetate (1.1 equivalents) and iodine (1 equivalent) in cyclohexane to give, in 83% yield, a mixture of the olefinic lactones 262, 263 and 264 in a 1:2:2 ratio.  40  It should be noted that a twenty fold excess of lead tetraacetate is often used in similar type reactions.  178  Based on the above studies one objective of this study was the synthesis of tertiary alcohols of a tricyclo[4.4.0.0 ' ]decane system from thujone and the subsequent 2 4  investigation of ring-opening of this system to give germacrane-type products.  179  3.2 Results and  Discussion.  For the study of the ring-opening of thujone-derived tertiary alcohols to ten-membered rings, the two alcohols 265 and 245 were conceived as the initial targets. The alcohols were envisaged to arise from known intermediates. The masked carbonyl group was expected to be used later in the manipulation of the ten-membered rings, produced by ringopening the alcohols 245 and 265, into known germacranes.  265  3.2.1.  245  The Synthesis of Alcohol 265.  The initial target molecule to be synthesised was the tertiary alcohol 265, which could be generated from the known ketal 266, synthesised previously in our laboratory (5a) from thujone (1).  265  266  The synthesis of 266 was repeated using modifications on the methods outlined by Kutney (5a) [Scheme 36]. However, since in the present study, various aspects of the reactions were altered and some of the by-products not previously isolated were characterised, a brief overview of this synthesis will be discussed.  180  266 Scheme 36  The synthesis of 235 from 1 by the method detailed in the previous publication (5a), gave 235 in 50-55% yield. Zoretic in 1975 (96) reported the acid-catalysed "Robinson annelation" of 1-methylcyclohexanone.  The yields were equivalent to the 'classic'  Robinson annelation, but the reaction was more convenient to perform. Applying this method to thujone using l-chloropentan-3-one (prepared from propionyl chloride by Friedel-Crafts reaction with ethylene (97)) and a catalytic amount of para-toluenesulphonic acid resulted in a mixture of 235 and the diketone 268 (235:268 = 1:4) in 60% yield. Further purification gave a pure sample of 268. The diketone 268 was characterised by the two carbonyl absorbances at 1740 and 1720 c m in the infrared spectrum, by the 1  molecular ion peak at m/z = 236 in the mass spectrum and by the *H nmr spectrum which indicated six protons a to a carbonyl functionality. The production of 268 was surprising, since in both the base-catalysed annulation of thujone, and in the acid-catalysed annulation  181  O  1  235  268 1:4 6 0 %  of 1-methylcyclohexanone, direct dehydration of the intermediate P-hydroxy ketone occurred to produce the enone products. Obviously, the presence of the cyclopropane functionality must retard the acid-catalysed aldol reaction. Since the yield of this reaction was not substantially higher than the previously performed base-catalysed annulation, and also a further step would be needed to convert 268 to 235, this route was not pursued further. The ketalisation of 235 using 2,2-dimethyl-l,3-propanediol was performed as reported (5a). However, it was found that the prolonged reaction time (20 h), which afforded complete consumption of the starting material, resulted in the production of 270 as a byproduct, that presented difficulty in separation from 266.  Reduction in the reaction time  270  182  to 9 h resulted in a maximum yield (80%) of the ketal (based on recovered starting material) without the production of the by-product 270.  The production of the by-product 270 on  prolonged heating of the ketal in the presence of acid was not unexpected. It has been reported (6a) that treatment of the ketal 271  with a catalytic amount of para-  toluenesulphonic acid in boiling toluene for 100 h, at which time only a trace of 271 remained, resulted in the production of 272 as the major isolated product in 25% yield.  271  272  The previous publication (5a) indicated the production of only one isomer 266 in the ketalisation of 235.  However, monitoring by G C , and isolation of the products, indicated  that a small amount (10%) of the isomer 269 was formed concurrently. Based on the *H nmr data, and the perceived thermodynamic stability of the two products, the major product has the methyl at C2 in an equatorial position. In the *H nmr spectrum of 269 a broadening of the signals at 5 3.24 (C2-H) and 1.78 (C4-He ) ppm indicated the presence q  of a W-coupling, which could only occur if the C2 proton was equatorial. An *H nmr nOe difference experiment performed on 266 showed, by an enhancement of the resonance due to the C6 methyl group on irradiation of the resonance corresponding to the C2 proton, an axial-axial relationship between the C6 methyl group and the C2 proton. Since the stereochemistry of the methyl group at C2 would be fixed for the following reactions, the separation of 266 from 269 was necessary. Only pure 266 reactions.  was used in subsequent  183  To obtain the tertiary alcohol 265 from 266 a "Markovnikov" directed addition of hydroxyl was required. Classically, this could be achieved by an oxymercuration reaction. Attempts at this transformation using a variety of mercury compounds and conditions (98) (Hg(OAc)2, Hg(02CCF3)2; NaBFLj.NaOH, NaBHU.AcOH) resulted, in most cases, in the recovery of starting material.  No tertiary alcohol was obtained in any of these  reactions. As oxymercuration proved unsuccessful, attention was turned to a two-step process, in which the double bond was epoxidised, and the epoxide opened regioselectively to give the tertiary alcohol. Epoxidation of 266 with meto-chloroperoxybenzoic acid in dichloromethane gave the epoxide 273 in 85% yield (use of dibasic sodium phosphate as a buffer increased this yield to 93%).  41  The upfield shift of the CIO proton from 8 5.34 ppm in 266 to 8 3.35 ppm in  273 in the *H nmr spectrum, and the absence of the absorbance in the infrared spectrum at 1645 c m  -1  corresponding to the double bond stretching frequency, was consistent with the  formation of the epoxide. The (3-stereochemistry of the epoxide was concluded by correlation to a related structure 290 (see page 190). The approach of the epoxidising agent to the double bond from the same side as the cyclopropane ring was at first surprising. However, molecular models indicated that the a face was just as hindered, if not more so, than the (3 face.  273 This modification was based on the advantages of buffering peroxytrifluoroacetic acid reactions (99).  184  Subsequent reactions by others in the laboratory have shown that hydrogenation of 266 proceeded from the P face of the double bond (shown by a subsequent X-ray crystal structure), as does reaction of 266 with alkaline potassium permanganate (shown by an 42  X-ray crystal structure)  4 3  This is opposite to the stereochemistry proposed by Kutney  (5a) for hydroboration of the double bond in 266. Hydride attack on the epoxide was expected to occur at the less hindered secondary centre. However, use of lithium aluminium hydride in refiuxing diethyl ether, or refiuxing tetrahydrofuran, did not effect any reaction. Even the use of Super-Hydride® (lithium triethylborohydride) (100), resulted in recovery of the epoxide. Obviously, the attack of the epoxide from the a face of the molecule, as required by an S 2-type mechanism, was N  hindered by the steric effects of the isopropyl group, and the slight concave nature of the molecule. Attention was turned to the reduction of the epoxide using Birch-type reduction, that is, a 'solvated' electron system obtained by dissolving lithium in amines. Using lithium in anhydrous ethylamine (101), the tertiary alcohol 265 was produced in 93% yield. A small amount (0.5% by GC) of the secondary alcohol 274 was detected. The tertiary alcohol 265 was characterised by the strong absorbance centred at 3490 c m ' in the infrared 1  spectrum. In the *H nmr spectrum the absence of any signal in the region 8 4.00-6.00 ppm and the presence of an A B system (8 1.45 and 2.43 ppm) corresponding to the two C10 protons, indicated that the epoxide ring had opened to give the tertiary alcohol. The presence of the three sets of signals at 8 0.14, 0.83, and 1.20 ppm, indicated that the cyclopropane ring had not been opened in the reaction.  N. Cheng, private communication. C. Cirera, private communication.  185  Prior to the use of lithium in ethylamine, Brown's method (102) of lithium in ethylenediamine reduction for hindered bicyclic epoxides was employed that resulted  in the isolation, in high yield (91%), of a mixture of secondary and tertiary alcohols, 274 and 265 respectively, which were inseparable by column chromatography. However, it was found that treatment of the mixture of alcohols with benzoyl chloride in pyridine with a catalytic amount of 4-dimethylaminopyridine (103), afforded selective benzoylation of the secondary alcohol, leaving the less reactive tertiary alcohol 265 intact. Separation of the benzoate 275 from 265 was accomplished by column chromatography on Florisil®. It was hoped that the ring-opening would have been more selective to give a higher percentage of the tertiary alcohol. However, Brown had shown that, while the lithiumethylenediamine reduction of the epoxide 276 generated from 2-methyl-2-norbornene  undergoes ring cleavage preferentially at the more accessible site, the epoxide 279 generated from 2-methyl-2-butene undergoes the opposite cleavage (102).  OH  <"0 71% 276  277  X  278  OH  OH 10%  279  OH  29%  280  90% 281  The secondary alcohol 274 produced in the epoxide ring-opening was characterised via its benzoate derivative.  The aromatic ester funtionality of 275 was shown by an  absorbance at 1705 c m in the infrared spectrum. The *H nmr spectrum showed three sets -1  of signals downfield of 5 7.20 ppm corresponding to the protons on the benzene ring, a quintet at 8 1.79 ppm corresponding to the C2 proton with a coupling of 7 Hz to the CI proton, and a doublet at 8 6.31 ppm (/ = 8 Hz) for the proton at C10 adjacent to the benzoate group. The stereochemistry of the benzoate group was inherent from the P-orientated epoxide ring system. Since the five-membered B ring was close to planarity, the coupling of 8 Hz between the CI proton at the A/B ring junction and the C10 proton indicated that the dihedral angle between the two protons was close to 0°, i.e. the A/B ring junction was cis in nature. It was found that when the mixture of 275 and 265 was chromatographed on silica gel, the benzoate decomposed to give a mixture of dienes 285 and 270 [Scheme 37].  187  285  286 Scheme 37  Characterisation of the dienes was made by the molecular ion peak at m/z = 304 in the mass spectrum, and by careful assignment of the complex H nmr spectrum. The *H nmr l  spectrum showed the absence of any signals corresponding to cyclopropane protons, and the appearance of two additional methyl signals, a singlet at 8 1.73 ppm for 270, and a doublet for 285 at 8 1.28 ppm. Coincident signals at 8 5.93 ppm corresponded to the olefinic proton in each diene. There were two sets of protons for the ketal group in the region 8 3.20 - 8 3.75. Further proof of the structure of these dienes, in particular the presence of a 6,5 ring system with the absence of a cyclopropyl group, was provided by  188  deketalisation of this mixture. This was effected, in 73% yield, by the use of 2M hydrochloric acid in tetrahydrofuran. Deprotection of the carbonyl group was accompanied by isomerisation of the double bonds into conjugation to give one major product 286. The *H nmr spectrum showed no evidence of cyclopropane protons. The isopropyl group was evident and there was a one proton singlet resonating at 5 6.20 ppm corresponding to the proton on the double bond at C9. The infrared spectrum indicated the conjugated carbonyl functionality by the absorbances at 1650, 1626, and 1595 cnr . Studies of the molecular 1  model of 286 indicated that the more thermodynamically stable product would have the C7 -methyl group a-orientated.  3.2.2.  The Synthesis of Tertiary Alcohol 245.  The synthesis of 245 [Scheme 38] proceeded via methods similar to those for the alcohol 265. The ketal 271 was synthesised by the method detailed by Kutney (5a). Alcohol 245 had been reported (5a) as a minor product in the hydroboration of 271 by our group. Epoxidation of ketal 271 with meta-chloroperoxybenzoic acid in dichloromethane gave 290 in 59% yield. The use of a buffer (dibasic sodium phosphate) (99) increased this yield to 8 9 % , and the use of a free radical scavenger (3-rerf-butyl-4-hydroxy-5methylphenylsulphide) (104) increased the yield to 91%. Using a combination of buffer and free radical scavenger, the yield of 290 was increased to 96%. The epoxide was characterised by the molecular ion at m/z = 264 in the mass spectrum, the absence of the absorbance at 1650 c m  -1  in the infrared spectrum corresponding to the double bond  stretching frequency, and by the shift of the C10 proton resonating at 8 5.38 ppm in the *H nmr spectrum of 271 to 8 3.29 ppm in 290.  Attempts were made to ring-open this epoxide with hydride reducing agents. The epoxide proved inert. Ring-opening was effected with electrons generated by dissolving lithium in amines. Yields with lithium in ethylenediamine were very poor, though in contrast to the products from the ring-opening of epoxide 273, only the tertiary alcohol was obtained. Yields were greatly increased by using lithium dissolved in ethylamine, the tertiary alcohol 245 being obtained as white crystals in 87% yield. Alcohol 245 was easily characterised by its molecular ion peak at m/z = 266 in the mass spectrum, and the broad absorbance centred at 3600 c m  -1  in the infrared spectrum,  190  corresponding to an O-H stretching frequency.  The *H nmr spectrum indicated the  cyclopropane ring was still intact by the A M X system of protons at 8 0.20,0.88, and 1.18 ppm. There were two sets of geminally coupled protons at 8 1.54 and 2.48 ppm for the CIO protons, and at 5 1.48 and 1.90 ppm for the C2 protons. The resonance at 8 2.48 ppm was shown to be the a proton at C10 by the small W-coupling to one of the cyclopropane protons at 5 0.20 (CS-R ). exo  The resonance at 8 1.90 ppm was shown by a  decoupling experiment to be coupled to the signal at 8 1.48 ppm, and also by a small Wcoupling to a C4 proton. For W-coupling to occur, the proton resonating at 8 1.90 ppm must be in the equatorial position. Fortuitously, after much effort at obtaining crystals suitable for X-ray analysis, it was found that, after one year of standing, an oil containing approximately 90% of 245 had grown a crystal of the alcohol approximately 2 cm long, with a cross-section of 1 cm. From a section taken from this crystal, an X-ray crystallographic structure was obtained [Figure 7]. This clearly shows the cis nature of the 6,5 ring junction. Relating back from this, it is clear that the epoxide 290  must possess the epoxide functionality in a  P-orientation (cis to the cyclopropane group). Since epoxidation of 271 and 266 afforded only one isomer each, it was logical to assume that the epoxide 273 also had the epoxide ring in a fi-orientation. The X-ray structure showed that the B ring, including the cyclopropane ring adopted a pseudo-boat conformation. The actual five-membered ring was distorted from planarity by approximately 20°. The atoms C6, C7, C9 and C10 were close to planarity (torsional angle between the bonds C6-C7, and C9-C10 was 4°). The X-ray structure also showed the concave nature of the molecule, due to the cis ring junction. Since the concave nature would have been present (though to a lesser extent) in the precursor  290,  191  Figure 7. Single crystal X-ray structure of alcohol 245.  192  this explained the difficulty experienced in epoxide ring-opening with the hydride reagents. Prior to the single crystal X-ray structure determination of 245 being obtained, efforts at establishing the stereochemistry of the alcohol were examined. A second chemical synthesis of 245 in which the stereochemistry could be established was undertaken (see Section 3.2.5). A H nmr spectroscopic determination, involving the differences in the l  chemical shifts of methyl group resonances between various solvents was investigated. It has been shown by Wenkert (105) that in compounds possessing vicinal methyl and hydroxyl groups, substantial downfield shifts of the methyl protons are noted in the *H nmr spectra taken in ds-pyridine (relative to di-chloroform) when the methyl group and the hydroxyl group are synclinal (5 hioroforrrr Spyridine c  =  ^Py  =  0-20 to -0.30 ppm), whereas  _  relatively little shift is observed when the groups are antiperiplanar ( A ppm).  py  = -0.01 to -0.05  In order to confirm that this method may be applied to 6-methylbicyclo-  [4.4.0]decan-l-ols, Ayer (106) measured the pyridine induced shifts of the tertiary methyl groups in 291, 292 and 293 and these were found to be -0.01, -0.14 and -0.18 ppm, respectively.  291  293  292  The *H nmr spectrum of 245 was recorded in di-chloroform and ds-pyridine.  The  axial ring junction methyl at C6 was clearly assigned at 8 1.03 ppm in di-chloroform, and at 8 1.31 ppm in ds-pyridine, indicating a shift of A  p y  = -0.28 ppm. Based on Wenkert's  193  work, this shift would indicate a cis A/B ring junction, which is in agreement with the stereochemistry determined from the X-ray crystal structure.  3.2.3.  The Ring-Opening of 245 to a Ten-membered Ring.  Having efficient syntheses of 245 and 265, attention was turned to investigating the ring-opening of these systems to give ten-membered rings.  Alcohol 245 was first  investigated as a model study in the synthesis of germacranes, since a preliminary report on the ring-opening of this compound was available.  44  Using the conditions employed by P.  Grice on 245 (lead tetraacetate, pyridine, hv, benzene), resulted in the recovery of a majority of the starting material along with a complex product mixture from which no major component could be isolated. Since Mihailovic (89a) had stated that pyridine was an inhibitor of the desired bond cleavage, it was decided to investigate the ring-opening using lead tetraacetate with the addition of other reagents. A common co-reagent in lead tetraacetate reactions is iodine. Irradiation of a solution of 245, lead tetraacetate and iodine in benzene with ultraviolet light using a Pyrex® filter (hv > 300 nm) gave a small amount (10%) of a mixture of alkenes 294 and 295, and the acetate 246 (28%). It was observed that the condition of the lead tetraacetate was important. Non-recrystallised lead tetraacetate (even though white in colour, and reactive in other reactions) did not effect a reaction, nor did recrystallised (from acetic acid) lead tetraacetate that had been dried under vacuum for more than one day. It seemed that lead tetraacetate moistened with acetic acid was the most effective. It has been noted that recrystallised lead tetraacetate contains two moles of acetic acid loosely associated with each molecule of lead  4 4  P. Grice, research report, U.B.C. (1979).  194  tetraacetate (89b).  Based on this observation, 245 was allowed to react with lead  tetraacetate in the presence of additional acetic acid. Irradiation of this mixture in benzene with ultraviolet light (hv > 300nm) gave the alkenes 294, 295, and 296 in 14% yield, along with the acetate 246 (28%).  45  The acetate 246 was obtained as a very viscous liquid after purification of the reaction mixture by column chromatography. The infrared spectrum showed a broad carbonyl In subsequent reactions the solvent was changed to toluene without affecting yields.  