Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The chemistry of thujone : enantioselective syntheses of drimanetype antifeedants and ambergris fragrances Chen, Yong-Huang 1992

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

Item Metadata

Download

Media
831-ubc_1992_spring_chen_yong-huang.pdf [ 5.46MB ]
Metadata
JSON: 831-1.0061731.json
JSON-LD: 831-1.0061731-ld.json
RDF/XML (Pretty): 831-1.0061731-rdf.xml
RDF/JSON: 831-1.0061731-rdf.json
Turtle: 831-1.0061731-turtle.txt
N-Triples: 831-1.0061731-rdf-ntriples.txt
Original Record: 831-1.0061731-source.json
Full Text
831-1.0061731-fulltext.txt
Citation
831-1.0061731.ris

Full Text

THE CHEMISTRY OF THIJJONE: ENANTIOSELECTIVE SYNTHESES OF DRIMANE TYPE ANTIPEEDANTS AND AMBERGRIS FRAGRANCES Yong-Huang Chen B.Sc., Amoy University, 1985 A THESIS SUBMflTED 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 UNWERSITY OF BRITISH COLUMBIA January 1992 ©Yong-Huang Chen, 1992  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 of  /  ,‘  The University of British Columbia Vancouver, Canada Date  DE-6 (2/88)  1/12/L(  I,  7a  Abstract This thesis is concerned with the development of thujone (3) as an effective chiral building block for natural product synthesis. Treatment of thujone (3) with ozone in solution gave thujonol (94) and thujonone (95) in a good total yield (70%) via oxidation of the tertiary carbon in the isopropyl side chain. This type of selective oxidation with ozone was generally applicable to a series of thujone derivatives, thus providing versatile intermediates for the syntheses of compounds of interest in the fields of insecticides and perfumery chemicals. Studies on acid promoted ring cleavage of cyclopropylcarbinols obtained from ozonation revealed three distinct pathways, depending on substrates and reaction conditions. Treatment of 97 with concentrated hydrochloric acid gave chloride 123 while heating alcohol 130, derived from thujone in five steps, in dioxane:water with a catalytic amount of p toluenesulfonic acid generated homoallylic alcohol 144. On the other hand, concentrated hydrobromic acid treatment of 120, obtained from thujonol (94) by Robinson annulation, resulted in bromide 322. Compound 322 was further reduced with tributyltin hydride to natural (+)-J3-cyperone (8), thus completing a new four step synthesis from thujone (3). In a projected synthesis of drimane antifeedants (-)-polygodial (2) and (-)-warburganal (10), a novel radical-mediated ring expansion from 123 to 126 was discovered when the former was treated with tributyltin hydride. However, when a related intermediate 132, derived by treatment of 130 with hydrochloric acid, was reacted in this manner, no rearrangement but simple reduction to 133 was observed. Clearly, the ring expansion process is critically dependent on the nature of functionality in ring A. Generation of 126 and, in turn, subsequent intermediates afforded a convenient route to the exclusion of the original isopropyl side chain in many thujone-derived compounds by ozonolysis.  11  An alternative route developed for the exclusion of the isopropyl side chain involved Baeyer-Villiger oxidation. For example, ketone 131 available from ozonation of alkane 128, when subjected to m-CPBA oxidation, provided acetate 160, which after hydrolysis to cyclopropanol 161 and treatment of the latter with ferric chloride yielded 13-chloroketone 162. The enone 163, obtained from dehydrochlorination of 162, was converted to dienone 168 with phenylselenenyl chloride and hydrogen peroxide. Birch reduction of 168 generated the crucial intermediate 64. Since enone 64 had been previously converted to (-)-polygodial (2) and (-)-warburganal (10), a formal enantioselective synthesis was thus completed. The enone 163 could also serve as an attractive intermediate for the synthesis of  (-)-  Ambrox® (179). Stereoselective conjugate addition of enone 163 with vinylmagnesium bromide and cuprous iodide yielded compound 245 which was further regioselectively methylated to 246. Introduction of a double bond into 246 via selenium chemistry as noted above furnished 250 which was reduced to the trans-fused decalone 251 by Birch reduction. L-Selectride treatment of 251 produced the axial alcohol 253 and subsequent hydroboration yielded the 1,5-diol 255. p-Toluenesulfonic acid catalyzed cyclization of the 1,5-diol 255 provided the potent ambergris odorants ()Ambrox® (179) and an interesting rearrangement compound 257 as major products. At lower temperature (80°C), 179 was the major product while 257 became predominant at higher temperature (100°C). Ring expansion of thujone was also investigated in order to explore alternative routes leading to the synthesis of (2) and (179). Reaction of thujone (3) with ethyl diazoacetate ’ (272). t generated 3-Ketoester 270, which upon decarboxylation furnished “homothujone Robinson annulation of compound 272 yielded enone 274. Alkane 291 was derived from 274 in three steps and its ozonation reaction was performed. Surprisingly, the normally observed attack at the tertiary carbon of the isopropyl side chain did not occur. Instead, ketone 292 was isolated as the major product.  111  Q”?ll  0’H  °i( 95  0  °H :97  x:::f::< 123  O1T_ 322  144  130  ç3 cE! =< 3  126  132  133  8  “r 131  128  CHO CHO  CI  cti  OR 160: R=Ac 161: R=H  162  163  2: R=H 10: R=OH  64  168  0 245  250  246  251  253  cE;c90o 255  0 OEt 270  179  272  257  274  291  iv  292  Table of Contents Abstract List  .  of Figures  ii  .ix  List of Schemes  xi  List of Tables  xiii  List of Abbreviations  xiv  Acknowledgements  xvii  Chapter 1. General Introduction  1  1.1. Synthesis of Enantiomerically Pure Compounds 1.2. Thujone as a Chiral Building Block  1 3  Chapter 2. Studies Directed to the Synthesis of (-)-Polygodial and (-)-Warburganal. .6 2.1. Introduction 6 . .  2.1.1.  Drimane-type Antifeedants  2.1.2.  Total Synthesis of Drimane-type Antifeedants  2.2. Discussion 2.2.1.  6 11 24  General Considerations about the Synthesis of (-) -Polygodial (2) and (-)-Warburganal (10) from Thujone (3)  24  2.2.2.  Ozonation of Thujone and Its Derivatives  25  2.2.3.  Stereochemistry of Hydrogenation of Thuj one-derived Tricyclic Enones  34  2.2.4.  Acid Promoted Ring Cleavage of Thujone-derived Cyclopropylcarbinols  41  2.2.5.  The Radical-mediated Rearrangement  45  2.2.6.  Failure of the Radical-mediated Ring Expansion Reaction  49  2.2.7.  Further Studies on the Acid-promoted Ring Cleavage of Cyclopropylcarbinols  58  2.2.8.  Baeyer-Villiger Oxidation of Cyclopropyl Ketones 2.2.9. Regioselective Ring Opening of the Cyclopropyl Alcohol 161 2.2.10. A Formal Enantioselective Synthesis of (-)-Polygodial (2) and (-)-Warburganal (10)  67 72 77  2.3. Experimental  81  2.3.1.  General  81  2.3.2.  Ozonation: thujone (3) to thujonol (94) and thujonone (95) Catalytic Hydrogenation: enone 7 to ketone 96  83  2.3.3.  V  84  2.3.4.  .85  2.3.5.  Ozonation: ketone 96 to ketol 97 and dione 98 Ozonation: dione 104 to hydroxydione 106 and trione 107  2.3.6.  Catalytic Hydrogenation: enone 113 to ketone 114  89  2.3.7.  Aldol Condensation: hydroxydione 106 to hydroxyenones 117 and 118  87  2.3.8.  89 Catalytic Hydrogenation: hydroxyenones 117 and 118 to ketol 120.91  2.3.9.  Methylation: ketol 97 and 120 to ketol 121  92  2.3.10. Robinson Annulation: thujonol (94) to hydroxyenone 122  93 2.3.11. Cyclopropane Ring Opening Reaction: ketol 97 to chloroketone 123.94 2.3.12. Radical-mediated Rearrangement: chioroketone 123 to enones 125 and 126  95  2.3.13. Methylation: ketone 96 to ketone 119 96 2.3.14. Woif-Kishner-Huang Minion Reaction: ketone 119 to aikane 128 .97 2.3.15. Ozonation: alkane 128 to alcohol 130 and ketone 131 2.3.16. Dehydration: alcohol 130 to alkene 138  98 100  2.3.17. Cyclopropane Ring Opening Reaction: alcohol 130 to chloride 132 .101 2.3.18. Ozonolysis: alkene 138 to ketone 131 2.3.20. Conversion of 138 to 133 via 139 2.3.21. Reduction by Bu SnH: chloride 132 to alkene 133 3 2.3.22. Cyclopropane Sliding Reaction: alcohol 130 to alcohol 144  101 103 104 104  2.3.23. Epoxidation: alcohol 144 to epoxyalcohol 147  105  2.3.24. Reductive Fragmentation by LAH: epoxyalcohol 147 to allylic Alcohol 151  106  2.3.25. Allylic Oxidation by Mn02: homoallylic alcohol 151 to enone 152 .107 2.3.26. Cyclopropane Sliding Reaction: alcohol 130 to acetates 153 and 154  108  2.3.27. Cyclopropane Sliding Reaction: ketol 117 to ketol 155 2.3.28. HOAc Promoted Ring Opening: ketol 117 to ketoacetates 156 and 157  110 111  2.3.30. Saponification: acetate 160 to cyclopropanol 161 113 2.3.31. Cyclopropane Ring Opening Reaction by FeCl : cyclopropanol 157 3 to 13-chloroketone 162 114 2.3.32. Dehydrochiorination: f3-chloroketone 162 to enone 163 115 2.3.33. Ring Opening Reaction by NBS: cyclopropanol 161 to -bromoketone 167  116  2.3.34. Dehydrogenation: enone 167 to dienone 168  117  vi  2.3.35. Birch Reduction: dienone 168 to enone 64 2.3.36. Dehydrogenation: ketone 171 to dienone 172 2.3.37. Birch Reduction: dienone 172 to enone 173 and ketone 174 Chapter 3. The Synthesis of Ambergris Fragrances 3.1. Introduction  118 119 119 122 122  3.1.1.  Ambergris Fragrances  122  3.1.2.  Structure and Activity Relationship of Ambergris Fragrances Synthesis of Ambrox®  125 128  3.1.3.  3.2. Discussion 3.2.1. 3.2.2. 3.2.3. 3.2.4.  137  Retrosynthetic Analysis for Synthesis of ()Ambrox® (179) from Enone 163  137  Studies on Conjugate Addition to Enone 163 and Subsequent Methylation of 245 Conversion of cis-fused y,-enone 246 to trans-fused y,ö—251 Synthesis of Diol 255 from trans-Fused y,6-Enone 251  139 148 153  3.3. Future Developments 3.4. Experimental 3.4.1. Conjugate Addition: cL,3-enone 163 to cis-fused y,6-enone 245 3.4.2. Methylation by LDA and Iodomethane: cis-fused y,ö-enone 245 to cis-fused y,-enone 246  164 167 167 168  3.4.3.  Dehydrogenation by PhSeCI/H : cis-fused ‘y,6-enone 246 0 2  3.4.4.  to dienone 250 Birch Reduction: dienone 250 to trans-fused y,ö-enone 251 Reduction by L-Selectride: trans-fused -enone 251 to  169 170  alcohol 253  171  3.4.5. 3.4.6.  Hydroboration: alcohol 253 to 1,5-diol 255 172 3.4.7. Cyclization: 1,5-Diol 255 to 179, 189, and 257 173 Chapter 4 Exploratory Studies of Different Strategies to Develop Thujone as a Chiral Building Block 176 4.1. Studies on ‘Homothujone” and Its Derivatives: a new strategy 176 4.1.1. Regioselective Ring Expansion of Thujone 178 4.1.2. Stereoselective Robinson Annulation of Homothujone (272) 184 4.1.3.  Attempted Generation of the trans-Fused Hydrocarbon 284 4.1.4. Ozonation of 291 4.2. Studies on Utilizing the C2-C3 Bond Cleavage Products: a strategy 4.3. A Formal Synthesis of(+)-13-Cyperone: a ].Q strategy vii  189 198 200 204  4.4. Concluding Remarks: prospect of thujone chemistry  213  4.5. Experimental  216  4.5.2.  Ring Expansion: thujone (3) to ketoester 270 Decarboxylation: ketoester 270 to homothujone (272)  216 217  4.5.3.  Robinson Annulation: homothujone (272) to enone 274  218  4.5.4.  Birch Reduction-CH3I Trapping and Birch Reduction-TMSC1  4.5.1.  Trapping-Simmons-Smith Reaction-Hydrolysis Sequences: enone 274 to ketone 277  219  4.5.5. 4.5.6.  Catalytic Hydrogenation: enone 274 to ketone 278 Methylation: a,3-enone 274 to 13,y-enone 282  221 222  4.5.7.  Wolf-Kishner-Huang Minlon Reaction: 3,y-enone 282 to  4.5.8. 4.5.9.  alkene 283  223  Hydroboration: alkene 283 to diol 285 and alcohol 286 Hydroboration: alkene 283 to diol 285 and alcohol 286  224  4.5.10. 0-Methylation: ketone 278/279 to methyl enol ether 280 4.5.11. Wolf-Kishner-Huang Minlon Reaction: ketone 277 to Alkane 291  225 226 .  .227  4.5.12. Ozonation: alkane 291 to ketone 292 and alcohol 293 4.5.13. Ketoacid 308  228  4.5.14. Methylation by Diazomethane: ketoacid 308 to ketoester 309 4.5.15. Ozonation: ketoester 309 to compound 310 4.5.16. Cyclopropane Ring Opening Reaction: thujonol (94)  230  to bromoenone 318 and carvacrol (319)  229 230 231  4.5.17. Cyclopropane Ring Opening Reaction: thujonol (94) to chloroenone 320 and carvacrol (319)  233  4.5.18. Cyclopropane Ring Opening Reaction: hydroxyenone 122 to bromodienone 322  233  Bibliography  235  Appendix 1 X-ray Structure Report on Epoxide 147  247  Appendix 2 X-ray Structure Report on Diol 285  259  viii  List of Figures Figure 1  Drimane Antifeedants  .7  Figure 2  Analogues of Drimane Antifeedants  9  Figure 3  Interaction between Drimane Antifeedants and the Insect’s Receptor: the first hypothesis  9  Interaction between Drimane Antifeedants and the Insect’s Receptor: the second hypothesis  10  Figure 4 Figure 5 Figure 6  Chiral Starting Materials for the Synthesis of (-)-Warburganal (10) Oxygen Insertion into Carbon-Hydrogen Bonds  22  Figure 7  32  Figure 8  Oxygen Insertion into Carbon-Carbon Bonds Decoupling Experiments on Ketone 96  Figure 9  Single Crystal X-ray Structure of 98 (PLUTO Drawing)  37 .40  31 36  Figure 10 Comparison of CD Spectra of 121 Prepared from Two Different Routes Figure 11 A Notation of Ring Cleavage Reactions Figure 12 Rationalization of HC1 Promoted Ring Cleavages  .  42 44  Figure 13 A Proposed Mechanism for the Novel Ring Expansion of 123 Figure 14 Rationalization of the “Methyl Effect”  46  Figure 15 Single Crystal X-ray Structure of Epoxide 147 (ORTEP Drawing) Figure 16 Mechanism of the Fragmentation of Epoxide 147  60 63 64  57  Figure 17 Mechanism of the “Cyclopropane Sliding Reaction” Figure 18 Regioselective Cleavage of Cyclopropanols Figure 19 FMO Interactions of Carbinyl and Oxyl Radicals with Cyclopropane C-C Bonds Figure 20 The Constituents of Ambergris Figure 21 Stereoisomers of ()Ambrox®  72 76 123 126  Figure 22 The Effect of the gem-Dimethyl Groups on the Ambergris Odor Activity Figure 23 Triaxial Rule of Ambergris Odor Sensation Figure 24 The Ambergris Triangle Rule Figure 25 Several Other Diterpene Starting Materials for ()Ambrox® Synthesis Figure 26 Potential Candidate Intermediates for the Stereoselective Conjugate Addition  ..  .126 127 127 132 142  Figure 27 Decoupling Experiments of 246  147  Figure 28 Stereochemistry of Phenylselenenylation of 246 Figure 29 Structural Analysis of Stereoselective Reduction Product 253  151  ix  154  Figure 30 1,4-Diols Utilized for Acid Catalyzed Cyclization to ()Ambrox® Figure 31 Mechanistic Analysis of Cationic Cyclizations of 202 and 235  157  Figure 32 Mechanism for the Formation of 257 Figure 33 The Conversion of 3-13-Friedelanol (263) into 13 (18)-Oleanene (264)  162 163  Figure 34 Decoupling Experiments of 272 Figure 35 Conformational Analysis of 270cx and 272x  181 183  Figure 36 Explanation for Regioselectivity of the Carbon Insertion Reaction  183  Figure 37 2D-HETCOR spectrum of 274  187  Figure 38 The ‘ C Broad Band Decoupling (BB) and APT Spectra of 274 3  188  Figure 39 The NOE Experiments of 274  190  Figure 40 Single Crystal X-ray Structure of 285 (ORTEP Drawing)  195  Figure 41 Novel Cyclopropane Ring Cleavage in the Hydroboration of 283  196  156  Figure 42 A Structural Misperception for 301 Figure 43 The Endo-type Cleavage Mechanism for the Formation of 318 and 319 Figure 44 The Ring Opening reaction of 122 via the Endo-type Cleavage Pathway  . .  . .  203 .208 .212  Figure 45 Incorporation of a Dimethylated Bicylo[3.3.0]octane unit Figure 46 Single Crystal X-ray Structure of Epoxide 147 (PLUTO Drawing) Figure 47 The Unit Cell Structure of Epoxide 147 (Packing Diagram)  213 249  Figure 48 Single Crystal X-ray Structure of Diol 285 (PLUTO Drawing)  261  Figure 49 The Unit Cell Structure of Diol 285 (Packing Diagram)  262  x  250  List of Schemes Scheme 1  Robinson Annulation of Thujone  .4  Scheme 2  Ozonation of Thujone and Its Derivatives  5  Scheme 3 Scheme 4  12 13  Scheme 5  de Groot’s Synthesis of (±)-Polygodial Cortes’ First Synthesis of (-)-Polygodial (2) Cortes’ Second Synthesis of (-)-Polygodial (2)  14  Scheme 7  Mon’s Synthesis of (+)-Polygodial (21)  16  Scheme 8 Scheme 9  He and Wu’s Synthesis of (-)-Polygodial (2) de Groot’s Synthesis of (-)-Polygodial (2)  17 18  Scheme 10 Kutney’s Synthesis of (-)-Polygodial (2) Scheme 11 Goldsmith’s Synthesis of (±)-Warburganal  19 20  Scheme 12 Kende’s Synthesis of (±)-Warburganal Scheme 13 Ohno’s Synthesis of (-)-Warburganal (10)  21 23  Scheme 14 The Overall Plan towards the Synthesis of (-)-Polygodial (2) and (-)-Warburganal (10)  25  Scheme 15 A Perceived Sequence to Utilize Alcohols Derived from Ozonation of Thujone Derivatives Scheme 16 An Ozonation Route to a trans-fused Decalone (Retrosynthetic Analysis) Scheme 17 Generality of Selective Ozonation of Thujone Derivatives  .  .  26 .27 29  Scheme 18 Attempted Catalytic Hydrogenation of Tricyclic Enones Scheme 19 Gao’s Synthesis of a (-)-Polygodial Analogue 127  48  Scheme 20 Gunning’s Synthesis of Rose Oil Fragrances  48  Scheme 21 A Revised Plan to an Enantiomerically Pure, trans-fused Decalone 65 Scheme 22 Radical-initiated Selective Ring Cleavage of a Vinylcyclopropane 136  49  Scheme 23 Radical-initiated Ring Cleavage of Vinylcyclopropane 138 Scheme 24 Precedents of the Endo-type Cleavage  55  39  54 58  Scheme 25 Generality of the Cyclopropane Sliding Reaction 66 Scheme 26 Utilization of Cyclopropyl Ketone 131 via Baeyer-Villiger and Cyclopropanol Cleavage Reactions  69  Scheme 27 The Preparation of Enantiomerically Pure Enone 64 from 163 Scheme 28 A Possible Sequence to a New (-)-Polygodial Analogue Scheme 29 Stoll and Hinder’s Synthesis of ()Ambrox® from Sclareol (187) Scheme 30 Nafs Synthesis of ()Ambrox® from Sclareol  77  Scheme 31 Christenson’s Synthesis of ()Ambrox® from Sclareol  131  xi  80 129 130  Scheme 32 Coste-Manere’s Synthesis of ()Ambrox® from Sciareol Scheme 33 Cortes’ Synthesis of ()Ambrox® from (-)-Drimenol (33) Scheme 34 Mori’s Synthesis of ()Ambrox® from Geranylacetone (219) Scheme 35 Matsui’s Synthesis of (±)Ambrox® from Dihydro-3-ionone (226)  132 133 134 135  Scheme 36 Buchi and Wuest’s Synthesis of (±)Ambrox® from Dihytho--ionone (226) Scheme 37 Retrosynthetic Analysis for Synthesis of ()Ambrox®  136 137  Scheme 38 Conjugate Addition of Organocopper Reagents to trans-Fused Octalones ...140 Scheme 39 Conjugate Addition of Organocopper Reagents to Cross-conjugated 140  Dienones Scheme 40 Conjugate Addition of Organocopper Reagents to a cis-Fused Octalone  .141  .. .  