195  absorption, and no hydroxyl absorptions. The molecular ion peak at m/z = 324 was detected in the mass spectrum. The presence of an acetate group was indicated by the singlet methyl resonance at 8 2.15 ppm in the *!! nmr spectrum and the sharp singlet at 8 170.9 ppm in the  1 3  C nmr spectrum. In the  1 3  C nmr spectrum the resonance at 8 202.6  ppm corresponded to the ketone functionality, that arose in conjunction with C1-C6 bond cleavage. The two resonances at 8 123.5 and 136.7 ppm indicated the presence of one double bond in the molecule. The *H nmr showed two doublets at 8 0.93 and 0.94 ppm, and a septet at 8 2.86 ppm, attributable to the isopropyl protons. The absence of any resonances upfield of 8 0.93 ppm indicated that the cyclopropane group had been cleaved. A proton and a methyl attached to a double bond were indicated by a triplet at 8 5.33 ppm and a singlet at 8 1.70 ppm, respectively. Both the H and C nmr spectra showed some l  1 3  some broad unresolved resonances. The broadening of various resonances presented difficulty in the full interpretation of the spectrum. It was thought that this poor resolution was due to recording the spectrum at a temperature that corresponded to the coalescence point for the signals resulting from various conformations (107). Since the exchange rate between conformations is related to temperature, then cooling a sample should separate the proton signals for the various conformations. Conversely, heating the sample would give sharp signals, corresponding to the average magnetic environments of the protons. The H nmr spectrum of 246 was also recorded in ds-toluene at normal probe l  temperature (= 30°C), 50°C and 80°C (Figure 8). The spectrum at 30°C (Figure 8a) showed some sharpening of the signals (relative to the spectrum recorded in di-chloroform) though broadening of the resonances was still evident. The resonance at 8 5.36 ppm (equivalent to 8 5.33 ppm in di-chloroform) showed as a very broad singlet. In the spectrum recorded at 80°C, however, all the resonances had sharpened leading to the  196  assignment of the majority of the spectrum (Figures 8c). The resonance at 8 5.36 ppm (assigned to the C7 proton) was a doublet of doublets with coupling of 10 Hz to one of the C8 protons resonating at 8 2.19 ppm and a coupling of 5 Hz to the other C8 proton resonating at 8 2.42 ppm, as shown by a decoupling experiment. The C8 protons were coupled geminally and with the C7 proton. Other resonances clearly assigned were the ketal protons at 8 3.45-3.57 ppm, the acetate methyl singlet at 8 1.90 ppm, the C6 methyl singlet at 8 1.54 ppm (indicative of a methyl on a double bond), and the isopropyl group with the septet proton at 8 2.85 ppm and the two methyl doublet at 8 0.76 and 0.84 ppm. Irradiation of the doublet at 8 3.07 ppm resulted in the collapse of the doublet at 8 2.22 ppm to a singlet, with the coupling of 14 Hz indicating these to be geminal methylene protons, either at C2 or C10. A pair of doublets at 8 2.54 and 2.66 ppm could also be assigned as geminal methylene protons, either at C2 or C10. The protons at C3 and C4 occurred as overlapping multiplets. A n nOe experiment was performed on 246 in ds-toluene at 80°C in order to establish conclusively that the acetate group was attached at the C9 position. Irradiation of the septet at 8 2.85 ppm lead to enhancement of the signals at 8 0.76, 0.84, and 1.90 ppm. Since this septet was clearly assigned as the isopropyl proton, then the acetate (8 1.90 ppm) must be at the C9 position to be in close proximity. The reverse experiment with the irradiation of the acetate singlet at 8 1.90 ppm gave enhancement of the signal corresponding to the isopropyl proton, as well as enhancement of the signals corresponding to the isopropyl methyls. Further evidence that the acetate was not at the other possible position (C6) was given by irradiation of the signal at 8 1.54 ppm, corresponding to the C6 methyl group. No enhancement of the acetate signal was detected. Enhancement of one C8 proton resonance (8 2.19 ppm), and one of the geminal protons at 8 2.54 ppm occurred. Since no  c)  80°C  A.  b)  50°C  A_  a)  30°C  I  5  4  ppm  3  Figure 8. Partial H nmr spectrum of acetate 246 recorded at 30°, 50° and 80°C (400 MHz, C7D8). J  2  198  enhancement of the C7 proton resonating at 8 5.36 ppm occurred, this indicated the double bond to be of the E-configuration. It has been noted (78a) that the typical lack of serious transannular non-bonded interactions in 1,4 or 1,5 cyclodecadiene systems makes it possible to consider only those confomers composed totally of minimum energy torsional fragments.  Analysis of  molecular models for 246, indicate that, if the C6 methyl, C6, C7, C8 carbon atoms and C8 proton are placed in the same plane, and the carbonyl placed parallel to this plane, then a conformation was derived (a chair-chair conformation) with minimal eclipsing interactions. From the molecular models, the proximity of the C6 methyl protons with the C8 proton was evident, as was the proximity of the C6 methyl protons with a proton at C2, indicating that the resonance at 8 2.54 ppm, that was enhanced on irradiation of the C6 methyl resonance, and by implication the resonance at 8 2.66 ppm, are due to the C2 protons. This implied that the resonances at 8 2.22 and 8 3.07 ppm can be attributed to the CIO protons. Clearly this conformation is solvent dependent, since the C7 proton resonance in di-chloroform was seen as an apparent triplet with coupling to the C8 protons of 8 Hz, while, in ds-toluene, it appeared as a doublet of doublets with coupling to the C8 protons of 10 and 5 Hz. The specific optical rotation of the acetate 246 was found to be [CC]D = +14.0°, indicating that there was a chiral excess of one enantiomer (though not proving that 246 was enantiomerically pure). Even though 245 is enantiomerically pure (since it is derived in an enantiospecific sequence from the enantiomerically pure intermediate 235 (5a)), both a chiral and non-chiral transition state for the ring-opening could be envisaged, leading to a non-racemic mixture of enantiomers.  199  Attention was turned to the use of iodobenzene diacetate to effect a similar reaction. Irradiation of 245 with iodobenzene diacetate and iodine in toluene with ultraviolet light (hv > 300 nm) resulted in the ring-opening of 245 to the inseparable alkenes 294 and 295 in 2 0 % yield, with the exocyclic double bond predominating (88:12).  Buffering the  solution with sodium acetate increased the yield of the reaction to 40% without changing the product ratio. Attempts at performing this reaction in cyclohexane with irradiation by a sunlamp (95) resulted in no reaction.  245  The alkenes 294  294  and 295,  88:12  295  formed in the photochemical reaction of 245  with  iodobenzene diacetate and iodine, or lead tetraacetate with iodine, were obtained as an inseparable mixture, even after chromatography on silver nitrate impregnated silica. The product ratio of 88:12 (or 85:15 for the lead tetraacetate reaction) was determined by G C and *H nmr analyses. The infrared spectrum showed a carbonyl stretching frequency at 1704 c m , indicative of a non-conjugated carbonyl group, as well as an olefinic stretching -1  frequency at 1625 c m . The ! H nmr spectrum of the mixture recorded in di-chloroform -1  showed three intense singlets at 8 1.53, 1.75, and 1.80 ppm, indicating three methyl groups attached to double bonds, which, along with the absence of the characteristic isopropyl methyl doublets, indicated the terra-substituted exocyclic double bond in 294. As with the acetate 246, broadening of some resonances was observed. An apparent triplet at 8 5.18 ppm corresponded to the C7 proton of 294.  A doublet at 8 1.04 ppm  200  integrating for six protons, and a multiplet at 8 5.82 ppm integrating for two protons, showed the presence of the isomeric alkene 295. The enone 296, found in the mixture of alkenes formed by the photochemical reaction of lead tetraacetate and acetic acid with 245, could also be obtained by the elimination of acetate from 246, using sodium methoxide. In this manner, 296 was obtained in 51% yield. 7  246  296  The molecular ion peak at m/z = 264 in the mass spectrum of 296, and the appearance of two further olefinic carbons in the  1 3  C nmr spectrum, coupled with the loss of the  resonance at 8 170.9 ppm, indicated the elimination of one mole of acetic acid from the acetate 246. The absorbance at 1680 c m in the infrared spectrum, and the appearance of -1  a singlet at 8 5.58 ppm in the *H nmr spectrum, indicated the presence of the a,(3unsaturated carbonyl functionality. As in the case of the acetate 246, and the alkenes 294 and 2 9 5 , the *H nmr spectrum of 2 9 6 recorded in di-chloroform at normal probe temperature showed some broadening of signals, although a doublet at 8 1.07 ppm integrating for six protons, a methyl singlet at 8 1.54 ppm, a septet at 8 2.25 ppm, and two olefinic signals at 8 5.39 ppm (a triplet corresponding to the C7 proton) and 8 5.58 ppm, could be detected. The H nmr spectrum was also recorded in ds-toluene at 30°C, 50°C, J  and 80°C. However no substantial improvement in signal resolution was observed. An *H nmr nOe difference experiment indicated that the newly created double bond had a Z configuration. Irradiation of the singlet at 8 5.52 ppm gave an enhancement of the  201  septet at 8 2.25 ppm and the doublet at 8 1.07 ppm corresponding to the proton and methyls of the isopropyl group respectively. No enhancement of the C8 protons was detected. The E configuration of the C6-C7 double bond was already clear from the spectrum of acetate 246. The ultraviolet spectrum of 296 showed a maxima at 236 nm, which is in close agreement with the calculated value (239 nm) for the 7C —> TC* transition of a,P-unsaturated ketones (107). The low extinction coefficient (e = 3700) indicated that the enone was distorted from planarity. Analysis of molecular models indicated that if the enone was planar, then very severe steric interactions of either the oxygen of the carbonyl, or the protons at C2, with the C6 or C7 carbon atoms would occur. At first glance, the production of 294 as the predominant alkene product in the photochemical reaction of 245 to the alkenes, and the non-conjugation of the double bond in 296 seemed odd. However, molecular models indicated that a fully conjugated dienone in a ten-membered ring would be highly strained. The stability of 294 over 295 or 296 was also evident from molecular models, since an exocyclic double bond minimises the transannular steric interaction. Since some preliminary studies on the isomerisation of the mixture of alkenes to one major product proved unsuccessful, the reaction of lead tetraacetate and acetic acid with 245 was investigated with a view to maximising the yield of 246. Variation in the time of the reaction was important. For a given ratio of reagents, the production of acetate 246 reached a peak and further irradiation did not increase the yield although more starting material was consumed [Table 4]. Photodecomposition of the products appeared to be occurring. The number of equivalents of lead tetraacetate used for a given reaction time was also important. Use of 2.1 equivalents of lead tetraacetate and irradiation with  202  Table 4. The effect of reaction time of 245 with lead tetraacetate on the isolated yields of alkenes 294, 295 and 296 and acetate 246.  Eq. ofLTA  Acetic acid: toluene  Time (min)  Isolated yields  245  Alkenes  Total recovery  246  1  4.2  1:9  60  57%  8%  14%  79%  2  4.2  1:9  100  14%  16%  28%  58%  3  4.2  1:9  180  1%  4%  24%  29%  4  2.1  1:9  100  43%  9%  16%  68%  5  2.1  1:9  180  20%  14%  28%  62%  The reactions were run with 0.064 M solutions of 245 in acetic acid/toluene irradiated with ultraviolet light (hv > 300nm) generated by a Hanovia high pressure mercury lamp.  Table 5. The effect of increasing the equivalents of lead tetraacetate in the reaction with 245 on the isolated yields of alkenes 294, 295 and 296 and acetate 246.  Eq. of L T A  Acetic acid: toluene  Time (min)  Isolated yields  245  Alkenes  Total recovery  246  1  2.1  1:9  100  43%  9%  16%  68%  2  4.2  1:9  100  14%  16%  28%  58%  3  6.3  1:9  100  16%  12%  23%  51%  The reactions were run with 0.064 M solutions of 245 in acetic acid/toluene irradiated with ultraviolet light (hv > 300nm) generated by a Hanovia high pressure mercury lamp.  203  ultraviolet light for 100 min resulted in a considerable consumption (57%) of starting alcohol and the production of acetate 246 in 16% yield. Use of 4.2 equivalents for the same reaction time gave the acetate in 28% yield, although the overall mass recovery was somewhat lower. Use of more equivalents (i.e. 6.3 equivalents) of lead tetraacetate did not change significantly the relative amounts of starting material consumed, or the amount of acetate produced [Table 5] . The use of 2.1 equivalents with irradiation for 180 min gave approximately the same ratio of starting alcohol and products as the use of 4.2 equivalents for 100 min [Entries 2 and 5, Table 4]. Increasing the ratio of acetic acid to substrate resulted in approximately the same yields of alcohol and acetate, but with a decrease in the yield of the alkenes [Table 6]. No reaction could be effected using exclusively acetic acid as the solvent [Entry 3].  Table 6. The effect of increasing the amount of acetic acid in the reaction of 245 with lead tetraacetate on the isolated yields of alkenes 294, 295 and 296 and acetate 246. Eq. of L T A  Acetic acid: toluene  Time (min)  Isolated yields  245  Alkenes  Total recovery  246  1  4.2  1:9  100  14%  16%  28%  58%  2  4.2  7:13  100  12%  3%  24%  39%  3  4.2  a  360  93%  —  —  93%  a) Acetic acid used as solvent. The reactions were run with 0.064 M solutions of 245 in acetic acid/toluene irradiated with ultraviolet light (hv > 300nm) generated by a Hanovia high pressure mercury lamp.  It was found that buffering the reaction by the addition of sodium acetate resulted in a reduction in the rate of the reaction, although no increase in the relative yields was observed  204  [Table 7]. Reaction of lead tetraacetate and acetic acid with alcohol 245 in the dark resulted in complete recovery of the starting material, indicating that the acid alone was not catalysing the reaction [Entry 3]. Table 7. The effect of buffering and the absence of light in the reaction of 245 with lead tetraacetate on the isolated yields of alkenes 294,295 and 296 and acetate 246. Eq. of L T A  Acetic acid: toluene  Isolated yields  Time (min)  245  Alkenes  Total recovery  246  1  4.2  1:9  100  14%  16%  28%  58%  2a  4.2  1:9  100  45%  15%  12%  72%  3b  4.2  1:9  1440  100%  —  —  —  c  a) 8 equivalents of sodium acetate added, b) The reaction was performed in the dark, c) GC yield. The reactions were run with 0.064 M solutions of 245 in acetic acid/toluene irradiated with ultraviolet light (hv > 300nm) generated by a Hanovia high pressure mercury lamp (except entry 3).  In summary, the optimum conditions found (Entry 2 and 5, Table 4), resulted in the production of the alkenes in 16% yield and the acetate in 28% yield. Based on recovered starting material, the overall yield ofring-openedproducts was 53%. Ten-membered carbocycles can be produced from tricyclic alcohols derived from thujone. However, the moderate yields obtained negate the use of the alcohol 245 as an intermediate to the germacranes, since extra manipulation is required to attach the C2 methyl group present in all germacranes.  3.2.4.  The Photo-induced Cleavage of the Cyclopropane Group in 265.  Having some success at generating the ten-membered ring from the tertiary alcohol 245, our attention was now turned to the tertiary alcohol 265, since it is this series that  205  would lead direcdy to the germacranes. However, treatment with lead tetraacetate, using the conditions which were successful in the ring-opening of 245, resulted in the recovery of starting material. Using lead tetraacetate in benzene under refiuxing conditions also resulted in recovery of the starting alcohol. When 265 was treated with iodobenzene diacetate and iodine in toluene, and irradiated with ultraviolet light, consumption of starting material did occur, although a multitude of products were formed, along with the production of some polar tarry material. After careful analysis, it was found that some of the minor products were artifacts obtained during the work-up, and were in fact derived from the major product 297 (isolated in 20-30% yield) of the reaction. OAc  265  Characterisation of 297 presented some difficulty. The infrared spectrum showed absorbances due to the hydroxyl group, a double bond and an ester group. The mass spectrum gave a molecular ion peak at m/z = 380, corresponding to an oxidative addition of an acetate group. The greatest difficulty was experienced in the interpretation of the H J  nmr spectrum. The proton spectrum run at 400 MHz in di-chloroform showed very little resolution (Figure 9). The *H nmr spectrum of acetate 297 in d2-dichloromethane was taken at room temperature, -20°C, -40°C, and -60°C. A sharpening of the very broad resonances in the region 8 4.5-5.5 ppm was shown at -60°C, although interpretation of the spectra was still difficult. Acquisition of data at temperatures above room temperature in d8-toluene were also performed. The room temperature spectrum in ds-toluene was much  206  207  better resolved than that taken in di-chloroform, and increasing the temperature to 50°C and 95°C gready sharpened the signals in the spectrum. It is clear from the spectrum taken at 95°C (Figure 10) that the cyclopropane ring has been opened, that an acetate group has been added (8 1.72 ppm for the OCOCH3 protons and 8 5.57 ppm for the proton adjacent to this group) and that a trisubstituted double bond is present (8 5.40 for the proton on the double bond). The  1 3  C nmr spectrum did not show the presence of any resonances in the region 8 190-  220 ppm corresponding to a ketone group. Since a ketone group would be generated by the proposed mechanism for ring-opening of the 6,5 ring system, then this indicated that the desired opening had not occurred. The broadband-decoupled  1 3  C nmr spectrum was  interesting in that certain resonances were the sharp singlet usually encountered, but that other resonances were broad, again indicative of the fluxional nature of the molecule. In the A P T  1 3  C nmr experiment, the broad resonances do not show at all.  At the time, the exact structure of the acetate 297 was unknown. This was resolved, however, by saponification of the ester functionality with sodium methoxide to give, in 65% yield, diol 298, on which a single crystal X-ray structure determination was made.  297  298  The infrared spectrum of 298 clearly showed the absence of the absorbance corresponding to an acetate functionality. The H nmr spectrum recorded in di-chloroform at room l  temperature was poorly resolved, as was the spectrum recorded in d8-toluene at normal  OAc  —r  r  i  5  4  3  r  PP  m  2  T  1  Figure 10. *H nmr spectrum of acetate 297recordedat 95°C (400 MHz, C7D8).  209  probe temperature. However, recording the spectrum in ds-toluene at 80°C (Figure 11) resolved most signals. Three singlets each integrating for three protons and three doublets each integrating for three protons could now be clearly seen, along with the quartet at 8 2.17 ppm, corresponding to the C2 proton, and a septet (partly hidden by residual solvent resonances) at 8 2.06 ppm, corresponding to the isopropyl proton. The proton attached to the double bond at C8 resonated at 8 5.46 ppm as a broad singlet. The X-ray structure (Figure 12) showed clearly the cis ring junction of the decalin system, explaining the fluxional nature of 297 and 298, since c/s-decalins are known to exist predominantly in two conformations. In 297 and 298 the substituents hinder rapid exchange between the various conformations, thus leading to the broadening of signals in the nmr spectra.  ds-decalin conformational flipping  The X-ray structure shows the two hydroxyl groups are cis to each other on the B ring, and that in the solid state, the A ring adopts a chair conformation, with the C6-methyl group axial and the C2-methyl group equatorial. The B ring adopts a flattened chair-like conformation. Diol 298 was converted back to 297, in 75% yield, by acetylation in the presence of 4dimethylaminopyridine, to confirm that no rearrangement had occurred in the conversion of 297 to 298. The cis ring junction of 297 was a surprise, since, before the X-ray structure of the tertiary alcohol 245 was obtained, it had been assumed that a trans ring junction existed in  210  211  Figure 12. Single crystal X-ray structure of diol 298.  212  245 and 2 6 5 , based on the earlier assumption made by Kutney (5a). This resulted in a proposal that the reaction of 2 6 5 with iodobenzene diacetate proceeded to give a tenmembered ring, which then closed to give the m-decalin system. However, since the initial alcohol is now believed to have a cis ring junction, it is possible that the reaction proceeds with only the cyclopropane ring-opening, though a mechanism in which the C l C6 ring junction is opened cannot be ruled out. The production of 2 9 7 was unfortunate.  However, 2 9 7 could be utilised in the  formation of germacranes since a Grob-type fragmentation would yield a ten-membered ring. Much work was performed on increasing the yield of the product 297, with a view to utilising 2 9 7 in such a way, as summarised in Table 8. The best reaction conditions discovered was the use of two equivalents of iodobenzene diacetate and two equivalents of iodine with 265 in toluene under irradiation with ultraviolet light for 2 h [Entry 2]. Entry 8 shows an experiment in which a second addition of two equivalents of each of the reagents was added after 14 h, which resulted in complete consumption of the stating material, but with no increase in the yield of the product (the compound actually isolated was 299, that has been shown to arise from 297 during the work-up procedure). Crown ether was used [Entry 5] to increase the solubility of the reagents, though with no significant effect on the yield of 297. With the use of cyclohexane, N,A/-dimethylformamide or tetrahydrofuran as the solvent no reaction was observed and the starting alcohol was recovered quantitatively. Under irradiation with ultraviolet light (hv > 240 nm) complete decompostion of the alcohol 265 to the enone 235 occurred. Sodium acetate was added to some reactions, since this had increased the yields slightly in the previous reactions of the other tertiary alcohol 245 with iodobenzene diacetate. However, in the reaction of 265 no increase in the yield of 2 9 7 was found.  213  Table 8. The effect of varying stoichiometry and reaction time on the isolated yield of 297 in the reaction of 265 with iodobenzene diacetate and iodine Eq. of PhI(OAc)2  Eq. ofI  Time(h)  Isolated yields  Total recovery  2  265  297  1  2  2  1  83%  3%  86%  2  2  2  2  31%  27%  58%  3  2  2  4  30%  28%  58%  4  2  3  24  20%  31%  51%  5a  2  3  4  6%  37%  43%  6  4  4  1  19%  19%  38%  7  4  4  4  14%  24%  38%  8  2+2  14+2  trace  26%  26%  2+2  b  The reactions were run with 0.031 M solutions of 265 in toluene irradiated with ultraviolet light (hv > 300nm) generated by a Hanovia high pressure mercury lamp, a) Crown-6 added, b) Isolated as 299  In the reaction of 265 with iodobenzene diacetate and iodine, other minor products were also formed. Identification of these products showed that they were derivatives of the acetate 297, and not ten-membered ring compounds. The yields of these by-products were variable, often depending on the work-up and purification procedures. With careful workup and purification, these by-products were found in less than 1% yield. Two of these byproducts isolated, 299 and 300, were also found in samples of 297 left in di-chloroform overnight prior to *H nmr analysis. Direct synthesis of these products from 297 was performed for characterisation. Treatment of 297 with sodium bicarbonate in refiuxing toluene affords 299 in 51% yield. Diene 299 was characterised by its maximum of  214  239 nm (calculated value = 239 nm (107)) in the ultraviolet spectrum. The *H nmr spectrum (recorded in ag-benzene, since use of di-chloroform as the solvent gave poor resolution), showed the two methyl singlets on the double bond (8 1.62 and 1.66 ppm) and the two olefinic protons at 8 5.30 and 6.40 ppm coupled to each other (/ = 9 Hz). OAc  Deketalisation of 299 with 2M hydrochloric acid led, in low yield (26%), to another byproduct of the photochemical reaction, ketone 300. The tertiary alcohol was still intact, as evidenced by the broad absorbance at 3576 cm in the infrared spectrum. The H nmr -1  l  spectrum in di-chloroform indicated two doublets for the olefinic protons (8 5.40 and 6.49 ppm), the two methyl groups on a double bond resonating at 8 1.73 and 1.82 ppm, and the C2 methyl group resonating as a doublet at 8 1.15 ppm. If 300 was left in the nmr tube, elimination and isomerisation to the conjugated ketone 301 occurred.  215  The moderate yields obtained in the production of 297 in the reaction of 265 with iodobenzene diacetate and iodine hindered the use of this product as an intermediate in the synthesis of germacranes. The surprising production of 297 has prompted investigations into the use of this product as an intermediate in the synthesis of other natural products to be pursued by co-workers.  3.2.5.  The Synthesis of Epoxy Ketone 302.  Prior to obtaining suitable crystals of the tertiary alcohol 245 and thereby undertaking an X-ray structure analysis, chemical proof for 245 was sought. The epoxides 302 and 303 were selected as intermediates for this purpose, since these compounds could also be used in another study involving photochemical rearrangements, based on the work of Jeger (108).  302  303  The intention was to synthesise the epoxide 302, with known stereochemistry, and then convert this epoxide to the tertiary alcohol 305 via a ketalisation and epoxide ring-opening sequence. A comparison of the spectral data of 305 with that from the alcohol 245 would establish the complete stereochemistry of 245.  302  304  305  216  Epoxide 302 was produced by a reduction of the known enone 288, followed by selective epoxidation and oxidation [Scheme 39]. The enone 288 was obtained in 68%  302  308 Scheme 39  yield from the non-conjugated ketone 289, by treatment with glacial acetic acid containing a drop of concentrated sulphuric acid (5a). Reduction of 288 using sodium borohydride  289  288  with the addition of cerium chloride (49) gave a quantitative 1,2 reduction to the two alcohols 306 and 307 (306:307  = 93:7), that could be separated by column  chromatography. Use of diisobutylaluminium hydride to reduce 288 resulted in a similar product ratio to the sodium borohydride reduction.  217  The alcohol 306 was characterised by its infrared spectrum that showed broad absorbances at 3327 cm' , corresponding to the presence of a hydroxyl group, and an 1  absorbance at 1675 cm' (double bond). The molecular ion peak at m/z = 206 indicates the 1  addition of only one mole of hydrogen, and the *H nmr spectrum shows a resonance at 8 5.10 ppm, corresponding to the proton on the double bond, and a broad resonance at 8 4.28 ppm, corresponding to the proton adjacent to the hydroxyl group. Stereochemical assignment of the hydroxyl group was not possible from the *H nmr spectrum. The acetate 309 was prepared in 70% yield from the alcohol 306, since often the coupling constants for the C3 proton are more readily measured in the H nmr spectrum of the acetates. The ]  acetate 309 was characterised by the presence of two absorbances at 1735 c m  -1  and 1690  cm" , the loss of the absorbance, corresponding to the O-H functionality, in the infrared 1  309  310  311/312  spectrum, and by the presence of a methyl singlet at 8 2.02 ppm in the *H nmr spectrum. The proton at C3 was a broad multiplet resonance at 8 5.37 ppm. Direct determination of the stereochemistry was still not possible, although comparison of the width at half height of this resonance in relation to the C3 proton resonance of the acetate 310 (produced from alcohol 306 using the Mitsunobu reaction (109)) indicated that the proton at C3 in 309  218  was in a pseudo-axial position. Confirmation of this assignment was obtained by an X-ray crystal structure performed on a compound obtained in a later reaction. A change in the work-up procedures (using 2 M HCl) for the sodium borohydride/ cerium chloride reduction of 288 resulted in poorer yields, with the production of the ethers 311 and 312 as major by-products. The structures of 311 and 312 are supported by the infrared spectrum, the molecular ion peak at m/z = 220 in the mass spectrum, and the presence of a methyl singlet at 8 3.32 ppm in the H nmr spectrum. l  Alcohol 306 was epoxidised in 80% yield, by the method of Sharpless (110) using vanadyl acetylacetonate and /-butyl hydroperoxide. This epoxidation proceeds on cyclic allylic alcohols to give directed epoxidation on the face of the double bond cis to the alcohol. The product 308, produced as a white solid and recrystallised from diethyl ether/petroleum ether, was characterised by the presence of O-H stretching frequencies at 3590 c m  -1  and 3450 c m , and by the absence of the double bond stretching frequency in -1  the infrared. The molecular ion peak at m/z = 222, corresponding to an increase in mass of 16 when compared to the molecular ion peak of alcohol 306, and the two resonances in the *H nmr spectrum at 8 3.14 and 8 3.95 ppm, corresponding to the proton attached to the epoxide and the proton adjacent to the hydroxyl functionality, respectively, further confirmed the structure. A single crystal X-ray structure (Figure 13) was obtained on the crystals of 308. This showed that the alcohol was indeed P (or cis to the angular methyl), and that the epoxidation had also occurred on the fj-face of the double bond. Alcohol 308 was oxidised by Jones' reagent (111) to give, in 76% yield, the epoxy ketone 302, that was characterised by the strong carbonyl absorbance at 1730 c m ' in the 1  infrared, and by the downfield shift of the resonances, due to the C4 proton, to 8 2.58 ppm in the *H nmr spectrum. The resonance at 8 2.58 ppm has the splitting pattern of a triplet of doublets, due to a geminal coupling of 14 Hz to the C4 equatorial proton, a  219  coupling of 14 Hz to the C5 axial proton, and a coupling of 5 Hz to the C5 equatorial proton. Direct epoxidation of the enone 288 using sodium hydroxide and hydrogen peroxide solution resulted in the production, in 58% yield, of the same epoxide 302, as shown by *H nmr spectroscopy. The direct epoxidation of 235 was also performed to give, in 51% yield, the epoxy ketone 303. Both epoxides 302 and 303 were to be used in another project. Since the structure of 245 was solved by an X-ray crystal structure, the conversion of 302 to 245 was not continued.  Figure 13. Single crystal X-ray structure of epoxide 308.  220  3 . 3 Experimental. See section 2.4.1. for General experimental.  3.3.1. (IS, 4 R , 5S)  4-MethyI-l-(l-methylethyl)-4-(pentan-3-one)bicyc!o[3.1.0]hexan-3-one  A solution of thujone  46  (268).  (943.8 mg, 5.9 mmol of thujone), l-chloropentan-3-one  (1.1 ml, 9.5 mmol) and /wa-toluenesulphonic acid monohydrate (47.4 mg, 0.25 mmol) in benzene (3 ml) was heated at reflux for 2 days. The reaction was cooled, poured into saturated sodium bicarbonate solution (20 ml) and extracted with diethyl ether (2 x 25 ml). The combined organic extracts were washed with water (20 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a red oil (1.7 g).  Purification by flash  chromatography using ethyl acetate/petroleum ether (1:19) as eluent gave a mixture (815 mg, crude yield = 60%) consisting of enone 235 (20%) and diketone 268 (80%). A pure sample of diketone 268 (219.2 mg) was obtained by further flash chromatography using ethyl acetate/petroleum ether (1:39) as eluent: [OC]D = + 3 7 . 2 ° (CHCI3, c = 0.33); ir v  m a x  (CHCI3): 2960, 1740, 1720 cm" ; *H nmr (400 MHz, CDCI3) 5: 0.00 (IH, dd, / = 6, 1  4 H z , C6-U ), endo  0.65 (IH, ddd, / = 8, 6, 3 Hz, C6 ), exo  0.93 (3H, d, / = 7 Hz,  C 7 - C H ) , 0.97 (3H, s, C 4 - C H ) , 1.04 (3H, d, J = 7 Hz, C 7 - C H ) , 1.05 (3H, t, 3  3  D i s t i l l e d from Western cedar leaf o i l , ca 95%  3  purity.  221  J = 7 Hz, C5'-H ), 1.21 (IH, dd, / = 8, 4 Hz, C5-H), 1.44 (IH, septet, / = 7 Hz, 3  C7-H), 1.73 (IH, m, Cl'-H), 1.83 (IH, m, Cl'-H), 2.10 (IH, d, / = 18 Hz, C 2 - H 2.40 (4H, m, C2'-H , C4'-H ), 2.64 (IH, dd, / = 18, 3 Hz, C2-B. ) 2  2  exo  endo  ),  ppm; mass  spectrum m/z: 236 (M ), 218, 203, 121, 57 (100); high resolution mass measurement: +  calculated for C15H24O2: 236.1776; found: 236.1775; anal.: calculated for C15H24O2: C 76.22; H 10.23; found: C 76.49; H 10.30. *H nmr decoupling experiments: irradiation of the doublet of doublets resonating at 8 0.00 ppm simplified the signal resonating at 8 0.65 ppm to a doublet of doublets (/ = 8, 3 Hz) and simplified the doublet of doublets resonating at 8 1.21 ppm to a doublet (7 = 8 Hz); irradiation of the signal resonating at 8 0.65 ppm simplified the doublet of doublets resonating at 8 0.00 ppm to a doublet (/ = 4 Hz) and simplified the doublet of doublets resonating at 8 2.64 ppm to a doublet (7=18 Hz); irradiation of the septet resonating at 8 1.44 ppm collapsed the doublets resonating at 8 0.93 and 1.04 ppm to singlets.  3.3.2. (2S, 6 R , 7S, 9S)  2,6-Dimethyl-3-(4,4-dimethyl-2,6-dioxane)-  9-(l-methylethyl)-tricyclo[4.4.0.0 ' ]dec-l(10)-ene 7  and  (2R, 6 R , 7S, 9S)  9  (266)  2,6-dimethyI-3-(4,4-dimethyl-2,6-dioxane)-  9-(l-methyIethyl)-tricyclo[4.4.0.0 > ]dec-l(10)-ene 7  9  (269).  The ketals 266 and 269 were produced in 80% overall yield (266:269 = 9:1) by a modification of the method of Kutney (5a) using a reduced reaction time (9 h) to minimise  222  the production of by-products. Purification of the reaction product by repeated flash chromatography using ethyl acetate/petroleum ether (1:39) as eluent afforded pure samples of ketals 266 and 269 as colourless oils. Ketal 266: [cc]22 = 142° (CHC1 , c = 0.500); ir v D  3  m a x  (thin film): 2954, 2868,  1645 cm- ; *H nmr (400 MHz, CDC1 ) 8: 0.22 (IH, t, / = 4 Hz, Ci-U ),  0.59 (IH,  1  3  endo  dd, J = 8.5, 4 Hz, C 8 - J W ) , 0.70 (3H, s), 0.87 (3H, d, 7 = 7 Hz, CII-CH3), 0.94 (3H, d, 7 = 7 Hz, CII-CH3), 1.09 (3H, s, C6-CH3), 1.12 (3H, d, 7 = 7 Hz, C2-CH ), 1.15 3  (IH, dd, / = 8.5, 4 Hz, C7-H), 1.17 (3H, s), 1.37 (3H, m, C4-H, C5-H ), 1.45 (IH, 2  septet, 7 = 7 Hz, Cll-H), 2.18 (IH, qd, 7 = 7, 1.5 Hz, C2-H), 2.69 (IH, m, C4-H), 3.28 (IH, dd, 7=11, 1 Hz), 3.31 (IH, dd, 7 = 11, 1 Hz), 3.58 (IH, d, 7 = 11 Hz), 3.77 (IH, d, 7 = 11 Hz), 5.34 (IH, br s, wh/ = 4 Hz, C10-H) ppm; 2  1 3  C nmr (75 MHz,  CDCI3) 8: 9.33, 18.96, 20.47, 20.63, 20.94, 22.26, 23.31, 23.85, 29.82, 30.86, 33.95, 36.39, 39.78, 40.30, 47.46, 69.80, 70.60, 100.43, 124.85, 149.16 ppm; mass spectrum m/z: 304 (M ), 289, 200, 141 (100); high resolution mass measurement: calculated for +  C20H32O2: J  304.2402; found: 304.2389.  H nmr decoupling experiments: irradiation of the doublet of doublets resonating at 8 0.59  ppm simplified the signal resonating at 8 0.22 ppm to a doublet (7 = 4 Hz) and simplified the signal resonating at 8 1.15 ppm to a doublet (7 = 4 Hz); simultaneous irradiation of the signals resonating at 8 1.12 and 1.15 ppm simplified the signal resonating at 8 0.22 ppm to a doublet (7 = 8.5 Hz), simplified the doublet of doublets resonating at 8 0.59 ppm to a doublet (7 = 8.5 Hz) and affected the signal resonating at 8 2.18 ppm; irradiation of the signal resonating at 8 2.18 ppm collapsed the doublet resonating at 8 1.12 ppm to a singlet and sharpened the signal resonating at 8 5.34 ppm; irradiation of the signal resonating at 8 5.34 ppm simplified the signal resonating at 8 2.18 ppm to a quartet (7 = 7 Hz).  223  *H nmr nOe difference experiments: irradiation of the signal resonating at 8 0.22 ppm caused an enhancement of the signal resonating at 8 0.59 ppm and of the signal resonating at 8 1.09 ppm (C6-CH3); irradiation of the signal resonating at  8 0.59 ppm caused an  enhancement of the signals resonating at 8 0.22, 1.15 and 1.45 ppm; irradiation of the signal resonating at 8 2.18 ppm caused an enhancement of the signals resonating at 8 1.09 and 1.12 ppm; irradiation of the signal resonating at 8 2.69 ppm caused an enhancement of the multiplet resonating at 8 1.37 ppm; irradiation of the signal resonating at 8 5.34 ppm caused an enhancement of the doublets resonating at 8 0.87,0.94 and 1.12 ppm. Ketal 269: ir v  (thin film): 2960, 1645 cm" ; H nmr (300 MHz, 1  m a x  (IH, t, / = 4 Hz, CS-H ), endo  (3H, d, /  = 7 Hz), 0.97  ]  0.63 (IH, dd, J = 9, 4 Hz, CZ-H ), eX0  (3H, d, J  = 1 Hz), 1.00  (3H, d, /  CDCI3) 8:  0.38  0.84 (3H, s), 0.87  = 7 Hz), 1.14  (3H, s),  1.20  (IH, m, C7-H), 1.20 (3H, s), 1.39 (IH, septet, 7 = 7 Hz, Cll-H), 1.44 (IH, dt, / = 12, 4 Hz, C5-pH), 1.64 (IH, td, / = 13, 4 Hz, C5-ocH), 1.78 (IH, m, C4-aH), 1.94 (IH, td, j =  14, 4 Hz, C4-0H), 3.24  (IH, br q, / = 7 Hz, C2-H), 3.37 (2H, d, J =  11  Hz), 3.51  (IH, d, / = 11 Hz), 3.65 (IH, d, J = 11 Hz), 5.40 (IH, s, C10-H) ppm; mass spectrum m/z: 304 (M ), 289, 141 (100). +  224  3.3.3. (IS, 2S, 6 R , 7S, 9S, 10R)  2,6-Dimethyl-3-(4,4-dirnethyl-2,6-  dioxane)-l,10-epoxy-9-(l-methylethyl)-tricyclo[4.4.0.0 ' ]decane 7  (273).  9  A solution of ketal 266 (3.153 g, 10.35 mmol), anhydrous sodium phosphate dibasic (2.506 g, 17.63 mmol) and mera-chloroperoxybenzoic acid (2.859 g, 16.57 mmol) in dichloromethane (100 ml) was stirred at 0 ° C for 15 min, then 3 h at room temperature. The reaction was quenched with saturated sodium bicarbonate solution (100 ml). The organic layer was washed with water (50 ml), dried (over MgSC<4), filtered and evaporated in vacuo to give a pale yellow oil (4.05 g). Purification by flash chromatography using ethyl acetate/petroleum ether (1:39) as eluent afforded epoxide 273 (3.096 g, 93%) as a colourless oil: [ a ] = +18.3° (CHCI3, c = 0.63); ir v ax (thin film): 2955, 2869 c n r ; D  1  m  ! H nmr (400 M H z , CDCI3) 5: 0.20 (IH, dd, / = 9, 5 Hz, C8-H), 0.72 (3H, s, ketal-CH ), 0.84 (3H, d, / = 7 Hz), 0.94 (3H, d, / = 7 Hz), 1.00 (3H, s, C6-CH3), 1.01 3  (3H, d, / = 7 Hz), 1.02 (IH, m), 1.12 (IH, t, J = 5 Hz), 1.18 (3H, s, ketal-CH ), 1.21 3  (IH, td, / = 14, 3 Hz), 1.45 (IH, dt, J = 14, 3 Hz), 1.56 (IH, septet, / = 7 Hz, Cll-H), 1.61 (IH, td, J = 14, 3 Hz), 1.97 (IH, q, / = 7 Hz, C2-H), 2.64 (IH, dt, J = 14, 3 Hz, C3-Heq), 3.26 (IH, dd, / = 11, 2 Hz), 3.33 (IH, dd, / = 11, 2 Hz), 3.35 (IH, s, C10-H), 3.56 (IH, d, / = 11 Hz), 3.76 (IH, d, / = 11 Hz) ppm; mass spectrum m/z: 320 (M+), 305, 209, 141 (100); high resolution mass measurement: calculated for C20H32O3: 320.2352; found: 320.2347; anal: calculated for C20H32O3: C 74.96; H 10.06; found: C 74.94; H 9.99.  