Scheme 41 The Formulation of “lonoxide”  158  Scheme 42 A Possible Shorter Route to Compound 257  164  Scheme 43 A Possible Synthesis of Compound 192  165  Scheme 44 A Possible Synthesis of (-)-epi-Ambrox (190)  166  Scheme 45 A Possible Synthesis of Ambraoxide (186) Scheme 46 The Potential of a Regioselective Ring Expansion Reaction Scheme 47 “Homothujone” Strategy for Syntheses of Various Natural Products Scheme 48 An Alternative Sequence to Hydrocarbon 284  166 177  Scheme 49 An Alternative Route to Ketone 277 Scheme 50 Ring Cleavage of seco-(C2-C3) Cyclopropylcarbinols  197 200  Scheme 51 A Novel Sequence to (-)-Polygodial (7)  201  178 193  202 Scheme 52 The Utilization of a seco-(C2-C3) Intermediate 308 Strategy to the Synthesis of (-)-Polygodial (7) .204 Scheme 53 The Final “seco/corro” 210 Scheme 54 A Formal Synthesis of (+)-13-Cyperone (8) .  Scheme 55 A Potential New jj Strategy  212  xii  List of Tables Table 1  Comparison of Dry and Wet Ozonation of Thujone  .27  Table 2  30 47  Table 4  The Wet Ozonation of 96 to 97 and 98 Yield Optimization for Conversion of 123 to 126 The Optimization of Baeyer-Villiger Reaction of Ketone 131  Table 5  Cyclization of the 1,5-Diol 255 under Different Conditions  161  Table 6  Final Atomic Coordinates (fractional) and Beq (A ) of Epoxide 147 2 Hydrogen Atom Coordinates (fractional) and B ) of Epoxide 147 2 0 (A  251  Table 3  Table 7 Table 8 Table 9  Bond Lengths  (A)  of Epoxide 147  Bond Angles (deg) of Epoxide 147 Table 10 Torsional or Conformational Angles (deg) of Epoxide 147 Table 11 Final Atomic Coordinates (fractional) and Beq (A ) of Diol 285 2 Table 12 Hydrogen Atom Coordinates (fractional) and B ) of Diol 285 2 0 (A Table 13 Bond Lengths (A) of Diol 285 Table 14 Bond Angles (deg) of Diol 285 Table 15 Torsional or Conformational Angles (deg) of Diol 285  xlii  70  252 253 254 255 263 264 265 266 267  List of Abbreviations 2D-HETCOR  two dimensional heteronuclear correlation spectroscopy  25 D 1  specific rotation recorded at 25 C using sodium D-line  Ac  acetyl  AIBN  2,2-azoisobutylnitrile  APT  attached proton test (in ‘ C NMR) 3  aq.  aqueous  ax  axial  BB Benz.  broad band decoupling (in C 13 NMR) benzene  bs  broad singlet  c CA  concentration  CD  circular dichroism  1 cm  wave number  conc.  concentrated  cont. 6  continue chemical shift  d  doublet  dd  doublet of doublets  DDQ  2,3-dichloro-5,6-dicyano- 1 ,4-benzoquinone molar circular dichroism  DEG  diethylene glycol  deg DHP  degree (angle)  DIBAL  diisobutylaluminum hydride  DMAP  4-dimethylaminopyridine  DME  dimethoxyethane  DMF  N,N-dimethylformamide  DMS DMSO  dimethyl sulfide dimethyl sulfoxide  dt  doublet of triplets  eq eqv  equatorial  .  o  .  Chemical Abstract  dihydropyran  equivalent  xiv  Et EVK  ethyl ethyl vinyl ketone  FMO  frontier molecular orbital  g GC  gram  HMPA hv  hexamethyiphosphoramide light radiation  HOMO  highest occupied molecular orbital (energetically)  Hz  Hertz  i-Pr  isopropyl  JR  infrared  J  coupling constant wavelength  L-Selectride  lithium tri-sec-butylborohydride  LAH LDA logE  lithium aluminum hydride  LTA  lead tetraacetate lowest unoccupied molecular orbital (energetically)  LUMO  gas-liquid chromatography  lithium diisopropylamide the log of extinction coefficient  M m M+  molar  m-CPBA  meta-chloroperbenzoic acid  m.p.  melting point  m/z max.  mass to charge ratio  Me  methyl  mg  milligram  MHz  mm i1  megahertz minute microliter  mmol  millimole  MS  mass spectrometry  MVK v  methyl vinyl ketone frequency  NB S  N-bromosuccinimide  multiplet molecular ion  maximum  xv  nm NMR  nanometer nuclear magnetic resonance  NOE  nuclear Overhauser effect  °C ORTEP  degree Celsius oak ridge themal ellipsoid program  PCC  pyridinium chlorochromate  Ph  phenyl  ppm  part per million  pyr. O  pyridine effipticity angle molar effipticity angle  [9] q r. t.  quartet room temperature  s  singlet  SOMO  singly occupied molecular orbital  t  triplet  TBDMS THF TMS Ts  t-butyldimethylsilyl tetrahydrofuran trimethylsilyl para-toluenesulfonyl  UV  ultraviolet  v/v  volume to volume ratio  A  Angstrom  xvi  Acknowledgements I wish to express my gratitude to my research supervisor, Dr. James Kutney, for his valuable guidance and advice both during the course of this work and in the preparation of this thesis. I am thankful to Professor Williams von egger Doering at Harvard University for initiating the Chemistry Graduate Program (CGP) between the People’s Republic of China and a group of North American Universities, to which I was admitted in 1986. I am also grateful to the late Professor Xuemin Gu, the former Head of Department of Chemistry at Amoy University, for giving me unforgettable encouragement and help during my preparation for the admission examination. Financial assistance from the State Education Commision of the People’s Republic of  China and the University of British Columbia are gratefully acknowledged. The technical expertise of the staff in the NMR, mass spectrometry, X-ray diffraction analysis, and microanalysis services as well as the glass blowing, mechanical, and electronic shops is very much appreciated. Thanks are also due to Huachun Zeng, Shichang Miao, Duson Milatovic, Edward Koerp, and Thomas Hu, whose friendships have made my life more pleasant and enjoyable in the past few years. It is a great pleasure to express my appreciation of past and present members of Dr. Kutney’s research group, Francisco Lopez, Mijo Samija, Kang Han, Francisco Kuri-Brena, Carles Cirera, Kai Li, and many others, for their advice, enlightening discussions, and unending help. I owe the secretary Pat much for her many little nice suggestions. I am indebted to Thomas Hu, Drs. Phil Gunning, and Terry Jarvis for meticulous proofreading and constructive criticism. Especial thanks go to my mother, whose caring, support, and exemplification give me strength and inspiration throughout these years of education.  xvii  -I  CD  -I.’  B  0  -t  0  B  CD  CD  If there is one way better than another, It is the way of Nature Aristotle  xviv  Chapter 1. 1.1.  General Introduction  Synthesis of Enantiomerically Pure Compounds The synthesis of a chiral compound in its enantiomerically pure form has become an  important goal for organic chemists in recent years. In addition to aesthetic reasons, the very 2 dictates pure dependence of various biological activities on absolute stereochemistry” enantiomers to be prepared and investigated in academic research and frequently only enantiomerically pure agents to be produced in industry. A racemic drug, (±)-thalidomide, had to be withdrawn from the market due to a serious side effect of one of the enantiomers. It was 3 that (R)-(+)-thalidomide (1), an effective sedative, had no teratogenic effects when reported administered to rats and mice even at high doses; but its enantiomer, (S)-(-)-thalidomide, was devoid of sedative effect and resulted in deformities in the animal tested. (-)-Polygodial (2), a drimane type sesquiterpene, showed a potent antifeedant activity against African army worms while its enantiomer (+)-polygodial and the racemic mixture exhibited an undesirable . 4 phytotoxic effect  :Q  1  2  Three basic methods are available to produce enantiomerically pure compounds, including resolution of racemates, application of asymmetic synthesis, and use of chiral : 5 materials as building blocks. Each method has its own advantages and drawbacks 1. Resolution The risk associated with a projected synthesis based upon a resolution is evident because of the empirical nature of resolution; resolution is potentially wasteful unless the  1  undesired enantiomer can be recycled; it has to be performed early in the synthetic sequence to avoid further waste of reagents and labor. However, many resolutions have been successfully carried Out. Resolution provides a rapid access to both enantiomers of a compound, which is desirable in biological studies. 2. Asymmetric Synthesis Asymmetric synthesis holds great promise in producing chiral molecules effectively, as reflected by the vigorous activities in this field in recent years. It has even greater efficiency if the chiral auxiliary is employed catalytically. However, only a few asymmetric syntheses can provide products of high enantiomeric excess reliably without resorting to further enantiomeric enrichment. 3. Chiral Building Blocks The third method is to utilize readily available chiral molecules, either naturally occurring products or their derivatives, as starting materials (i.e., chiral templates). If these chiral building blocks are enantiomerically pure and racemization is avoided by choosing reaction conditions carefully, the method is the safest way of obtaining enantiomerically pure compounds.  However, the initial chiral molecules have to be consumed during their  incorporation into target molecules. Moreover, because of the limited spectrum of readily available chiral compounds, substantial chemistry has to be implemented for their conversion into viable enantiomerically pure intermediates, the preparation of which in racemic form is simpler in perception and / or execution. All these methods are in some way related to the ‘chiral pooi derived from Nature. In resolution, a chiral compound is used to convert enantiomers into two diastereomers or to differentiate them through chemical reactions (i.e., kinetic resolution); in asymmetric synthesis, a chiral auxiliary is employed either catalytically or stoicheometrically to introduce diastereomeric transition states; as a building block, a chiral molecule becomes an integrated part of the target. Therefore, additions to this ‘chiral pool’ by the introduction of new enantiomerically pure compounds and modification of existing ones are always welcomed. 2  1.2.  Thujone as a Chiral Building Block The occurrence of thujone (3) in Western red cedar (Thuja plicata Donn) was reported 7  as early as in 1939 and its absolute stereochemistry was assigned 8 in 1964. This natural product is actually a mixture of two epimers, (-)-x-thujone (4) and (+)-f3-thujone (5) (4:5=10:1). Of these two epimers, ct-thujone is slightly more stable. Treatment of thujone (3) with potassium hydroxide in ethanol 9 gave an equilibrium mixture containing c-thujone and  f3-  thujone in a ratio of approximately 2 to 1.  7 8  4  3  5  The logging practice of Western red cedar which is abundant in British Columbia forests generates a waste product, generally called “slash”. The slash consists of the left-over branches, and leaves. It often must be removed for reforestation and elimination of the potential fire hazard. Alternatively, “on-site” steam distillation of the slash produces an essential oil containing thujone up to 88%, thus providing an inexpensive source of thujone while also serving as a means of removing the left-over slash , Although the oil obtained can 7 be sold for use in the perfumery industry, higher grade chemical products originating from it are well sought after in recent years from the viewpoints of both economic opportunity and environmental concern. In fact, recent synthetic work in our laboratories has proven that thujone is a viable chiral starting material for the enantioselective synthesis of biologically active natural products and their analogues including juvenile hormone analogues’ , pyrethroid 0 , aryl terpenoids’ 11 insecticides , sesquiterpenes 2 , steroids 13 , and insect antifeedants’ 14 5  3  The novelty of thujone chemistry stems from its unique structural features. The bicyclo[3.l.Ojhexane moiety is a cisfusion# of the two smallest odd-rnernbered rings (i.e., 3membered ring and 5-membered ring). The inherent cyclopropane ring should manifest the close relevance of thujone chemistry to the chemistry of cyclopropyl group 17 since its transformation is a necessity for most synthetic efforts. The carbonyl group would lend its versatility to a great range of synthetic elaborations. The Robinson annulation of thujone (3) is a pivotal transformation in which a aJ (Scheme 1). 3 mannerl quaternary carbon center was generated in a highly stereoselective 8 Presumably, the approach of Michael acceptors (e.g., MVK and EVK) took place exclusively from the less hindered convex side of the more stable enolate (i).  Subsequent aldol  condensation of the products (ii) generated tricyclic enones 6 and 7.  base R  3  R  (i)  (ii)  6: R=H 7: R=CH 3 R  Scheme 1 Robinson Annulation of Thujone  This newly generated quaternary center became a reference point in correlating the thujone structure with chosen target molecules, for example, (+)--cyperone (8)13a,  polygodial (2)15, and the steroid analogue  9•1413  There is no organic molecule known to possess a trans-fused bicyclo[3. 1 .O]hexane moiety (see ref. 16). 4  (  CHO H0  9  2  8  Another important transformation recently discovered is the selective functionalization of the tertiary carbon at the isopropyl side chain of thujone and its derivatives by ozonation (Scheme 2). The importance of such a functionalization lies in the possible utilization of two operations required in the synthesis of different target molecules by using intermediates derived from the ozonation process. These two operations are the exclusion of the isopropyl side chain and the regioselective ring cleavage of the cyclopropyl group. The ozonation reaction has been , and more recently, 46 applied to synthesis of drimane antifeedants’ , rose oil fragrances 5 ambergris fragrances (to be described in this thesis).  03 +  Scheme 2 Ozonation of Thujone and Its Derivatives  This thesis is divided into three parts: the first part deals with a formal enantioselective synthesis of (-)-polygodial (2), a potent insect antifeedant; the second part describes the synthetic studies leading to products of ambergris fragrance, one of the most valuable animal fragrances; the third part discusses some exploratory studies of new strategies to develop thujone as a more versatile chiral building block.  5  Chapter 2.  2.1. 2.1.1.  Studies Directed to the Synthesis of (-)-Polygodial and (-)- Warburganal  Introduction Drimane-type Antifeedants Along with herbicides, insecticides are presently the most useful agrochemicals to  c. 1 a 9 cropsl O 9 . Insecticides can be divided into two groups. The first group protect food and 2 includes synthetic organochiorines, organophosphates, dinitrophenols, formamidines, carbamates, and pyrethroids together with the naturally occuring nicotine, rotenone (Derris), sabadilla, and ryania. They have wide applications due to their effectiveness, low cost, and easy usage. However, there are certain disadvantages associated with them: many are quite toxic to vertebrates, fish or beneficial lower forms of life, and are non-selective, killing both pest insects and beneficial insects; some are extremely persistent in the environment and are likely to accumulate in animals; higher and possibly phytotoxic dosages are required because of the developed resistance of some pest insects. The second group of insecticides consist of repellents, attractants, pheromones, insect growth regulators, oviposition inhibitors, and antifeedants. They promise to overcome drawbacks of the first group agents and are attracting the attention of researchers worldwide . 2 ’ 21 2 These compounds are readily degradable and thus friendly to the environment. They are highly specific in insect-plant, insect-insect relations or certain processes within the insect, and therefore less toxic to human beings and useful insects. Many of them are mimics of or even the same as compounds which are essential in the life processes of the pest insect, thus making it more difficult for the insect to restrict the uptake and detoxify such molecules than in the case of synthetic insecticides. In other words, the development of resistance from the target insect is less likely to happen. a, an antifeedant is defined as a chemical which does not kill the 23 According to Munaka  insect directly but inhibits feeding, the insect often remaining near the treated plant material and  6  possibly dying through starvation. An antifeedant is different from an olfactory repellent which is usually a volatile compound that repels the insect before it starts to eat. The exact mode of action of these antifeedants is still largely speculative . 2 ” 23 3 They are believed to play . Probably 24 a major role in the ever continuing battle for survival between insects and plants all plants contain one or more substances which are unpalatable to insects. Plants selection  programmes in the evolution process have often chosen varieties with higher contents of antifeedants. The use of these compounds as crop protection agents seems to bear considerable . 