225  *H nmr nOe difference experiment: irradiation of the singlet resonating at 8 0.72 ppm caused an enhancement of the singlet resonating at 8 1.18 ppm and an enhancement of the signals resonating at 8 3.26, 3.33, 3.56 and 3.76 ppm.  3.3.4. (IS, 2S, 6R, 7S, 9S)  2,6-Dimethyl-3-(4,4-dimethyl-2,6-dioxane)-  l-hydroxy-9-(l-methylethyl)-tricyclo[4.4.0.0 ' ]decane 7 9  (265)  Method A: To a solution of electrons, formed by dissolving lithium (322 mg, 46.4 mg-atoms) in anhydrous ethylamine (30 ml) at -78°C, was added dropwise epoxide 273  (1.947 g, 6.075 mmol) in anhydrous tetrahydrofuran (20 ml) at a rate so as not to  discharge the blue colour. After addition was complete, ethanol (1 ml) was added to discharge the colour and the reaction diluted with diethyl ether (150 ml). The organic solution was washed with cold 2M hydrochloric acid until the washings remained acidic, then water (50 ml), dried (over MgSC»4), filtered and evaporated in vacuo to leave a yellow oil (2.4 g). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent afforded alcohol 265  (contaminated with 0.5% of secondary alcohol  (1.8264 g, 93%) as a white solid.  Sublimation gave 265  274)  as fine white needles: mp  93-95°C; [oc] = +17.6° (CHCI3, c = 0.17); ir Vmax (CHCI3): 3489(br), 2955 cnr*; H l  D  nmr (400 M H z , CDCI3) 8: 0.14 (IH, ddd, / = 8, 4, 1.5 Hz, C$-H ), exo  ketal-CH ), 0.83 (IH, dd, / = 8, 4 Hz, C 8 - H 3  endo  0.70 (3H, s,  ) , 0.87 (3H, d, J = 7 Hz, CII-CH3),  226  0.89 (IH,  s, exchangeable, O-H), 0.92 (3H, d, J = 7 Hz, C I I - C H 3 ) , 1.04 (3H, s,  C6-CH3), 1.11 (3H, d, / = 7 Hz, C2-CH3), 1.12 (IH, td, J = 14, 5 Hz, C3-H ), 1.19 ax  (3H, s, ketal-CH ), 1.20 (IH, m, C7-H), 1.28 (IH, septet, J = 7 Hz, C l l - H ) , 1.45 3  (4H, m), 2.43 (IH, br d, / = 14 Hz, C10-aH), 2.53 (IH, dt, J = 14, 3 Hz, C3-H ), eq  3.30 (2H, m), 3.55 (IH, d, / = 11 Hz), 3.74 (IH, d, / = 11 Hz) ppm; H nmr (400 MHz, l  d -pyridine) 5 : 0.28 (IH, ddd, / = 8, 5, 1.5 Hz, C8-H ), 5  exo  0.60 (3H, s), 0.93 (3H, d,  / = 7 Hz, C I I - C H 3 ) , 0.95 (IH, dd, / = 8, 5 Hz ), 1.02 (3H, d, / = 7 Hz, C I I - C H 3 ) ,  1.23 (IH, m), 1.24 (3H, s), 1.34 (3H, s), 1.35 (IH, m, Cll-H), 1.51 (IH, m), 1.53 (3H, d, J = 7 Hz, C2-CH ), 1.63 (IH, td, / = 14, 3 Hz), 1.87 (2H, m), 2.04 (IH, q, J = 3  7 Hz, C2-H), 2.60 (IH, dt, / = 14, 3 Hz, C3-H ), 2.71 (IH, d, / = 14 Hz, C10-H), eq  3.32 (2H, m), 3.55 (IH, d, / = 11 Hz), 3.77 (IH, d, / = 11 Hz) ppm; mass spectrum m/z: 322(M ), 304, 200, 141 (100); high resolution mass measurement: calculated for +  C20H34O3: 322.2508; found: 322.2511; anal.: calculated for C 2 0 H 3 4 O 3 :  C 74.49; H  10.63; found: C 74.62; H 10.53. *H nmr decoupling experiments  (CDCI3):  irradiation of the signal resonating at 5 0.14  ppm simplified the signal resonating at 8 0.83 ppm to a doublet (J = 4 Hz), affected the signal resonating at 5 1.20 ppm and sharpened the signal resonating at 8 2.43 ppm; irradiation of the signal resonating at 8 2.43 ppm affected the signal resonating at 8 1.45 ppm and simplified the signal resonating at 8 0.14 ppm to a doublet of doublets (/ = 8, 4 Hz); irradiation of the septet resonating at 8 1.28 ppm collapsed the doublets resonating at 8 0.87 and 0.92 ppm to singlets; irradiation of the signal resonating at 8 2.53 ppm affected the signals resonating at 8 1.45 and 1.12 ppm. *H nmr decoupling experiments (ds-pyridine): irradiation of the signal resonating at 8 2.04 ppm collapsed the doublet resonating at 8 1.53 ppm to singlet; irradiation of the signal resonating at 8 0.28 ppm affected the signals resonating at 8 0.95, 1.23 and 1.87 ppm.  227  Method B: To a solution of epoxide 273  (260 mg, 0.811 mmol) in anhydrous  ethylenediamine (1.4 ml) was added lithium (20 mg, 3 mg-atoms). The reaction was stirred at 50°C for 3 h to obtain a persistent blue colour. The reaction was quenched by the addition of water (5 ml) and extracted with tetrahydrofuran (20 ml). The organic layer was dried (over MgSC«4), filtered and evaporated in vacuo to give a viscous oil (264 mg). Purification of part of this crude oil (252 mg) by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent afforded an inseparable mixture of alcohols 265 and 274 (265:274 = 60:40, 227 mg, 91%) To a solution of the above mixture (64 mg, 0.20 mmol) in pyridine (1 ml) was added benzoyl chloride (23 ul, 0.20 mmol). After stirring for 19 h, 3 M sodium hydroxide solution (5 ml) was added, followed by extraction with diethyl ether (20 ml). The organic layer was washed with 2 M hydrochloric acid (10 ml), water (10 ml), dried (over MgSC<4), filtered and evaporated in vacuo. Purification by flash chromatography on Florisil® using ethyl acetate/petroleum ether (1:4) as eluent afforded benzoate 275 (22 mg, 26%) as a glass. Further elution gave tertiary alcohol 265 (32 mg, 50%) as a white solid. Benzoate 275:  [ a ] = -52° (CHC1 , c = 0.23); ir v D  3  m a x  (CHCI3): 2950, 1705 cm-l; *H  nmr (400 M H z , CDCI3) 5: 0.41 (IH, dd, / = 8, 5 Hz), 0.70 (3H, s), 0.88 (3H, d, / = 7 Hz, CII-CH3), 0.88 (IH, m), 0.94 (3H, d, / = 7 Hz, CII-CH3), 0.95 (3H, d, / = 7 Hz, C2-CH ), 1.01 (IH, dd, J = 8, 5 Hz), 1.08 (3H, s), 1.11 (IH, td, / = 14, 3  3 Hz, C4-H ), 1.18 (3H, s), 1.40 (IH, br d, / = 14 Hz, C5-He ), 1.55-1.70 (3H, m, ax  q  C l l - H , C l - H , C5-H ), 1.79 (IH, quintet, / = 7 Hz, C2-H), 2.60 (IH, dt, J = 14, 3 Hz, ax  C4-H ), 3.28 (IH, dd, / = 11, 2 Hz), 3.44 (IH, dd, / = 11, 2 Hz), 3.58 (IH, d, J = 11 eq  Hz), 3.76 (IH, d, J = 11 Hz), 6.31 (IH, d, J = 8 Hz, C10-H), 7.43 (2H, t, J = 8 Hz,  228  C3"-H, C5"-H), 7.53 (IH, t, / = 8 Hz, C4"-H), 8.05 (2H, d, / = 8 Hz, C2"-H, C6"-H) ppm; mass spectrum m/z: 426 (M ), 343, 304, 141 (100); high resolution mass +  measurement: calculated for C27H38O4: 426.2770; found: 426.2767; anal.: calculated for C27H38O4: C 76.02; H 8.98; found: C 76.02; H 9.00. ! H nmr decoupling experiments: irradiation of the signal resonating at 8 0.41 ppm simplified the doublet of doublets resonating at 8 1.01 ppm to a doublet (/ = 5 Hz) and affected the multiplet resonating at 8 0.88 ppm; irradiation of the signal resonating at 8 6.31 ppm affected the multiplet resonating at 8 1.55-1.65 ppm; irradiation of the signal resonating at 8 1.79 ppm affected the multiplet resonating at 8 1.55-1.65 ppm and collapsed the doublet resonating at 8 0.95 ppm to a singlet; irradiation of the signal resonating at 8 2.60 ppm affected the multiplet resonating at 8 1.55-1.65 ppm and affected the signal resonating at 8 1.11 ppm. A repetition of this procedure using epoxide 273 (2.2124 g, 6.63 mmol) without intermediate purification afforded an oil, after benzoylation, that was purified by flash chromatography on silica using ethyl acetate/petroleum ether (1:9) as eluent to give an inseparable mixture of dienes 270 and 285 (1:1, 551 mg, 27%) as an oil. Further elution afforded benzoate 275 (77 mg, 3%) as an oil, followed by alcohol 265 (1.289 g, 60%) as a white solid. Dienes 270 (300 M H z ,  and 285  (1:1): ir v  CDCI3) 8:  m a x  (thin film):  2955, 2867, 1626 c m " ; *H nmr 1  0.67 (6H, s), 0.95 (3H, s), 1.00 (3H, s), 1.02 (3H, d,  / = 7 Hz), 1.05 (3H, d, / = 7 Hz), 1.08 (3H, d, / = 7 Hz), 1.14 (3H, s), 1.15 (3H, s), 1.22 (3H, d, / = 7 Hz), 1.24 (IH, m), 1.28 (3H, d, / = 7 Hz), 1.32-1.60 (5H, m), 1.70 (6H, br s, wh/2 = 4 Hz), 1.73 (3H, br s, wh/ = 4 Hz), 2.45-2.75 (6H, m), 3.19-3.38 2  (4H, m), 3.44 (IH, d, / = 11 Hz), 3.51 (IH, d, / = 11 Hz), 3.75 (2H, br d, / = 11 Hz), 5.93 (2H, s) ppm; mass spectrum m/z: 304 (M+), 276, 218, 191 (100).  229  3.3.5. (6R, 7R) 8-(l-MethyIethyl)-2,6,7-trimethylbicydo[4.3.0]nona-l(2),8-dien-3-one (286).  To a mixture of dienes 284 and 285 (1:1, 409 mg, 1.34 mmol) in tetrahydrofuran (10 ml) was added 2 M hydrochloric acid (10 ml). The reaction was stirred at room temperature for 19 h, then diluted with diethyl ether. The aqueous layer was neutralised with 3M sodium hydroxide and extracted with diethyl ether (20 ml). The combined organic extracts were washed with water (20 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a yellow oil.  Purification by flash chromatography using ethyl  acetate/petroleum ether (1:9) as eluent gave diene 286 (214 mg, 73%) as a colourless oil: ir Vmax (thin film): 2966, 1650, 1626, 1595 cm" ; U nmr (400 MHz, CDCI3) 6: 1.05 1  l  (3H, s, C6-CH3), 1.07 (3H, d, J = 1 Hz), 1.09 (3H, d, / = 7 Hz), 1.20 (3H, d, / = 7 Hz), 1.79 (3H, s, C2-CH ), 1.85 (IH, td, J = 13, 5 Hz, C5-H ), 2.01 (IH, ddd, 3  ax  / = 13, 5, 2 Hz, C5-H ), 2.45 (IH, ddd, / = 13, 5, 2 Hz, C4-He ), 2.50-2.63 (3H, m), eq  q  6.20 (IH, br s, wh/ = 5 Hz, C9-H) ppm; mass spectrum m/z: 218 (M+), 203, 175 (100). 2  *H nmr decoupling experiments: irradiation of the signal resonating at 5 2.01 ppm simplified the signal resonating at 8 2.45 ppm to a doublet of doublets (/ = 13, 5 Hz), affected the signal resonating at 8 1.85 ppm and affected the multiplet resonating at 8 2.502.63 ppm; simultaneous irradiation of the signals resonating at 8 1.05, 1.07 and 1.09 ppm affected the multiplet resonating at 8 2.50-2.63 ppm; irradiation of the signal resonating at  230  8 6.20 ppm affected the multiplet resonating at 8 2.50-2.63 ppm; irradiation of the signal resonating at 8 1.20 ppm affected the multiplet resonating at 8 2.50-2.63 ppm.  3.3.6. (IS, 6R, 7S, 9S, 10R)  3-(2,5-Dioxolane)-l,10-epoxy-6-methyl-  9-(l-methylethyl)-tricyclo[4.4.0.0 > ]decane (290). 7  f  9  1  To a solution of ketal 271 (965.4 mg, 3.886 mmol) in dichloromethane (25 ml) at 47  0 ° C was added anhydrous sodium phosphate dibasic (2.00 g, 14.1 mmol), 3-f-butyl-4hydroxy-5-methylphenylsulphide (32 mg, 0.089 mmol) and mefa-chloroperoxybenzoic acid (1.13 g, 6.55 mmol).  The reaction was stirred for 30 min, then quenched with  saturated sodium bicarbonate (25 ml). The organic layer was washed with water (25 ml), dried (over MgSC>4), filtered and evaporated in vacuo to give a colourless oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent recovered starting material (15.3 mg, 2%). Further elution gave 290 (967.5 mg, 94%) as a colourless oil: [oc] = +27° (CHC1 , c = 0.11); ir v ax (thin film): 3079, 2956, 2874 D  3  m  cm- ; H nmr (300 MHz, CDCI3) 8: 0.22 (IH, m, C8-H), 1.00 (3H, s, C6-CH3), 1.01 1  A  (3H, d, / = 7 H z , C I I - C H 3 ) , 1.02 (3H, d, / = 7 Hz, C I I - C H 3 ) , 1.11 (2H, m, C8-H,  C7-H), 1.35 (IH, dd, / = 14, 2 Hz, C2-H ), 1.60 (3H, m, C4-He , C5-He , Cll-H), eq  q  q  1.67 (IH, td, J = 13, 4 Hz, C5-H ), 1.83 (IH, td, / = 13, 4 Hz, C4-H ), 1.98 (IH, d, ax  ax  Synthesised in 3 steps from thujone in 43% overall yield as described in reference 5a.  231  J = 14 Hz, C2-H ), 3.29 (IH, s, C10-H), 3.92 (4H, br s, wh/ = 3 Hz, ketal protons) ax  ppm; mass spectrum m/z:  2  264 (M ), 249, 246, 99 (100); high resolution mass +  measurement: calculated for C16H24O3: 264.1726; found: 264.1726; anal: calculated for C16H24O3: C 72.69; H 9.15; found: C 72.70; H 9.20. *H nmr decoupling experiments: irradiation of the doublet of doublets resonating at 8 1.35 ppm collapsed the doublet resonating at 8 1.98 ppm to a singlet and affected the multiplet resonating at 8 1.60 ppm; irradiation of the triplet of doublets resonating at 8 1.83 ppm affected the signal resonating at 8 1.67 ppm and affected the multiplet resonating at 8 1.60 ppm; irradiation of the signal resonating at 8 1.98 ppm simplified the doublet of doublets resonating at 8 1.35 ppm to a doublet (7 = 2 Hz); irradiation of the signal resonating at 8 0.22 ppm affected the multiplet resonating at 8 1.11 ppm.  3.3.7. ( I S , 6 R , 7 S , 9S)  3-(2,5-Dioxolane)-l-hydroxy-  6-methyl-9-(l-methyIethyl)-tricyclo[4.4.0.0 > ]decane 7  9  (245).  To a solution of electrons, formed by dissolving lithium (100 mg, 14.4 mg-atoms) in anhydrous ethylamine (10 ml) at -78°C, was added dropwise epoxide 290 (205 mg, 0.775 mmol) in anhydrous tetrahydrofuran (15 ml) at a rate so as not to discharge the colour. Ethanol (5 ml) was added to quench the reaction followed by water (50 ml). The aqueous layer was extracted with diethyl ether (2 x 50 ml), neutralised with 2M hydrochloric acid and extracted with a further portion of diethyl ether (50 ml). The  232  combined organic layers were washed with water (20 ml), dried (over MgSC<4), filtered and evaporated in vacuo to leave yellow crystals (241 mg).  Purification by flash  chromatography using ethyl acetate/petroleum ether (1:4) as eluent afforded alcohol 245 (180.0 mg, 87%) as a white solid: mp 92-93.5°C (sublimed); [cc] = -6.0° (CHC1 , D  c = 2.65); ir v  m a x  3  3605, 2958, 2872 cm-1; * H nmr (400 M H z ,  (CHCI3):  CDCI3)  0.22 (IH, ddd, / = 9, 4, 1.5 Hz, C 8 - I W , 0.86 ( 3 H , d, / = 7 Hz, C I I - C H 3 ) , 0.88 dd, /  = 9, 4 Hz, CS-Hendo),  0.96  ( 3 H , d, /  = 7 Hz, C I I - C H 3 ) ,  1.00  5:  (IH,  (IH, br s,  wh/2 = 30 Hz, exchangeable, OH), 1.03 ( 3 H , s, C 6 - C H 3 ) , 1.18 (IH, t, J = 4 Hz, C7-  H), 1.36 (IH, septet, / = 7 Hz, C l l - H ) , 1.48 (IH, d, / = 14 Hz, C 2 - H ) , 1.54 (IH, d, ax  / = 14 Hz, C10-pH), 1.55-1.70 ( 4 H , m), 1.90 (IH, dd, / = 14, 2 Hz, C2-He ), 2.48 q  (IH, br d, / = 14 Hz, C10-aH), 3.85-4.00 ( 4 H , m, ketal protons) ppm; (400 M H z , d -pyridine) 5  8: 0.30 (IH, ddd, J = 9, 4, 1.5 Hz, CS-H ), exo  C I I - C H 3 ) , 0.98 (IH, dd, J = 9, 4 Hz, CS-H ), endo  1.31  (3H,  s,  C6-CH3),  1.39  0.93 ( 3 H , d, / = 7 Hz,  1.06 ( 3 H , d, / = 7 Hz, CII-CH3),  (IH, septet, J = 1 Hz, C l l - H ) , 1.60 (IH, dt, J = 14, 3 Hz,  C 5 - H ) , 1.67 (IH, dq, / = 14, 3 Hz, C4-He ), 1.76 (IH, td, / = 14, 3 Hz), 1.85-1.90 q  e q  (2H, m), 1.93 (IH, d, / = 14 Hz, C10-pH), 2.06 (IH, d, / = 14 Hz, C 2 - H ) , 2.25 (IH, ax  dd, / = 14, 2 Hz, C 2 - H ) , 2.83 (IH, br d, / = 14 Hz, ClO-aH), 3.80-3.95 ( 4 H , m, eq  ketal protons) ppm; mass spectrum m/z:  266 ( M ) , 251, 248, 186, 99 (100); high +  resolution mass measurement: calculated for  C16H26O3:  266.1882; found: 266.1884;  anal.: calculated for C 1 6 H 2 6 O 3 : C 72.14; H 9.84; found: C 72.09; H ! H nmr decoupling experiments  (CDCI3):  9.95.  irradiation of the signal resonating at 5 0.22  ppm simplified the doublet of doublets resonating at 5 0.88 ppm to a doublet (7 = 4 Hz) and simplified the triplet resonating at 5 1.18 ppm to a doublet (7 = 4 Hz); irradiation of the septet resonating at 8 1.36 ppm collapsed the doublets resonating at 8 0.86 and 0.96 ppm to singlets; irradiation of the doublet of doublets resonating at 8 1.90 ppm collapsed the  233  doublet resonating at 8 1.48 ppm to a singlet and affected the multiplet resonating at 8 1.551.70 ppm; irradiation of the triplet resonating at 8 1.18 ppm simplified the doublet of doublets resonating at 8 0.88 ppm to a doublet (/ = 9 Hz) and simplified the signal resonating at 8 0.22 ppm to a doublet of doublets (/ = 9, 1.5 Hz); irradiation of the broad doublet resonating at 8 2.48 ppm simplified the signal resonating at 8 0.22 ppm to a doublet of doublets (/ = 9,4 Hz) and affected the multiplet resonating at 8 1.55-1.70 ppm. !H nmr decoupling experiments (ds-pyridine): irradiation of the doublet resonating at 8 2.25 ppm affected the signal resonating at 8 2.06 ppm; irradiation of the signal resonating at 8 2.06 ppm collapsed the signal resonating at 8 2.25 ppm to a singlet. Single crystal X-ray structure data available in Appendix 2.  3.3.8. 9-Acetoxy-3-(2,5-dioxoIane)9-(l-methylethyl)cycIodec-6-en-l-one (246). 7  Method A: Ten dry 50 ml Pyrex® round-bottomed flasks were each charged with alcohol 245 (200 mg, 0.75 mmol), recrystallised lead tetraacetate (1.45 g, 3.27 mmol) and glacial acetic acid (1.3 ml, 23 mmol) in anhydrous toluene (11 ml). These solutions were irradiated for 80 min with ultraviolet light produced from a Hanovia high pressure mercury lamp. The reactions were combined and diluted with diethyl ether (250 ml) and water (200 ml). The aqueous layer was neutralised with sodium hydroxide solution and  234  extracted with diethyl ether (50 ml). The combined organic extracts were washed with saturated sodium bicarbonate solution (200 ml), water (200 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a dark oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:4) as eluent recovered alcohol 245 (842.8 mg, 42%).  Further elution afforded a mixture of isomeric alkenes 294, 296 and 295  (294:296:295 = 52:28:20, 415.2 mg, 21%) as a pale yellow oil. Further elution using ethyl acetate/petroleum ether (2:3) as eluent afforded acetate 246 (590.0 mg, 24%) as a pale yellow viscous oil. Mixture of alkenes 294, 296 and 295: ir v  m a x  (CHCI3): 2977, 1725 (br) cm" ; ! H nmr 1  (400 M H z , C6D6) [partial signals corresponding to 296] 8: 0.85 (6H, d, J = 1 Hz, C11-(CH ) ), 1.51 (3H, s, C6-CH3), 5.30 (IH, t, / = 7 Hz, C7-H), 5.44 (IH, s, C10-H) 3  2  ppm, [partial signals corresponding to 294] 8: 1.50 (3H, s, C6-CH3), 1.59 (3H, d, J = 7 Hz, C11-CH ), 1.62 (3H, d, J = 1 Hz, CII-CH3), 5.21 (IH, t, J = 7 Hz, C7 H) ppm, V  3  [partial signals corresponding to 295] 8: 0.95 (6H, d, / = 7 Hz, C11-(CH )2), 1.54 (3H, 3  s, C6-CH3), 5.84 (IH, br s, wh/ = 8 Hz), 5.92 (IH, br s, wh/ = 8 Hz) ppm; mass 2  spectrum m/z:  2  264 (M+), 246, 222, 86 (100). Further data representing 294 and 296 are  given in sections 3.3.10. and 3.3.9., respectively. Acetate 246: [ a ] (300 M H z ,  D  = +14° (CHCI3, 1.00); ir v  C D C I 3 ) 8:  0.93  (3H,  m a  x (CHCI3): 2973, 1726 cm" ; *H nmr  d, / = 7 Hz, C I I - C H 3 ) , 0.94  1  (3H,  d, / = 7 Hz,  CII-CH3), 1.70 (3H, s, C6-CH3), 1.80 (IH, m), 1.95-2.10 (3H, m), 2.15 (3H, s, OCOCH3), 2.33 (4H, m), 2.75 (IH, br s, wh = 24 Hz), 2.86 (IH, septet, J = 7 Hz, /2  Cll-H), 3.11 (IH, br d, / = 14 Hz), 3.85-4.00 (4H, m, ketal protons), 5.33 (IH, t, / = 8 Hz, C7-H) ppm; (400 MHz, dg-toluene, 80°C) 8: 0.76 (3H, d, J = 1 Hz, CII-CH3),  0.84 (3H, d, / = 7 Hz, CII-CH3), 1.54 (3H, s, C6-CH3), 1.87 (IH, m), 1.90 (3H, s,  OCOCH3), 1.97 (IH, br d, / = 12 Hz), 2.03 (IH, dd, / = 14, 2 Hz), 2.19 (IH, dd,  235  / = 14, 10 Hz, C8-H), 2.22 (IH, d, / = 14 Hz, C10-H), 2.23 (IH, m), 2.42 (IH, dd, / = 14, 5 Hz, C8-H), 2.54 (IH, d, J = 17 Hz, C2-H), 2.66 (IH, d, / = 17 Hz, C2-H), 2.85 (IH, septet, / = 7 Hz, Cll-H), 3.07 (IH, d, J = 14 Hz, C10-H), 3.45-3.57 (4H, m, ketal protons), 5.36 (IH, dd, / = 10, 5 Hz, C7-H) ppm,  1 3  C nmr (75 MHz, CDC1 ) 5: 3  16.4, 17.6, 18.6 (br), 22.5, 31.1, 31.6 (br), 33.7, 45.2 (br), 51.4, 63.7, 64.4, 91.3 (br), 108.6, 123.5, 136.7 (br), 170.9, 202.6 (br) ppm; mass spectrum m/z: 324 (M ), 296, +  281, 86 (100); high resolution mass measurement: calculated for C18H28O5: 324.1937; found: 324.1933. *H nmr decoupling experiments (ds-toluene, 80°C): irradiation of the doublet resonating at 8 3.07 ppm collapsed the signal resonating at 8 2.22 ppm to a singlet; irradiation of the doublet of doublets resonating at 5 5.36 ppm simplified the doublet of doublets resonating at 5 2.42 ppm to a doublet (/ = 14 Hz) and simplified the doublet of doublets resonating at 8 2.19 ppm to a doublet (J = 14 Hz). l  H nmr nOe difference experiments (d8-toluene, 80°C): irradiation of the septet resonating  at 8 2.85 ppm enhanced the doublets resonating at 8 0.76 and 0.84 ppm and enhanced the singlet resonating at 8 1.90 ppm; irradiation of the singlet resonating at 8 1.90 ppm enhanced the doublets resonating at 8 0.76 and 0.84 ppm and enhanced the septet resonating at 8 2.85 ppm; irradiation of the singlet resonating at 8 1.54 ppm enhanced the doublet of doublets resonating at 8 2.19 ppm and enhanced the doublet resonating at 8 2.54 ppm.  Method B: A solution of alcohol 245 (187.6 mg, 0.71 mmol), recrystallised lead tetraacetate (776 mg, 1.7 mmol) and iodine (337 mg, 1.3 mmol) in anhydrous benzene (30 ml) was irradiated with ultraviolet light (hv > 300 nm), generated by a Hanovia high pressure lamp  236  fitted with a Pyrex® filter, for 2 h. The reaction was filtered, diluted with diethyl ether (30 ml), washed with water (30 ml), dried (over MgS04), filtered and evaporated in vacuo to leave a dark oil. Purification by flash chromatography using ethyl acetate/ petroleum ether (3:7) as eluent afforded a mixture of isomeric alkenes 294 and 295 (294:295 = 85:15,19.7 mg, 10%) as a pale yellow oil. Further elution afforded acetate 246 (61.4 mg, 28%) as a pale yellow viscous oil.  3.3.9. 3-(2,5-DioxoIane)-9-(l-methylethyI)cyclodec-6,9-dien-l-one  (296).  7  To a solution of acetate 246 (169.1 mg, 0.521 mmol) in anhydrous methanol (10 ml) at - 2 0 ° C was added sodium methoxide (0.8 M solution in methanol, 5 ml, 0.4 mmol). The reaction was warmed to room temperature and stirred for 20 h. Ammonium chloride solution (5 ml) was added and the organic solvent evaporated in vacuo. The residue was dissolved in diethyl ether (25 ml) and water (25 ml). The aqueous layer was extracted with diethyl ether (2 x 25 ml). The combined organic layers were dried (over MgSCU), filtered and evaporated in vacuo to leave a yellow oil (158 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent afforded enone 296 (70.6 mg, 51%) as a colourless oil: ir Vmax (CHCI3): 2940, 1680, 1645 cm- ; uv ?im 1  ax  (MeOH):  236 nm (e = 3700); H nmr (400 M H z , CDCI3) 5: 1.07 (6H, d, / = 7 Hz, C l l l  ( C H ) ) , 1.54 ( 3 H , s, C 6 - C H 3 ) , 1.70 (IH, m), 1.90 (IH, br s, wh/ = 44 Hz), 2.05 3  2  2  (IH, br s, wh/2 = 48 Hz), 2.25 ( I H , septet, / = 7 Hz, C l l - H ) , 2.30 (IH, br s,  237  wh/2 = 40 Hz), 2.60 (IH, br s, wh/ = 40 Hz), 2.74 (IH, br s, wh/ = 52 Hz), 2.94 2  2  (IH, br s, wh/ = 36 Hz), 3.19 (IH, br s, w h / = 5 6 H z ) , 2  2  3.92 (4H, br s,  wh/ = 28 Hz, ketal protons), 5.39 (IH, t, / = 7 Hz, C7-H), 5.58 (IH, s, C10-H) ppm; 2  !3C nmr (75 M H z , CDC1 ) 8: 16.0, 21.0, 21.2, 29.9, 30.6, 33.7, 35.9, 51.5, 63.9, 3  64.4, 109.3, 124.7, 125.2, 137.3, 150.2, 203.6 ppm; mass spectrum m/z: 264 (M+), 246, 160, 143, 99 (100); high resolution mass measurement: calculated for C 1 6 H 2 4 O 3 : 264.1725; found: 264.1725. Partial *H nmr data recorded in C^Ds are given in section 3.3.8. *H nmr nOe difference experiment: irradiation of the singlet resonating at 8 5.52 ppm enhanced the septet resonating at 8 2.25 ppm and enhanced the doublet resonating at 8 1.07 ppm.  3.3.10. 3-(2,5-Dioxolane)-9-(l-methyIethylidene)cyclodec-6-en-l-one (294) and  3-(2,5-dioxolane)-9-(l-methylethyl)cyclodec-6,8-dien-l-one 7  (295).  7  A solution of alcohol 245 (353.0 mg, 1.33 mmol), iodobenzene diacetate (971.5 mg, 3.02 mmol), iodine (1.0133 g, 3.99 mmol) and sodium acetate (1.0723 g, 13.1 mmol) in toluene (50 ml) was irradiated with ultraviolet light (hv > 300 nm), generated by a Hanovia high pressure lamp fitted with a Pyrex® filter, for 30 min. The reaction was filtered and the solvent evaporated in vacuo to leave a yellow wax. Purification by flash  238  chromatography using ethyl acetate/petroleum ether (1:4) as eluent afforded alkenes 294 and 295 (294:295 = 88:12, 139.2 mg, 40%) as a pale yellow oil: ir  Vmax (CHCI3):  2933,  1704, 1625 cm- ; *H nmr (400 MHz, CDCI3) [signals corresponding to 294] 8:  1.53  1  (3H, s, C 6 - C H 3 ) , 1.75 (3H, s, CII-CH3), 1.80 (3H, s, CII-CH3), 1.85-1.95 (2H, m), 2.15-2.20 (2H, m), 2.70-3.10 (6H, m), 3.90-4.00 (4H, m, ketal protons), 5.18 (IH, t, 7 = 8 Hz, C7-H) ppm; mass spectrum m/z: 264 ( M ) , 246, 121 (100); high resolution +  mass measurement: calculated for C16H24O3: 264.1726; found: 264.1723. P a r t i a l ^ nmr data recorded in C6E>6 is given in section 3.3.8.  3.3.11. (IS, 2S, 6R, 7S)  7-Acetoxy-2,6-dimethyl-3-(4,4-dimethyI-  2,6-dioxane)-l-hydroxy-9-(l-methylethyl)-bicyclo[4.4.0]dec-8-ene  (297).  OAc  Method A: A solution of alcohol 265 (254.0 mg, 0.788 mmol), iodobenzene diacetate (519 mg, 1.61 mmol) and iodine (409 mg, 1.61 mmol) in anhydrous toluene (25 ml) was irradiated, for 2 h, with ultraviolet light (hv > 300 nm), generated by a Hanovia high pressure mercury lamp fitted with a Pyrex® filter. The organic solution was washed with sodium thiosulphate solution until colourless, water (50 ml), dried (over MgSC<4), filtered and evaporated in vacuo at 10°C to give a dark oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent recovered starting alcohol (79 mg, 31%).  239  Further elution afforded acetate 297 (82.0 mg, 27%) as a pale yellow viscous oil: ir  Vmax  (thin film): 3580, 3448, 2961, 1724, 1677 c n r ; *H nmr (400 MHz, dg-toluene, 95°C) 1  5: 0.55 (3H, s), 0.95 (3H, d, / = 7 Hz), 0.95 (3H, s), 0.97 (3H, d, / = 7 Hz), 1.18 (3H, s), 1.24 (3H, br d, / = 7 Hz), 1.35 (3H, m), 1.65 (IH, m), 1.72 (3H, s), 2.13 (IH, m), 2.27 (IH, q, J = 7 Hz, C2-H), 2.32 (IH, d, / = 18 Hz, C10-H), 2.55 (IH, brd, / = 18 Hz), 3.11 (IH, dd, / = 11, 2 Hz), 3.20 (IH, dd, J = 11, 2 Hz), 3.28 (IH, d, J = 11 Hz), 3.35 (IH, d, J = 11 Hz), 3.53 (IH, br s, wh/ = 50 Hz), 5.40 (IH, br s, wh/ = 10 Hz), 2  5.57 (IH, br s, wh/ = 14 Hz) ppm; 2  2  nmr (75 MHz, C D , 60°C) 5: 6  6  17.1, 20.7,  21.2, 21.3, 22.3, 23.1, 23.7, 28.3, 29.9, 34.9, 36.9, 42.0, 42.1, 46.6, 69.7, 70.2, 74.6, 75.2, 100.7, 117.0, 146.0, 170.0 ppm; mass spectrum m/z: 380 (M ), 362, 337, 320, +  302, 171, 141, 85 (100); high resolution mass measurement: calculated for C22H36O5: 380.2562; found: 380.2558.  Method B: To a solution of alcohol 298 (31.2 mg, 0.092 mmol) and 4-dimethylaminopyridine (DMAP, 20.6 mg, 0.167 mmol) in anhydrous dichloromethane (1.5 ml) was added acetic anhydride (12 ul, 0.13 mmol). The reaction was heated at reflux for 3 days, then cooled and diluted with diethyl ether (20 ml). The reaction was washed with saturated ammonium chloride solution (10 ml), water (10 ml), dried (over MgSQ*), filtered and evaporated in vacuo to give a yellow oil (98.8 mg).  Purification by flash  chromatography using ethyl acetate/petroleum ether (3:17) as eluent afforded acetate 297 (26.4 mg, 75%) as a colourless viscous oil.  240  3.3.12. (IS, 2S, 6R) 2,6-DimethyI-3-(4,4-dimethyl-2,6-dioxane)l-hydroxy-9-(l-methy!ethylidene)-bicyclo[4.4.0]dec-7-ene (297).  Method A: Repetition of the procedure in section 3.3.11 (Method A) with removal of toluene in vacuo at 50°C resulted, after purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent followed by purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent, in recovery of starting alcohol 265 (95 mg, 39%) and the isolation of diene 299 (54 mg, 23%) as a yellow oil: [oc] = -71.6° D  (CHC1 , c = 0.405); uv X 3  max  (MeOH) 239 nm (e = 7900); ir Vmax (CHCI3): 3573, 3462,  2959, 1655 c m ' ; H nmr (400 MHz, C D ) 5: 0.39 (3H, br s, wh/ = 23 Hz) 1.13 1  J  6  6  2  (3H, br s, wh/ = 16 Hz), 1.19 (3H, s), 1.19 (IH, m), 1.35 (2H, br s, wh/ = 44 Hz), 2  2  1.50 (2H, br d, 7 = 12 Hz),  1.56 (2H, br s, wh/ = 30 Hz), 2  1.62 (3H, s,  wh/ = 5 Hz, C I I - C H 3 ) , 1.66 (3H, br d, 7 = 1 Hz, wh/ = 15 Hz, C I I - C H 3 ) , 2.33 2  2  (IH, br s, wh/ = 70 Hz), 2.42 (IH, q, / = 7 Hz, C2-H), 2.81 (2H, m, CIO-H2), 3.13 2  (IH, dd, 7 = 11, 2 Hz), 3.21 (IH, dd, 7 = 11, 2 Hz), 3.30 (IH, d, 7 = 11 Hz), 3.49 (IH, br d, / = 11 Hz, wh/ = 28 Hz), 5.30 (IH, d, 7 = 9 Hz, C7-H), 6.40 (IH, d, 2  7 = 9 Hz, C8-H) ppm; mass spectrum m/z: 320 (M ), 302, 216, 171, 141, 85 (100); +  high resolution mass measurement:  calculated for C20H32O3:  320.2351; found:  320.2356; anal.: calculated for C20H32O3: C 74.96; H 10.06; found: C 75.00; H 10.20.  241  Method B: To a solution of acetate 297 (63.0 mg, 0.17 mmol) in toluene (1 ml) was added sodium bicarbonate (150 mg, 1.79 mmol). After heating at reflux for 24 h, the solvent was evaporated in vacuo. Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent gave diene 299 (27.0 mg, 51%) as a pale yellow oil.  3.3.13. (IS, 2S, 6R) 2,6-Dimethyl-l-hydroxy-9(l-methylethy!idene)-bicyclo[4.4.0]dec-7-en-3-one (300).  To a solution of diene 299 (28.5 mg, 0.089 mmol) in tetrahydrofuran (3 ml) was added 2 M hydrochloric acid (3 ml). After stirring for 5 min, the reaction was washed with sodium bicarbonate solution (5 ml), water (5 ml), dried (over MgSC<4), filtered and evaporated in vacuo to give a yellow oil. Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent gave ketone 300 (5.4 mg, 26%) as a colourless oil: [a]  D  = +9.6° (CHC1 , c = 0.23); uv X 3  max  (MeOH):  246 nm (e = 9900); ir v  m a  x (CHCI3):  3576, 3023, 1708, 1639 cm" ; ' H nmr (400 MHz, CDCI3) 5: 1.15 (3H, d,J = 7 Hz, 1  C2-CH3), 1.36 (3H, s, C6-CH3), 1.57 (IH, s, exchangeable, OH), 1.62 (IH, td, J = 14, 5 Hz, C5-H ), 1.73 (3H, s, CII-CH3), 1.82 (3H, d, J = 1 Hz, CII-CH3), 1.83 (IH, ax  m, C5-H ), 1.91 (IH, d, J = 16 Hz, C10-H), 2.32 (IH, ddd, / = 14, 5, 2 Hz, C4-He ), eq  q  2.54 (IH, td, J = 14, 7 Hz, C4-H ), 2.59 (IH, d, / = 16 Hz, C10-H), 2.94 (IH, q, / = 7 ax  Hz, C2-H), 5.40 (IH, d, / = 9 Hz), 6.49 (IH, d, / = 9 Hz) ppm; mass spectrum m/z: 234  242  (M ), 216, 201, 145, 135; high resolution mass measurement: calculated for C15H22O2: +  234.1619; found: 234.1621.  3.3.14. (IS, 2S, 6R, 7S)  2,6-Dimethyl-3-(4,4-dimethyl-2,6-dioxane)-  l,7-dihydroxy-9-(l-methylethyl)-bicyclo[4.4.0]dec-8-ene (298). OH  To a solution of acetate 297 (65.6 mg, 0.172 mmol) in anhydrous methanol (10 ml) was added sodium methoxide (0.043 M, 9.0 ml, 0.39 mmol). The reaction was stirred for 1 day at room temperature, then quenched by the addition of water (2 ml) and saturated ammonium chloride solution (2 ml). The organic solvent was evaporated in vacuo and the residue diluted with water (20 ml) and diethyl ether (20 ml). The organic layer was washed with water (10 ml), dried (over MgSO.4), filtered and evaporated in vacuo to leave a white residue. Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent gave diol 298 (38.2 mg, 65%) as a white solid. Recrystallisation from ethyl acetate/ petroleum ether gave crystals suitable for X-ray analysis: mp = 148-149°C; [O:]D = 55.3° (CHCI3, c = 0.13); ir v ax (CHCI3): 3600, 3454 (br), 2961 cm" ; *H nmr (400 MHz, 1  m  CDC1 ) [partial signals] 6: 0.72 (3H, br s, wh/ = 26 Hz), 1.03 (3H, d, / = 7 Hz), 1.05 3  2  (3H, s), 1.05 (3H, d, / = 7 Hz), 3.33 (2H, br s, wh/ = 32 Hz), 3.60 (2H, br s, 2  wh/ = 80 Hz), 2  4.50 (IH,  br s, wh/ = 108 H z , C7-H), 2  5.62  (IH,  br s,  wh/ = 26 Hz, C8-H) ppm; (400 MHz, d -toluene, 80°C) 8: 0.46 (3H, s), 0.97 (3H, 2  8  243  d, / = 7 Hz, CII-CH3), 0.98 (3H, d, / = 7 Hz, CII-CH3). 1.01 (3H, s), 1.17 (3H, s), 1.19 (3H, d, / = 7 Hz, C2-CH3), 1.30-1.55 (3H, m), 2.06 (2H, m), 2.17 (IH, q, / = 7 Hz, C2-H), 2.78 (2H, br s, wh/ = 52 Hz, CIO-H2), 3.10 (IH, dd, / = 11, 2 Hz), 3.16 2  (IH, dd, / = 11, 2 Hz), 3.31 (IH, d, / = 11 Hz), 3.40 (IH, d, / =11 Hz), 3.65 (IH, br s, wh/ = 20 Hz, C7-H), 5.46 (IH, br s, wh/ = 10 Hz, C8-H) ppm; mass spectrum m/z: 2  2  338 (M ), 320, 234, 141, 85 (100); high resolution mass measurement: calculated for +  C20H34O4: 338.2457; found: 338.2462; anal.: calculated for C20H34O4: C 70.97; H 10.12; found: C 70.80; H 10.00. *H nmr decoupling experiments (ds-toluene, 80°C): irradiation of the broad singlet resonating at 8 3.65 ppm sharpened the singlet resonating at 8 5.46 ppm; simultaneous irradiation of the signals resonating at 8 0.97, 0.98 and 1.01 ppm affected the multiplet resonating at 8 2.06 ppm; irradiation of the broad singlet resonating at 8 5.46 ppm affected the multiplet resonating at 8 2.06 ppm. Single crystal X-ray structure data available in Appendix 3.  244  3.3.15. (IR, 2R, 6R, 7S, 9S)  l,2-Epoxy-6-methyl-  9-(l-methylethyl)-tricyclo[4.4.0.0 » ]decan-3-one 7  9  (302).  7 8  9  o  o  11  10  Method A: To a solution of enone 288 (1.024 g, 5.01 mmol) in methanol (75 ml) stirred at 48  0°C was added, dropwise and simultaneously, hydrogen peroxide solution (30%, 11.0 ml, 108 mmol) and sodium hydroxide solution (4 M, 7 ml, 28 mmol). The reaction was stirred at 4°C for 7.5 h, warmed to room temperature and stirred for a further 3 h. The reaction was quenched with saturated sodium sulphite (15 ml) and brine (30 ml), then extracted with diethyl ether (3 x 150 ml). The combined organic extracts were washed with water (100 ml), dried (over MgSC»4), filtered and evaporated in vacuo to give a yellow oil (766.0 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded epoxide 302 (642.7 mg, 58%) as a colourless oil: [OC]D = +116° (CHCI3, c = 1.40); ir v ax (thin film): 2975, 2940, 1730 cm" ; *H nmr (300 MHz, 1  m  CDCI3) 8: 0.72 (2H, m, C 8 - H 2 ) , 0.88 (3H, d, / = 7 Hz, CII-CH3), 0.93 (3H, d, / = 7 Hz, CII-CH3), 0.96 (IH, dd, / = 9, 5 Hz, C7-H), 1.09 (3H, s, C 6 - C H 3 ) , 1.25 (IH, septet, J = 7 Hz, Cll-H), 1.80 (IH, d, / = 15 Hz, C10-pH), 1.92 (IH, dd, / = 15, 1.5 Hz, C10-aH), 1.92-2.10 (3H, m, C4-H , C5-H ), 2.58 (IH, td, J = 14, 5 Hz, C4-H ), eq  2  ax  3.03 (IH, s, C2-H) ppm; mass spectrum m/z: 220 (M ), 205, 105 (100); high resolution +  4 8  Enone 288 was obtained in 68% yield (75% based on recovered starting material) from ketone 289 by  the procedure outlined by Kutney (5a).  245  mass measurement:  calculated for C14H20O2: 220.1463; found: 220.1465; anal.:  calculated for C14H20O2: C 76.33; H 9.15; found: C 76.28; H 9.15. *H nmr decoupling experiments: irradiation of the doublet resonating at 8 1.80 ppm affected the signal resonating at 8 1.92 ppm; irradiation of the triplet of doublets resonating at 8 2.58 ppm affected the multiplet resonating at 8 1.92-2.10 ppm.  Method B: To a solution of epoxy alcohol 308 (31.7 mg, 0.143 mmol) in acetone (5 ml) was added dropwise Jones' reagent until an orange colour persisted (ca 65 ml). The reaction was neutralised with sodium bicarbonate solution (1 ml) and extracted with diethyl ether (2 x 10 ml). The combined organic extracts were washed with water (10 ml), dried (over MgSC»4), filtered and evaporated in vacuo to give an oil. Purification by flash chromatography using ethyl acetate/petroleum ether (3:17) as eluent afforded epoxy ketone 302 (24.0 mg, 76%) as a colourless oil.  3.3.16. (IR, 2R, 6R, 7S, 9S) 2,6-DimethyI-l,2-epoxy9-(l-methylethyl)-tricyclo[4.4.0.0 > ]decan-3-one (303). 7  9  To a solution of enone 235 (344.3 mg, 1.58 mmol) in methanol (30 ml) stirred at 0 ° C was added, dropwise and simultaneously, hydrogen peroxide solution (30%, 0.7 ml, 6.9 mmol) and sodium hydroxide solution (4 M, 1 ml, 4 mmol). The reaction was stirred  246  at 4 ° C for 3 days, warmed to room temperature and stirred for a further 3 days. The reaction was quenched with saturated sodium sulphite (15 ml) and brine (30 ml), then extracted with diethyl ether (3 x 150 ml). The combined organic extracts were washed with water (100 ml), dried (over MgSC«4), filtered and evaporated in vacuo to give a yellow oil (403.0 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded epoxide 303 (188.3 mg, 51%) as a colourless oil: [CC]D = +142° (CHC1 , c = 0.85); ir v ax (thin film): 2975, 2940, 1710 cnr*; H l  3  CDCI3) 5:  m  nmr (400 MHz,  0.74 (2H, m, C8-H ), 0.91 (3H, d, / = 7 Hz, CII-CH3), 0.93 (3H, d, 2  / = 7 Hz, CII-CH3), 0.93 (lH,m), 1.08 (3H, s, C6-CH3), 1.23 (IH, septet, J = 7 Hz, C l l - H ) , 1.37 (3H, s, C2-CH ), 1.64 (IH, d, J = 15 Hz, C10-BH), 1.88 (IH, dt, 3  J = 14, 3 Hz), 1.96 (IH, td, / = 14, 4 Hz), 2.10 (IH, dd, / = 15, 1.5 Hz, C10-aH), 2.14 (IH, dt, / = 14, 3 Hz), 2.72 (IH, td, / = 14, 4 Hz, C4-H ) ppm; mass spectrum ax  m/z:  234 (M ), 218, 105 (100); high resolution mass measurement: calculated for +  C14H20O2: 234.1620; found: 234.1620; anal: calculated for C15H22O2: C 76.88; H 9.46; found: C 77.11; H 9.40.  3.3.17. (3S, 6R, 7S, 9S)  3-Hydroxy-6-methyl-9-(l-methylethyl) tricyclo[4.4.0.0 > ]dec-l-ene  (306).  To a solution of enone 288 (1.034 g, 5.06 mmol) and cerium (III)  chloride  7  9  Method A:  heptahydrate (1.778 g, 4.78 mmol) in methanol (12 ml) at 0 ° C was added sodium  247  borohydride (200 mg, 5.26 mmol).  The reaction was allowed to warm to room  temperature and stirred for 15 min. The reaction was quenched by the addition of saturated ammonium chloride solution (20 ml) and extracted with diethyl ether (2 x 30 ml). The organic extracts were washed with water (25 ml), dried (over MgSC>4), filtered and evaporated in vacuo to give a yellow oil (1.1167 g) that was shown by G C to consist of a mixture of two alcohols 307:306 = 7:93. Purification by flash chromatography using ethyl acetate/petroleum ether (3:17) as eluent gave alcohol 307 (30.0 mg, 3%) as a colourless oil. Further elution afforded a mixture of 307 and 306 (120.5 mg, 12%) followed by a pure sample of 306 (890.4 mg, 85%) as a colourless oil. Alcohol 306: [ a ]  D  = +46.4° (CHC1 , c = 2.45); ir v 3  m a x  (thin film): 3327, 2957, 2870,  1675 cm- ; *H nmr (400 MHz, CDCI3) 8: 0.71 (2H, m, C8-H ), 0.86 (3H, d, / = 7 Hz, 1  2  CII-CH3), 0.87 (3H, d, J = 1 Hz, CII-CH3), 0.91 (IH, dd, / = 9, 5 Hz, C7-H), 0.97 (3H, s, C6-CH3), 1.23 (IH, septet, / = 7 Hz, Cll-H), 1.32 (IH, br s, wh/ = 14 Hz, 2  exchangeable, OH), 1.61 (2H, m), 1.83 (IH, m), 2.06 (IH, dd, / = 16, 1 Hz, C10-H), 2.08 (IH, m), 2.29 (IH, d, / = 16 Hz, C10-H), 4.28 (IH, br s, wh/ = 20 Hz, C3-H), 2  5.10 (IH, br s, wh/ = 8 Hz, C2-H) ppm; mass spectrum m/z: 2  173, 163, 145 (100); high resolution mass measurement:  206 (M+), 191, 188,  calculated for C14H22O:  206.1670; found: 206.1667; anal: calculated for C14H22O: C 81.50; H 10.75; found: C 81.51; H 10.70. *H nmr decoupling experiments: irradiation of the broad singlet resonating at 8 4.28 ppm sharpened the singlet resonating at 5 5.10 ppm and affected the signals resonating at 8 1.62 and 2.08 ppm; simultaneous irradiation of the signals resonating at 8 2.06 and 2.08 ppm sharpened the broad singlet resonating at 8 5.10 ppm to a doublet (/ = 3 Hz), simplified the doublet resonating at 8 2.29 ppm to a singlet and affected the signals resonating at 8 1.61, 1.83 and 4.28 ppm.  248  Repetition of this procedure on 288 (622.1 mg, 3.05 mmol) using 2 M hydrochloric acid in the work-up gave, after purification, in addition to 306 (242.0 mg, 39%), the byproducts 311 and 312 (311:312 = 31:69, 129.1 mg, 19%) as an inseparable mixture: ir Vmax (thin film): 3055, 2960, 2815, 1678 cm- ; H nmr (400 M H z , CDC1 ) [partial 1  l  3  signals corresponding to 311] 8: 1.23 (IH, septet, / = 7 Hz, C l l - H ) , 2.29 (IH, d, 7 = 16 Hz, C10-H), 3.32 (3H, s, OCH3), 3.90 (IH, br s, wh/ = 20 Hz, C 3 - a H ) , 2  5.17 (IH, br s, wh/ = 8 Hz, C2-H) ppm; [partial signals corresponding to 312] 8: 0.84 2  (3H, s, C6-CH3), 0.98 (IH, dd, / = 9, 5 Hz), 1.28 (IH, septet, J = 7 Hz, C l l - H ) , 2.12 (IH, dd, 7 = 16, 1 Hz, C10-ccH), 2.35 (IH, d, J = 16 Hz, C10-pH), 3.32 (3H, s, OCH3), 3.61 (IH, br s, wh/ = 11 Hz, C3-pH), 5.20 (IH, br s, wh/ = 8 Hz, C22  2  H) ppm; mass spectrum m/z: 220 (M+), 205, 177, 145 (100).  Method B: To a solution of enone 288 (385.6 mg, 1.89 mmol) in anhydrous tetrahydrofuran (25 ml) at -78°C was added dropwise DIBAL-H (1 M in hexanes, 2.1 ml, 2.1 mmol). After 30 min at -78°C, a further portion of DIBAL-H (1 M, 0.5 ml, 0.5 mmol) was added. After 30 min at -78°C, the reaction was warmed to -10°C and stirred for 15 min. A further portion of DIBAL-H (1 M, 1 ml, 1 mmol) was added to the reaction which was stirred for a further 15 min. To the reaction was added diethyl ether (20 ml) and water (20 ml) followed by 2 M hydrochloric acid (100 ml). The aqueous layer was extracted with diethyl ether (3 x 50 ml). The combined organic layers were washed with water (50 ml), dried (over MgSC«4), filtered and evaporated in vacuo to leave a yellow oil (456 mg). Purification by flash chromatography using ethyl acetate/petroleum ether (1:9) as eluent gave enone 288 (23.8 mg, 6%), alcohol 307 (27.9 mg, 7%) and alcohol 306 (327.3 mg, 84%) as colourless oils.  249  3.3.18. (3S, 6R, 7S, 9S)  3-Acetoxy-6-methyl-9-(l-methylethyl)tricyclo[4.4.0.0 > ]dec-l(2)-ene 7  9  (309).  To a solution of alcohol 306 (31.0 mg, 0.15 mmol) in pyridine (1 ml) was added acetic anhydride (100 ul, 1.06 mmol). After stirring for 20 h, the reaction was diluted with diethyl ether (10 ml) and added to cold 2 M hydrochloric acid (10 ml). The organic layer was washed with 2 M hydrochloric acid (5 ml), water (10 ml), dried (over MgSC>4), filtered and evaporated in vacuo to leave a yellow oil.  