2 ’ 22 promise 5  12 11  1  CHC 6 14  2 R  3 1 R = 2 H R 2: = 10: R =OH, 3 1 = 2 R = I-1 R =H, 3 2 R = OH 11: R =OH, R 1 =OAc, R 2 =H 3 12: R =OH, R 1  13  Figure 1 Drimane Antifeedants  Several sesquiterpenes, mostly of the drimane type, were isolated from the bark of East African medical plants Warburgia ugandensis and W. stuhmannii (Cannellaceae). Some of these sesquiterpenes (Figure 1)26, (-)-polygodial (2), (-)-warburganal (10), 3hydroxywarburganal (11), ugandensidial (12), and muzigadial (13), possess strong antifeedant activity against the African army worms, Spodoptera exempta and S. littoralis. More recently, (-)-polygodial (2) and (-)-warburganal (10) were shown to inhibit feeding and colonization of the aphid of Myzus persicae and to decrease the transmission of different plant  7  viruses by the aphid . Among many other biological activities exhibited by these antifeedants, 27 the hot taste to humans appears the most interesting. Kubo and Ganjain 28 suggested a correlation between the antifeedant activity and the pungent taste experienced by human beings. It should be noted that some of these compounds had also been isolated from other sources even before they were tested positive of antifeedant activity, including (-)-polygodial (2) from a) and Drimys lanceolata 29 marsh pepper Polyonum hydropiper (Polygonaceae) °, muzigadial (13) identical with canellal from Cane ha winterana 29 (Winteraceae) , and ugandensidial (12) identical with cinnamodial isolated from Cinnmosma 29 (Winteraceae) eS 29 fragrans (Canellaceae) In order to elucidate the relationship between structure and antifeedant activity, a large number of compounds, either isolated from plants or prepared by chemical synthesis, have been evaluated (Figure 2). The fact that epi-polygodial (14), polygodial derivatives 15, 16, 17, 18, cinnamolide (19), and betadiennolide (20), are devoid of any activity, reveals the necessity of both the C-9f equatorial aldehyde and the enal moiety for the activity in the A/Btrans-fused series . The (+)-polygodial (21) is as active as naturally occurring 26  (-)-  polygodial (2)’, although, earlier, 21 was shown to be highly phytotoxic. Of the two cis fused analogues 22 and 23, only 23 which has a C-9c aldehyde group, has strong activity; this apparent inversion compared with the corresponding trans-fused 2 and 14 was & (see below). The structure and activity of the diastereoisorners saccalutal (24) 3 rationalized and isosaccalutal (25) parallel that of (-)-polygodial (2) and epi-polygodial (14): compound 24 like compound 2 is active while compound 25 like compound 14 is inactive. Taking the very active muzigodial (13) into consideration, it is apparent that modification in the ring A exerts little effect on the activity. Hydroxylation at C9ct enhances the activity, while the introduction of an acetoxy group at C6 reduces the activity, possibly by steric hindrance.  8  14: R =H, 3 1 = 2 R = CHO R  20  19  15: 2 = 1 R = 3 O CH H, RR =H 2  16: 2 R1=CH O H, R =H, R 2 =CHO 3 17: 2 = 1 R = 3 H CO R ,R =H 2 18: 2 = 1 R = 3 M CO R e, R =H 2 12  CHO  .CHO  LJ 22: 1 R = CHO, R =H 2 23: R =H, R 1 =CHO 2  21  24: R=CHO, R =H 2 25: R =H, R 1 =CHO 2  Figure 2 Analogues of Drimane Antifeedants Based on the above studies of relationship between structure and antifeedant activity, experiments l a, and biomimetic studies electrophysiological 3 , two main hypotheses 30 concerning the actual molecular mechanism were brought forward to correlate structure to activity. The first suggests (Figure 3) that the enal moiety may act as a thiol acceptor of the insect’s chemoreceptor membranes in a way similar to the Michael reaction and inhibits the CHO  CHO HO  HS-R  Figure 3 Interaction between Drimane Antifeedants and the Insect’s Receptor: the first hypothesis  9  la. The lack of activity of epi-polygodial (14) is explained on the basis of 3 stimuli transduction the steric hindrance of its C-9a aldehyde to the approaching free thiol function on the receptor surface lb 3 The second hypothesis, suggested by Sodano et al. , assumes that pyrrole formation 30 by reaction of the C8 and C9 dialdehyde moieties with a primary amine of the receptor site is responsible for the antifeedant activity (Figure 4). Under biomimetic conditions, only the active (-)-913-polygodial (2) instead of the epi-polygodial (14) reacts with a variety of amines, Oa,b. 3 including L-cysteine, L-lysine, L-alanine, and methyl amine, to give pyrrole derivatives The shorter distance between the C-8 and C-9 aldehyde groups in (-)-polygodial (2) relative to epi-polygodial (14) is considered to be responsible for its ease of pyrrole ring closure and oa. The strong activity of the cis-fused dial 23 with an c 3 therefore the antifeedant activity aldehyde group at C-9, in contrast to the cis-fused dial 22 with a  13  aldehyde group at C-9, is  Oc. 3 rationalized in a similar way using their more stable steroid-like conformations  CHO  ‘OH  A CHO  C/OA 2  14  Figure 4 Interaction between Drimane Antifeedants and the Insect’s Receptor: the second hypothesis Based on these two earlier hypotheses, a new proposal was brought forward lc, which assumes that a three-pronged interaction between the receptor and the 3 recently  10  substrate, involving pyrrole formation, Michael addition of the thiol group to the enal moiety, and van der Waals interactions involving the ring A (especially at Cl), is responsible for the antifeedant activity. 2.1.2.  Total Synthesis of Drimane-type Antifeedants The discovery of the antifeedant activity of drimane-type sesquiterpenes has stimulated  a surging interest in their synthesis in the past decade. An excellent review by de Groot and T. b summarizes all studies prior to 1987. More recently, an updated version by de 9 A. van Beekl Groot and Jansen’ 91 describes in detail all published synthetic work prior to early 1990. So far, the syntheses of polygodial , 3 ’ 32 37 in 3 warburganal , 3 ’ 34 , and muzigodial 36 5 cinnamidial their racemic and enantiomerically pure forms have been achieved by different research groups. The following discussion will focus on the enantioselective synthesis of (-)-polygodial and  (-)-  warburganal. A few racemic syntheses of these two compounds will also be discussed since they have direct relevance to our strategy towards the synthesis of drimane-type antifeedants. 2.1.2.(a).  Polygodial  A synthesis of (±)-polygodial and (±)-warbuganal was developed by de Groot et al. 32 starting from the decalone 26 (Scheme 3). The carbonyl function at C7 was used first to introduce the necessary functionalized carbon atoms at C8 and C9 and subsequently to generate the C7, C8 double bond. The fonnylation of 26 and the subsequent dehydrogenation gave the unsaturated keto-aldehyde 27, which underwent a stereoselective conjugate addition by cyanide to 28. The resulting keto-aldehyde 28 was converted to the unsaturated aldehyde 29 by reduction of its (n-butylthio)methylene derivative, followed by mild hydrolysis. Protection of the aldehyde group in 29 and the reduction of the nitrile group in 30 gave compound 31. The epimerization of 31 to 32 was effected by potassium t-butoxide in t-butanol. Acidic hydrolysis of 32 then provided (±)-polygodial.  11  a, b  26  28  27 CN  d, e  CHO  f  29  g  30  31  h  32  (±)-polygodial  a) NaH, HCOOMe; b) PhSeC1, H ; c) CN; d) H, HSBu; e) NaBH 0 2 , H, H 4 0; 2 KOtBu, HOtBu; ; g) DIBAL; h) ) 2 (HOCH 0. 2 i) H H  Scheme 3 de Groot’s Synthesis of (±)-Polygodial  Since the conversion of 32 to (±)-warbuganal had been accomplished by MoO 5  -  hydroxylation of the enolate from 32 followed by the hydrolysis of the resulting b, 3 b 32 compound 4 the above sequence also provided a route to (±)-warburganal.  12  CHO  1) LDA, MoO , HMPT 5 2) H, H 0 2 32  (±)-warburganal  The conversion of (-)-drimenol (33) into the natural enantiomer (-)-polygodial (2) was a1 (Scheme 4). Oxidation of 33 and the subsequent protection of the b reported by Cortes et 33 aldehyde gave 34, which was then oxidized with a catalytic amount of selenium dioxide and bis-(4-methoxyphenyl) selenoxide as co-oxidant to give 35 in 45% yield. Deprotection of 35 generated (-)-polygodial (2) in an overall yield of 30%.  OH 2 CH  a, b  33  c  34  35  (-)-drimenol e 2: (-)-polygodial  0H, H; c) Se0 3 ) 2 a) PCC; b) HO(CH SeO; d) H, H 2 0. 2 , (4-MeOPh) 2  Scheme 4 Corte& First Synthesis of (-)-Polygodial (2) An alternative sequence from (-)-drimenol (33) was published by the same group of c (Scheme 5). Acetylation of (-)-drimenol (33) provided acetate 36 which was then 33 authors oxidized to produce the allylic alcohol 37. Saponification of 37 by means of potassium  13  carbonate in methanol resulted in diol 38. Swern oxidation of the diol gave (-)-polygodial (2) in almost quantitative yield. The overall yield of (-)-polygodial was 30% from (-)-drimenol (33). OAc  b  a  33  c  36  37 CHO  OHd  38  2: (-)-polygodial  SeO; c) KOH, MeOH; d) (COd) 2 0, Pyr.; b) Se0 2 , DMSO. 2 2 (cat.), (4-MeOPh) a) Ac Scheme 5 Cortes’ Second Synthesis of (-)-Polygodial (2) a 33 A synthesis of both enantiomers of polygodial has been developed by Mon et al. starting from (S)-3-hydroxy-2,2-dimethylcyclohexanone (40), which was obtained by reduction of dione 39 using Baker’s yeast. The (S)-ketol 40 was converted to diene 42 as indicated in Scheme 6.  The Diels-Alder reaction of this diene 42 with dimethyl  acetylenecarboxylate yielded a 1:1 mixture of 43 and 44. The diastereomeric diesters 43 and 44 were separated in 32% and 35% yield by HPLC. They were then utilized as starting materials for the preparation of both enantiomers of polygodial. Diester 43 was reduced to trans-fused diester 45. Treatment of 45 with 2 I SO 3 CF C and DMPA eliminated the axial hydroxyl group. Hydrogenation of 46 over Pd-C and  14  reduction of the diester functions gave diol 38. Swem oxidation of this diol 38 provided the natural (-)-polygodial in an overall yield of 3.0%.  a  b,c,d [H  o)ço  HO)(O  39  -  I  -  e,f  SiO  41  40  o  0 C.QMe 0  COMe 0  g, h  C0Me0 i  :o_c0MMe  OH 2 CH  j, 1 bCH20H  38  k  CHO m  ;:J_CH0  2: (-)-polygodial  C 3 Me S iC1; Me c) Mel, LDA; d) HCCNa; e) CuSO ; f) H 4 , Pd-CaCO 2 a) baker’s yeast; b) 2 ; 3 C--C0 MeO M e; h) HF, CH CN; i) DBU; i) H 3 , Pd-C; k) 2 2 g) 2 SO 3 CF C I, DMAP; 1) LAH; m) (COC1) , DMSO. 2  Scheme 6 Mon’s Synthesis of (-)-Polygodial (2) The diastereomeric diester 44 was transformed into the unnatural (+)-polygodial through a slightly different route (Scheme 7). Base-catalyzed isomerization of 44 to a  15  conjugated diene was followed by elimination of the hydroxyl group to give triene 47. Hydrogenation of this triene afforded the trans-fused diester 48 in 70% yield. LAH reduction of the diester functions produced diol 49 which was then converted to the unnatural (÷)polygoclial 21 by Swern oxidation in an overall yield of 2.9%.  ,,0 cOMe 0  -  ‘0Me  a,b 44  c  48  47 OH 2 CH OH 2 ,CH  CHO HO  e  21: (+)-polygodial  SO 3 CF C 1, DMAP; a) DBU; b) 2  , Pd-C; d) LAH; e) (COd) 2 , DMSO. 2 c) H  Scheme 7 Mon’s Synthesis of (+)-Polygodial (21) A sequence to (-)-polygodial (2) involving an intramolecular Diels-Alder reaction was d (Scheme 8). f3-Ionone (50) was treated with sodium hypobromite 33 developed by He and Wu to produce 51. Reduction of 51 with LAH, followed by condensation with maleic acid mono 1-menthyl ester gave 52 in 36% overall yield.  Refluxing of 52 in xylene afforded  diastereomers 53 and 54 in 79% yield at a ratio 1.75:1. The lactone 53 was then reduced to a diol which was cyclized to lactone 55 by p-toluenesulfonic acid in benzene in 76% yield. Oxidative cleavage of 55 with Cr03 furnished 56 in 65% yield which was then hydrogenated to 57. The carbonyl group in 57 was converted into an olefinic double bond as shown in 58 by a three-step sequence in 66% overall yield. LAH treatment of 58 produced the diol 38  16  b. The overall yield 33 which was finally converted to (-)-polygodial (2) by Swern oxidation was 4.1% from f3-ionone (50).  Men 2 CO a  b, c  d  52  51  50  +  h, i,  j  55  H0  1  58  38  2: (-)-polygodial  C-CC-CO HO M en, DCC ; d) xylene, reflux, N a) NaClO; b) LAH, 0°C; c) 2 ; e) p-TsOH, 2 j) MeOH; h) ; 4 NaBH i) C 2 MeSO 1, benzene; f) Cr0 ; g) H 3 Pyr.; DMSO; k) LAH; , Pd-C, 2 1) DMSO, (COd) . 2 Scheme 8 He and Wu’s Synthesis of (-)-Polygodial (2)  17  An enantioselective synthesis of (-)-poiygodiai (2) using (-)-carvone (59) as the building block was reported by de Groot et al.(Scheme 9)33e. Robinson annulation of 59 with MVK produced ketol 60 in 55% yield which was dehydrated to 61 in 87% yield. Dimethylation of 61 afforded 62 in 93%yield which was transformed to conjugated diene 63 by the Huang Minion modification of Woiff-Kishner reaction reduction in 70% yield. Selective ozonolysis of diene 63 provided enone 64 which was then further reduced to enantiomerically pure 65 by lithium and ammonia in an overall yield of 70%. Since the racemic mixture of 65, i.e., 26, has been transformed to (±)-po1ygodia1 32 (Scheme 3), 65 can be converted into (-)-polygodial (2).  a  b  c  60 (-)-dihydrocarvone f 0  62  64  63  2: (-)-polygodial  65 OH, heating; c) CH 3 a) MVK, KOH, 0°C; b) KOH, CH I, KOLBu; d) 2 3 NN NH , KOH, 200°C; e) 03; 1) Li, NH . 3  Scheme 9 de Groot’s Synthesis of (-)-Polygodial (2)  18  Another method to prepare enantiomerically pure 65 by using thujone as the chiral starting material was published recently by Kutney et al. 15 (Scheme 10). (+)-3-Cyperone (8) prepared from thujone  (3)13a  was methylated to a mixture of dienones 67 and 68 in 61%  yield. The mixture was then converted into pure dienone 68 by iodine in refluxing hexane in 86% yield.  Subsequent reduction of 68 produced diene 63 in 85% yield which was  ozonolyzed to enone 64 following the previous conditions by de Groot et al. Compound 64 was further transformed into enantiomerically pure 65 by Birch reduction. Br  b, c, d  a  e  322  7  g  8  68  67  1  64  63  C(CH ) 3 (CH O ; H) c) KMnO ; d) HBr (aq); 4 a) EVK, KOH, EtOH; b) H, 2 DMSO; g) ‘2, hexane; h) KOH, 2 NH DEG; i) O3 NN ,  e) Bu SnH; 1) CH 3 I, NaOMe, 3  Scheme 10 Kutney’s Synthesis of (-)-Polygodial 2.1.2.(b). Warburganal Two total syntheses of (±)-wargburganal were achieved starting from 5,5,8a-trimethylg. 34 trans-fused-1-decalone (70). Scheme 11 shows the synthesis by Goldsmith et a1  19  Formylation of 70 and subsequent dehydrogenation afforded the unsaturated keto-aldehyde 71 in high yield. Selective protection of the aldehyde group and the addition of methyllithium produced tertiary alcohol 72, which was dehydrated using the Burgess reagent. Osmylation of diene 73 provided diol 74. Oxidation, followed by hydrolysis of the acetal group afforded (±)-polygodial. 0  a,b  ‘CHOc,d  71  70  72  CHO  f  73  (±)-warburganal  74  ; c) H, 2 0 2 a) NaH, HCO Et; b) PhSeC1, Pyr./H 2 (CH CH O ; H) d) CH Li; e) 3 3 CN-SO 2 McO N ; Et 1) 0s0 , Pyr.; g) DMSO, Dcc; h) H, H 4 0 2 Scheme 11 Goldsmith’s Synthesis of (±)-Warburganal h reported the synthesis outlined in Scheme 12. Decalone 70 was 34 Kende et al. converted into the selectively protected unsaturated ketone 75 by formylation, dehydrogenation with DDQ and reaction with 1 ,3-propanediol. The hindered carbonyl function in 75 did not react with several ylides, but addition of substituted organometallic reagents can be accomplished in good yield. Thus, addition of [methoxy(trimethylsilyl)-methyl]lithium gave a diastereomeric mixture of alcohols 76, which underwent elimination of trimethylsilanol to afford a 1:3 mixture of (E) and (Z) enol ethers 77 and 78. Epoxidation of the (E) isomer 77  20  gave the x-epoxide 79, which could be hydrolyzed under mild acidic condition to (±)warburganal. Epoxidation of the (Z) isomer 78 yielded a 4:1 mixture of the 3-and a-epoxides  80 and 81, which were hydrolyzed to (±)-epiwarburganal and (±)-warburganal. 3 OCH  a, b, c  70  76  CHO  f  g  +  H  H  OCH  OCH  I  g  f  a) NaH, HCO Et; b) DDQ; c) H, 2 2 OH) d) (MeO)(Me (CH ; Si)CHLi; e) KH; 1) m-CPBA; 3 g)H, H 0. 2  Scheme 12 Kende’s Synthesis of (±)-Warburganal Enantioselective synthesis of (-)-warburganal (10) has been accomplished by degradation of diterpenes, abietic acid (82)35b and royleanone (86)”, and triterpene 85’  21  and transformation of functionalized drimanes, drimenol  (33)35a,  (+)-confertifoline (83)35c,  and diene 8435e (Figure 5).  OH 2 CH 0  H 2 CO  33  82  84  83  86  85  Figure 5 Chiral Starting Materials for the Synthesis of (-)-Warburganal (10)  synthesis of (-)-warburganal (10) from (-)-abietic acid (82) is outlined in b The first 35 Scheme 13. The regioselective osmylation of the double bond of the C ring of 82, followed by esterification of the acid function, afforded a diastereomeric mixture of diols 87. The ester group was transformed into a methyl group by the procedure indicated in the Scheme 13. The mixture of diols 88 was cleaved with Pb(OAc)4 to give ketoaldehyde 89 and the aldehyde function was protected as its acetal. The regioselective formation of silylenol ether 90, followed by ozonolysis gave aldehyde 91. Compound 91 was subject to the silyl enol ether formation and ozonolysis again to provide aldehyde 32. x-Hydroxylation of the aldehyde (32) and the removal of the protective group furnished (-)-warburganal (10).  22  OH  c,d  OH  e,f,g O2Ho2Me  87  82  88  II h  CHO  HO  k  89  1, k  91  90 CHO OH  m,  ‘  32  CHO  10: (-)-warburganal  NO; b) CH 3 ; c) DHP, H; d) LAH; e) PCC; t) H, H N 2 a) 0s0 , Me 4 0; g) 2 2 NH KOH; NH , h) Pb(Ac) ; i) H, 2 4 (CH CH O ; H) j) LDA, TMSC1, HMPA; k) 03, Me S; I) LDA, TMSCI; 2 m) LDA, MoO 5  Scheme 13 Ohno’s Synthesis of (-)-Warburganal (10)  23  2.2.  Discussion  2.2.1.  General Considerations about the Synthesis of (-)-Warburganal (10) from Thujone (3)  (-) -Polygodial (2) and  As discussed in the Introduction, three published sequences, shown in Schemes 1, 9, and 10, to (±)-polygodial and (±)-warburganal utilized trans-fused decalones as their starting materials. The essential feature of these studies is to utilize the existing carbonyl groups in the decalones effectively for the introduction of all necessary carbons and functional groups required in the target molecule. Therefore, we set as our first goal to convert the thujone-derived enone 7 into some enantiomerically pure, functionalized trans-decalones such as 65, 92 and 93 (Scheme 14).  2$ 8 )  1’  0 etc  65  92  CHO .CHO  2 (-)-polygodial  10 (-)-warburganal  Scheme 14 The Overall Plan towards the Synthesis of (-)-Polygodial (2) and (-)-Warburganal (10)  24  Elaboration of enantiomerically pure decalones into (-)-polygodial (2) and (-)-warburganal  (10) would be completed in a later stage. In formulating our synthetic approach, we recognized two necessary and likely associated operations: the exclusion of the isopropyl side chain arid the regioselective cleavage of the internal C-C bond (i.e., C7-C9 bonds ) of the cyclopropyl group. There is no functional group nearby to be used to achieve these aims. Substantial chemistry is thus dictated.  2.2.2.  Ozonation of Thujone and Its Derivatives Ozonation of saturated hydrocarbons into alcohols and ketones by inserting oxygen into  C-H bonds has been . 39 Usually tertiary carbons are preferentially ’ 38 well-documented attacked. However, the low solubility of ozone in organic solvents 40 (—0.1-0.3% by weight at -78°C) requires a long reaction time, leading to over-oxidation and poor selectivity. A practical improvement came from “dry ozonation” in which silica gel rather than organic solvents is used . At -78°C, the silica gel pre-adsorbed with the substrate (—1% by 41 as the reaction medium weight) was saturated with ozone; the mixture was then allowed to warm slowly to room temperature. Since silica gel adsorbs ozone efficiently at low temperature 42 (—4.5% by weight at -78°C), a complete oxidation of tertiary carbon-hydrogen bonds of cyclic hydrocarbons with high selectivity may be achieved under the reaction condition. The selective “dry ozonation” of thujone at the tertiary carbon of the isopropyl side chain was first observed in our laboratories by Dr. K. Piotrowska in a study related to the preparation of steroid analogues’ . Both ketol 94 (i.e., “thujonol’) and dione 95 (i.e., 5 “thujonone”) were obtained in a ratio of 2 to 1, resembling the selectivity previously observed in the “dry ozonation” of isopropropyl cyclopropane . However, the low overall conversion 43  £ Numbering for thujone-derived tricyclic intermediate 7 and latter related intermediates is kept similar to that for drimane sesquiterpenes such as (-)-polygidial (2) and (-)-warburganal (10) in order to provide facile  comparison.  25  (—40%) and the inconvenience of handling a large amount of silica gel during scaling up discouraged further exploration of this reaction.  03 +  3  95 thujonone  94 thujonol  The use of ozonation reaction in projected syntheses of trans-fused decalones was easily perceived.  The ring cleavage of cyclopropylcarbinols by acids has been well-  documented in the 1iterature. For example, treatment of cyclopropylcarbinols with aqueous hydrohalides generated homoallyic halides in good yields. If the ring cleavage of ozonation R  —c  HX OH  derived fused cyclopropylcarbinols occurred in the desired direction as shown in Scheme 15, the homoallylic halides produced would have an isopropylene side chain which could be oxidatively cleaved to provide a carbonyl group. In short, the ozonation of thujone and its derivatives could provide an entry to both required operations mentioned earlier (Scheme 14).  x HX  [0]  Scheme 15 A Perceived Sequence to Utilize Alcohols Derived from Ozonation of Thujone Derivatives  *  The ratio of silica gel to substrate in weight is usually 100 to 1 in order to observe a complete reaction in terms of attack at the tertiary carbon-hydrogen bond of the cyclic hydrocarbon according to the original dry ozonation procedure . 41  26  In summary, the following synthetic pathway to a trans-fused decalone was thus envisaged (Scheme 16):  x  x  / Scheme 16 An Ozonation Route to a trans-fused Decalone (Retrosynthetic Analysis) Thus, efforts were directed to finding a better ozonation condition. Finally, the relatively neglected solution ozonation (“wet ozonation”) was found to be satisfactory. The “wet ozonation” is easier to scale up (up to 30g scale), more reproducible, and easier to monitor. Complete conversion by wet ozonation can be easily achieved. A comparison of dry and wet ozonation of thujone (3) is shown in Table 1. Table 1 Comparison of Dry and Wet Ozonation of Thujone Dry Ozonaton Method  Wet Ozonation Method  sample preparation  solvent evaporation of the slurry of silica gel in thujone-petroleum ether solution  dissolution of thujone in EtOAc  condition  7 hrs at -25 C  workup  8 hrs at -78°C, then warm up to r.t. extraction with diethyl ether  conversion  43%  water and sodium bicarbonate (aq.) extraction complete  yield  70%  65-70%  94:95  2:1  1.5:1  27  0  (94)* revealed the molecular ion peak The mass spectrum of thujonol at m/z 168 while the JR spectrum indicated intense absorption peaks at 3100-3700 and 1730 cm, 1 corresponding to the hydroxyl and carbonyl stretching absorptions. The ‘H-NMR spectrum displayed two methyl singlets& at 6 1.22 and 1.32 ppm corresponding to the two methyl groups of the isopropyl side chain and a methyl doublet& at 6 1.18 ppm (J=7.6 Hz) corresponding to the methyl group at C4. A one-proton broad singlet# at 6 1.60 ppm was assigned to the proton of the hydroxyl group. The mass spectrum of thujonone (94) showed the molecular ion peak at m!z 152 while , corresponding 1 the JR spectrum revealed two intense absorption peaks at 1740 and 1685 cm to absorptions of the C3 carbonyl and the acetyl carbonyl groups. The ‘H-NMR spectrum indicated a methyl doublet at 6 1.22 ppm (3=8.4 Hz) and a methyl singlet at 6 2.09 ppm, correponding to the methyl at C4 and the methyl of the acetyl group respectively. This selective ozonation was generally applicable to other thujone derivatives. For example, the ozonation of 99 and 102 has been applied in the syntheses of drimane antifeedant . The ozonation of 105 was explored in an attempted 46 45 and rose oil fragrances ’ 5 analogues’ synthesis of steroid ana1ogues. Diketol 106 and trione 107 were obtained in 36% and 28% z 238 while t yield respectively. The mass spectrum of 106 showed the molecular ion peak at m,  the JR spectrum displayed the hydroxyl stretching absorption at 3450 cm1 and the two carbonyl stretching absorptions at 1730, 1710 cm . The 1 1 H-NMR spectrum of 106 revealed four methyl singlet signals at 6 1.00, 1.17, 1.33, 2.15 ppm. The mass spectrum of 107 indicated the molecular ion peak at m/z 222 while the JR spectrum showed three carbonyl *  Thujone used in this studies was a mixture of a and 3 diastereomers in a ratio of 10:1 as indicated from GC. Accordingly, thujonol and thujonone were mixtures of their a and diastereomers in a similar ratio as analyzed from GC. All spectral data were recorded for these diastereomeric mixtures. The 1 H-NMR spectral data presented here should belong to a diastereomers only since signals of f3 diastereorners were hardly observable from the spectra. & A methyl singlet is a singlet signal corresponding to a methyl group while a methyl doublet is a doublet  signal corresponding to a methyl group. a signal consisting of one proton. Accordingly, a signal consisting of m protons is  # A one-proton signal is called a m-proton signal.  This work carried out by this author is not described in this thesis. 28  absorption peaks at 1735, 1705, and 1685 cm1 respectively. The ‘H-NMR spectrum of 107 revealed three methyl singlet signals at ö 1.04, 2.09, and 2.12 ppm.  +  EtOAc -40°C  :101  100  99  1  +  EtOAc 102  104  103  25C  °H  105  106  °1 107 0  Scheme 17 Generality of Selective Ozonation of Thujone Derivatives  A more detailed study on the ozonation of the cis-fused ketone 96, prepared from enone 7 by catalytic hydrogenation, was carried out in order to find out factors influencing the wet ozonation reaction. The compound 96 was chosen since it was readily available (see the discussion on its preparation and stereochemistry in Section 2.2.3.).  29  03  , Pd-C 2 H EtOH  96 :980  Table 2 The Wet Ozonation of 9 to 97 and 98 Experimentsa  #1  #2  #3  #4  #5  Solventb  EtOAc  EtOAc  EtOAc  EtOAc  C1 CH 2  Temp. (°C)  -78  -40  -25  0  -40  Time(hrs)  7  7  5  3  7  %Recoveryof 96 %Total Yield 97:98c  90%  12%  0%  0%  10%  70%  68%  62%  55%  50%  1.55:1  1.50:1  1.40:1  1.20:1  1.00:1  a) 200 mg of 96 was used for every experiment; b) 50 ml of solvent was used for every experiment; c) the molar ratio of 97 and 98 was revealed by comparison of integrations at 6 0.35-0.70 ppm and 6 2.06 ppm of the mixture N1vIR spectrum; the signal at 6 0.35-0.70 ppm were due to two of the three cyclopropane protons in 97 while the signal at 6 2.06 ppm was from the three methyl protons of the acetyl group in 98.  As shown in Table 2, when the temperature increased, the total yield of the two products and the ratio of 98 to 97 dropped down. Changing solvents from ethyl acetate to methylene chloride decreased the total yield as well as the ratio of 98 to 97. It is of interest to understand these results in terms of the mechanistic proposals of ozonation. For the insertion of oxygen into carbon-hydrogen bonds, a unified proposal was put forward by Hamilton et al. 47 According to this proposal (Figure 6), the transition state (I) can either convert to produce ROH and a singlet oxygen directly by path (2) or collapse to a  30  hydrothoxide R000H which then decomposes to product ROll and a singlet oxygen by path (1) or degrades to a triplet oxygen, a hydroxyl radical, and an alkyl radical (II) which undergoes a chain reaction via an alkoxyl radical (III) to afford ROH by path (3). The occurrence of the different reaction paths depends on structural environments near the carbonhydrogen bonds and the reaction conditions. Carbon-hydrogen bonds adjacent to heteroatoms, such as the c carbon-hydrogen bonds of alcohols, ethers, and amines, and the carbon-  hydrogen bond of aldehyde groups favor path (1) because of the greater contribution of the resonant structure (Ib) to the transition state  (I)48.c.  Clear evidence for hydrotrioxides has  dl. Carbon48 been obtained only with ozonation of alcohols, ethers, amines and aldehydes hydrogen bonds not activated by adjacent heteroatoms will go through path (2) to produce ROH directly in the liquid phase and at low temperature (<0°C) with the retention of configuration being usually observed . In the vapor phase and at high temperature (>25°C), 47 the radical-mediated path (3) becomes dominant ’. The mechanism of dry ozonation was 48 presumed the same as that in liquid phase . 40  R—H  +  H.... 0 I Re  03  0 H.... ——  (1)  R  R000H  eQ__c  (I-a)  ROH  R•  +  OH  +  (I-b)  02 (triplet)  ROH (retention of  configuration) O3 HR  ROe  +  02 (singlet)  (III)  Figure 6 Oxygen Insertion into Carbon-Hydrogen Bonds  31  +  02 (singlet)  For the production of ketones from tertiary carbons through cleavage of C-C bonds, a similar insertion mechanism has been proposed for the liquid phase reactions (Figure The transition  state  7)4349.  (V) was assumed to collapse into a trioxide by cleaving one carbon-carbon  bond. The further decomposition of the trioxide provided a ketone and a hydroperoxide. The alternative mechanism, the fragmentation of alkoxyl radical (IV) generated from the oxygen insertion into the carbon-hydrogen bond following the Hamilton mechanism in Figure 6, was considered only possible at higher temperature (—25°C) . 43  [  — 1 R  I  R +  2 R  H —C—R 1 R 3 -  —C--O—O—R 1 R 3  2 R (V)  1 R +  3 HO—OR  2 R  Figure 7 Oxygen Insertion into Carbon-Carbon Bonds The selective ozonation of thujone (3) and its derivatives at the tertiary carbonhydrogen bond of the isopropyl side chain is perhaps due to lower energies of transition states (I) in Figure 6 and (V) in Figure 7 resulting from the participation of the cyclopropyl group in these two transition states. The cyclopropyl group is known to stabilize neighboring positive a. The oxidation of cx-rnethylene groups of 7 charge in a way similar to an olefinic groupl Oa. 5 bicyclo[n.1.O]allcanes to carbonyl groups was also reported  32  Increase of temperature in the ozonation reaction appeared to encourage oxidation in other carbon-hydrogen and carbon-carbon bonds, therefore causing the total yield of 97 and 98 to drop. The accompanying increase of the overall reaction rate and the decrease of the 97:98 ratio seemed to follow the general relationship between the selectivity of two ’. Changing solvents from ethyl acetate to methylene 501 competative reactions and temperature chloride had a dramatic effect on the total yield and the 97:98 ratio. This may reflect the participation of solvents in transition states (I) and (V). Ketone 98 might be produced directly from alcohol 97. To test this assumption, a solution of 97 in EtOAc was treated with ozone at -40°C for 7 hours. A new polar spot H-NMR spectrum, this spot contained several appeared on TLC plates. As revealed from the 1 compounds which were not characterized further. Apparently, ketone 98 was not directly generated from alcohol 97. The fact that 97 and 98 were produced at an almost constant ratio of approximately 1.5:1 from the beginning to the end of the reaction, as indicated by GC analysis, supported this conclusion.  7  0 97  98  108  A small amount of olefin 108 could be isolated from the reaction. Its molecular ion peak appeared at mlz 218 in the mass spectrum while the JR absorptions of carbonyl and . The ‘H-NMR spectrum 1 carbon-carbon double bonds were observed at 1710 and 1630 cm showed a characteristic vinylic methyl singlet at 6 1.60 ppm and two overlapped one-proton  33  singlets at  4.65-4.80 ppm corresponding to the two olefinic protons. This result indicated  that ketone 98 might be produced from 97 via the ozonolysis of the dehydration product 108. The strong acidity accumulated during the reaction could promote the dehydration of 97 especially at higher temperature (see the following paragraph). During our studies, a basic workup was found to be necessary to ensure that the alcohol product could be isolated intact; direct evaporation of the ethyl acetate mixture without neutralization with sodium bicarbonate aqueous solution led to serious decomposition of alcohols. A strong acidic medium was produced in the ozonation reaction; the water extract of the final reaction mixture had a pH value close to 1  .  To test if the acidic by-products were  formed from substrates or solvents, ozone was passed through blank solvents, ethyl acetate and methylene chloride, at -40°C for the same period of time as in the regular ozonation (i.e., 7 hours). Water extracts of the resulting solutions showed a similar strong acidity. It appeared that the acidic by-products were mainly generated by the oxidation of solvents or impurities present in them. 2.2.3.  Stereochemistry of Hydrogenation of Thujone-derived Tricycli c  Enones As shown in the Schemes 14 and 16, a trans-fused A/B ring junction was needed in developing a sequence to (-)-polygodial (2). Therefore, we hoped the reduction of enone 7  (see Scheme 14) would provide a trans-fused tricyclic compound 110.  110  96  34  109  Catalytic hydrogenation of enone 7 by 10% Pd-C in ethanol gave a major product 96 in  95% yield and a minor product 109 (2%) instead of the desired trans-fused ketone 110. The minor product 109 (2%), the epimer of 96 at C4, was very labile. It epimerized to 96 completely in CDC1 3 at room temperature overnight. The ‘H-NMR spectrum of ketone 109 showed a two-proton multiplet at 6 0.30 ppm, three methyl doublets at 8 0.88 (J= Hz), 0.96 (J= Hz), 1.06 (J= Hz) ppm, a methyl singlet at 6 0.99 ppm, a two-proton multiplet at 6 2.002.30 ppm, and a one-proton multiplet at 62.45 ppm. Ketone 96 had its molecular ion at m/z 220 in the mass spectrum. The carbonyl stretching frequency appeared at 1710 cm1 in its IR spectrum while the H-NMR spectrum revealed three methyl doublets at 6 0.85 (J=6.8 Hz), 0.91 (3=6.8 Hz), and 0.94 (7.2 Hz) ppm,  a methyl singlet at 6 1.23 ppm, a triplet (1H, J=ll.5 Hz) at 6 1.29 ppm, and one-proton multiplets at 6 1.34, 1.72, 2.15, 2.42, 2.58 ppm. Assignment of protons in the ‘H-NMR spectrum of 96 was accomplished by the following experiments (Figure 8). Decoupling by irradiation at the 6 2.58 ppm signal caused the methyl doublet (J=7.2 Hz) at 6 0.94 ppm to collapse to a singlet and a simplification of a  one-proton multiplet at 6 1.72 ppm. Thus, the methine proton at C4, the methyl at C4, and the methine proton at CS were assigned to the signals at 6 2.58, 0.94, and 1.72 ppm respectively. The only methyl singlet at 6 1.23 ppm in the off-resonance spectrum was obviously from the methyl at ClO. This ClO methyl signal was very close to a one-proton nuiltiplet at 8 1.34 ppm and a one-proton triplet (3=11.5 Hz) at 6 1.29 ppm. Irradiation at 6 1.23 ppm, which actually affected the multiplet and the triplet simultaneously, led to the collapsing of two methyl doublets (both J=6.8 Hz) at 6 0.85 and 0.91 ppm and the simplification of the C5 proton multiplet. Therefore, the multiplet at 8 1.34 ppm must be from the methine proton in the isopropyl side chain and the triplet must be due to one of the methylene protons at C6. The  methylene proton must be opposite to the C5 proton with regard to the cyclopentyl ring since the coupling constant (J=1 1.5 Hz) for the triplet was relatively large. The closeness of these proton signals and the complication of six possible conformational structures (two each from  35  11 1  9  2 15 12  A. 1 V’LjL b)  a) I  I  I  I  2.5  2.0  1.5  1.0  Figure 8 Decoupling Experiments on Ketone 96  a) off resonance spectrum. b) homonuclear spin decoupling at 1.23 ppm. c) homonuclear spin decoupling at 2.58 ppm.  36  I  0.50  3 (ppm)  H H  H  C  H  H H  H  H  0 H  C  H  H  a) 66 Figure 9 Single Crystal X-ray Structure of 98 (PLUTO Drawingl  37  trans-fused 110, cis-fused 96, and cis-fused 109 to be considered in the analysis discouraged further NOE experiments in order to elucidate the stereochemistry of ketone 96. Fortunately, separation of ozonation products 97 and 98 (see p. 30) by column chromatography with a mixed solvent system (hexanes:methylene chloride:methanol=10: 1:1) gave fractions containing 98 which crystalized readily upon slow evaporation of the solvent upon standing. The crystals were suitable for X-ray diffraction analysis. The crystals were also prepared from hexanes by Z. Gao in our laboratories and submitted for analysis . The 45 X-ray structure of 98 clearly showed an AIBcis-fused ring junction and an ce-orientation of the methyl at C4 (Figure 9). The stereochemistry of 96 was thus established. An attempt to obtain trans-fused compound 110 by Birch reduction using lithium and ammonia failed . 45 The similar phenomenon was observed in a previously published study related to b. Either catalytic or Birch reduction of 111, followed by acetic anhydride 4 steroid synthesisl treatment, gave the same cis -fused enol lactone 112.  