Purification by flash  chromatography using ethyl acetate/petroleum ether (1:19) as eluent afforded pure acetate 309  (28.3 mg, 76%) as a colourless oil:  iHnmr  (400 M H z ,  ir v  m a  x (thin film): 2970, 1735, 1690 cm" ; 1  C D C 1 ) 8: 0.71 (2H, m, C8-H ), 0.84 (3H, d, / = 7 Hz, 2  3  CI I-CH3), 0.86 (3H, d, / = 7 Hz, CII-CH3), 0.92 (IH, dd, J = 8, 5 Hz, C7-H), 0.99 (3H, s, C6-CH3), 1.23 (IH, septet, / = 7 H z , Cll-H), 1.65 (IH, m), 1.73 (IH, m), 1.84 (IH, dt, / = 12, 3 Hz, C5-H ), 2.02 (3H, s, OCOCH3), 2.08 (2H, m, C10-H, C4-H), eq  2.31 (IH, d, J = 16 Hz, C10-H), 5.09 (IH, br s, wh/ = 8 Hz, C2-H), 5.37 (IH, m, 2  C3-H) ppm; mass spectrum m/z: 248 (M+), 205, 188, 173, 145 (100); high resolution mass measurement: calculated for C 1 6 H 2 4 O 2 : 248.1776; found: 248.1768.  250  3.3.19. (3R, 6R, 7S, 9S) 3-Acetoxy-6-methyl-9-(l-methylethyl) tricyclo[4.4.0.0 > ]dec-l(2)-ene 7  9  (310).  A solution of diethyl azodicarboxylate (DEAD, 137.2 mg, 0.79 mmol) in anhydrous tetrahydrofuran (1 ml) was added to a solution of alcohol 306 (101.8 mg, 0.49 mmol), triphenylphosphine (200.8 mg, 0.77 mmol) and glacial acetic acid (40 (il, 0.7 mmol) in anhydrous tetrahydrofuran (2 ml). After stirring for 16 h, the solvent was evaporated in vacuo to leave a viscous oil. Purification by flash chromatography using ethyl acetate/ petroleum ether (1:19) as eluent afforded acetate 310 (77.0 mg, 63%) contaminated by 309 (15%): irvmax (thin film): 2940, 1730, 1680 cm" ; *H nmr (400 MHz, CDC1 ) 1  3  [signals corresponding to 310] 8: 0.70 (2H, m, C8-H ), 0.86 (3H, d, / = 7 Hz, C l l 2  CH3), 0.86 (3H, s, C6-CH3), 0.88 (3H, d, J = 7 Hz, CII-CH3), 0.92 (IH, m, C7-H), 1.28 (IH, septet, / = 7 Hz, Cll-H), 1.59 (IH, m), 1.66 (IH, m), 1.75 (IH, m), 2.01 (3H, s, OCOCH3), 2.02 (IH, m), 2.10 (IH, dt, / = 16, 3 Hz, C10-ccH), 2.34 (IH, d, J =  16 H z , C10-f3H),5.10  (IH, br s, wh/ = 6 H z , C2-H), 5.13 (IH, 2  br s,  wh/2 = 12 Hz, C3-H) ppm; mass spectrum m/z: 248 (M+), 205, 188, 173, 145 (100).  251  3.3.20. (IR, 2R, 3S, 6R, 7S, 9S)  l,2-Epoxy-3-hydroxy-6-methyl-  9-(l-methyIethyl)tricyclo[4.4.0.0 > ]decane (308). 7  9  To a solution of alcohol 306 (92.0 mg, 0.446 mmol) and vanadyl acetylacetonate (5.0 mg, 0.019 mmol) in dichloromethane (2 ml) was added, dropwise, rm-butylhydroperoxide (70%, 100 ul, 0.73 mmol). After stirring at room temperature for 3 h, the reaction was quenched by the addition of saturated ammonium chloride solution (5 ml). After the addition of dichloromethane (20 ml) and water (10 ml), the organic layer was dried (over MgS04), filtered and evaporated in vacuo. Purification by flash chromatography using ethyl acetate/petroleum ether (3:7) as eluent afforded 308 (79.4 mg, 80%) as a white solid: mp 77-80°C (ether/petroleum ether); [ a ] = +53.5° (CHCI3, c = 0.535); ir D  Vmax (CHCI3): 3590, 3450 (br), 2970, 2890 cm" ; *H nmr (400 MHz, CDCI3) 5: 0.65 1  (2H, m, C8-H2), 0.86 (IH, dd, / = 9, 5 Hz, C7-H), 0.89 (3H, d, / = 7 H z , CII-CH3), 0.90 (3H, s, C6-CH3), 0.91 (3H, d, J = 7 Hz, CII-CH3), 1.21 (IH, septet, 7 = 7 Hz, C l l - H ) , 1.30 (2H, m), 1.59 (IH, dd, J = 12, 3 Hz), 1.68 (IH, m), 1.74 (IH, d, 7 = 14 Hz, C10-H), 1.83 (IH, d, J = 14 Hz, C10-H), 1.93 (IH, br s, wh/ = 36 Hz, exchangeable, OH), 3.14 (IH, br s, wh/ = 4 Hz, C2-H), 3.95 (IH, 2  2  ddd, / = 8, 6, 1.5 Hz, C3-H) ppm; mass spectrum m/z: 222 (M+), 204, 189, 175, 161, 133, 105, 91 (100); high resolution mass measurement:  calculated for C14H22O2:  222.1619; found: 222.1619; anal.: calculated for C14H22O2: C 75.65; H 9.97; found: C 75.59; H 9.92. Single crystal X-ray structure data available in Appendix 4.  252  4.  References.  1.  D . WHTTTAKER and D . V . BANTHORPE. Chem. Rev. 72, 305 (1972).  2.  a) J . P. KUTNEY. J . BALSEVICH, R. CARRUTHERS, A . MARKUS. M . J . MCGRATH, R. N.YOUNG, and B. R. WORTH. Bioorg. Chem. 7, 289 (1978); b) J . P. KUTNEY, M . J . MCGRATH, R. N . YOUNG, andB. R. WORTH. Can. J. Chem. 57, 3145 (1979).  3.  a) J . P. KUTNEY, M . K . CHOUDHURY, J . M . DECESARE, H. JACOBS, A. K . SINGH, and B. R. WORTH. Can. J . Chem. 59, 3162 (1981); b) A . BECALSKI, W . R. CULLEN, M . D . FRYZUK, G . HERB, B. R. JAMES, J . P. KUTNEY, K . PIOTROWSKA, and D . TAPIOLAS. Can. J . Chem. 6 6 , 3108 (1988) .  4.  C. CARVALHO, W . R. CULLEN, M . D . FRYZUK, H. JACOBS, B. R . JAMES, J . P. KUTNEY, K . PIOTROWSKA, and V . K . SINGH. Helv. Chim. Acta, 72, 205 (1989) .  5.  a) J . P. KUTNEY, J . BALSEVICH, and P. GRICE. Can. J . Chem. 5 8 , 2641 (1980); b) J . P. KUTNEY and A. K . SINGH. Can. J . Chem. 60, 1842 (1982); c) J . P. KUTNEY and A . K . SINGH. Can. J . Chem. 61, 1 111 (1983); d) J . P. KUTNEY and A . K . SINGH. Can. J . Chem. 62, 1407 (1984); e) J . P. KUTNEY and A. K . SINGH. Can. J . Chem. 62, 2813 (1984).  6.  a) J . P. KUTNEY, P. GRICE, K . O. PASCOE, K . PIOTROWSKA, S. J . RETTIG, J . SZYKULA, and J . TROTTER. Helv. Chim. Acta, 65, 2345 (1982); b) J . P. KUTNEY, P. GRICE, K . PIOTROWSKA, S. J . RETTIG, J . SZYKULA, J . TROTTER, and L . V . CHU. Helv. Chim. Acta, 66,1821 (1983); c) J . P. KUTNEY, P. GRICE, K . O. PASCOE, K . PIOTROWSKA, S. J . RETTIG, J. SZYKULA, and J . TROTTER. Helv. Chim. Acta, 66, 1806 (1983).  7.  J . P. KUTNEY, K . PIOTROWSKA, Y - H . CHEN, K-P. N . CHENG, Z . G A O , S. J . RETTIG. Can. J . Chem. 68, 1698 (1990).  8.  V . HACH, R. W . LOCKHART, E . C. MCDONALD, and D . M . CARTLIDGE. Can. J . Chem. 4 9 , 1762 (1971).  9.  S. P. ACHARYA, H. C. BROWN, A . SUZUKI, S. NOZAWA, and M . ITOH. J . Org. Chem. 34, 3015 (1969).  10.  Fragrance Chemistry, The Science of the Sense of Smell, edited by E . T .  11.  a) J . E . AMOORE, J . W . JOHNSTON and M . RUBIN. Scientific American, 210(2), 42 (1964).  THEIMER, Academic Press (1982) and references therein.  253  b) A recent paper describes the discovery of 18 genes responsible for coding in the odour receptor proteins: L. B U C K and R. A X E L . Cell, 6 5 , 175 (1991). Summarised in Time, 15th April 1991. 12.  J . V . BRAUN. Ber. Dtsh. Chem. Ges. 7 0 , 1250 (1937).  13.  Perfumery and Flavouring Synthetics. 2nd Edition by P. Z. BEDOUKIAN,  14.  G. OHLOFF. R. H. Wright Award Public Lecture, Simon Fraser University (1989)  15.  a) G. F. RUSSELL and J . I. HILLS. Science, 172, 1043 (1971); b) L. FRIEDMAN and J . G. MILLER. Science, 172, 1044 (1971).  16.  E. DEMOLE, P. ENGGIST, U. SAUBERLI, M . STOLL, and E. KOVATS. Helv. Chim. Acta, 5 3 , 541 (1970).  17.  W. RENOLD, R. NAF-MULLER, U. KELLER, B. WILLHALM and G. OHLOFF. Helv. Chim. Acta, 5 7 , 1301 (1974).  18.  E. DEMOLE and D. BERTHET. Helv. Chim. Acta, 54, 681 (1971).  19.  a) C. FEHR and J . GALINDO. J . Am. Chem. Soc. 110, 6909 (1988).  20.  a) E. KOVATS, E. DEMOLE, G. OHLOFF, and M . STOLL. U. S. Patent 4,187,863 (1980). For a related synthesis see: b) S. TORE, K . UNEYAMA, and H. ICHIMURA. J . Org. Chem. 44, 2292 (1979).  21.  a) R. L. SNOWDEN, S. M . LINDER, B. L. MULLER, and K . H. SCHULTE-ELTE. Helv. Chim. Acta, 7 0 , 1858 (1987). b)R. L. SNOWDEN and S. M . LINDER. Helv. Chim. Acta, 71, 1587 (1988).  22.  a) C. FEHR and J . GALINDO. Helv. Chim. Acta, 69, 228 (1986); b) G. BUCHI and H. WUEST. Helv. Chim. Acta, 54, 1767 (1971). For a related synthesis using thioesters see: c) P. GOSSELIN. Tetrahedron, 44, 1979 (1988).  23.  a) K . H. SCHULTE-ELTE, H. STRICKLER, F. GAUTSCHI, W. PICKENHAGEN, M . GADOLA, J . LIMACHER, B. L. MULLER, F. WUFFLI, and G. OHLOFF. Liebigs. Ann. Chem. 484 (1975). b) Y. NAKATANI, K . KUBOTA, R. TAHARA, and Y. SHIGEMATSU. Agr. Biol. Chem. 3 8 , 1351 (1974). For related syntheses see: a) R. PELLICIARI, E. SISANI, and R. FRTNGUELLI. Tetrahedron Lett. 4039 (1980); b) H. J . Liu, H. K . HUNG, and G. L. MHEHE. Tetrahedron Lett. 4129 (1976);. c) H. J . Liu, H. K . HUNG, G. L. M H E H E , and M . L. D . WEINBERG. Can. J . Chem. 5 6 , 1368 (1978);  Elsevier Publishing (1967).  254  24.  L . COLUMBI, A. BOSSHARD, H. SCHINZ and C. F. SEIDEL. Helv. Chim. Acta, 3 4 , 265 (1951).  25.  a) S . ISOE, S. KATSUMURA, and T. SAKAN. Helv. Chim. Acta, 5 6 , 1514 (1973). For a related synthesis see: b) W . Y . LEE, J . M . PARK, K . H. NAM, S. Y . JANG, and O. S. PARK. Daehan Hwahak Hwoejee, 3 0 , 483 (1986).  26.  W . G . DAUBEN, A. P. KOZIKOWSKI and W . T. ZIMMERMAN. Tetrahedron Lett. 515 (1975);.  27.  a) E. AZZARI, C. FAGGI, N. GELSOMINI, and M . TADDEI. J . Org. Chem. 5 5 , 1106(1990); For related syntheses see: b) A. J . A. VAN DER WEERDT, W . APELDOORN, M . BOELENS, W . M . B. KONST, and R. TER HEIDE. J . Royal Netherlands Chem. Soc. 100, 447 (1981); c) G . BUCHI and J . C. VEDERAS. J . Am. Chem. Soc. 9 4 , 9128 (1972); d) C. DUPUY and J . L. LUCHE. Tetrahedron, 4 5 , 3437 (1989); e) K . H. SCHULTE-ELTE, B. L. MULLER, and G . OHLOFF. Helv. Chim. Acta, 5 6 , 310 (1973).  28.  F. N A F (sic) and R. DECORZANT. Helv. Chim. Acta, 57, 1317 (1974).  29.  For a related asymmetric synthesis see a) S. YAMADA, M . SHIBASAKI, and S. TERASHIMA. Tetrahedron Lett. 377 (1973);. b) M . SHTBASAKI, S. TERASHIMA, and S. YAMADA. Chem. Pharm. Bull. 23, 279 (1975).  30.  C. FEHR and J . GALINDO. J . Org. Chem. 5 3 , 1830 (1988).  31.  F. NAEF and R. DECORZANT. Tetrahedron, 42, 3245 (1986).  32.  a) E. KOVATS, E. DEMOLE, G . (1975) ; b) E. KOVATS, E. DEMOLE, G . (1976) ; c) E. KOVATS, E. DEMOLE, G . (1980); d) K . H. SCHULTE-ELTE. U . S. e) E. KOVATS, E. DEMOLE, G . (1976).  OHLOFF, and M . STOLL. U . S. Patent 3,928,456 OHLOFF, and M . STOLL. U . S. Patent 3,975,310 OHLOFF, and M . STOLL. U . S. Patent 4,226,892 Patent 3,892,809 (1975). OHLOFF, and M . STOLL. U . S. Patent 3,931,326  33.  R. H. EASTMAN and A. V . WINN. J . Am. Chem. Soc. 8 2 , 5908 (1960).  34.  C. L. STEVENS and A. J . WEINHEIMER. J . Am. Chem. Soc. 8 0 , 4072 (1958).  35.  Modern Synthetic Reactions. 2nd Edition by H. O. HOUSE, W . A. Benjamin Publishing (1972).  36.  a) L. N. MANDER and S. P. SETHI. Tetrahedron Lett. 5425 (1983);  255  b) S. R. CRABTREE, W . L. A. CHU, and L. N. MANDER. Synlett, 169 (1990). 37.  a) G . STORK, A . BRIZZOLARA, H. L A N D E S M A N , J. SZMUSZKOVICZ and R. TERRELL. J. Am. Chem. Soc. 85, 207 (1963); b) M . E. KUEHNE. Synthesis, 510 (1970).  38.  Y . TANIGUCHI, J. INANAGA, and M . YAMAGUCHI. Bull. Chem. Soc. Jpn., 54, 3229 (1981).  39.  F. E. ZEIGLER, J . - M . FANG, and C. C. TAM. J. Am. Chem. Soc. 104, 7174 (1982);  40.  a) D . A. EVANS, L. K. TRUESDALE and G . L. CARROLL. J. Chem. Soc. Chem. Comm. 55 (1973); b) D . A. EVANS and L. K. TRUESDALE. Tetrahedron Lett. 4929 (1973); c) M . ODA, A. YAMAMURO and T. WATABE. Chem. Lett. 1427 (1979).  41.  D . A. EVANS, G . L. CARROLL and L. K. TRUESDALE. J. Org. Chem. 39, 914 (1974).  42.  N. S. NUDELMAN, Z . GATTO and L. BOHE. J. Org. Chem. 49, 1540 (1984).  43.  a) Z . COHEN, E. KEINAN, Y . MAZUR, and T. H. VARKONY. J. Org. Chem. 40, 2141 (1975); b) For a review of ozonation see Ozonation in Organic Chemistry Vol II, edited by P. S. BAILEY, Academic Press (1982).  44.  a) D . TAL, E. KEINAN and Y . MAZUR. J. Am. Chem. Soc. 101, 502 (1979); b) L. V. RUBAN, S. K. RAKOVSKI and A. A. Povov. Izv. Akad. Nauk. SSSR, Ser. Khim. 9, 1950 (1976).  45.  a) M . JULIA, S. JULIA, and R. GUEGAN. Bull. Soc. Chim. Fr. 1072 (1960); b) S. SAREL, J. YOVELL and M . SAREL-IMBER. Angew. Chem. Int. Ed. Engl. 7, 577 (1968); c) S. F. BRADY, M . A. ILTON and W . S. JOHNSON. J. Am. Chem. Soc. 90, 2882 (1968); d) J. B. LAMBERT, J. J. NAPOLI, K. K. JOHNSON, K. N. TABA and B. S. PACKARD. J. Org. Chem. 50, 1921 (1985).  46.  W . P. NEUMANN. Synthesis, 665 (1987).  47.  a) M . CASTAING, M . PEREYRE, M . RATIER, P. M . BLUM, and A. G . DAVIES. J . Chem. Soc. Perkin II. 287 (1979); See also: b) P. S. MARIANO and E. BAY. J. Org. Chem. 45, 1763 (1980).  48.  E. WINTERFELDT, Synthesis, 617 (1975).  49.  a) J.-L. LUCHE. J. Am. Chem. Soc. 100, 2226 (1978); b) A. L. GEMAL and J.-L. LUCHE. J. Am. Chem. Soc. 103, 5454 (1981).  50.  a) H. M . B E L L and H. C. BROWN. J. Am. Chem. Soc. 88, 1473 (1966);  256  b) H. M. BELL, C. W . VANDERSLICE, and A. SPEHAR. J . Org. Chem. 34, 3923  (1969); c) R. O . HuTCfflNS, D . HOKE, J. KEOGH, and D . KOHARSKI. Tetrahedron Lett. 3495 (1969). 51.  C. H. EUGSTER, L. LEICHNER and E. JENNY. Helv. Chim. Acta, 46, 543 (1963).  52.  P. CANONNE, G. B . FOSCOLOS and G. LEMAY. Tetrahedron Lett. 155 (1980).  53.  F . J. WEIBERTH and S. S. HALL. J. Org. Chem. 52, 3901 (1987).  54.  V . GUAY and P. BRASSARD. Synthesis, 294 (1987).  55.  J.-P. PICARD, G. NZABAMWITA, F . LACOSTE, R. CALAS and J. DUNOGUES.  56.  Synthetic Comm. 19, 135 (1989).  G. I. Poos, G. E. ARTH, R. E. BEYLER and L. H. SARETT. J . Am. Chem. Soc.  75, 422 (1953).  57.  G. PIANCATELLI, A. ScETTRi and M. D'AURIA. Synthesis, 245 (1982).  58.  E. J. COREY and G. SCHMIDT. Tetrahedron Lett. 399 (1979).  59.  a) A . J . MANCUSO, S.-L. HUANG, and D . SWERN. J. Org. Chem. 43, 2480  (1978);  b) K . OMURA and D . SWERN. Tetrahedron, 34, 1651 (1978).  60.  A. J . FATIADI. Synthesis, 65 (1976).  61.  L. FITJER and U . QUABECK. Synthetic Comm. 15, 855 (1985).  62.  E. W . KUEMMERLE, T. A. RETTIG, and J. K . STILLE. J . Org. Chem. 40, 3665  63.  Advanced Organic Chemistry. 3rd Edition by J. MARCH, McGraw-Hill Publishing (1977)  64.  T. HOLM. Acta Chem. Scand. 18, 1577 (1964).  65.  D . C. WIGFIELD. Tetrahedron, 35, 449 (1979).  66.  W . NAGATA, M. YOSHIOKA and M. MURAKAMI. Org. Synth. 52,96 (1972).  67.  C. H. DEPUY and R. W . KING. Chem. Rev. 60, 431 (1960).  68.  J. ATTENBURROW, A. F . B . CAMERON, J. H. CHAPMAN, R. M. EVANS,  (1975);  B . A . HEMS, A . B . A. JANSEN, and T. WALKER. J . Chem. Soc. 1094 (1952).  69.  L. HORNER and E. H. WINKELMANN. Newer Methods of Preparative Organic Chemistry, 3, 151 (1964).  257  70.  D . W . BROOKS, H . S. BEVINAKATTI, E. KENNEDY, and J . HATHAWAY. J . Org. Chem. 50, 628 (1985);  71.  J . B . HEATHER, R. S. D . MITTAL, and C. J . SIH. J . Am. Chem. Soc. 98, 3662 (1976).  72.  R. H . EASTMAN, P. M . ILOFF and H . ISAACS. J . Am. Chem. Soc. 80, 1704 (1958).  73.  W . COCKER, P. V . R. SHANNON, and M . DOWSETT. J . Chem. Soc. Perkin II. 1527 (1988).  74.  W . C. STILL, M . K A H N and A. MITRA. J . Org. Chem. 43, 2923 (1978).  75.  Purification of Laboratory Chemicals 2nd Edition by D . D . PERRIN W. L.  76.  I. OGRIGANOV, D . IVANOV, V . HEROUT, M . HORAK, J . PLIVA, and F. SORM. Chem. Ind (London). 820 (1957).  77.  For reviews on the germacranes see: (a) F. SORM. Fortschr. Chem. Org. Naturst. 19, 1 (1961); (b) T. G . HALSALL and D . W . THEOBALD. Quart. Rev. 16,101 (1962); (c) N. H . FISCHER, E. J . OLIVIER, and H . D . FISCHER. Fortschr. Chem. Org. Naturst. 38, 47 (1979); (d) J . S. ROBERTS. Terpenoids and Steroids, 2, 92 (1972); 10, 51 (1981); 11, 52 (1982); 12, 104 (1983); (e) J . S. ROBERTS and I. BRYSON. Natural Product Reports, 1,144 (1984); (f) B . M . FRAGA. Natural Product Reports, 2, 150 (1985); 3, 279 (1986); 4, 481 (1987); 5, 304 (1988); 7, 67 (1990).  78.  (a) W . C. STILL. J . Am. Chem. Soc. 101, 2493 (1979); (b) W . C. STILL, S. MURATA, G . REVIAL, and K . YOSHIHARA. J. Am. Chem. Soc. 105, 625 (1983); (c) W . C. STILL. J . Am. Chem. Soc. 99, 4186 (1977). For related synthesis see: (d) S. RAUCHER, K - W . CHI, K - J . HWANG, and J . E. BURKS. J . Org. Chem. 51, 5503 (1986); (e) T. NAKAMURA, H HIROTA, C. KURODA, and T. TAKAHASHI. Chem. Lett. 1897 (1986); (f) M . IGUCHI, A . NISHIYAMA, S. YAMAMURA, and Y . HIRATA. Tetrahedron Lett 4295 (1969); (g) A. GOPALAN and P. MAGNUS. J . Org. Chem. 49, 2317 (1984); J . Am. Chem. Soc. 102, 1756 (1980); (h) K . KATO, Y . HIRATA, and S. YAMAMURA. J . Chem. Soc. Chem. Comm. 1324 (1970); (i) P. A . GRIECO and M . NISHIZAWA. J . Org. Chem. 42, 1717 (1977); (j) L. A . PAQUETTE, D . T. DERUSSY, and C. E. COTTRELL. J . Am. Chem. Soc. 110, 890 (1988).  Armarego and D . R. PERRIN, Pergamon Press (1980).  258  79.  a) M . KODAMA, Y . MATSUKI, and S. ITO. Tetrahedron Lett. 1121 (1976);. (b) M . KODAMA, S. YOKOO, H. YAMADA, and S. ITO. Tetrahedron Lett. 3121 (1978).  80.  (a) J . A . MARSHALL, J . LEBRETON, B. S. D E H O F F and T. M . JENSON. J . Org. Chem. 52, 3883 (1987). For related synthesis see: (b) T. TAKAHASHI, H. NEMOTO, Y . KANDA, J . TSUJI, Y . F U K A Z A W A , T. OKAJiMAand Y . FUJISE. Tetrahedron, 43, 5499 (1987); (c) T. TAKAHASHI, H. NEMOTO, Y . KANDA, and J . TSUJI. Heterocycles, 25, 139 (1987).  81.  (a) P. A . W E N D E R and J . C. LECHLEITER. J . A m . Chem. Soc. 102,  6340  (1980) . For related synthesis see: (b) P. A . WENDER and J . C. LECHLEITER. J . Am. Chem. Soc. 99, 267 (1977); (c) G . L. LANGE, S. SO, M . LAUTENS, and K . LOHR. Tetrahedron Lett. 3 1 1 (1981) ; (d) G . L. LANGE, M . A. HUGGINS, and E. NEIDERT. Tetrahedron Lett. 4409 (1976) ; (e) S. R. WILSON, L. R. PHILLIPS, Y . PELISTER, J . C. HUFFMAN. J . Am. Chem. Soc. 101, 7 3 7 3 (1979). 82.  E. J . COREY and A. G . HORTMANN. J . Am. Chem. Soc. 87, 5 7 3 6 (1965).  83.  M . WATANABE and A. YOSHIKOSKI. J . Chem. Soc. Chem. Comm. 698 (1972).  84.  (a) Y . FUJIMOTO, T. SHIMIZU, M . OHMORI, and T. TATSUNO. Chem. Pharm. Bull. 27, 9 2 3 (1979); (b) Y . FUJIMOTO, T. SHIMIZU, and T. TATSUNO.Tetrahedron Lett. 2041 (1976).  85.  C. A. GROB. Angew. Chem. Int. Ed. Engl. 8, 543 (1969).  86.  J . A. MARSHALL. Synthesis, 2 2 9 (1971). For a related synthesis see P. S. WHARTON, C. E. SUNDIN, D . W . JOHNSON, and H. C. KLEUNDER. J . Org. Chem. 37, 3 4 (1972).  87.  J . A. MARSHALL, W . F. HUFFMAN, and J . A. RUTH. J . Am. Chem. Soc. 94, 4691 (1972).  88.  (a) A . G . G O N Z A L E Z , A. GALINDO, H. M A N S I L L A , A . GUTIERREZ, J . A. PALENZUELA, M . A. GOMEZ-RODRIQUEZ, M . MARTTNEZ-RIPOLL, and S. GARCIABLANCO. J . Org. Chem. 50, 5 8 5 6 (1985). For related syntheses see: (b) P. S. WHARTON, Y . SUMI, and R. A. KRETCHMER. J . Org. Chem. 34, 2 3 4 (1965); (c) M . MIYASHITA and A. YOSHIKOSHI. J . Am. Chem. Soc. 9 6 , 1 9 1 7 (1974); (d) J . M . BROWN, T. M . CRESP, and L. N . MANDER. J . Org. Chem. 42, 3984 (1977) .  259  89.  (a) M . L. MIHAILOVIC and Z . CEKOVIC. Synthesis, 209 (1970). See also: (b) K . H E U S L E R and J. K A L V O D A . Angew. Chem. Int. Ed. Engl. 3, 5 2 5 (1964); (c) K . HEUSLER. P. WIELAND, and C. MEYSTRE. Org. Synth. 45, 57 (1965).  90.  M . L. MIHAILOVIC M . JAKOVLJEVIC, V . TRIFUNOVIC, R. VUKOV and Z . CEKOVIC. Tetrahedron, 24, 6 9 5 9 (1968).  91.  a) S. M O O N and J. M . LODGE. J. Org. Chem. 29, 3453 (1964);  b) M . L. MIHAILOVIC, S. KONSTANTINOVIC, A . MILOVANOVIC, J. JANKOVIC, Z . CEKOVIC, and D . JEREMIC. J. Chem. Soc. Chem. Comm. 2 3 6 (1969). 92.  (a) M . L. MIHAILOVIC, L. LORENC, M . GASIC, M . ROGIC, A . M E L E R A and M . STEFANOVic.Tetrahedron, 22, 2345 (1966); (b) M . L. MIHAILOVIC, M . STEFANOVIC, L. LORENC, and M . GASIC. Tetrahedron Lett. 1867 (1964).  93.  H . SUGINOME, H . WASHIYAMA, and S. YAMADA. Bull. Chem. Soc. Jpn. 60, 1071 (1987).  94.  (a) A . VARVOGLIS. Chem. Soc. Rev. 10, 377 (1981); (b) A . VARVOGLIS. Synthesis, 7 0 9 (1984); (c) R. M . MORIARTY and O. PRAKASH. ACC. Chem. Res., 19, 2 4 4 (1986).  95.  a) P. A R M A S , J. I. CONCEPCION, C. G . FRANCISCO, R. HERNANDEZ, J . A. SALAZAR, and E. SUAREZ. J. Chem. Soc. Perkin Trans. I, 4 0 5 (1989); b) J. I. CONCEPCION, C. G . FRANCISCO, R. HERNANDEZ, J. A. SALAZAR, and E. SUAREZ. Tetrahedron Lett. 1953 (1984); c) R. FRIERE, J. J. MARRERO, M . S. RODRIGUEZ, and E. SUAREZ. Tetrahedron Lett. 383 (1986).  96.  P. A . ZORETIC, B. BRANCHAUD and T. MAESTRONE. Tetrahedron Lett. 5 2 7 (1975) .  97.  T. G . HALSALL, D . W . THEOBALD and K . B. WALSHAW.J. Chem. Soc. 1029 (1964)  98.  a) S. J. CRISTOL, J. S. PERRY and R. S. BECKLEY. J. Org. Chem. 41, 1912 (1976) ; b) H . C. BROWN and P. GEOGHEGAN. J. Am. Chem. Soc. 89, 1522 (1967); c) H . C. BROWN and P. GEOGHEGAN. J. Org. Chem. 35, 1844 (1970).  99.  W . D . EMMONS and A . S. PAGANO. J. Am. Chem. Soc. 7 7 , 89 (1955).  100.  a) H . C. BROWN and S. KRISHNAMURTHY. J. Am. Chem. Soc. 95, 1669 (1973); b) S. KRISHNAMURTHY, R. M . SCHUBERT and H . C. BROWN. J. Am. Chem. Soc. 95, 8486 (1973).  260  101.  a) A . S . HALLSWORTH and H . B . HENBEST. J . Chem. Soc. 4604 (1960); b) A . S . HALLSWORTH and H . B . HENBEST. J . Chem. Soc. 3571 (1960).  