112 H 2 CO  A range of enones were hydrogenated under catalytic conditions (Scheme 18). Reduction of the known compound 113 13a gave the cis-fused 114 in 90% yield. Its molecular ion peak in the mass spectrum appeared at m/z 206 while its carbonyl absorption was  observed at 1710 cm. The ‘NMR spectrum of 114 showed a one-proton doublet of doublets 1 (J=4.8 and 8.0 Hz) at doublets at  0.23 ppm, a one-proton thplet (J=4.8 Hz) at  0.45 ppm, two methyl  0.86 (J=6.4 Hz) and 0.93 (J= 6.4 Hz) ppm, a methyl singlet at  four-proton multiplet at  1.20 ppm, a  2.10-2.55. In order to establish the stereochemistry of 114, it was  38  subjected to methylation by treatment with potassium t-butoxide and iodornethane in t-butanol. The major product obtained in 70% yield was identical to the cis-fused ketone 119 prepared from the methylation of 96 (see p. 49) in all spectroscopic data. Thus, the cis fusion in 114 was confirmed. Two known compounds 115 and 116 with carbon-carbon double bonds at C5 and C6’ 3 were hydrogenated. Complex product mixtures were obtained as indicated from GC chromatograms and ‘H-NMR spectra. The complication might be due to the cleavage of conjugated cyclopropyl groups.  , Pd-C 2 H  114  113  116  115  , Pd-C 2 H  b  ttg  122  97  Scheme 18 Attempted Catalytic Hydrogenation of Tricyclic Enones A mixture of 117 and 118, obtained from the pyrrolidine catalyzed aldol condensation of ozonation product 106 (Scheme 17), was reduced to the cis -fused ketol 120 in 70% yield.  39  The mass spectrum of 120 showed the molecular ion peak at m/z 222 while its JR spectrum indicated stretching absorptions of the hydroxyl and carbonyl groups at 3100-3700 and 1710 . The ‘H-NMR spectrum displayed a one-proton doublet of doublets (3=4.0 and 5.4 Hz) 1 cmat 8 0.44 ppm, a one-proton doublet of doublets of doublets (3=1.2, 5.4, and 8.6 Hz) at 80.63 ppm, three methyl singlets at 6 1.14, 1.21, and 1.25 ppm, a complex four-proton multiplet at 6 2.12-2.52 ppm. The cis A/B ring junction of 120 was established by correlating it with ketol 97 (p. 30) chemically. Thus, compound 120 was converted into a dimethylated compound in 60% yield by treatment with iodomethane and potassium t-butoxide in t-butanol. This compound was identical in all spectroscopic data to the cis-fused ketol 121, prepared in 75% yield by treating 97 similarly. The comparison of their CD spectra* is shown in Figure 10.  0  I  I  I  I.  a)  b)  1. I I  I  I  I  I  t  (nm)  I  I  I  40g  Figure 10 Comparison of CD Spectra of 121 Prepared from Two Different Routes a) Ketol 121 prepared from 118. b) Ketol 121 prepared from 97. *  We are indebted to Dr. Ian Clark who gave us most helpful guidance in running the CD spectrometer. t The observed ellipticity angle 0 is expressed in a relative scale. Curve (a) is moved one devision up vertically in order to facilitate comparison. Since both measurements were made at 25°C in the same concentration (10.4 mg/mi), solvent (dioxane), and cell, there is no need to convert 0 into molar ellipticity angle [0] or molar circular dichroism Ae.  40  The mass spectrum of 121 showed the molecular ion peak at m/z 250 while the JR spectrum indicated the absorptions of the hydroxyl and carbonyl groups at 3100-3650 and  1705 cm. The ‘H-NMR spectrum revealed a one-proton triplet (J=4.8 Hz) at 1  0.41 ppm, a  one-proton doublet of doublets (J=4.8 Hz) at ö 0.58, five methyl singlets at ö 0.96, 1.12, 1.22, 1.24, and 1.34 ppm, two one-proton multiplets at  2.17 and 2.70 ppm.  The catalytic hydrogenation of 12211 (see Section 4.4. for its stereochemistry), prepared from the Robinson annulation of thujonol (94) with ethyl vinyl ketone in 35% yield, produced a single compound 97 which was again identical to the ozonation product 97 previously obtained from 96. Difficulties encountered in the direct preparation of trans-fused compounds by hydrogenation can be understood from a different perspective. The easy access to the 6,5fused enones and the need to generate a trans-fused /D portion in steroid synthesis provided ’ or 5 examples about reduction of these compounds. In fact, either catalytic hydrogenation Birch reduction 52 generally gave only or predominantly the cis-fused products. Our tricyclic enones derived from thujone (3) indeed behaved quite similarly. However, this outstanding problem has been remedied to some extent by recently developed hydroxyl-directed catalytic hydrogenation using homogeneous catalysts . 53 Unable to find a simple and efficient way to obtain trans-fused series of compounds directly, we decided that further effort in this direction would be terminated. The alternative, requiring two extra steps, was to carry on the sequence in the cis-fused series and to correct the stereochemistry at the ring junction in a later stage. From the point of view of preparing diverse types of analogues, the stereochemical correction alternative has its own advantage. 2.2.4. Acid Promoted Ring Cleavage of Thujone-derived Cyclopropylcarbinols  11 The tertiary hydroxyl group of the isopropyl side chain of 122 may serve as a directing group for the desired direct a face hydrogenation, although not explored in present studies . 53  41  It is well known that cyclopropylcarbinols can be cleaved through the pathway as shown in Scheme 15. When the reaction is applied to a non-symmetrically substituted cyclopropylcarbinols with an achiral center at  X  position, two different compounds are  expected to be generated, depending on which C-C bond is cleaved. In our specific system, we propose the following notation for the convenience of discussion: the endo-type cleavage will lead to a 6-membered ring homoallylic halide (X=halides); the exo-type 1 will result in formation of a 5-membered homoallylic halides (Figure 11). It is obvious that the endo-type cleavage is desirable for our purpose. The novel exo-type 2 cleavage is presented here in advance for the completeness of the notation and later discussion will indicate this mode of fragmentation (see section 2.2.7.). cleavage  exo- type 1 cleavage  x endo- cleavage  Figure 11 A Notation of Ring Cleavage Reactions Treatment of the ketol 97 in either methylene chloride or diethyl ether with concentrated HBr solution (48%) gave a mixture of starting material and a less polar fraction which could not be identified by 1 H-NMR spectrum. The same result was obtained when the reaction was  carried out with anhydrous MgBr2 in refluxing ether . It was considered that the complication 55 might arise from the relatively weak C-Br bond of ring cleavage products and their consequent  42  decomposition. If this was the case, the corresponding chioro compounds may be stable enough to allow purification and characterization. In fact, treatment of 97 with concentrated HC1 gave a stable major compound 123 rather than 124 in approximately 75% yield after column chromatographic purification.  :  124 In summary, compound 123 arises from the exo-type 1 cleavage. The IR spectrum of 123 showed the absence of an absorption corresponding to the hydroxyl stretching frequency. The parent ions at m/z 256 (0.6%) and 254 (2.2%) in the mass spectrum of 123 were consistent with two isotopic peaks C1 27 H 15 (C 3 0 3 and C1). 15 C 2 H 3 0 3 The ‘H-NMR spectrum of 123 showed two methyl singlets in 6 1.71 and 1.60 ppm, indicating the presence of an isopropylidene group; a multiplet (octet) at 6 3.55 ppm, characteristic of the A/B portion of an ABX system, corresponded to the methylene attached to the chlorine. The  f3 orientation of the chioromethyl side chain was verified since a sample of the  isolated product under stirring overnight in methylene chloride and silica gel regenerated the starting ketol 97 exclusively.  43  CI  silica gel C1 CH , 2  r.t. 97  Whether the solvolysis reaction takes place stepwise through a cyclopropylcarbinyl cation or in a concerted manner through a SN2’ like transition state has not been established (Figure 12). The regioselectivity is often explained by the SN2 mechanism involving stereoelectronic and steric factors. Considering the concerted mechanism, the rotation of the isopropyl side chain allows the hydroxyl to align antiparallel to either of the C-C bonds which may undergo cleavage. Inspection with molecular models revealed seemingly equal steric hindrance to these two alignments. Therefore, the hindrance to the incoming group, Cl- in this case, is likely playing an important role. In the case of the stepwise mechanism, this factor seems to be able to differentiate endo- and exo-type 1 cleavage paths. In any event, the preference to this exo-type 1 cleavage is likely due to the more exposed and accessible nature of the methylene compared to the methine in the cyclopropyl ring.  C1  Cl-  effed  stepwise  concerted  Figure 12 Rationalization of HC1 Promoted Ring Cleavages  44  2.2.5.  The Radical-mediated Rearrangement Although the major product from concentrated HC1 (aq.) treatment was initially  determined to be the structure 123, we felt that further evidence about this structure could be provided by a simple reduction. Reduction of 123 using tributyltin hydride as reducing agent was carried out with the expectation that the reduction product 125 would show a doublet H-NMR spectrum. Surprisingly, in corresponding to the newly generated methyl group in its 1 addition to the expected product 125 which showed an extra methyl doublet (J=7.2 Hz) at 6 0.92 ppm and absence of the two-proton multiplet at 6 3.5 ppm in 123, another major product was isolated in 50% yield. Its ‘H-NMR spectrum showed a methyl doublet (J=6.4 Hz) at 6 1.02 ppm, a methyl singlet at 6 1.26 ppm, and two vinyl methyl singlets at 6 1.64 and 1.66 ppm. Thus, this major compound was assigned to be 126. Its mass spectrum confirmed that it had a parent ion at m/z 254 corresponding to the molecular formula 0 14 C 2 H 5 of 126.  Benz:n:,reflux<  123  +  125  Apparently, a ring expansion took place during the reduction. mechanism was proposed to rationalize this novel reaction (Figure 13).  45  126  The following  LI  SnC1 3 R  Sn. 3 R E  126  SnH 3 R (a)  SnH 3 R (c)  Sn. 3 R  SnH 3 R Sn. 3 R 125  Figure 13 A Proposed Mechanism for the Novel Ring Expansion of 123 In this mechanism, a cyclopropylcarbinyl radical (b) generated from cyclization of the initial radical (a) is postulated as the intermediate to the final ring-expanded radical (c). An alternative pathway could involve the apparent direct 1, 2-shift from (a) to (c).  The  thermodynamic driving force for the radical rearrangement from (a) to (c) is probably the  46  greater stability of the secondary radical (c) in comparison with the primary radical (a). The cyclization step from (a) to (b) is analogous to the cyclization of chloride 123 to 97 (p. 44). We were not able to isolate any compound 96, a possible product resulting from the quenching of (b) during the reaction. A literature survey revealed that the postulation of a cyclopropyl carbinyl radical as an intermediate in the rearrangement of homoallylic radicals has b.l. More recent 57 a and verified by product studies and labelling experiments 57 been proposed ’. 571 studies are focusing on the quantitative aspect of this rearrangement To improve the yield of the ring expansion product 126, the direct quenching of radical (a) had to be suppressed. A longer life time for the initial radical (a) by decreasing the concentrations of both substrate 123 and reducing agent tributyltin hydride should allow it more likely to undergo a series of rearrangements and therefore improve the yield of 126. In fact, further experiments verified this postulate (Table 3).  Table 3 Yield Optimization for Conversion of 123 to I 26k’ 123  SnH 3 Bu  AIBN  benzene  126:12 5  total yield  50.4mg  82 I.tl  3.2mg  20m1  2.8:1  80%  50.4 mg  82 p1  3.2 mg  4.0 ml  2.4:1  74%  : Refluxin was continued for two days for both reactions.  In summary, despite the undesirable exo-type 1 cleavage to a hydroindane system in the acid-promoted ring cleavage reaction, the novel radical-mediated ring expansion provided us with a method to prepare the desired decalin system. Using the ozonation, acid-promoted ring cleavage, and radical-mediated ring expansion reactions as key steps, 127, an analogue of (-)polygodial (2), was prepared from thujone by Z. . 45 (Scheme 19; see also Scheme 17 ’ 5 Gao’ for preparation of 100 via ozonation).  47  .cI HC1  SnH 3 Bu AIBN, benzenc  CHO  clcç  03  127  Scheme 19 Gao’s Synthesis of a (-)-Polygodial Analogue 127  The same sequence was also successfully applied to the synthesis of the rose oil fragrances, 13-damascone and f3-damascenone, from thujone by Philip Gunning . (Scheme 46 20; see also Scheme 17 for preparation of 103 via oznation).  HC1  SnH 3 Bu .-  C1 CH 2  .  AIBN, benzene  NC  103  L.JI\  13-damascone  NC  f3-damascenone  Scheme 20 Gunning’s Synthesis of Rose Oil Fragrances  48  2.2.6.  Failure of the Radical-mediated Ring Expansion Reaction Having established the above sequence on the model compound 96, we tried to apply it  towards the synthesis of natural (-)-polygodial (2). The plan is shown in Scheme 21. A cis fused alkane 128 would be derived from ketone 119 which could be obtained by methylation of 96. Applying the established sequence to 128 would generate cis-fused decalone 129, from which a stereochemical correction into A/Btrans-fused decalone 26 would be carried out. The racemate of 26 (i.e., 65) was used as a starting material in the synthesis of (±)-polygodial and (±)-warburgamal by de Groot et al.(Scheme 3). During the course of our study, an enantioselective synthesis of 26 was completed by the same group from (-)-dihydrocarvone (Scheme 9).  119  128  1) 03, 2) HC1 SnH 3 3) Bu  129  65  Scheme 21 A Revised Plan to an Enantiomerically Pure, trans-fused Decalone 65  Following a standard method , cis-fused ketone 96 was refluxed with iodomethane 58 and potassium t-butoxide in t-butanol to give the gem-dimethyl ketone 119 in 85% yield. The mass spectrum of 119 revealed the molecular ion peak at m/z=234. Its IR spectrum showed absorption peaks at 3060 cm , characteristic of carbon-hydrogen stretching of the cyclopropyl 1  49  group, and 1700 cm, corresponding to the carbonyl stretching frequency. Its ‘H-NMR 1 spectrum showed two methyl doublets (6 0.85 ppm, J=6.6 Hz; 6 0.90 ppm, J=6.6 Hz) corresponding to the two methyl groups at the isopropyl side chain. Three methyl singlets (6 0.97 ppm, 1.22 ppm, 1.32 ppm) were observed which corresponded to the gem-dimethyl groups at C4 and the angular methyl at ClO. A two-proton multiplet appeared at 6 2.15-2.70 ppm, corresponding to the methylene at C2.  HOtBu  KOH, DEG NH NH 2  96  119  128  Wolf-Kishner reduction 59 of 119 gave alkane 128 in 70% yield. The mass spectrum showed the molecular ion peak at m/z 220. Its JR spectrum was characterized by the absence of the carbonyl stretching absorption and an absorption peak at 3060 cn , resulting from the 1 stretching of carbon-hydrogen bonds in the cyclopropyl group. In the 1 H-NMR spectrum, no signals above 6 1.80 ppm were noted. Two one-proton multiplets at high field, one at 6 0.04 ppm (dd, J=4.5 and 7.5 Hz) and the other at 6 0.40 ppm (t, J=4.5 Hz) corresponded to two of the three protons in the cyclopropyl group. 03, EtOAc -40°C 128  130  131 Li, THF 3 CH -40°C  When alkane 128 was treated with ozone at 400C in ethyl acetate for 8 hours, alcohol 130 and ketone 131 were obtained in 42% and 27% respectively. To obtain a maximal yield  50  of the alcohol, ketone 131 was treated with methyl lithium in THF at -40°C to give alcohol 130 in 70% yield. Therefore, the desired alcohol 130 was obtained in 61% overall yield from alkane 128. In the mass spectrum, alcohol 130 revealed its molecular ion peak at m/z 236 and a fragment ion peak at mlz 220 due to the dehydration of the parent molecule. Its JR spectrum 1 and a carbonwas characterized by a broad hydroxyl stretching absorption near 3400 cmhydrogen stretching absorption at 3060 cm1 due to the C-H bonds in the cyclopropyl group. Its 1 H-NMR spectrum showed five methyl singlets at 80.72, 0.92,1.05, 1.10, and 1.19 ppm. There was a complex two-proton multiplet at high field 60.40-0.55 ppm due to two protons in the cyclopropyl group. The collapse of the two separate one-proton signals originally noted in the ‘H-NMR spectrum of 128 into this multiplet was probably due to the electronic effect of the newly introduced hydroxyl group at the isopropyl side chain. The mass spectrum of ketone 131 showed a molecular ion peak at m/z 220. The IR spectrum had an intense absorption at 1675 cm1 due to the carbonyl stretching frequency. This bathochromic shift when compared to usual saturated carbonyl absorptions  (j—  1700 cm ) 1  ° and this 6 was the result of conjugation between the carbonyl and cyclopropyl groups phenomenon was also observed in diketone 98. The ‘H-NMR spectrum had four methyl singlets at 6 0.83, 1.00, 1.15, and 2.00 ppm.  The methyl singlet at  2.00 ppm was  apparently due to the methyl group at the methyl ketone side chain.  Treatment of alcohol 130 with concentrated hydrochloric acid in methylene chloride for 30 minutes produced homoallylic chloride 132 in 85% by the expected exo- type I cleavage. .01  conc. HC1 C1 CH 2 132  130  51  Compound 132 had a mass spectrum showing molecular peaks at m/z 256 (4.8%) and 17 C 2 H 6 and Cl. 17 C 2 H 6 Its JR m/z 254 (14.8%) corresponding to two isotopic isomers Cl spectrum was devoid of 0-H stretching absorption and the usual C-H stretching absorption  Hfrom the cyclopropyl group due to the absence of both groups in this new compound. Its 1 NMR spectrum revealed five methyl singlets: three of them at higher field, 6 0.84, 1.04, and  1.22 ppm; two of them at lower field, 6 1.63 and 1.70 ppm, resulting from the two vinylic methyl groups of the isopropylene side chain. There was a two-proton octet at 6 3.40-3.75 ppm, corresponding to the methylene group carrying the chlorine function. Using the condition previously established (Table 3), tributyltin hydride reduction of chloride 132 in refluxing benzene for 48 hours, generated in this instance only the simple reduction product 133 rather than the expected ring expansion product 134. Compound 133 had a peak at m/z 220 corresponding to the molecular ion in the mass spectrum. Its ‘H-NMR spectrum showed five methyl singlets at 6 0.84, 0.87, 1.02, 1.58, 1.62 ppm and a methyl doublet at 6 1.05 ppm (J=6 Hz).  