102.  H . C. BROWN, S . IKEGAMI and J . H . KAWAKAMI. J . Org. Chem. 35, 3243 (1970).  103.  G. HOKLE, W . STEGLICH and H . VORBRUGGEN. Angew. Chem. Int. Ed. Engl. 17, 569 (1978).  104.  Y . KISHI, M . ARATANI, H . TANINO, T . FUKUYAMA, T. GOTO, S . INOUE, S . SUGIURA and H . KAKOI. J . Chem. Soc. Chem. Comm. 64 (1972).  105.  P. V . DEMARCO, E. FARKAS, D . DODDRELL, B . L . MYLARI, and E. WENKERT. J . Am. Chem. Soc. 90, 5480 (1968).  106.  W . A . A Y E R and M . G. PAICE. Can. J . Chem. 54, 910 (1976).  107.  Spectroscopic  Identification  of Organic Compounds. 3rd Edition by  D . H . WILLIAMS and I. FLEMING, McGraw-Hill ( U K ) Publishing (1980).  108.  H . WEHRLI, C. LEHMANN, P. KELLER, J . J . BONET, K . SCHAFFNER and O. JEGER. Helv. Chim. Acta, 49, 2218 (1966).  109.  O. MTTSUNOBU. Synthesis, 1 (1981).  110.  a) K . B . SHARPLESS and R. C. MICHAELSON. J . Am. Chem. Soc. 95, 6136 (1973); b) B . E. ROSSJTER, T. R. VERHOEVEN and K . B . SHARPLESS. Tetrahedron Lett. 4733 (1979).  111.  A. BOWERS, T. G. HALSALL, E. R. H . JONES and A. J . LEMIN. J . Chem. Soc. 2548 (1956).  Appendix  1.  X-ray Crystal Structure Report on Acetate 156. A. Crystal data.  Compound  156  Formula  C  Molecular weight  209.29  Crystal colour, habit  Colourless, irregular  Crystal size,  0.30 x 0.30 x 0.40 mm  Crystal system  Monoclinic  1 2  H  1 9  N0  2  No. reflections used for unit cell determination (26 range)  25 (79.5-91.1°)  Omega scan peak width at half height  0.38  Space group  P2i/c (#14)  Lattice parameters:  a=  8.271 (2)  A  b=  11.035 (2) A  c=  13.726(1) A  (3= 105.102(9)° 1209.5 (3) A3  V= Z value  4  ^calc.  1.149 g/cm  F(000)  456  HCu-Ka)  5.87 cm"  3  1  B. Intensity Measurements Diffractometer  Rigaku AFC6S  Radiation  Cu-K (k = 1.54178 A) a  262  Temperature  21°C  Takeoff angle  6.0°  Detector aperature  6.0 mm horizontal 6.0 mm vertical  Crystal to detector distance  285 mm  Scan type  co-29  Scan width  (1.21 +0.20 tan 6 ) °  Scan rate  32.0°/min  2^max  155.20  No. of reflections measured  Total: 2553 Unique: 2377 (R t = 0.016) in  Corrections  Lorentz-polarisation Absorption (trans, factors: 0.89-1.00) Secondary extinction (coefficient: 0.14513 E-04)  C. Structure Solution and Refinement Structure solution  Direct methods  Refinement  Full-matrix least-squares  Function minimised  Zw(IF l-IF l) 0  c  2  where w = F / o ( F ) 0  2  2  0  2  R = EIIF I-IF II/ZIF I 0  R  w  C  0  = (Iw(IF l-IF l) /SwlF l ) 0  c  2  gof=[X(IFJ-IF l) /(m-«)] c  p-factor  0.03  Anomalous dispersion  All non-hydrogen atoms  2  0  1/2  2  1/2  No. of reflcns with / > 3a(7)  1675  No. of variables  213  Reflection/parameter ratio  7.86  Residuals: R; R  0.047; 0.062  w  Goodness of fit indicator  2.84  Max shift/error in final cycle  0.01  Maximum peak in final diff. map  +0.18  e/A  Minimum peak in final diff. map  -0.14  e/A  Values given for R, R  w  3  3  and gof are based on those reflections with / > 3o(7).  Table 9. Final atomic coordinates (fractional) and B  (A ) 2  eq  [Compound 156].  z  Beq  atom  X  y  0(1)  0.0567 (2)  0.1412 (1)  0.1628 (1)  4.75 (5)  0(2)  -0.0904 (2)  -0.0295 (2)  0.1136 (1)  7.49 (9)  N(l)  0.2310(3)  -0.1373 (2)  0.2638 (2)  6.19 (9)  C(l)  0.1745 (2)  0.0966 (2)  0.2524 (1)  3.90 (7)  C(2)  0.3454 (2)  0.1574 (2)  0.2526 (1)  4.20 (7)  C(3)  0.4714 (3)  0.1292 (2)  0.3536 (2)  5.06 (9)  C(4)  0.4115 (3)  0.1657 (3)  0.4445 (2)  5.9 (1)  C(5)  0.2448 (3)  0.1073 (3)  0.4420 (1)  5.6 (1)  C(6)  0.1100 (3)  0.1322 (2)  0.3444 (1)  4.47 (8)  C(7)  0.3227 (4)  0.2950 (2)  0.2389 (2)  5.6 (1)  C(8)  0.4110 (4)  0.1089 (2)  0.1660 (2)  5.5 (1)  C(9)  0.1974 (2)  -0.0369 (2)  0.2525 (1)  4.35 (8)  C(10)  -0.0542 (3)  0.0721 (3)  0.3457 (2)  6.1 (1)  C(ll)  -0.0691 (3)  0.0763 (2)  0.1038 (1)  4.98 (9)  C(12)  -0.1760(5)  0.1563 (3)  0.0260 (2)  6.9 (1)  264  Table 10. Bond lengths (A) with estimated standard deviations [Compound 156]. O(l)  C(l)  1.442 (2)  C(l)  C(6)  1.545 (3)  0(1)  C(ll)  1.347 (2)  C(l)  C(9)  1.485 (3)  0(2)  C(ll)  1.193 (3)  C(6)  C(5)  1.528 (3)  N(l)  C(9)  1.143 (2)  C(6)  C(10)  1.516 (3)  C(2)  C(l)  1.564 (3)  C(5)  C(4)  1.514 (3)  C(2)  C(3)  1.534 (3)  C(4)  C(3)  1.513 (3)  C(2)  C(7)  1.536 (3)  C(ll)  C(12)  1.486 (3)  C(2)  C(8)  1.527 (3)  Table 11. Bond angles (deg) with estimated standard deviations [Compound 156]. atom  atom  atom  angle  atom  atom  atom  angle  C(l)  0(1)  C(ll)  125.2 (2)  C(6)  C(l)  C(9)  108.8 (2)  C(l)  C(2)  C(3)  108.4 (2)  C(l)  C(6)  C(5)  110.0 (2)  C(l)  C(2)  C(7)  110.1 (2)  C(l)  C(6)  C(10)  113.5 (2)  C(l)  C(2)  C(8)  110.7 (2)  C(5)  C(6)  C(10)  111.3 (2)  C(3)  C(2)  C(7)  110.1 (2)  C(6)  C(4)  C(4)  113.3 (2)  C(3)  C(2)  C(8)  109.7 (2)  C(5)  C(5)  C(3)  111.1 (2)  C(7)  C(2)  C(8)  107.9 (2)  C(2)  C(6)  C(4)  113.6 (2)  0(1)  C(l)  C(2)  105.3 (1)  N(l)  C(9)  C(l)  171.1 (2)  O(l)  C(l)  C(6)  108.1 (1)  0(1)  C(ll)  0(2)  124.5 (2)  0(1)  C(l)  C(9)  113.4(1)  0(1)  C(ll)  C(12)  109.8 (2)  C(2)  C(l)  C(6)  113.2 (1)  0(2)  C(ll)  C(12)  125.8 (2)  C(2)  C(l)  C(9)  108.1 (2)  265  Table 12. Torsion or conformation angles (in deg) [Compound 156]. (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  0(1)  C(l)  C(2)  C(3)  171.8 (1)  C(l)  C(2)  C(3)  H(7)  177 (1)  0(1)  C(l)  C(2)  C(7)  51.3 (2)  C(l)  C(2)  C(7)  H(8)  -178 (1)  0(1)  C(l)  C(2)  C(8)  -67.9 (2)  C(l)  C(2)  C(7)  H(9)  -59 (2)  0(1)  C(l)  C(6)  C(5)  -169.9 (2)  C(l)  C(2)  C(7)  H(10)  59 (1)  (Xl)  C(l)  C(6)  C(10)  64.7 (2)  C(l)  C(2)  C(8)  H(ll)  -57 (2)  0(1)  C(l)  C(6)  H(l)  -55 (1)  C(l)  C(2)  C(8)  H(12)  57 (2)  O(l)  C(l)  C(9)  N(l)  174 (1)  C(l)  C(2)  C(8)  H(13)  179 (2)  O(l)  C(ll)  C(12) H(17)  -44 (2)  C(l)  C(6)  C(5)  C(4)  53.4 (3)  (Xl)  C(ll)  C(12) H(18)  74 (3)  C(l)  C(6)  C(5)  H(2)  176 (2)  0(1)  C(ll)  C(12) H(19)  -158 (2)  C(l)  C(6)  C(5)  H(3)  -67 (1)  0(2)  C(ll)  0(1)  -6.5 (3)  C(l)  C(6)  C(10) H(14)  177 (1)  0(2)  C(ll)  C(12) H(17)  136 (2)  C(l)  C(6)  C(10) H(15)  -65 (2)  0(2)  C(ll)  C(12) H(18)  -107 (3)  C(l)  C(6)  C(10) H(16)  61 (2)  0(2)  C(ll)  C(12) H(19)  22 (2)  C(6)  C(l)  O(l)  C(ll)  -95.3 (2)  N(l)  C(9)  C(l)  C(2)  58 (1)  C(6)  C(l)  C(2)  C(3)  53.9 (2)  N(l)  C(9)  C(l)  C(6)  -66 (1)  C(6)  C(l)  C(2)  C(7)  -66.5 (2)  C(2)  C(l)  O(l)  C(ll)  143.5 (2)  C(6)  C(l)  C(2)  C(8)  174.3 (2)  C(2)  C(l)  C(6)  C(5)  -53.7 (2)  C(6)  C(5)  C(4)  C(3)  -54.6 (3)  C(2)  C(l)  C(6)  C(10)  -179.1 (2)  C(6)  C(5)  C(4)  H(4)  66(1)  C(2)  C(l)  C(6)  H(l)  61 (1)  C(6)  C(5)  C(4)  H(5)  -177 (2)  C(2)  C(3)  C(4)  C(5)  56.0 (3)  C(5)  C(6)  C(l)  C(9)  66.5 (2)  C(2)  C(3)  C(4)  H(4)  -65 (2)  C(5)  C(6)  C(10) H(14)  52(1)  C(2)  C(3)  C(4)  H(5)  179 (2)  C(5)  C(6)  C(10) H(15)  171 (2)  C(l)  C(2)  C(ll)  C(12)  173.2 (2)  C(5)  C(6)  C(10) H(16)  -64 (2)  C(l)  C(2)  C(3)  C(4)  -54.9 (2)  C(5)  C(4)  C(3)  H(6)  -65 (1)  C(l)  C(2)  C(3)  H(6)  67 (1)  C(5)  C(4)  C(3)  H(7)  179 (2)  C(l)  angle  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  266  Table 12. Torsion or conformation angles (in deg) [Compound 156]. (1)  (2)  (3)  (4)  C(4)  C(5)  C(6)  C(10)  C(4)  C(5)  C(6)  C(4)  C(3)  C(4)  angle  continued angle  (1)  (2)  (3)  (4)  -179.9 (2)  C(8)  C(2)  C(7)  H(8)  -57 (2)  H(l)  -62 (1)  C(8)  C(2)  C(7)  H(9)  62 (2)  C(2)  C(7)  65.6 (3)  C(8)  C(2)  C(7)  H(10)  179 (1)  C(3)  C(2)  C(8)  -175.8 (2)  C(9)  C(l)  O(l)  C(ll)  25.5 (2)  C(3)  C(2)  C(l)  C(9)  -66.7 (2)  C(9)  CQ)  C(6)  CQO)  -58.9 (2)  C(3)  C(2)  C(7)  H(8)  63 (2)  C(9)  C(l)  C(6)  H(l)  -179 (1)  C(3)  C(2)  C(7)  H(9)  -179 (2)  C(10) C(6)  C(5)  H(2)  -57 (2)  C(3)  C(2)  C(7)  H(10)  -61 (1)  C(10) C(6)  C(5)  H(3)  60(1)  C(3)  C(2)  C(8)  H(ll)  63(2)  H(l)  C(6)  C(5)  H(2)  61 (2)  C(3)  C(2)  C(8)  H(12)  177 (2)  HQ)  C(6)  C(5)  H(3)  178 (2)  C(3)  C(2)  C(8)  H(13)  -62 (2)  HQ)  C(6)  C(10) HQ4)  -64 (2)  C(3)  C(4)  C(5)  H(2)  -176 (2)  HQ)  C(6)  CQO) HQ5)  54 (2)  C(3)  C(4)  C(5)  H(3)  66(1)  HQ)  C(6)  C(10) HQ6)  180 (2)  C(7)  C(2)  C(l)  C(9)  172.8 (2)  H(2)  C(5)  C(4)  H(4)  -56 (2)  C(7)  C(2)  C(3)  H(6)  -172(1)  H(2)  C(5)  C(4)  H(5)  62 (2)  C(7)  C(2)  C(3)  H(7)  -62 (2)  H(3)  C(5)  C(4)  H(4)  -174 (2)  C(7)  C(2)  C(8)  H(ll)  -177 (2)  H(3)  C(5)  C(4)  H(5)  -56 (2)  C(7)  C(2)  C(8)  H(12)  -63 (2)  H(4)  C(4)  C(3)  H(6)  173 (2)  C(7)  C(2)  C(8)  H(13)  58 (2)  H(4)  C(4)  C(3)  H(7)  58 (2)  C(8)  C(2)  C(l)  C(9)  53.7 (2)  H(5)  C(4)  C(3)  H(6)  57 (2)  C(8)  C(2)  C(3)  H(6)  -53 (1)  H(5)  C(4)  C(3)  H(7)  -58 (2)  C(8)  C(2)  C(3)  H(7)  56 (2)  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  Figure 14. Single crystal X-ray structure of acetate 156 (stereo view).  Appendix  2.  X-ray Structure Report on Alcohol  245.  A. Crystal data. 245  Compound Formula  C  16 26°3 H  Molecular weight  266.38  Crystal colour, habit  Colourless, prism  Crystal size,  0.350 x 0.400 x 0.450 mm  Crystal system  Orthorhombic  No. reflections used for unit cell determination (29 range)  25 (80.0-120.0O)  Omega scan peak width at half height  0.36  Space group  P2 2 2 (#19)  Lattice parameters:  a=  12.936(1) A  b=  13.974(1) A  c=  8.5388 (8) A  V=  1543.6 (4) A  1  1  1  Z value  4  ^calc.  1.146 g/cm  F(000)  584  HCa-Ka)  5.81 cm"  3  3  1  B. Intensity Measurements Diffractometer  Rigaku AFC6S  Radiation  Cu-A^ (k = 1.54178 A)  Temperature  21°C  269  Takeoff angle  6.0°  Detector aperature  6.0 mm horizontal 6.0 mm vertical  Crystal to detector distance  285 mm  Scan type  (0-26  Scan width  (1.31+0.30 tan 6 ) °  Scan rate  32.0°/min 155.1°  2®max  No. of reflections measured  Total: 1889  Corrections  Lorentz-polarisation Absorption (trans, factors: 0.93-1.00) Secondary extinction (coefficient: 0.60504 E -04)  C. Structure Solution and Refinement Structure solution  Direct methods  Hydrogen atom treatment  Refined or included in calculated positions (dc-H = 0.98 A )  Refinement  Full-matrix least-squares  Function minimised  Zw(IF l-IF l) 0  c  2  where w = F / a ( F ) G  2  2  0  2  R = H\F \-\F \\mF \ c  0  R  w  Q  = (Ew(IF l-IF l) /EwlF l ) / 0  c  2  0  gof=[Z(IF l-IF l) /(m-/i)] / 0  p-factor  0.03  c  2  1  2  2  1  2  270  Anomalous dispersion  All non-hydrogen atoms  No. of reflcns with / > 3o(/)  1636  No. of variables  177  Reflection/parameter ratio Residuals: R; R  9.24 0.040; 0.055  w  Goodness of fit indicator  2.41  Max shift/error in final cycle  0.15  Maximum peak in final diff. map  +0.16 e/A  Minimum peak in final diff. map  -0.14 e/A  Values given for R, R  w  3  3  .  and gof are based on those reflections with / > 3a(/).  H21 H22  Figure 15. Single crystal X-ray structure of alcohol 245 (stereo view).  Table 13. Final atomic coordinates (fractional) and Beq (A ) 2  [Compound 245].  atom  X  y  0(1)  0.2026 (2)  0.1042 (1)  0.4220 (2)  6.5 (1)  0(2)  0.2356 (2)  0.2523 (1)  0.3300 (2)  5.73 (8)  0(3)  0.2906 (1)  0.0996 (1)  -0.1159 (2)  4.67 (6)  C(l)  0.3122 (1)  0.1574 (1)  0.0196 (2)  3.60 (7)  C(2)  0.2425 (2)  0.1206 (1)  0.1500 (2)  4.00 (8)  C(3)  0.2658 (2)  0.1555 (1)  0.3126 (3)  4.60 (9)  C(4)  0.3793 (2)  0.1412 (2)  0.3504 (3)  5.4 (1)  C(5)  0.4458 (2)  0.1921 (2)  0.2302 (3)  5.2 (1)  C(6)  0.4306 (1)  0.1572 (2)  0.0608 (2)  4.31 (8)  C(7)  0.4757 (2)  0.2339 (2)  -0.0445 (3)  5.5 (1)  C(8)  0.4465 (2)  0.2375 (2)  -0.2141 (3)  6.5 (1)  C(9)  0.3922 (2)  0.3008 (2)  -0.0961 (3)  4.9 (1)  C(10)  0.2917 (1)  0.2612 (1)  -0.0314 (2)  3.95 (7)  C(ll)  0.4084 (2)  0.4077 (2)  -0.1043 (4)  6.3 (1)  C(12)  0.3286 (3)  0.4552 (2)  -0.2077 (5)  8.3 (2)  C(13)  0.4132 (4)  0.4523 (2)  0.0540 (5)  9.7 (2)  C(14)  0.4812 (2)  0.0594 (2)  0.0398 (3)  6.0 (1)  C(15)  0.1637 (2)  0.1688 (2)  0.5344 (3)  6.1 (1)  C(16)  0.2028 (3)  0.2639 (2)  0.4863 (4)  6.9 (1)  z  Beq  272  Table 14. Bond lengths (A) with estimated standard deviations in brackets [Cpd. 245] atom atom  1.415 (2)  atom atom C(9) C(ll) C(8) C(7) C(7) C(6)  C(16) C(l)  1.410 (4) 1.438 (2)  C(6) C(6)  C(l) C(l) C(l) C(10)  C(10) C(6) C(2) C(9)  1.538 (3) 1.572 (3)  C(5) C(4)  C(9) C(9)  C(8) C(7)  1.523 1.517 1.513 1.494  C(ll) C(12) C(ll) C(13)  0(1) O(l) 0(2)  C(3) C(15) C(3)  0(2) 0(3)  distance 1.433 (3) 1.410 (3)  (3) (3) (4) (3)  C(5) C(14)  C(4) C(3) C(3) C(2) C(16) C(15)  distance 1.510 (3) 1.497 (4) 1.516 (3) 1.539 (3) 1.526 (3) 1.516 1.516 1.502 1.480 1.512 1.490  (4) (4) (3) (4) (5) (6)  273  Table 15. Bond angles (deg) with estimated standard deviations in brackets. atom  atom  atom  angle  atom  atom  atom  angle  C(3)  O(l)  C(15)  109.1 (2)  C(l)  C(6)  C(5)  109.5 (2)  C(3)  0(2)  C(16)  107.0 (2)  C(l)  C(6)  C(14)  113.2 (2)  0(3)  C(l)  C(10)  105.6 (2)  C(7)  C(6)  C(5)  106.5 (2)  0(3)  C(l)  C(6)  111.6 (2)  C(7)  C(6)  C(14)  113.5 (2)  0(3)  C(l)  C(2)  106.5 (2)  C(5)  C(6)  C(14)  109.9 (2)  C(10)  C(l)  C(6)  103.5 (2)  C(6)  C(5)  C(4)  114.5 (2)  C(10)  C(l)  C(2)  115.0 (2)  C(5)  C(4)  C(3)  110.1 (2)  C(6)  C(l)  C(2)  114.4 (2)  0(1)  C(3)  0(2)  104.6 (2)  C(l)  C(10)  C(9)  107.4 (2)  0(1)  C(3)  C(4)  110.3 (2)  C(10)  C(9)  C(8)  115.3 (2)  O(l)  C(3)  C(2)  109.0 (2)  C(10)  C(9)  C(7)  106.5 (2)  0(2)  C(3)  C(4)  111.8 (2)  C(10)  C(9)  C(ll)  119.7 (2)  0(2)  C(3)  C(2)  110.6 (2)  C(8)  C(9)  C(7)  59.5 (2)  C(4)  C(3)  C(2)  110.4 (2)  C(8)  C(9)  C(ll)  118.9 (2)  C(l)  C(2)  C(3)  116.6 (2)  C(7)  C(9)  C(ll)  122.2 (2)  0(2)  C(16)  C(15)  105.2 (2)  C(9)  C(8)  C(7)  59.5 (2)  0(1)  C(15)  C(16)  105.3 (2)  C(9)  C(7)  C(8)  60.8 (2)  C(9)  C(ll)  C(12)  111.5 (2)  C(9)  C(7)  C(6)  109.9 (2)  C(9)  C(ll)  C(13)  112.2 (3)  C(8)  C(7)  C(6)  120.1 (2)  C(12)  C(ll)  C(13)  112.0 (3)  C(l)  C(6)  C(7)  103.9 (2)  274  Table 16. Torsion or conformation angles (in deg) [Compound 245]. (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  angle  O(l)  C(3)  0(2)  C(16)  29.6 (3)  0(3)  C(l)  C(2)  H(ll)  -71  0(1)  C(3)  C(4)  C(5)  177.6 (2)  0(3)  C(l)  C(2)  H(12)  47  0(1)  C(3)  C(4)  H(9)  57  C(l)  C(10) C(9)  C(8)  48.6 (3)  0(1)  C(3)  C(4)  H(10)  -62  C(l)  C(10) C(9)  C(7)  -15.2 (2)  0(1)  C(3)  C(2)  C(l)  -172.2 (2)  C(l)  C(10) C(9)  C(ll)  -159.1 (2)  O(l)  C(3)  C(2)  H(ll)  67  C(l)  C(6)  C(7)  C(9)  20.9 (2)  0(1)  C(3)  C(2)  H(12)  -51  C(l)  C(6)  C(7)  C(8)  -46.1 (3)  O(l)  C(15)  C(16) 0(2)  16.0 (3)  C(l)  C(6)  C(7)  H(6)  159  0(1)  C(15)  C(16) H(13)  -103  C(l)  C(6)  C(5)  C(4)  51.1 (3)  O(l)  C(15) C(16) H(14)  135  C(l)  C(6)  C(5)  H(7)  172  0(2)  C(3)  0(1)  C(15)  -19.2 (3)  C(l)  C(6)  C(5)  H(8)  -70  0(2)  C(3)  C(4)  C(5)  -66.5 (2)  C(l)  C(6)  C(14) H(17)  -63  0(2)  C(3)  C(4)  H(9)  173  C(l)  C(6)  C(14) H(18)  57  0(2)  C(3)  C(4)  H(10)  54  C(l)  C(6)  C(14) H(19)  177  0(2)  C(3)  C(2)  C(l)  73.4 (2)  C(l)  C(2)  C(3)  C(4)  -50.8 (2)  0(2)  C(3)  C(2)  H(ll)  -48  C(10) C(l)  0(3)  H(l)  -179 (3)  0(2)  C(3)  C(2)  H(12)  -166  C(10) C(l)  C(6)  C(7)  -29.1 (2)  0(2)  C(16)  C(15) H(15)  -103  C(10) C(l)  C(6)  C(5)  84.4 (2)  0(2)  C(16)  C(15) H(16)  135  C(10) C(l)  C(6)  C(14)  0(3)  C(l)  C(10)  C(9)  -89.8 (2)  C(10) C(l)  C(2)  C(3)  0(3)  C(l)  C(10) H(2)  30  C(10) C(l)  C(2)  H(ll)  45  0(3)  C(l)  C(10) H(3)  150  C(10) C(l)  C(2)  H(12)  163  0(3)  C(l)  C(6)  C(7)  84.0 (2)  C(10) C(9)  C(8)  C(7)  -95.0 (2)  0(3)  C(l)  C(6)  C(5)  -162.5 (2)  C(10) C(9)  C(8)  H(4)  14  0(3)  C(l)  C(6)  C(14)  -39.5 (2)  C(10) C(9)  C(8)  H(5)  156  0(3)  C(l)  C(2)  C(3)  167.6 (2)  C(10) C(9)  C(7)  C(8)  110.1 (2)  -152.6 (2) -75.7 (2)  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  275  Table 16. Torsion or conformation angles (in deg).  continued  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  C(10) C(9)  C(7)  C(6)  -4.0 (3)  C(8)  C(7)  C(9)  C(ll)  -107.1 (3)  C(10) C(9)  C(7)  H(6)  -142  C(8)  C(7)  C(6)  C(5)  -161.7 (2)  C(10) C(9)  C(ll)  CQ2)  -60.8 (4)  C(8)  C(7)  C(6)  CQ4)  C(10) C(9)  C(ll)  CQ3)  65.8 (4)  C(7)  C(9)  C(10) H(2)  -135  C(10) C(9)  C(ll)  H(20)  -177  C(7)  C(9)  CQO) H(3)  104  (1)  angle  77.2 (3)  C(9)  C(10) C(l)  C(6)  27.6 (2)  C(7)  C(9)  C(8)  H(4)  109  C(9)  C(10) C(l)  C(2)  153.0 (2)  C(7)  C(9)  C(8)  H(5)  -109  C(9)  C(8)  C(7)  C(6)  97.2 (2)  C(7)  C(9)  C(ll)  CQ2)  161.0 (3)  C(9)  C(8)  C(7)  H(6)  -108  C(7)  C(9)  C(ll)  CQ3)  -72.4 (4)  C(9)  C(7)  C(8)  H(4)  -109  C(7)  C(9)  CQl)  H(20)  45  C(9)  C(7)  C(8)  H(5)  109  C(7)  C(8)  C(9)  CQl)  112.5 (3)  C(9)  C(7)  C(6)  C(5)  -94.7 (2)  C(7)  C(6)  C(l)  C(2)  -155.0 (2)  C(9)  C(7)  C(6)  CQ4)  144.2 (2)  C(7)  C(6)  C(5)  C(4)  162.8 (2)  C(9)  C(ll)  C(12) H(21)  -178  C(7)  C(6)  C(5)  H(7)  -76  C(9)  C(ll)  C(12) H(22)  -58  C(7)  C(6)  C(5)  H(8)  42  C(9)  C(ll)  C(12) H(23)  62  C(7)  C(6)  CQ4) HQ7)  179  C(9)  C(ll)  C(13) H(24)  57  C(7)  C(6)  CQ4) HQ8)  -61  C(9)  C(ll)  C(13) H(25)  177  C(7)  C(6)  C(14) HQ9)  59  C(9)  C(ll)  C(13) H(26)  -63  C(6)  CQ)  0(3)  HQ)  69 (3)  C(8)  C(9)  C(10) H(2)  -71  C(6)  C(l)  CQO) H(2)  147  C(8)  C(9)  CQO) H(3)  168  C(6)  C(l)  CQO) H(3)  -92  C(8)  C(9)  C(7)  C(6)  -114.1 (3)  C(6)  CQ)  C(2)  C(3)  C(8)  C(9)  C(7)  H(6)  108  C(6)  CQ)  C(2)  HQl)  165  C(8)  C(9)  C(ll)  CQ2)  90.5 (3)  C(6)  CQ)  C(2)  HQ2)  -77  C(8)  C(9)  CQl)  CQ3)  -142.9 (3)  C(6)  C(7)  C(9)  C(ll)  138.8 (3)  C(8)  C(9)  C Q l ) H(20)  -26  C(6)  C(7)  C(8)  H(4)  43.9 (2)  -12  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  276  Table 16. Torsion or conformation angles (in deg).  continued  (1)  (2)  (3)  (4)  angle  -153  C(2)  C(l)  C(6)  C(14)  81.5 (2)  C(3)  -60.2 (3)  C(2)  C(3)  0(1)  C(15)  -137.5 (2)  C(4)  H(9)  60  C(2)  C(3)  0(2)  C(16)  146.8 (2)  C(5)  C(4)  H(10)  180  C(2)  C(3)  C(4)  H(9)  -63  C(5)  C(6)  C(l)  C(2)  -41.5 (3)  C(2)  C(3)  C(4)  H(10)  177  C(5)  C(6)  C(7)  H(6)  43  C(14) C(6)  C(7)  H(6)  -78  C(5)  C(6)  C(14) H(17)  60  C(14) C(6)  C(5)  H(7)  47  C(5)  C(6)  C(14) H(18)  180  C(14) C(6)  C(5)  H(8)  165  C(5)  C(6)  C(14) H(19)  -60  C(ll)  C(9)  C(10) H(2)  81  C(5)  C(4)  C(3)  C(2)  57.0 (2)  C(ll)  C(9)  C(10) H(3)  -39  C(4)  C(5)  C(6)  C(14)  -73.8 (3)  C(ll)  C(9)  C(8)  H(4)  -138  C(4)  C(3)  0(1)  C(15)  101.1 (2)  C(ll)  C(9)  C(8)  H(5)  3  C(4)  C(3)  0(2)  C(16)  -89.8 (3)  C(ll)  C(9)  C(7)  H(6)  1  C(4)  C(3)  C(2)  H(ll)  -172  C(12) C ( l l )  C(13) H(24)  -176  C(4)  C(3)  C(2)  H(12)  70  C(12) C ( l l )  C(13) H(25)  -56  C(3)  0(1)  C(15) C(16)  2.0 (3)  C(12) C ( l l )  C(13) H(26)  64  C(3)  O(l)  C(15) H(15)  121  C(13) C ( l l )  C(12) H(21)  55  C(3)  0(1)  C(15) H(16)  -117  C(13) C ( l l )  C(12) H(22)  175  C(3)  0(2)  C(16) C(15)  -28.5 (3)  C(13) C ( l l )  C(12) H(23)  -65  C(3)  0(2)  C(16) H(13)  91  H(4)  C(8)  C(7)  H(6)  143  C(3)  0(2)  C(16) H(14)  -148  H(5)  C(8)  C(7)  H(6)  1  C(3)  C(4)  C(5)  H(7)  179  H(7)  C(5)  C(4)  H(9)  -61  C(3)  C(4)  C(5)  H(8)  61  H(7)  C(5)  C(4)  H(10)  C(2)  C(l)  0(3)  H(l)  -56 (3)  H(8)  C(5)  C(4)  H(9)  C(2)  C(l)  C(10) H(2)  -87  H(8)  C(5)  C(4)  H(10)  -60  C(2)  C(l)  C(10) H(3)  33  H(13) C(16) C(15) H(15)  137  (1)  (2)  (3)  (4)  C(6)  C(7)  C(8)  H(5)  C(6)  C(5)  C(4)  C(6)  C(5)  C(6)  angle  59 -179  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  277  Table 16. Torsion or conformation angles (in deg). (1)  (2)  (3)  (4)  C(15)  H(16)  16  H(20)  C ( l l ) C(12) H(23)  178  H(14) C(16) C(15)  H(15)  16  H(20)  C ( l l ) C(13) H(24)  -59  H(14) C(16) C(15)  H(16)  -105  H(20)  C ( l l ) C(13) H(25)  60  H(20) C ( l l )  C(12)  H(21)  -62  H(20)  C ( l l ) C(13) H(26)  -180  H(20) C ( l l )  C(12)  H(22)  58  H(13) C(16)  angle  (1)  (2)  continued (3)  (4)  angle  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  278  Appendix  3.  X-ray Structure Report on Diol 298. A. Crystal data.  Compound  298  Formula Molecular weight  338.49  Crystal colour, habit  Colourless, prism  Crystal size,  0.500 x 0.450 x 0.300 mm  Crystal system  Orthorhombic  No. reflections used for unit cell determination (20 range)  25 (86.6-89.8°)  Omega scan peak width at half height  0.35  Space group  P2 2 2 (#19)  Lattice parameters:  a =  1  b=  l  1  9.598 (2) A 23.435 (2) A  c=  8.708 (2) A  V=  1958.7 (7) A  Z value  4  Dcalc.  1.148 g/cm  F(000)  744  U(Cu-/i:a)  5.88 cm"  3  3  1  B. Intensity Measurements Difrractometer  Rigaku AFC6S  Radiation  Cu-K (k = 1.54178 A)  Temperature  21°C  a  279  Takeoff angle  6.0°  Detector aperature  6.0 mm horizontal 6.0 mm vertical  Crystal to detector distance  285 mm  Scan type  u)-26  Scan width  (1.25 + 0.35 tan 6)0  Scan rate  16.0°/min  26max  145.3° Total: 2331  No. of reflections measured  Unique: 2007 (Ri = 0.030) nt  Lorentz-polarisation Corrections  Absorption (trans, factors: 0.90-1.00) Secondary extinction (coefficient: 0.27296 E -04)  C. Structure Solution and Refinement Structure solution  Direct methods  Hydrogen atom treatment  Refined or included in calculated positions (dc-H = 0.95 A )  Refinement  Full-matrix least-squares  Function minimised  Lw(IF l-IF l) 0  c  2  where w = F / o ( F ) 0  2  2  0  2  R = LIIF I-IF II/EIF I 0  R  C  0  = (XWOFQI-IFCI^/ZWIFOI ) / 2  w  gof=[Z(IF l-IF l) /(m-n)] / 0  p-factor  0.03  c  2  1  2  1  2  Anomalous dispersion  All non-hydrogen atoms  No. of reflcns with / > 3a(/)  1824  No. of variables  235  Reflection/parameter ratio  7.76  Residuals: R; R  0.045; 0.069  w  Goodness of fit indicator  3.39  Max shift/error in final cycle  0.02  Maximum peak in final diff. map  +0.22 e/A  Minimum peak in final diff. map  -0.24 e/A  Values given for R, R  w  3  3  and gof are based on those reflections with / > 3a(/).  Figure 16. Single crystal X-ray structure of diol 298.  281  Table 17. Final atomic coordinates (fractional) and Beq (A ) [Compound 298]. 2  atom  X  0(1)  0.2820 (2)  0.78494 (7)  0.8010 (3)  4.07 (8)  0(2)  0.0407 (2)  0.77147 (8)  0.7991 (3)  4.12 (8)  0(3)  0.1542 (2)  0.95564 (7)  0.6187 (3)  4.43 (9)  0(4)  0.2175 (3)  0.9684 (1)  0.3158 (3)  6.0 (1)  C(l)  0.2134 (3)  0.8987 (1)  0.6202 (3)  3.5 (1)  C(2)  0.1274 (3)  0.8617 (1)  0.7328 (3)  3.8 (1)  C(3)  0.1557 (3)  0.7969 (1)  0.7212 (3)  3.6 (1)  C(4)  0.1632 (4)  0.7760 (1)  0.5552 (4)  4.6 (1)  C(5)  0.2500 (4)  0.8146 (1)  0.4517 (3)  4.7 (1)  C(6)  0.1972 (3)  0.8768 (1)  0.4535 (3)  4.0 (1)  C(7)  0.2860 (4)  0.9138 (1)  0.3466 (4)  5.1 (1)  C(8)  0.4283 (4)  0.9260 (1)  0.4013 (5)  5.8 (2)  C(9)  0.4649 (3)  0.9232 (1)  0.5469 (5)  5.4 (2)  C(10)  0.3668 (3)  0.9022 (1)  0.6671 (4)  4.3 (1)  C(ll)  0.6098 (4)  0.9395 (2)  0.6060 (9)  8.7 (3)  C(12)  0.7019 (6)  0.8953 (3)  0.621 (1)  12.2 (5)  C(13)  0.6214 (9)  0.9939 (4)  0.658 (2)  12.4 (7)  C(13B)  0.684 (2)  0.9847 (8)  0.512 (3)  13 (1)  C(14)  0.0458 (4)  0.8796 (1)  0.3970 (4)  5.5 (2)  C(15)  0.1383 (4)  0.8826 (1)  0.8992 (4)  5.4 (2)  C(16)  0.3068 (3)  0.7261 (1)  0.8307 (4)  4.7 (1)  C(17)  0.1868 (3)  0.6987 (1)  0.9164 (4)  4.3 (1)  C(18)  0.0558 (3)  0.7119 (1)  0.8272 (4)  4.4 (1)  C(19)  0.1782 (4)  0.7231 (1)  1.0795 (4)  5.5 (2)  C(20)  0.2096 (4)  0.6341 (1)  0.9230 (5)  6.0 (2)  y  z  Beq  282  Table 18. Bond lengths (A) with estimated standard deviations in brackets [Cpd. 298] atom  atom  length  atom  atom  length  0(1)  C(16)  1.424 (3)  C(6)  C(7)  1.530 (4)  0(1)  C(3)  1.425 (3)  C(6)  C(14)  1.535 (4)  0(2)  C(3)  1.425 (3)  C(7)  C(8)  1.474 (5)  0(2)  C(18)  1.426 (3)  C(8)  C(9)  1.318 (6)  0(3)  C(l)  1.450 (3)  C(9)  C(10)  1.492 (5)  0(4)  C(7)  1.463 (4)  C(9)  C(ll)  1.531 (5)  C(l)  C(10)  1.530 (4)  C(ll)  C(12)  1.368 (7)  C(l)  C(6)  1.548 (4)  C(ll)  C(13)  1.36 (1)  C(l)  C(2)  1.549 (4)  C(ll)  C(13B)  1.51 (2)  C(2)  C(15)  1.533 (4)  C(16) C(17)  1.515 (4)  C(2)  C(3)  1.546 (3)  C(17) C(18)  1.510 (4)  C(3)  C(4)  1.528 (4)  C(17) C(19)  1.534 (5)  C(4)  C(5)  1.525 (4)  C(17) C(20)  1.531 (4)  C(5)  C(6)  1.545 (4)  283  Table 19. Bond angles (deg) with estimated standard deviations in brackets. atom  atom  atom  angle  atom  atom  atom  angle  C(16)  O(l)  C(3)  114.9 (2)  C(14)  C(6)  CQ)  112.5 (3)  C(3)  0(2)  C(18)  114.4 (2)  C(5)  C(6)  CQ)  106.8 (2)  0(3)  C(l)  C(10)  109.4 (2)  0(4)  C(7)  C(8) *  107.7 (3)  0(3)  C(l)  C(6)  104.9 (2)  0(4)  C(7)  C(6)  110.9 (3)  0(3)  C(l)  C(2)  108.2 (2)  C(8)  C(7)  C(6)  115.4 (3)  C(10)  C(l)  C(6)  111.4 (2)  C(9)  C(8)  C(7)  123.3 (3)  C(10)  C(l)  C(2)  111.9 (2)  C(8)  C(9)  CQO)  121.5 (3)  C(6)  C(l)  C(2)  110.7 (2)  C(8)  C(9)  CQl)  123.6 (4)  C(15)  C(2)  C(3)  111.3 (2)  C(10)  C(9)  C(ll)  114.8 (4)  C(15)  C(2)  C(l)  112.5 (2)  C(9)  C(10)  CQ)  116.0 (3)  C(3)  C(2)  C(l)  114.6 (2)  C(12)  C(ll)  C(9)  115.5 (4)  O(l)  C(3)  0(2)  110.2 (2)  C(12)  C(ll)  C(13B)  O(l)  C(3)  C(4)  111.0 (2)  C(13)  C(ll)  C(9)  114.9 (5)  O(l)  C(3)  C(2)  108.1 (2)  C(13)  C(ll)  CQ2)  128.8 (6)  0(2)  C(3)  C(4)  110.7 (2)  C(13B) C Q l )  C(9)  114.8 (8)  0(2)  C(3)  C(2)  104.1 (2)  0(1)  C(16)  C(17)  111.8 (2)  C(4)  C(3)  C(2)  112.6 (2)  C(16)  C(17)  CQ9)  109.9 (3)  C(5)  C(4)  C(3)  113.3 (2)  C(16)  C(17)  C(20)  109.1 (3)  C(4)  C(5)  C(6)  112.0 (3)  C(18)  C(17)  CQ6)  107.0 (2)  C(7)  C(6)  C(14)  108.0 (3)  C(18)  CQ7)  CQ9)  110.8 (3)  C(7)  C(6)  C(5)  110.2 (2)  C(18)  CQ7)  C(20)  109.9 (3)  C(7)  C(6)  C(l)  109.1 (2)  C(20)  CQ7)  C(19)  110.0 (3)  C(14)  C(6)  C(5)  110.3 (3)  0(2)  CQ8)  CQ7)  111.9 (2)  106 (1)  284  Table 20. Torsion or conformation angles (in deg) [Compound 298]. (1)  (2)  (3)  (4)  (1)  (2)  (3)  (4)  (3) (3) (3) (3)  0(3)  CQ)  0(3) 0(3) 0(4)  CQ) CQ) C(7)  C(2) C(2) C(2)  44.19 163.06 -76.4 (3) -164.64  0(4) 0(4) 0(4) 0(4)  C(7) C(7) C(7)  H(3) C(15) C(3) H(9) C(9) CQ4)  -50.0 (3) 79.0 (3) 53.5 (3) -78.47 40.40 161.0 (2) -47.53 67.1 (3)  C(l) C(l) C(l) CQ) CQ) CQ)  CQ) CQ) CQ) CQ) CQ) CQ) C(2) C(2)  0(1) 0(1) 0(1) O(l)  C(16) C(17) C(16) C(17) C(16) C(17) C(3) 0(2) C(3) C(4) C(3) C(4) C(3) C(4) C(3) C(2)  0(1) (Xl) 0(2) 0(2) 0(2) 0(2) 0(2) 0(2)  C(3) C(3) C(3) C(3) C(3) C(3) C(3) C(3)  C(2) C(2) 0(1) C(4) C(4) C(4) C(2) C(2)  C(l) C(16) H(4) H(5) C(5) H(3) C(15)  0(2) 0(2) 0(2) 0(2) 0(3) 0(3) 0(3) 0(3) 0(3) 0(3)  C(3) C(18) C(18) C(18) C(l)  C(2) C(17) C(17) C(17) C(10) C(10) C(10) C(6) C(6) C(6)  -163.8 (2) C(l) -53.2 (3) C(16) C(20) -171.6 (3) 66.6 (3) C(19) 45.12 H(10) 163.22 H(ll) -75.8 (3) C(9) 64.6 (3) C(7) C(14) -55.1 (3) C(5) -176.3 (2)  0(1) 0(1) CXI) 0(1)  C(l) C(l) C(l) C(l) C(l)  C(18)  angle  C(20) C(19) C(18) H(4) H(5) C(5) H(3) C(15)  52.5 171.4 -67.9 -53.9  CQ) CQ)  C(2) C(2)  C(7) CQO) CQO) C(6) C(6) C(6) C(6) C(6) C(6) C(6) C(6) C(2) C(2) C(2) C(2) C(3) C(3) C(3) C(3)  C(8) C(8) C(6) C(6) C(6) C(9) C(9) C(7) C(7) CQ4) C(14) CQ4) C(5) C(5) C(5) CQ5) CQ5) C(15) C(3) 0(1) 0(2) C(4) C(4)  C(5) CQ) C(8)  angle 50.71 -64.5 (3) 167.0 (2) -77.48 102.5 (4) 44.8 (3) 165.4 (3) -77.6 (3)  -15.0 (4) 166.4 (3) CQl) 165.14 H(8) 45.2 (4) C(8) 59.71 HQ9) H(21) -60.32 H(20) 179.70 H(6) -58.00 -177.22 H(7) 62.4 (3) C(4) -71.04 HQ6) HQ7) 48.98 HQS) 168.97 C(4) -43.9 (3) CQ6) 166.6 (2) CQ8) -169.5 (2) H(4) 165.50 H(5) -75.63  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  285  Table 20. Torsion or conformation angles (in deg).  continued  (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  C(2)  C(3)  C(4)  C(5)  C(l) C(l) C(l) C(l) C(l) C(l) C(l)  0(3) H(l) C(10) H(10) C(10) H(ll) C(10) C(9) -178.8 (3) C(6) C(7) 61.4 (3) C(6) C(14)  C(4) C(4)  C(3)  C(2) C(2) C(2) C(2) C(2) C(2) C(2)  44.9 (4) 92(2) -74.82 43.28 164.2 (2)  C(2) C(2)  C(3) C(3) C(3) C(3) C(3) C(3) C(3) C(3)  0(1) O(l) 0(1) 0(2) 0(2) 0(2) C(4) C(4)  C(16) C(16) C(16) C(18) C(18) C(18) C(5) C(5)  C(3) C(3) C(3) C(3) C(3) C(3) C(4) C(4) C(4) C(4)  C(4) C(2) C(2) C(2) C(2) C(2) C(5) C(5) C(3) C(3)  C(6)  C(5)  -59.7 (3)  H(12) 65.13 H(13) -175.61 -55.2 (3) C(17) H(14) 176.69 -64.05 H(15) 56.3 (3) C(17) 64.37 H(6) -176.42 H(7) -56.0 (4) C(5) C(6) 59.17 C(15) H(16) 179.18 C(15) H(17) -60.82 C(15) H(18) -72.4 (3) C(l) C(10) C(l) C(6) 52.6 (3) -179.2 (3) C(6) C(7) C(6) C(14) -60.1 (3) -69.4 (3) 0(1) C(16) 0(2) C(18) 69.3 (3)  C(5) C(5)  C(3) C(6) C(6) C(6) C(6) C(6) C(6)  C(6) C(6) C(6) C(6) C(6) C(6) C(6) C(6)  C(7) C(7) C(7) C(5) C(5) CQ) CQ) CQ)  C(6) C(6) C(6) C(7) C(7) C(7) C(7) C(7) C(7)  CQ) CQ) CQ) C(8) C(8) C(6) C(6) C(6) C(6)  C(7)  C(6)  C(5) C(5) C(5) C(5)  (4)  angle  72.39 H(3) C(15) -173.0 (3) 48.16 C(7) H(8) -71.8 (3) C(7) C(8) C(14) HQ 9) 178.87 58.84 C(14) H(21) CQ4) H(20) -61.14 CQ) 0(4) C(8) C(8) C(4) C(4) 0(3) C(10) C(10) C(10) C(2) C(2) C(9) C(9) CQ4) CQ4) CQ4) C(5) C(5)  CQO) H(2) H(9) C(9) H(4) H(5) HQ) HQO) HQl) C(9) H(3) CQ5) CQO)  65.5 (3) 52 (3) 158.04 -21.9 (5) -176.60 64.53 -149 (2) 160.58 -81.32  39.6 (3) -63.72 -178.9 (3) 5.5 (5) CQl) -176.0 (3) HQ 9) -60.68 H(21) 179.29 H(20) 59.31 60.37 C(6) -58.84 H(7)  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  286  Table 20. Torsion or conformation angles (in deg). (2)  continued (3)  (4)  angle  (1)  (2)  (3)  (4)  angle  C(7)  C(6)  CQ)  CQO)  -53.5 (3)  CQO) C(9)  CQl)  C(13B)  C(8)  C(9)  CQO) HQO)  -135.95  CQO) CQ)  0(3)  HQ)  C(8)  C(9)  CQO) H Q l )  105.96  CQO) CQ)  C(6)  CQ4)  C(8)  C(9)  CQl)  H(35)  146.98  CQO) CQ)  C(2)  H(3)  C(8)  C(9)  CQl)  H(28)  -5  CQO) CQ)  C(2)  CQ5)  56.1 (3)  C(8)  C(9)  CQl)  CQ3)  95 (1)  CQ6) CQ7)  CQ8) H Q 4)  -173.56  C(8)  C(9)  CQl)  CQ2)  -94.9 (7)  CQ6)  CQ7)  CQ8) H Q 5)  67.19  C(8)  C(9)  CQl)  C(13B)  29(1)  CQ6)  CQ7)  C(20)  H(25)  179.63  C(8)  C(7)  0(4)  H(2)  CQ6) CQ7)  C(20)  H(26)  -60.37  C(8)  C(7)  C(6)  CQ4)  167.7 (3)  CQ6)  CQ7)  C(20)  H(27)  59.62  C(9)  C(8)  C(7)  H(8)  -141.88  CQ6)  CQ7)  CQ9) H(22)  -61.18  C(9)  C(ll)  CQ3) H(33)  -134.15  CQ6)  CQ7)  CQ9) H(23)  58.83  C(9)  CQl)  CQ3) H(34)  -12.36  CQ6) CQ7)  CQ9) H(24)  178.83  C(9)  CQl)  CQ3) H(32)  105.66  CQ8)  CQ7)  CQ6) HQ2)  -67.82  C(9)  C(ll)  CQ3) H(35)  90.89  CQ8)  CQ7)  CQ6) H Q 3)  172.90  C(9)  C(ll)  CQ2) H(30)  55.82  CQ8)  CQ7)  C(20)  H(25)  -63.27  C(9)  C(ll)  CQ2) H(31)  175.88  CQ8)  CQ7) C(20)  H(26)  56.73  C(9)  CQl)  CQ2) H(29)  -64.18  CQ8)  CQ7)  C(20)  H(27)  176.73  C(9)  CQl)  C(13B)H(38)  67.40  CQ8)  CQ7)  CQ9) H(22)  -179.27  C(9)  CQl)  C(13B)H(36)  -162.52  CQ8) CQ7)  CQ9) H(23)  -59.26  C(9)  CQl)  C(13B)H(37)  -49.24  CQ8) CQ7)  CQ9) H(24)  60.74  C(10)  C(9)  C(8)  H(9)  -174.47  CQ4)  C(6)  C(7)  H(8)  -72.37  C(10)  C(9)  CQl)  H(35)  -34.40  CQ4)  C(6)  C(5)  H(6)  179.48  C(10)  C(9)  C ( l l ) H(28)  178.57  CQ4) C(6)  C(5)  H(7)  60.27  C(10)  C(9)  CQl)  CQ3)  -87 (1)  CQl)  C(9)  C(8)  H(9)  4.06  C(10)  C(9)  CQl)  C(12)  83.7 (7)  CQl)  C(9)  CQO) HQO)  -75 (3)  (1)  -152(1) -30 (2) -173.3 (2) 171.30  45.40  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  287  Table 20. Torsion or conformation angles (in deg). (1)  (2)  C(ll)  C(9)  (3)  angle  (4)  (1)  (2)  continued (3)  (4)  angle  CQO) H ( l l )  -72.70  C(13) C ( l l )  C(13B)H(38)  -37.81  C(19) C(17) C(18) H(14)  -53.76  C(13) C ( l l )  C(13B)H(36)  92.27  C(19) C(17) C(18) H(15)  -173.01  C(13) C ( l l )  C(13B)H(37)  -154.45  C(19) C(17) C(16) H(12)  171.78  C(13B)C(11) C(13) H(33)  -29.01  C(19) C(17) C(16) H(13)  52.50  C(13B)C(11) C(13) H(34)  92.79  C(19) C(17) CQO) H(25)  59.00  C(13B)C(11) C(13) H(32)  -149.20  C(19) C(17) CQO) H(26)  179.00  C(13B)C(11) C(12) HQO)  -72.64  C(19) C(17) CQO) H(27)  -61.00  C(13B)C(11) C(12) H Q l )  47.42  C(20) C(17) C(18) H Q 4)  68.03  C(13B)C(11) C(12) H(29)  167.37  CQO) C(17) C(18) H(15)  -51.22  H(2)  0(4)  C(7)  CQO) C(17) C(16) H(12)  51.07  H(3)  C(2)  C(15) H(16)  173.78  CQO) C(17) C(16) H(13)  -68.21  H(3)  CQ)  C(15) H(17)  -66.20  CQO) C(17) C(19) H(22)  59.01  H(3)  CQ)  C(15) H(18)  53.79  CQO) C(17) C(19) HQ3)  179.03  H(4)  C(4)  C(5)  H(6)  -56.21  CQO) C(17) C(19) HQ4)  -60.98  H(4)  C(4)  C(5)  H(7)  63.01  H(8)  169.51  C(12) C ( l l )  C(13) H(33)  57.05  H(5)  C(4)  C(5  H(6)  -175.08  C(12) C ( l l )  C(13) H(34)  178.85  H(5)  C(4)  C(5)  H(7)  -55.86  C(12) C ( l l )  C(13) H(32)  -63.14  H(8)  C(7)  C(8)  H(9)  38.10  C(12) C ( l l )  C(13) H(35)  -77.91  HQ8) C ( l l )  C(13) H(33)  -39.32  C(12) C ( l l )  C(13B)H(38)  -163.78  HQ8) C ( l l )  CQ3) H(34)  82.47  C(12) C ( l l )  C(13B)H(36)  -33.71  HQ8) C ( l l )  C(13) H(32)  -159.51  C(12) C ( l l )  C(13B)H(37)  79.58  HQ8) C ( l l )  C(13) H(35)  -174.28  C(12) C ( l l )  C(13B)C(13)  -126.0 (9)  HQ8) C ( l l )  C(12) H(30)  -39.07  C(13) C ( l l )  C(12) H(30)  -135.44  HQ8) C ( l l )  C(12) H Q l )  80.99  C(13) C ( l l )  C(12) H(31)  -15.38  HQ8) C ( l l )  C(12) H(29)  -159.07  C(13) C ( l l )  C(12) H(29)  104.57  HQ8) C ( l l )  C(13B)H(38)  124.26  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  288  Table 20. Torsion or conformation angles (in deg). (1)  (2)  (3)  (4)  angle  (1)  Q)  continued (3)  H(34) CQ3) C Q l )  (4)  angle  H(35)  -103.25  H(28) C Q l ) C(13B)H(36)  -105.66  H(28) C Q l ) C(13B)H(37)  7.63  H(35) C Q l ) C(13B)H(38)  -50.43  H(29) CQ2) C Q l )  H(35)  53.93  H(35) C Q l ) C(13B)H(36)  79.65  H(30) CQ2) C ( l l )  H(35)  173.92  H(35) C ( l l ) C(13B)H(37)  -167.07  H Q l ) CQ2) C ( l l )  H(35)  -66.01  H(35) C(13) C(13B)H(38)  154.77  H(32) CQ3) C ( l l )  H(35)  14.77  H(35) C(13) C(13B)H(36)  -94.38  H(33) CQ3) C ( l l )  H(35)  134.96  H(35) C(13) C(13B)H(37)  69.47  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  289  Appendix 4.  X-ray Structure Report on Epoxide 308. A. Crystal data.  Compound  308  Formula  C  Molecular weight  222.33  Crystal colour, habit  Colourless, prism  Crystal size,  0.23 x 0.25 x 0.46 mm  Crystal system  Orthorhombic  Space group  P2 2{l  Lattice parameters:  a =  14 22°2 H  x  x  6.3248 (5) A  b=  11.6243 (9) A  c=  17.835 (1) A  V= 1311.2 (2) A Z value  4  ^calc.  1.129 g/cm  F(000)  488  3  3  0.69 c m '  1  B. Intensity Measurements Diffractometer  Enraf-Nonius CAD4-F  Radiation  Mo-*: (kK a  XK  a2  al  = 0.70930,  = 0.71359 A)  Temperature  21°C  Takeoff angle  2.7°  Detector aperature  (2.0 + tan 0) x 4.0 mm  290  Crystal to detector distance  173 mm  Scan type  co-20  Scan width  (0.65 + 0.35 tan 0 ) °  Scan rate  0.8-10.0°/min  20  60°  m a x  Crystal decay  5.8%  No. of unique reflections  2189  C. Structure Solution and Refinement Structure solution  Direct methods  Function minimised  Ew(IF l-IF l) 0  c  2  where w = F / a ( F ) 0  2  2  0  2  R = 2IF I-IF II/XIF I 0  R  C  0  = (ZwOFol-IFJ^/ZwIFol ) / 2 1  w  gof=[Z(IF l-IF l) /(m-«)] 0  No. of reflcns with / > 30(1)  1187  No. of variables  145  Residuals: R; R  c  2  0.045; 0.046  w  Goodness of fit indicator  2.34  Max shift/error in final cycle  0.01  Maximum peak in final diff. map  +0.16 e/A  Minimum peak in final diff. map  -0.16 e/A  3  3  Values given for R, R and gof are based on those reflections with / > 3a(/). w  1/2  2  291  Table 21. Final atomic coordinates (fractional) and Beq (A ) [Compound 308]. 2  z  atom  X  y  Oil)  0.3815 (3)  0.7720 (1)  0.8770 (1)  54  0(2)  0.0327 (4)  0.6927 (2)  0.9738 (1)  92  C(l)  0.2817 (4)  0.8347 (2)  0.8167 (1)  44  C(2)  0.1654 (4)  0.7456 (2)  0.8571 (1)  50  C(3)  0.0005 (5)  0.7723 (2)  0.9142 (1)  63  C(4)  0.0066 (5)  0.8975 (3)  0.9382 (1)  68  C(5)  0.0234 (4)  0.9748 (2)  0.8710 (2)  64  C(6)  0.2354 (4)  0.9630 (2)  0.8295 (1)  51  C(7)  0.2199 (5)  1.0085 (2)  0.7506 (2)  62  C(8)  0.4171 (6)  1.0243 (3)  0.7035 (2)  81  C(9)  0.2854 (5)  0.9179 (1)  0.6938 (1)  59  C(10)  0.3634 (5)  0.8160 (1)  0.7382 (1)  57  C(ll)  0.1664 (6)  0.9011 (2)  0.6210 (2)  77  C(12)  -0.0428 (7)  0.8430 (4)  0.6332 (2)  122  C(13)  0.2965 (7)  0.8363 (4)  0.5638 (2)  99  C(14)  0.4107 (5)  1.0203 (2)  0.8751 (2)  69  UeqX 10  3  292  Table 22. Bond lengths (A) with estimated standard deviations in brackets [Cpd. 308].  atom  atom  0(1)  C(l)  0(1)  length  atom  atom  1.444 (2)  C(6)  C(7)  1.506 (4)  C(2)  1.445 (3)  C(6)  C(14)  1.528 (4)  0(2)  C(3)  1.424 (3)  C(7)  C(8)  1.515 (4)  C(l)  C(2)  1.461 (3)  C(7)  C(9)  1.505 (4)  C(l)  C(6)  1.537 (3)  C(8)  C(9)  1.484 (4)  C(l)  C(10)  1.508 (3)  C(9)  C(10)  1.525 (3)  C(2)  C(3)  1.490 (4)  C(9)  C(ll)  1.516 (4)  C(3)  C(4)  1.518 (4)  C ( l l ) C(12)  1.502 (5)  C(4)  C(5)  1.503 (4)  C ( l l ) C(13)  1.512 (4)  C(5)  C(6)  1.537 (4)  length  Figure 17. Single crystal X-ray structure of epoxide 308.  293  Table 23. Bond angles (deg) with estimated standard deviations in brackets [Cpd. 308]. atom  atom  atom  C(l)  O(l)  C(2)  0(1)  C(l)  0(1)  angle  atom  atom  atom  angle  60.77 (14)  C(5)  C(6)  C(7)  111.2 (2  C(2)  59.62 (14)  C(5)  C(6)  C(14)  109.7 (2  C(l)  C(6)  117.6 (2)  C(7)  C(6)  C(14)  113.0 (2~  0(1)  C(l)  C(10)  117.9 (2)  C(6)  C(7)  C(8)  120.5 (2  C(2)  C(l)  C(6)  121.3 (2)  C(6)  C(7)  C(9)  111.8 (2  C(2)  C(l)  C(10)  121.9 (2)  C(8)  C(7)  C(9)  58.9 (2  C(6)  C(l)  C(10)  110.1 (2)  C(7)  C(8)  C(9)  60.2 (2  0(1)  C(2)  C(l)  59.61 (14)  C(7)  C(9)  C(8)  60.9 (2  O(l)  C(2)  C(3)  116.8 (2)  C(7)  C(9)  C(10)  106.3 (2  C(l)  C(2)  C(3)  122.8 (2)  C(7)  C(9)  C(ll)  122.6 (2  0(2)  C(3)  C(2)  105.9 (2)  C(8)  C(9)  C(10)  113.9 (2  0(2)  C(3)  C(4)  114.1 (2)  C(8)  C(9)  C(ll)  119.7 (2  C(2)  C(3)  C(4)  112.0 (2)  C(10)  C(9)  C(ll)  119.5 (2  C(3)  C(4)  C(5)  110.5 (2)  C(l)  C(10)  C(9)  104.9 (2  C(4)  C(5)  C(6)  113.1 (2)  C(9)  C(ll)  C(13)  112.3 (3  C(l)  C(6)  C(5)  108.9 (2)  C(9)  C(ll)  C(12)  112.2 (3  C(l)  C(6)  C(7)  102.4 (2)  C(13)  C(ll)  C(12)  110.7 (3  C(l)  C(6)  C(14)  111.3 (2)  294  Table 24. Torsion or conformation angles (in deg) [Compound 308].  (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  angle  C(2)  O(l)  C(l)  C(6)  111.9 (2)  C(4)  C(5)  C(6)  C(l)  48.6 (3;  C(2)  O(l)  C(l)  C(10)  -112.5 (2)  C(4)  C(5)  C(6)  C(7)  160.6 (2;  C(l)  O(l)  C(2)  C(3)  -114.1 (2)  C(4)  C(5)  C(6)  C(14)  -73.5 (3;  0(1)  C(l)  C(2)  C(3)  104.1 (2)  C(l)  C(6)  C(7)  C(8)  -74.1 (3;  C(6)  C(l)  C(2)  0(1)  -105.8 (2)  C(l)  C(6)  C(7)  C(9)  -8.4 (3;  C(6)  C(l)  C(2)  C(3)  -1.7 (3)  C(5)  C(6)  C(7)  C(8)  169.7 (2;  C(10) C(l)  C(2)  0(1)  106.0 (2)  C(5)  C(6)  C(7)  C(9)  -124.6 (2;  C(10) C(l)  C(2)  C(3)  -150.0 (2)  C(14) C(6)  C(7)  C(8)  45.7 (3;  0(1)  C(l)  C(6)  C(5)  -84.3 (2)  C(14) C(6)  C(7)  C(9)  111.5 (2;  O(l)  C(l)  C(6)  C(7)  157.8 (2)  C(6)  C(7)  C(8)  C(9)  98.5 (2;  0(1)  C(l)  C(6)  C(14)  36.8 (3)  C(6)  C(7)  C(9)  C(8)  -113.4 (3  C(2)  C(l)  C(6)  C(5)  -14.9 (3)  C(6)  C(7)  C(9)  C(10)  -4.7 (3  C(2)  C(l)  C(6)  C(7)  -132.7 (2)  C(6)  C(7)  C(9)  C(ll)  138.1 (3  C(2)  C(l)  C(6)  C(14)  106.3 (3)  C(8)  C(7)  C(9)  C(10)  108.7 (3  C(10) C(l)  C(6)  C(5)  136.8 (2)  C(8)  C(7)  C(9)  C(ll)  -108.5 (3  C(10) C(l)  C(6)  C(7)  18.9 (3)  C(7)  C(8)  C(9)  C(10)  -96.1 (2  C(10) C(l)  C(6)  C(14)  -102.1 (2)  C(7)  C(8)  C(9)  C(ll)  113.1 (3  0(1)  C(l)  C(10) C(9)  -161.0 (2)  C(7)  C(9)  CQO) C(l)  16.1 (3  C(2)  C(l)  C(10) C(9)  129.2 (2)  C(8)  C(9)  C(10) C(l)  81.0 (3  C(6)  C(l)  C(10) C(9)  -22.2 (3)  C ( l l ) C(9)  C(10) C(l)  -128.1 (3  0(1)  C(2)  C(3)  0(2)  -69.3 (3)  C(7)  C(9)  C ( l l ) C(13)  162.1 (3  0(1)  C(2)  C(3)  C(4)  55.7 (3)  C(7)  C(9)  C ( l l ) C(12)  -72.5 (4  C(l)  C(2)  C(3)  0(2)  -138.9 (2)  C(8)  C(9)  C(12) C(13)  89.5 (3  C(l)  C(2)  C(3)  C(4)  -13.9 (3)  C(8)  C(9)  C ( l l ) C(12)  -145.1 (3  0(2)  C(3)  C(4)  C(5)  166.5 (2)  C(10) C(9)  C ( l l ) C(13)  -59.7 (4  C(2)  C(3)  C(4)  C(5)  46.2 (3)  C(10) C(9)  C ( l l ) C(12)  65.7 (4  C(3)  C(4)  C(5)  C(6)  -67.0 (3)  The sign is positive if when looking from atom 2 to atom 3 a clockwise motion of atom 1 would superimpose it on atom 4.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059759/manifest

Comment

Related Items