133  132  134  Changing the reducing agent to triphenyltin hydride and the solvent to toluene did not result in any significant change. We also treated alcohol 130 with concentrated hydrobromic acid in order to obtain a different substrate 135 for the radical-mediated ring expansion.  52  Unfortunately, a complex mixture was obtained after column chromatography. A direct tributyltin hydride treatment of the mixture from hydrobromic acid solvolysis without column separation produced compound 133 in addition to a large portion of an inseparatable mixture.  Br  135  Therefore, the desired ring expansion for 132 had failed. Guided by the mechanistic proposal in Figure 13, we decided to approach the problem in a different way. According to this proposal, a cyclization to a cyclopropylcarbinyl radical (see (a) to (b) in Figure 13) from the initial primary radical was required before this cyclopropylcarbinyl radical opened to give the final radical (c), Figure 13). If, for some steric reason, the cyclization step did not take place, a simple reduction would be observed. On the other hand, if we could deliberately generate a cyclopropylcarbinyl radical centered at the tertiary carbon of the isopropyl side chain before opening the cyclopropane ring, the desired endo- type cleavage might be possible. The way to generate such a cyclopropylcarbinyl radical from a vinylcyclopropane was la. In their sequence, a regioselective 6 reported by Wender et al. in the synthesis of (±)coriolin cleavage of the vinylcylopropane in 136 (Scheme 22) was accomplished by thiophenol C which then 61 addition. In this reaction, thiophenol provided phenyl sulphuryl radical (PhS.) added to the double bond to generate a cyclopropylcarbinyl radical. The radical was selectively cleaved to produce intermediate 137. Paquette et al. also used thiophenol to cleave lb. 6 vinylcyclopropanes  53  OH  HSPh  (±)-coriolin  heating 137  136  Scheme 22 Radical-initiated Selective Ring Cleavage of a Vinylcyclopropane 136 After refluxing alcohol 130 and a catalytic amount of pyridinium tosylate in benzene for 30 minutes, vinylcyclopropane 138 was separated in 95% yield. Its molecular ion at m/z 218 was revealed from the mass spectrum. The JR spectrum indicated the absence of 0-H stretching absorption and a weak absorption peak at 1630 cm1 due to the stretching of the terminal carbon-carbon double bond. Its ‘H-NMR spectrum showed two proton signals at high field due to the protons in the cyclopropyl group: one at 6 0.52 (dd, J=7.2 and 4.8 Hz) and the other at 6 0.68 ppm (t, J=4.8 Hz). Four methyl singlets appeared at 6 0.81, 1.00, 1.13, and 1.65 ppm; the latter signal at 6 1.65 ppm was due to the vinylic methyl group in the side chain. Two one-proton broad singlets at 6 4.65 and 4.85 ppm corresponded to the two terminal olefinic protons.  TsOH  benzene, reflux 130  138  Refluxing of vinylcyclopropane 138 and thiophenol in benzene produced a rather complex inseparable mixture which may be expected in the form of four geometric isomers, two each of 139 and 140 (Scheme 23). The mixture was then subjected to lithium/ammonia hydrogenolysis at -33°C. In fact, column purification gave a colorless oil in 70% yield based on vinylcyclopropane 138. The oil was composed of 70% 133 and 30% 128 as revealed by GC and ‘H-NMR comparison with pure samples of these two compounds. Apparently, the  54  deliberately generated cyclopropylcarbinyl radical (i) cleaved mainly in the exo-type 1 manner to give geometric isomers of 139. The unexpected product 128 was probably derived from two diastereomers of 141, which were produced by quenching radical (i) with thiophenol.  Li, NH 3  HSPh benzene  138  133  E x SPh  L I 141  140  Li, NH 3  SPh  128  Scheme 23 Radical-initiated Ring Cleavage of Vinylcyclopropane 138  Comparison of the radical-mediated ring expansion reaction of 123, 132, and the homoallylic chloride derived from 100 (see Scheme 17) revealed the dramatic effect induced by an extra methyl group in ring A.  The radical-initiated ring cleavage reaction of  vinylcyclopropane 138 by thiophenol indicated that a cyclopropylcarbinyl radical could not  55  necessarily guarantee the endo-type cleavage. This again indicated that the additional methyl group in ring A played an important role in determining the overall course of the reaction. This subtle “methyl effect” could be rationalized in terms of the intermediate cyclopropylcarbinyl radical. The reaction of 132 with tributyltin hydride was assumed to involve a cyclopropylcarbinyl radical but the unidirectional cleavage of this radical in a way  similar to the radical i) in Scheme 23 resulted in the observed exo-type 1 cleavage product 133. Inspection with molecular models revealed that cis-fused annulated thujone derivatives can have chair-chair and chair-boat conformations as shown in Figure 14. In the chair-chair conformation, the methyl group at ClO and the C4[3 substituent are equatorially oriented with respect to ring A; the plane C6-C5-C1O is below plane C6-C7-C9-Cl0, making the bicyclo[3.1.O]hexane portion chair-like.  The major destabilizing factors are eclipsing  interactions of the equatorial C6-H bond with the C7-C1 1 bond and the CI-ClO bond with the C9-H bond, and the non-bonded interaction between the isopropyl side chain and the axial methyl group at C4. In the chair-boat conformation, the methyl group at ClO and the C4f3 substituent are axially oriented; the plane C6-C5-C1O is above the plane C6-C7-C9-Cl0, making the bicyclo[3.1.0] hexane moiety boat-like. The eclipsing interactions are greatly diminished. The seemingly important non-bonded interaction between the axial methyl group at ClO and the axial C4f3 substituent is actually small because the flattening nature of plane C6C5-C1O-C9 (torsional angel <C6-C5-C10-C9 estimated  250)*62  and the cis ring junction of  the A and B rings leads to a spreading apart of these two groups . In short, the chair-boat 63 conformation is greatly preferred regardless if the C413 substituent is either hydrogen or methyl. This conclusion is well supported by the X-ray diffraction analysis of dione 98 (Figure 9) and compound 147 (Figure 15 and Appendix 1), the negative Cotton effect of Ketol 121 (Figure 10), and structural studies of substituted bicylo[3. 1.01 hexanes 62  *  The estimation follows the average value provided by the studies on a series of bicylo[3. 1.0] hexane .  .  .  .  . 62 compounds  56  1 8  3  R=H or CH 3  R  chair-chair .  10  exo-type 1 cleavage  3  chair-boat  endo-type cleavage Figure 14 Rationalization of the “Methyl Effect’ In the endo-type cleavage, the immediate product 64 from the active reactant chair-boat conformer should have a torsional angle <C6-C5-C1O-C9 close to 55; the methyl group at ClO and the substituent at C4f3 approaches each other during the cleavage, causing an increase  in the energy of the transition state. If the substituent is methyl, the even greater increase in the transition state energy will probably forbid the endo-type cleavage from happening. In the exo type 1 cleavage, the immediate product has a cis-fused hydroindene conformation. There should be little change in the <C6-C5-C1O-C9 and therefore the distance between the ClO methyl group and the C43 substituent. Thus, change from hydrogen to methyl for the C413 substituent will not cause much difference for this exo-type 1 pathway.  # 550  is the average value for the torsional angle of a saturated cyclohexane ring.  $ 30° is the average value for the torsional angle of a saturated cyclopentane ring.  57  2.2.7.  Further Studies on the Acid-promoted Ring Cleavage of  Cyclopropylcarbinols The unsuccessful efforts with the radical-mediated ring expansion and cleavage reactions required a return to studies on the acid-promoted ring cleavage reaction of thujone  derived cyclopropylcarbinols in greater detail. There are examples in the literature showing the 65 on the preparation of vitamin D analogues, the use of other solvolysis conditions. In studies conversion of compound 142 into the trienes 143Z and 143E, which are geometric isomers  with regard to the newly formed double bond, was reported (Scheme 24).  Obviously,  compound 142 bears a close structural similarity to our thujone-derived cyclopropylcarbinols. The poor nucleophilicity of the attacking groups (e.g., H 0 and HOAc) in this set of 2 conditions may allow the ring cleavage reaction to occur in a less synchronized mechanism in which the C-C bond cleavage occurs faster (see also Figure 12). The endo-type cleavage proceeds through a more stable transition state because the tertiary nature of C5 accommodates the partial charge developed better than the primary nature of C6. Therefore, the endo-type cleavage prevails. R H  a) orb) +  143Z  142  a) H 0/dioxane, HOTs, 55°C 2 b) HOAc, 55°C  Scheme 24 Precedents of the Endo-type Cleavage  58  143E  Ri=OH Ri=OAc  The orientation of the newly introduced group (OH or OAc) at C5 agrees well with concertedness of the nucleophilic attack and the C5-C1 bond cleavage. Treatment of 130 in dioxane:H2O (1:1) with a catalytic amount of p-toluenesulfonic acid at 80°C for 1 hour generated a novel rearrangement product 144 in 85% yield rather than either of the ring cleavage products 145 and 146. The absence of any signals at 6 3.0-4.0 ppm in the NMR spectrum clearly revealed the product obtained cannot be a primary or secondary alcohol. The homoallylic tertiary alcohol 144 was characterized by its mass, JR and  ‘H-NMR spectra. Its mass spectrum indicated a peak at m/z 236 corresponding to the 0. The IR spectrum showed 2 molecular ion and a fragment ion peak at nl/z 218 due to loss of H 1 corresponding to the hydroxyl stretching frequency a broad absorption at 3100-3650 cm  H-NMR spectrum contained five methyl singlets at 6 0.87, 1.01, 1.17, 1.21, and while the 1 1.22 ppm, a four-proton multipiet at 6 2.10-2.40 ppm corresponding to protons of two allylic methylene groups, and a one proton broad singlet at 6 5.33 ppm corresponding to the olefinic proton.  HOTs  144 OH +  130  145  146  The most convincing evidence about the structure of 144 came from the X-ray diffraction analysis of its epoxide derivative 147. Treatment of 144 with rn-CPBA in  59  methylene chloride for one hour produced 147 in 90% yield, which was crystai.ized from methylene chloride. The structure of 147 established by X-ray analysis is shown in Figure 15 (See Appendix 1). The cis A/B ring junction in 147 and the  I face epoxidation are revealed.  The mass spectrum of 147 showed its molecular ion peak at m/z 252 and a fragment peak at  m/z 234 due to the loss of H20. Its IR spectrum was characterized by a strong hydroxyl  0  0  C  C  147 Figure 15 Single Crystal X-ray Structure of Epoxide 147 (ORTEP Drawing’ ’) 661  60  absorption at 3700 cm1 (CHC1 ). The ‘H-NMR spectrum indicated a one-proton singlet at ö 3 2.85 ppm corresponding to the proton on the epoxide ring, and five methyl singlets at ö 0.80, 0.98, 1.20, 1.24, and 1.31 ppm.  m-CPBA  )cc;  C1 CH 2  147  144  Before the crystal structure of 147 was revealed by X-ray analysis, the allylic alcohol 148 was mistakenly assumed as the ring cleavage product, since the spectral data noted above could be consistent with such a proposal. Mechanistically, the formation of 148 from 130 via 146 by some familiar rearrangement steps was also perceivable. Based on the structure 148, a sequence shown below was proposed to obtain the cis fused decalone 128. The epoxidation of 148 would generate 149, which should give glycol  150 by hydride attack from the less substituted carbon upon lithium aluminium hydride . The latter would then be cleaved to 128 by lead tetraacetate. 66 treatment  148  149  4 Pb(OAc)  128  61  150  Therefore, epoxide 147, mistaken as 148, was treated with LAH in THF at 70°C for 2 hours. To our surprise, allylic alcohol 151 was obtained in almost quantitative yield. The mass spectrum showed its molecular ion peak at m/z 194, corresponding to a loss of an acetone molecule (m/z=58) from 147 or 148. The JR spectrum displayed absorptions at 3 100-3650, 1 corresponding to hydroxyl, olefinic carbon-hydrogen, and carbon-carbon 3060, and 1650 cmdouble bond stretching frequencies. The 1 H-NMR spectrum indicated oniy three methyl singlets at 6 0.82, 1.02, and 1.14 ppm, a complex two-proton multiplet at 6 2.20-2.60 ppm corresponding to the allylic methylene protons, a one-proton singlet at 6 3.80 ppm corresponding to the allyic tertiary proton a. to the hydroxyl group, and two olefinic one-proton singlets at 6 5.06 and 5.21 ppm. The 3 orientation of the hydroxyl group was assigned based on a mechanistic argument shown in Figure 16. OH 2 Mn0 C1 CH 2  147  151  152  To confirm the structural assignment of 151, it was subjected to allylic oxidation by manganese dioxide 67 in methylene chloride at room temperature for two days. The enone 152 was obtained in a 70% yield. The mass spectrum showed the molecular ion peak at m/z 192. Its UV spectrum in methanol displayed an intense absorption at 235 nrn (log e=4.0) and a weaker one at 278 nm (loge=2.5). The JR spectrum indicated a carbonyl absorption at 1710 , and a carbon-carbon double bond absorption at 1635 cm 1 cm . 1  62  Li 0 2 H  147  151  Figure 16 Mechanism of the Fragmentation of Epoxide 147  ’ (lithium t We also treated 147, still then mistaken as 148, with ‘superhydride triethylaluminium hydride) in order to see if the the desired reduction rather than the fragmentation would take place. However, the same compound 151 was obtained as the only product. This puzzling fragmentation is finally understood when the structure of 147 was elucidated by X-ray analysis. Since the epoxide ring is on the convex side of the carbon framework, the nucleophilic ring opening of the epoxide by hydride has to take place from the concave side.  The unusually severe hindrance promotes the other pathway, that is,  fragmentation. The deprotonation of the tertiary hydroxyl group with hydride is proposed to results in the intermediate alkoxide first and the latter then undergoes the fragmentation shown in Figure 16. The novel rearrangement from 130 to 144 involved the insertion of the cyclopropane methylene into the position between the cyclopentyl ring and the isopropyl side chain. The cleavage of the carbon-carbon bond (i.e., the exo-type 2 cleavage, see the notation in Figure 11) in the original cyclopropyl ring was observed for the first time. The mechanism in Figure 17 is proposed to rationalize the reaction. Cyclopropylcarbinyl cation (i) is first formed by a proton-catalyzed elimination of the hydroxyl function in 130. The 1,3-shift of the methylene can result in another cyclopropylcarbinyl cation (ii). Further cleavage of (ii) in a selective fashion to form a more stable homoallylic cation (iii) occurs and the latter, upon reaction with water, converts to 144. The transformation between two cyclopropylcarbinyl cations in a  63  manner similar to that between (i) and (ii) was termed as a “cyclopropane sliding reaction” by H. Shirahama, who studied this type of transformation in greater detail with his system . The 68 mechanistic proposals involving this novel “sliding reaction” are scattered through the . 69 literature  (ii)  (i)  130  0 2 H  143  (iii)  Figure 17 Mechanism of the “Cyclopropane Sliding Reaction’  Treatment of 130 with acetic acid at 85°C for one hour produced the exo-type 2 cleavage product 153 in 60% yield in addition to the exo-type 1 cleavage product 154 in 6% yield. The competition of exo-type 1 cleavage is likely because 154 could only slowly convert to the cyclopropylcarbinyl cation (i) shown in Figure 17 once it is formed. The exo-type 1 product 145 could not be isolated in the previous reaction since it likely converts back to (i) rapidly under the acid catalysis. The mass spectrum of 153 showed an intense fragment peak at m!z 218 due to the loss  of an acetic acid molecule from the parent molecule (m/z=278). The chemical ionization mass spectrum using ammonia as carrier gas showed the protonated molecular ion (M+Hj peak at m/z 279. The JR spectrum indicated carbonyl and carbon-carbon double bond stretching  64  absorptions at 1735 and 1650 cm1 respectively. In the ‘H-NMR spectrum, six methyl singlets were observed at 6 0.85, 1.00, 1.15, 1.38, 1.45, and 1.97 ppm. The lowest field methyl singlet was due to the methyl protons of the acetate group. A complex multiplet at 6 2.02-2.62 ppm integrating for four protons was assigned to the two allylic methylene groups. There was  a one-proton singlet at 65.26 ppm, corresponding to the olefinic proton. The minor product 154 had its mass spectrum showing the molecular ion at m/z 278 and a fragment ion at m/z 218 due to loss of an acetic acid molecule. Its IR spectrum displayed . In the 1 1 H-NMR spectrum, six methyl singlets a carbonyl stretching absorption at 1730 cmwere observed at 6 0.85, 1.03, 1.14, 1.61, 1.69 and 2.01 ppm. The two singlets at 6 1.61 and 1.69 ppm were assigned to the two vinylic methyl groups of the isopropylidene group and the signal at 6 2.01 was clearly due to the methyl of the acetate group. A two-proton multiplet at 6 2.10-2.32 was due to the allylic methylene while a one-proton triplet (J=5.6 Hz) at 8 2.39 ppm was from the allylic methine proton. A two-proton multiplet at 6 3.92-4.25 ppm, which had a shape characteristic of the A/B portion of an ABX system, was assigned to the methylene attached to the acetate group. To test the generality of the cyclopropane sliding reaction, ketol 120 was employed as substrate under the two conditions previously used (Scheme 25). The endo-type cleavage product 155 was obtained in 87% yield underp-toluenesulfonic acid catalysis in dioxane:water (1:1) mixture. It was characterized by its ion molecular peak at m/z 222 and JR absorptions at 3050-3650, 1700, and 1650 cm due to hydroxyl, carbonyl, and carbon-carbon double bond stretching frequencies. Three methyl singlets, one at 6 1.20 ppm and two at 8 1.23 ppm, and an olefinic one-proton broad singlet at 6 5.20 ppm were observed in its 1 H-NMR spectrum.  65  OAc HOAc +  153  130  154  oxan:HO  HOAc QAc  QAc  +  +  Scheme 25 Generality of the Cyclopropane Sliding Reaction  Treatment of 120 with acetic acid gave mainly the cleavage products 156 (56%, exo type 1 cleavage) and 157 (14%, exo-type 2 cleavage). Although a very minor peak at ö 5.17 ppm in the 1 H-NMR spectrum of 156 indicated a probable presence of 158, the very minor amount present prevented its isolation. Presumably, the faster rate of these more direct cleavage reactions and perhaps the higher stability of the acetate products, prevent the formation of a cyclopropylcarbinyl cation like (i) in Figure 17 and therefore the sliding reaction from taking place. The keto-acetate 156 was characterized by its molecular ion peak at m/z 264, carbonyl stretchings at 1735 and 1705 cm t in the JR spectrum, and a two-proton multiplet at  3.95-4.20 ppm corresponding to the methylene attached to the acetate group in  the NMR spectrum. The electron impact mass spectrum of 157 revealed a fragment ion peak  66  at m/z 204 due to loss of a molecule of acetic acid; The chemical ionization mass spectrum using ammonia as carrier gas showed (M+I[H) at m/z 282 and (M+H) at m/z 265. The IR spectrum exhibited a carbonyl stretching absorption at 1710 cm. Two vinylic methyl singlets 1 appeared at 8 1.65 ppm and 1.72 ppm. A methyl singlet at 6 2.10 ppm corresponding to the methyl of the acetate group and a one-proton doublet of doublets at  5.19 ppm (J=4.2 and  10.2 Hz) corresponding to the methine attached to the acetate group were observed. The acetoxyl group was assumed to have 3-orietation, following the observed stereochemistry for the ring cleavage of related systems under similar conditions and the argument presented for this observation (Scheme 24). 2.2.8.  Baeyer-Villiger Oxidation of Cyclopropyl Ketones Our other efforts on applying the alcohols derived from ozonation of thujone  derivatives were to consider alternatives to the synthetic sequence shown in Scheme 21 by rearranging some steps involved. The successful radical-mediated ring expansion product 126 might be methylated to 159, which could be then decarbonylated and ozonolyzed to give 129. I in anhydrous t-butanol did not 3 Unfortunately, methylation of 126 by KOtBu and CH proceed at all at room temperature. Heating up the mixture gave a complex mixture. Although a protection of the methylene at C2 and the use of a strong base like LDA may eventually allow methylation proceed as desired, the added extra steps seemed to give very little advantage to such an effort. The other alternative sequence involved the use of 122A, resulting from methylation of 122. The mass spectrum of 122A showed the molecular ion peak at m/z 248 while the JR spectrum indicated absorptions of the hydroxyl and carbonyl groups at 3100-3700 and 1705 cm . The ‘H-NMR spectrum revealed a one-proton triplet at 3 0.44 ppm (J=4.4 1 Hz), a one-proton doublet of doublets at 6 1.04 ppm (3=4.4 and 8.0 Hz), five methyl singlets at 6 1.16, 1.19, 1.23, 1.26, and 1.28 ppm, two one-proton multiplets at 3 2.50 and 2.70 ppm, and a one-proton broad singlet at 6 5.62 ppm. Unfortunately, treatment of 122A with hydrochloric acid in methylene chloride gave an intractable mixture rather than the desired 67  122B. The low yield (30%) of 122, obtained from Robinson annulation of 94 with EVK, also discouraged further effort in this direction.  126  122A  159  129  122B  65  Therefore, the application of alcohols derived from ozonation of thujone derivatives to the synthesis of natural (-)-polygodial (2) had not met with any success.  Our next  consideration then to cyclopropyl ketones derived from ozonation, especially 131. It was perceived that a Baeyer-Villiger reaction ’ would cleave the side chain in a regioselective 7 manner to give 160 (Scheme 26). The preferential insertion of oxygen into the cyclopropyl group side during the Baeyer-Villiger reaction of methyl cyclopropyl ketone had been observed . The cyclopropanol 161 from saponification of 160 would be cleaved via the 70 previously internal carbon-carbon bond (i.e., endo-type 1 cleavage) by ferric chloride to generate the b chioroketone 162. It was recorded that the more substituted bonds of cyclopropanols were cleaved preferentially . Subsequent elimination of HC1 would afford enone 163 and the latter 73 could be elaborated to intermediates 65 in Scheme 9 by standard methods, thereby completing a formal synthetic sequence to (-)-polygodial (2).  68  -  -  160  131  162  161  163  Scheme 26 Utilization of Cyclopropyl Ketone 131 via Baeyer-Villiger and Cyclopropanol Cleavage Reactions  For this purpose, compound 131 was treated with m-CPBA in methylene chloride at room temperature for 2 days to produce 160 in 79% yield based on the recovery of 46% starting material 131. An optimization of the reaction was carried out according to Table 4. p-Toluenesulfonic acid had little catalytic effect on the m-CPBA oxidation reaction. Under refluxing conditions, the reaction appeared to accelerate at the beginning but slowed down quickly after a few hours and started to afford some unidentified by-products. When the concentration of the substrate was increased to 0.82 M, the reaction was quite complete after refluxing in methylene chloride for 12 hours and the yield of 160 was 82%. Although the use of trifluoroperacetic acid 70 improved the yield to as high as 94%, the reaction seemed not easily reproducible. This was likely due to the instability of trifluoroperacetic acid. Therefore, the preferred procedure for the preparation of 160 was to employ a high concentration of substrate 131 with m-CPBA as the oxidizing agent in refluxing methylene chloride.  69  Table 4 The Optimization of Baever-Villier Reaction of Ketone 131 Experiment  1  2  3  4  5  131  176mg  176mg  176mg  93mg  1.80g  Peracids’  m-CPBA 346 mg (2.0 eqv.)  m-CPBA 346mg (2.0 eqv.)  m-CPBA 346mg (2.0 eqv.)  CO 3 CF H 312 il (2.7M) (2.0 eqv.)  m-CPBA 4.45 g (2.5 eqv.)  C1 CH 2  5.0 ml  5.0 ml  5.0 ml  2.0 ml 10 ml  HOTs(mg)  0  39  0  0  0  Temp.  r.t.  r.t.  reflux  r.t.  reflux  Time (hrs)  48  48  24  48  12  %recovery of 131  46  46  30  48  5  %yieldof 160  79  78  60  94  82%  CO was preDared CF H : m-CPBA (80-85% mire) was used without further yurification while 3 in situ according to ref. 70a.  Acetate 160 had its mass spectrum showing the molecular ion peak at m!z 236. The JR H-NMR spectrum displayed four 1 while the 1 spectrum had a carbonyl absorption at 1735 cm methyl singlets at  0.80, 0.97, 1.05, and 2.10 ppm. The methyl singlet at  2.10 ppm was  obviously due to the acetate group, which unambiguously demonstrated the insertion of oxygen into the position between the quaternary cyclopropyl carbon (C7) and and the carbonyl carbon. According to the accepted mechanism of the Baeyer-Villiger reaction, a tetrahedral “Criegee intermediate” rearranges to ester products after it is formed by the peracid addition to the ketone carbonyl . (I), the transition state of this rearrangement step, can be described by 72 four resonance structures, Ia, Ib, Ic, and Id. The structure Ic implies that the preferred migration group will be the one that best accommodates a positive charge. Methyl cyclopropyl  70  ketone was unreactive to m-CPBA; only the much more reactive agent peroxytrifluoroacetic ob. This is probably 7 acid could make the oxygen insertion proceed to give cyclopropyl acetate because the cyclopropyl group cannot accommodate a positive charge well. The smooth reaction of 131 with m-CPBA at room temperature shows that the transition state involved is likely stabilized by the fused cyclopentyl group, which can presumably stabilize a positive charge better than the cyclopropyl group.. H  H 3 RCO  Ri  —C—R 1 R 2  (I)  1’IcQ_cQR  H 2 RCO —C--0R 1 R 2  R21  Criegee intermediate OH  OH I  I R  L  OH  1  I CR 2 O—O  Ta  2 R  O O CR 2  Tb  OH  CR 1 II CR 2 O O  1 CR  2 R  I  OCR  Id  Ic  As mentioned previously, ketone 131 was obtained in only 25% yield from the ozonation of alkane 128. Therefore, a sequence was developed to convert alcohol 130 to ketone 131 in order to optimize the yield of 131. Vinylcyclopropane 138, prepared by dehydration of 130 (p. 54), was ozonized to 131 in only 60% yield. A two-step procedure, involving the treatment of 138 with potassium permanganate in 1:1 t-butanol:water and the subsequent oxidative cleavage of the non-purified crude mixture of diols by lead tetraacetate in benzene, was then developed. In this case, 131 was obtained in 83% yield from 130. Thus, the overall yield of 131 from alkane 128 was improved to 65%.  4 KMnO  4 Pb(OAc)  138  131  71  2.2.9.  Regioselective Ring Opening of the Cyclopropyl Alcohol 161 Regioselective ring opening of alkyl substituted cyclopropanols by ferric chloride has  . The more substituted C-C bond is preferentially 73 been studied extensively by DePuy cleaved. The reaction (Figure 18) involves a cyclopropoxyl radical (i) which undergo a homolytic  scission of the more substituted C-C bond to give a carbinyl radical (ii). The  subsequent abstraction of a chlorine from ferric chloride produces the 3-chloroketone. This reaction was successfully applied to ring expansion of cyclic ketones via their derivatives 1. More recently, a different reagent, iodosobenzene, 74 trimethylsilyloxybicyclo[n.1.Ojalkanes was developed to effect the same ring expansion of cyclic ketones and lactones via similar . 75 derivatives 2 + HCI FeCI (iii)  RL 3 FeC1 R  OH  R  R  (i)  (ii)  3 eCl R0  2 FeC1 Figure 18 Regioselective Cleavage of Cyclopropanols Saponification of 160 was carried out in dilute potassium hydroxide-ethanol solution for 30 minutes. The rather polar 161 was obtained in almost quantitative yield. The reaction had to be worked up immediately because 161 could be further converted  into  164. The mass  spectrum of 161 indicated the molecular ion peak at m/z 194. Its IR spectrum showed the hydroxyl stretching frequency at 3050-3650 cm 1 and the absence of ester carbonyl absorption. The ‘H-NMR spectrum displayed three methyl singlets at ö 0.80, 0.96, and 1.01 ppm, a two-  72  proton multiplet at 5 1.98 ppm. The spectrum was contaminated by ketone 164 resulting from rapid decomposition of 161. The lability of cyclopropanol 161 dictated its immediate application to the next reaction. The mixture of anhydrous ferric chloride and 161 in anhydrous N,N-dimethylformamide were agitated under nitrogen at room temperature for 24 hours. The major product 162 was isolated in addition to a small amount of 164. The 13-chloroketone 162 had its mass spectrum showing molecular ion peaks at m/z 3 Its 1 5 C1. 1 C 2 H 3 3 and 0 1 230 (0.2%) and 228 (0.6%) corresponding to formulas C1 17 C 2 H 3 0 JR spectrum displayed carbonyl stretching absorption at 1720 cnr 1 while the ‘H-NMR spectrum indicated three methyl singlets at 5 0.76, 0.90, and 1.25 ppm, a complex four-proton multiplet at 5 2.10-3.00 ppm, and a one-proton doublet of doublets at  4.70 ppm (J=6.0 and  12 Hz) corresponding to the methine proton attached to the chlorine bearing carbon. The orientation of the chlorine function in 162 was uncertain. The mass spectrum of 164 indicated the molecular ion peak at rn/z 194 while the IR . The ‘H-NMR spectrum displayed 1 spectrum revealed the carbonyl absorption at 1725 cm two methyl singlets at 5 0.80 and 0.99 ppm, a methyl doublet at ö 0.93 ppm (J=7.2 Hz), a one-proton triplet at 5 1.77 ppm (J=8.0 Hz), a one-proton quartet at  2.06 ppm (J=7.2 Hz),  and a complex one-proton multiplet at 52.15-2.35 ppm.  3 FeC!  KOH  DMF  160  161  :0  +  164  162  73  The -chloroketone 162 underwent elimination of hydrogen chloride to give enone 163 easily even without addition of any base. The crude product from the above ring cleavage reaction was treated with sodium acetate in refluxing methanol for a few hours. The overall yield of 163 from acetate 160 was 80%, equivalent to a 93% yield for each step. The ketone 163 was a white solid with a m.p. of 64-66°C (literature value m.p. 68°C )76. The mass spectrum of 163 indicated its molecular ion peak at mlz 192. The JR spectrum showed an intense conjugated carbonyl stretching frequency at 1664 cm. Three methyl singlets appeared 1 H-NMR spectrum. A complex two-proton multiplet at 8 at 60.77,0.96, and 1.22 ppm in the 1 2.50-2.80 ppm corresponded to the methylene cx to of the carbonyl function. Two doublets at 6 5.95 and 6.27 ppm with a coupling constant, J=9.6 Hz were assigned to the two olefinic protons. The racemate of 163 was prepared previously by an interesting photochemical . The ether 165, prepared from the trans-fused isomer of 163, was irradiated 76 epimerization to generate 166 which then, upon hydrolysis, afforded (±)-163. The transformation from 165 to 166 was mediated by an achiral triene and therefore the chirality of starting material 165 was lost completely during the reaction. In other words, this method is inherently not enantioselective.  hv  O 3 H  3 OCH 165  166  /  74  (±)-163  The cyclopropylcarbinyl radical (i) and the cyclopropoxyl radical (ii) appears to have distinctly different cleavage pathways. Why are not the conformational factors previously considered in the cleavage of carbinyl radical (i) in Figure 14 playing any major role in the cleavage of the oxyl radical (ii).  carbinyl radical 0)  oxyl radical (ii)  A beautiful frontier molecular orbital (FMO) rationalization offered by Mariano and 77 is adopted here (Figure 19). As we know, the SOMO of the oxygen-centered radical Bay has much lower energy than that of the carbon-centered radical due to the greater electronegativity of oxygen. In general, the oxyl radical SOMO and the HOMO of a cyclopropane C-C bond are closer in energy than the SOMO-LUMO pair. Therefore, the SOMO-HOMO interaction contributes more to the stabilization of the transition state. A more alkyl substituted C-C bond has higher HOMO energy due to the electron donating nature of alkyl groups 78 and therefore has an enhanced SOMO-HOMO interaction. As a result, the more substituted C-C bond is preferentially cleaved. In the case of the oxyl radical (ii), such a SOMO-HOMO stabilizing interaction for the more substituted internal C-C bond overrides those unfavorable conformatiomal factors considered in the case of carbinyl radical (i) (Figure 14). Therefore, the endo-type cleavage is observed for the oxyl radical (ii). Using a similar argument, the kinetically controlled cleavage of cyclopropylcarbinyl radical (i) will go through the exo-type 1 pathway. The thermodynamically favorable endo-type cleavage cannot materialize even under the most favorable condition (i.e., high dilution and slow reaction rate) probably because of the great transition barrier present in this pathway (see Figure 14).  75  LUMO  SOMO  SOMO  HOMO  0 I  -C-..  -0:::  I  Figure 19 FMO Interactions of Carbinyl and Oxyl Radicals with Cyclopropane C-C Bonds  The preferential cleavage of the less substituted C-C bond in cyclopropanols by other reagents, which has a complementary regioselectivity to the ferric chloride reaction were also . We were curious to see if the exo-type 1 cleavage of the cyclopropanol 161, 73 recorded which represents the cleavage of the less substituted C-C bond, could be effected by using similar conditions. Indeed, the 13-bromoketone 167 was obtained in 60% yield after 161 was treated with NBS in DMSO:CHC1 3 (1:1) at room temperature. The mass spectrum of 167 revealed two isotopic molecular ion peaks at mlz 274 (2.4%) and 272 (2.5%) while the JR spectrum showed the carbonyl stretching absorption at 1730 cm. The ‘H-NMR spectrum 1 indicated three methyl singlets at 6 0.82, 1.00, and 1.14 ppm, a two-proton multiplet at 6 2.32 ppm, a triplet at 6 2.55 ppm (J=5.4 Hz), and a complex two-proton multiplet at 6 3.35-3.65 ppm. Br NBS  0 3 DMSO:CHC1 161  167  76  Therefore, the cyclopropyl ketone 131, much less considered than the other ozonation derived compound — cyclopropylcarbinol 130, turned out to be the more versatile intermediate for further elaboration. The cyclopropanol group in 161 may give a better control of regioselective ring cleavages than the cyclopropylcarbinol group in 130. In retrospect, we felt satisfied with what the ozonation method had brought us in terms of excluding the isopropyl side chain and controlling the regioselectivity of cyclopropane ring cleavage. 2.2.10.  A Formal Enantioselective Synthesis of (-)-Polygodial (2) and  (-)-  Warburganal (10) de Groot et al. have synthesized enantiomerically pure 64 from (-)-dihydrocarvone (Scheme  9)33e.  Ketone 64 was then converted to natural (-)-polygodial (2) and  (-)-  warburganal (10) using a sequence previously developed (Scheme 3). Consequently, If the cis-fused enone 163 were transformed into enone 64, a formal enantioselective synthesis of ()-Polygodial (2) and (-)-warburganal (10) from thujone was at hand.  8  2  Li, NH 3  1)LDA  17  2) PhSeC1 0 2 3) H  3  163  168  64  Scheme 27 The Preparation of Enantiomerically Pure Enone 64 from 163 To this end, LDA and phenylselenenyl chloride treatment of compound 163 in THF, followed by hydrogen peroxide oxidation, generated dienone 168 in very good yield (92%) (Scheme 27). Dienone 168, in its mass spectrum, showed the molecular ion peak at mlz 190 while its UV spectrum displayed a broad absorption peak at 2. 241 nm (log e=4.0). The JR spectrum indicated a conjugated carbonyl stretching absorption at 1660 cm 1 and a weak C=C absorption at 1620 cm. Three methyl singlets appeared at 8 1.22, 1.30, and 1.35 ppm in the 1  77  ‘H-NMR spectrum. The olefinic proton at C8 was a doublet of doublets at ö 6.14 ppm due to the couplings with the proton at C9 (J=9.9 Hz) and with the proton at C6 (W coupling, 1=0.2 Hz). The proton at C9 was a doublet at 6 6.25 ppm due to coupling with the the proton at C8 (J=9.9 Hz). The proton at C6 appeared as a doublet at 6 6.70 ppm (J=0.2 Hz) resulting from the above mentioned W coupling with the C8 proton. Birch reduction of dienone 168 without adding any proton donor gave the desired enone 64 in 70% yield. The specific rotation of compound 168 ([cL] =-l00, c=l.00, 5 ) is in close agreement to the value reported by de Groot ([a] 3 CHC1 =-105, c=l.0, 5 e. 3 ) 3 CHC1 3 This kind of selective reduction of the less substituted double bond of an . Presumably the higher reduction potential 79 analogous dienone 169 was observed previously  Li, NH 3  170  169  of the less substituted double bond led to a faster reaction. The spectroscopy data of 64 e. Thus, a formal synthetic 33 obtained by us was identical to that reported by de Groot sequence of (-)-polygodial (2) and (-)-warburganal (10) was completed. The complete sequence from thujone (3) to enone 64 is summarized in the following scheme. This sequence consisting of 11 steps is considerably longer than the 5-step sequence developed by de Groot (Scheme 9). However, our sequence can be simplified by carrying out several continuous steps without purification of intermediates. Specifically, steps from b) to f), steps from g) to h), and steps from i) to  j)  have been performed in this manner. For the  sake of completing a formal synthesis, we purposely intercepted enone 64 by conversion of enone 163. Consequently, the A/B cis fusion became a complete handicap. In other words, the real strength of enone 163 as a chiral template and therefore thujone as a chiral building block could not be shown. As will be demonstrated in the synthesis of ambergris fragrances  78  a  e, f  b, c, d  128  g, h, i,  131  j  0  64  163  , Pd-C; c) Mel, KOtBu, LBUOH; d) 2 2 NH KOH, DEG; e) 03; NH , a) EVK, KOH, EtOH; b) H , DMF; j) NaOAc, MeOH; 3 f) 4 /Pb(OAc) g) m-CPBA; h) KOH, MeOH; i) FeC1 KMnO ; . 3 k) LDA, PhSeC1/H ; 1) Li, Nil 0 2  (Chapter 3), the direct application of the cis-fused enone 163 as a chiral template is much more advantageous*. Ketone 171, which was an intermediate used in the preparation of a (-)-polygodial analogue (Scheme 17), was converted to 173 and 174 using a similar dehydrogenation reduction sequence. Dienone 172 was obtained in 80% yield by treatment of 161 with DDQ in refluxing dioxane . This product was characterized by its molecular ion peak at mlz 176 in 80 the mass spectrum, a conjugated carbonyl and carbon-carbon double bond absorptions at 1650 and 1620 cm-’ in the JR spectrum as well as typical ‘H-NMR signals. In the latter spectrum, a methyl doublet at  1.14 ppm (J=6 Hz), a methyl singlet at  1.27 ppm, a one-proton septet  (J=6 Hz) corresponding to the methine proton at C4, a singlet at the olefinic proton at C6, a doublet at  6.11 ppm corresponding to  6.21 ppm (J=9.0 Hz) corresponding to the olefinic  proton at C8, and another doublet at ö 6.78 ppm (J=9.0 Hz) corresponding to the proton at C9 were observed. Birch reduction of 172 gave both enone 173 (42%) and the saturated ketone 174 (25%). The double reduction of 172 was probably due to the presence of a trace amount of *  Following the synthetic plan presented there (Section 3.2.1.), one may also envisage a new route to (-)polygoclial and (-)-warburganal, starting with the cis-fused enone 163. 79  water which could protonate the enolate of 173, generated in the initial reduction of the less substituted carbon-carbon double bond, to produce 173 in situ. The further reduction of 173 yielded 174. The mass spectrum of 173 indicated the molecular ion peak at m/z 178. The JR spectrum showed a conjugated carbonyl stretching absorption at 1660 cm1 and a carboncarbon double bond stretching absorption at 1610 cm . The NMR spectrum displayed a 4 methyl doublet at 6 1.06 (J=6 Hz), a methyl singlet at 6 1.25 ppm, a complex three-proton  multiplet corresponding to the allylic C4 proton and the methylene group a to the carbonyl function, and one singlet for the olefinic proton at 65.79 ppm. 25  The saturated ketone 174 had a specific rotation [aID =-39.7 (c=1.00, CHC13), which is in good agreement with the reported value ([a]=-39.0, c=1.0, . 80 Its mass ) 3 CHC1 spectrum showed the molecular ion peak at m/z 180. In the JR spectrum, the carbonyl H-NMR spectrum indicated a methyl stretching absorption appeared at 1702 cm. The 1 1 doublet (J=6 Hz) at 6 0.81 ppm, a methyl singlet at 6 1.05 ppm, and a complex four-proton multiplet at 6 2.00-2.55 ppm corresponding to the two methylene groups a to the carbonyl group. 9 8  2  DDQ  Li, NH 3  ‘7  3  171  172  174  173  175  Scheme 28 A Possible Sequence to a New (-)-.Polygodial Analogue Transformation of 173 and 174 into another (-)-polygodial analogue 175 using a sequence similar to that developed by de Groot (Scheme 3) can be perceived (Scheme 28).  80  2.3. 2.3.1.  Experimental  General Solvents as provided from the Chemistry Store were used for chromatography without  further purification. Petroleum ether refers to the fraction boiling in the range of 30-60°C. Anhydrous diethyl ether, tetrahydrofuran, and benzene were prepared by distillation from a mixture containing sodium and benzophenone. Anhydrous methylene chloride, chloroform, and n-pentane were prepared by distillation from phosphorus pentoxide.  Anhydrous  isopropylamine, HMPA, DMF and DMSO were prepared by distillation from calcium hydride and stored in the presence of molecular seives (3 A) under nitrogen. Anhydrous methanol and ethanol were distilled from magnesium. Commercial reagents were purified, when necessary, by procedures described in Perrin and Perrin’ . n-Butyllithium, LDA, and vinylmagnesium bromide solutions were 62 standardized by titration against sec-butanol in benzene using 1,10-phenanthroline as indicator under nitrogen’ . Borane in THF and L-Selectride were standardized by measuring hydrogen 65 released from their reaction with 1:1 glycerol:water solution . Thujone was distilled from 130 Western red cedar leaf oil which was generously donated by Intrinsic Research and Development Incorporated. Syringes and needles were oven-dried at 120°C for a minimum of 4 hours and stored in a desiccator. Unless stated otherwise, all reactions were carried out under a positive pressure of dry nitrogen. Reactions at -7 8°C, -40°C, -25°C, and 0°C were performed with dry ice! acetone, dry ice/acetonitrile, dry ice/carbon tetrachioride, and ice/water cooling baths respectively. Air-sensitive materials were transferred inside a glove bag filled with nitrogen during weighing. All glassware was assembled under nitrogen immediately after being oven dried. Alternatively, it was flame-dried with nitrogen flowing through the reaction setup. Reactions were monitored by thin layer chromatography (TLC) and/or gas chromatography (GC). Analytical TLC was carried out on aluminium-backed silica gel plates  81  254 Visualization was realized by ultraviolet light and/or by heating F (Merck Silica Gel 60 ). after spraying with 10% ammonium molybodate in 10% sulfuric acid. Gas chromatography was performed on a Hewlett-Packard 5890A gas chromatograph, using a flame ionization detector and a 14.5 m x 0.252 mm fused silica capillary column coated with cyanopropyl phenyl silicone gum (DB 1701). Unless otherwise stated, all reaction products were purified by “flash chromatography” using silica gel (230-400 mesh) supplied by E. Merck Co. with air pressure to obtain a suitable flow . 163 Melting points were measured 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 path length with the concentration (in g/100 ml) given in brackets. The ultraviolet spectra were recorded on Cary 15 or Perkin-Elmer Lambda 4B UV/VIS spectrometers using quartz cells of 1 cm path length. The infrared spectra were recorded on Perkin-Elmer 710, 7 lOB, and 1710 spectrometers in chloroform solution using NaCl cells of 0.1 mm path length or as thin film using NaCl plates. The ‘H-NMR spectra were obtained from Bruker WH-400, AE-200 or Varain XL-300 spectrometers with duteriochloroform as solvent and the chemical shifts are reported in the delta (6) scale in ppm relative to tetramethylsilane. The 13 C spectra were taken on Bruker AE-200, or XL-300 spectrometers and chemical shifts are reported in the delta (6) scale in ppm relative to tetramethylsilane. The low and high resolution mass spectra were recorded on AEI-MS-9 or KRATOS-MS-50 spectrometers using the electron impact ionization method while the chemical ionization mass spectra were recorded on a Delsi Nermag RiO-i OC spectrometer using ammonia as carrier gas. CD spectra were recorded on a JASCO J-20 automatic recording spectropolarimeter. Elemental analyses were performed 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 on a Rigaku AFC6S or Enraf-Nonius CAD4-F diffractometers.  82  All compounds are named in accordance with IUPAC and CA rules. For compounds of the tricyclo[4.4.0.0 ]decane skeleton (i.e., the cyclopra[a]indene skeleton), their von 9 ’ 7 Baeyer names are also included in order to facilitate comparison with other similar compounds previously prepared and named by our group. However, the numbering system employed in all Introduction and Discussion sections follows the normal conventions of terpenoid and steroid literature in order to have convenient comparison with natural products and with themselves. 2.3.2.  Ozonation:  thujone (3) to thujonol (94) and thujonone (95)  [1R-( 1 c,4c/f3,5cc)] 1 -(1 -hydroxyl- 1 -methylethyl)-4-methyl-bicyclo{3. 1.01 hexan-3-one (94) [1 R-( 1 c,4a/f3,5 ct)] 1 -acetyl-4-methyl-bicyclo[3. 1 .0]hexan-3-one (95)  94  0H  95  I1  0  Thujone (3) (10.00 g, 65.8 mmol) dissolved in EtOAc (500 ml) was subjected to a stream of ozone-oxygen at -25°C for 10 hours. After the ozonizer was turned off, the gas flow was allowed to continue for 15 minutes to remove the residual ozone. After addition of dimethyl sulfide (5 ml), the reaction mixture was warmed to room temperature with stirring for 15 minutes, washed with water (100 ml) and saturated sodium bicarbonate solution (2X50 ml), and dried over magnesium sulfate. Solvent evaporation in vacuo gave an oil which was chromatographed using a mixture of isopropanol:hexanes (3:7) to afford compound 94 (4.70 g, 47%) and 95 (2.32 g, 23%). The physical properties of 94 are as follows*: JR (film) vmax.: 3 100-3700 (0-H stretching), 1730 (C=0 stretching). *  All spectral data were taken from spectra of the mixture containing a and I diastereomers at a ratio of 10:1 as analyzed by OC. The 1 H-NMR spectral signals should be those of the major a diastereomer since they can be  83  ) 6: 0.11(1H, 3 ‘H-NMR (400 MHz, CDC1  t,  J=4.8 Hz), 1.13 (1H, m), 1.18(3H, d, J=7.6  Hz), 1.22 (3H, s), 1.32 (3H, s), 1.35 (1H, dd, J=4.0 and 8.0 Hz), 1.60 (1H, bs), 2.19 (1H, d, J=16.4 Hz), 2.29 (1H, q, J=7.6 Hz), 2.79 (1H, dm, J=16.4 Hz). MS m/z: 168 (M, 10.0%), 150 (4.0%), 107 (69.5), 43 (100.0%). High resolution mass measurement: calculated for 1 O C 6 O: H 168.1150; found: 168.1146. The physical properties of 95 are as follows*: JR (film) Vmax.: 1740 (C=O stretching), 1685 (C=O stretching). ) 6: 0.75 (1H, 3 ‘H-NMR (400 MHz, CDC1  t,  J=4.8), 1.22 (3H, d, J=8.4 Hz), 1.86-1.98  (2H, m), 2.09 (3H, s), 2.30-2.4 1(2H, m), 3.25 (1H, m). MS m/z: 152 (M, 35.0%), 137 (11.0%), 124 (32.0%), 109 (100.0%). High resolution mass measurement: calculated for 0: 12 152.0837; found:. 152.0839. H 9 C 2.3.3.  Catalytic Hydrogenation:  enone 7 to ketone 96  [1 aS-(1 act, 1b13,5a,5a13,6aa)] 1 a, lb,2,3,5,5a,6,6a-Octahydro- lb,5-dimethyl-6a-( 1methylethyl)cycloprop[a] inden-4( 1H)-one (96) or [1R,2S,6S ,7S ,9Rj 2,6-Dimethyl-9-( 1.]decan-3-one (96) 9 ’ 7 methylethyl)tricyclo [4.4.0.0  96 Enone 7 (62.00 g, 282 mmol) was dissolved in ethanol (500 ml). 10% palladium2 for 8 charcoal catalyst (1.50 g) was added. The mixture was vigorously stirred under 1 atm H hours and filtered through a thick Celite cake. Evaporation of ethanol gave 96 as a colorless oil (62.06 g, 99.1%). readily recognized from the integrations. the spectrum. See footnote at p. 28.  The signals of the minor 3 diastereomer were hardly observable from  84  The physical properties of 96 are as follows: ). 3 [a]=+61.5 (c=1.00, CHC1 IR (film)  vmax.:  3050 (C-H stretching of the cyclopropyl group), 1710 (C=O stretching) cm-’.  ) & 0.20 ( 1H, J=4.8 and 8.0 Hz), 0.42 (111, 3 ‘H-NMR (400 MHz, CDC1  t,  J=4.8 Hz), 0.85  (3H, d, J=6.8 Hz), 0.87-1.00 {7H, including 0.91 (3H, d, J=6.8 Hz) and 0.94 (3H, d, J=7.2 Hz)}, 1.10-1.40{5H, m, including 1.24 (3H, s)}, 1.45-1.90 (4H, m), 2.15 (1H, m), 2.42 (1H, m), 2.58 (1H, m). MS m/z: 220 (M, 8.0%), 205 (5.1%), 159 (%), 93 (75.2%), 86 (100.0%). High resolution mass measurement: calculated for C 0: 24 220.1821; found:. 220.1815. H 15 Elemental analysis: calculated for 0: 14 C 2 H 5 C 81.76, H 10.98; found: C 81.67, H 11.00 2.3.4.  Ozonation:  ketone 96 to ketol 97 and dione 98  [1 aR-( lax, 1bf,5ct,5a,6ac)] 1 a, lb,2,3,5,5a,6,6a-Octahydro-6a-( 1 -hydroxyl- 1methylethyl)-Cycloprop[a]indene-4( 1H)-one (97) or [1R,2S ,6S ,7R,9S] 2,6-Dimethyl-9-( 1hydroxyl- 1 -methylethyl)tricyclo[4.4.0.0 .Jdecan-3-one (97) 9 ’ 7 [[laR-( 1 ao, 1bf,5cc5af3,6ax)] 6a-Acetyl- 1 a, lb,2,3,5,5a,6,6a-octahydro- 1 ,5-dimethylcycloprop [a]indene-4( 1H)-one (98) or [1 R,2S,6S,7R,9S1 9-Acetyl-2,6-dimethyltricyclo .Jdecan-3-one (98) 9 ’ 7 [4.4.0.0  Method A: Ketone 96 (1.03 g, 4.68 mmol) in EtOAc (100 ml) was cooled to -40°C. A stream of ozone-oxygen was passed for 10 hours. The oxygen flow continued for another 15 minutes to  85  remove residual ozone. After dimethyl sulfide (1.0 ml) was added, the mixture was warmed slowly to room temperature with stirring, washed with water (50 ml), saturated sodium bicarbonate solution (2x50 ml), and brine (30 ml), dried palladium. Solvent evaporation gave an oil which was chromatographed with isopropanol:hexanes (1:10) to give compounds 97 (0.42 g, 40%) and 98 (0.27 g, 28%) in a total yield 68% in addition to starting material 96 (0.06 g, 6%). Method B: Compound 122 (100 mg, o.427 mmol) in ethanol (10 ml) was treated with 10% palladium-charcoal catalyst (10 mg) and stirred under 1 atm hydrogen for 1 hour. Filtration of the reaction mixture and concentration of the filtrate gave compound 97 (95 mg, 95%) as an oil. The physical properties of 97 are as follows: m.p.: 45-47°C. [cL]=+1.36x102 (c=l.00, CHC1 ). 3 JR (film) Vmax.: 3000-3650 (0-H stretching), 3050 (C-H stretching of the cyclopropyl group), 1710 (C=0 stretching) cm . 1 H-NMR (400 MHz, CDC1 1 ) & 0.42 (1H, t, J=4.4), 0.62 (1H, dd, J=4.4 and 8.0), 0.95 3 (3H, d, J=6.4), 1.10-1.35  { 1OH,  including 1.15( 3H, s), 1.23 (3H, s) and 1.26 (3H, s)},  1.41(1H, m), 1.59(1H, bs), 1.65-1.95 (4H, m), 2.21 (1H, m), 2.44 (1H, m), 2.61 (1H, m). MS m/z: 236 (M, 2.3%), 218 (17.8%), 203(10.7%), 178 (35.4%), 161 (24.5%), 147 (26.0%0,133 (100.0%). High resolution mass measurement: calculated for 0 14 C 2 H : 2 5 236.1776; found:. 236.1778. The physical properties of 98 are as follows: m.p.: 100-102°C. [c]=+1.72X 102 (c=1.00, CHC1 ). 3  86  IR Vm (film): 3020 (C-H stretching of the cyclopropyl group), 1713 (C=O stretching), 1680 (conjugated C=O stretching). ‘H-NMR (400 MHz, CDC1 ) 6: 0.99 (3H, d, J=7.2), 1.05 (1H, t, J=6.0), 1.31 (311, s), 1.38 3 (1H, dd, J=6.0 and 8.8), 1.65-2.02 (8H, m), 2.06 (3H, s), 2.17 (1H, m), 2.44 (1H, m), 2.62 (1H, m). MS m/z: 220 (M, 15.3%), 205 (3.2%), 192 (5.1%), 177 (10.4%), 43 (100.0%). High resolution mass measurement: calculated for 0 12 C 2 H : 2 4 220.1463; found:. 220.1461. Elemental Analysis: calculated for 0 12 C 2 H : 2 4 C 76.33, H 9.15; found: C 76.28, H 9.13. 2.3.5.  Ozonation:  dione 105 to hydroxydione 106 and trione 107  [1R-( 1 a,4oL,5ct)J 1 -(1 -Hydroxyl- 1 -methylethyl)-4-methyl-4-(3-oxobutyl)bicyclo[3. 1.01 hexan-3-one (106) [1R-( 1 x,4a,5cx)] 1 -Acetyl-4-methyl-4-(3-oxobutyl)-bicyclo[3. 1 .0]hexan-3-one (107)  107  106  0  Diketone 105 (1.00 g, 4.50 mmol) in ethyl acetate (100 ml) was cooled to -25°C and passed with a stream of ozone-oxygen for 10 hours. After the continuation of oxygen flow for another 15 minutes, the mixture was treated with dimethyl sulfide (1.0 ml) and warmed slowly to room temperature. Washing with water and saturated sodium bicarbonate solution and evaporation of solvent gave an oil which was chromatographed with a mixed solvent system isopropanol:hexanes (3:7) to give 106 (0.39 g, 36%) and 107 (0.28 g, 28%). Compound 106 was also prepared from ketol 94 in the following way:  87  The solution of thujonol 94 (52 mg, 0.31 mmol) in toluene (5.0 ml) was mixed with distilled water (5.0 ml), methyl vinyl ketone (77 .tl, 0.93 mmol), potassium hydroxide (93 mg, —80% pure, 1.3 mmol), and tetrabutylammonium iodide (28 mg, 0.076 mmol) under nitrogen. This mixture was stirred for 10 hours at room temperature. After the mixture was saturated with sodium chloride, the organic layer was separated and concentrated to give a yellowish oil. Column chromatography of this oil afforded compound 106 (45 mg, 62%). The physical properties of 106 are as follows: []5=44•7 (c=1.09, CHC1 ). 3 JR (film) vmax.: 3450 (0-H stretching), 1730, 1710 cm . 1 ‘H-NMR (400 MHz, CDC1 ) & 0.00 (1H, m), 0.96 (1H, m), 1.00 (3H, s), 1.17 (3H, s), 3 1.33 (3H, s), 1.41 (1H, dd, J=4.2 and 8.4 Hz), 1.59 (1H, bs), 1.77 (2H, m), 2.10-2.25 {4H, including 2.15 (3H, s)}, 2.51 (2H, m), 2.97 (1H, m). MS m/z: 238 (M, 0.2%), 220 (4.0%), 202 (1.4%), 43 (100.0%). High resolution mass measurement: calculated for 0 12 C 2 H : 3 4 238.1569; found: 238.1568. The physical properties of 107 are as follows: [a]=+15.5 (c=1.03, CHC1 ). 3 JR (film) vmax.: 1735 (cyclopentanone C=0 stretching), 1705 (aliphatic C=0 stretching), 1680 (conjugated C=0 stretching).cm . 1 ‘H-NMR (400 MHz, CDC1 ) 6: 0.62 (1H, t, J=5.6 Hz), 1.04 (3H, s), 1.70-1.90 (3H, m), 3 2.03 (1H, dd, J=5.6 and 8.6 Hz), 2.09 (3H, s), 2.12 (3H, s), 2.30-2.45 (3H, m), 3.30 (1H, dd, J=2.4 and 19.0 Hz). MS m/z: 222 (M, 11.6%), 207 (1.7%), 179 (11.9%), 164 (100.0%). High resolution mass measurement: calculated for 0 C 1 H : 3 3 222.1256; found: 222.1261. 8  88  2.3.6.  Cataiytic Hydrogenation:  enone 113 to ketone 114  [1 aS-( 1 acx, 1b13,5aJ3,6acz)] 1 a, lb,2,3,5,5a,6,6a-Octahydro- lb-methyl-6a-( 1 -methylethyl) cycloprop[ajinden-4( 1H)-one (114) or [1R,6S,7S,9R] 6.-Methyl-9-( 1 -methylethyl)tricyclo .]decan-3-one (114) 9 ’ 7 [4.4.0.0  114 Enone 113 (0.45 g, 2.2 mmol) in methylene chloride (20 ml) was treated with 5% palladium-charcoal (0.76 g) at room temperature. The mixture was stirred under 1 atm hydrogen for 12 hours, filtered, and concentrated in vacuo. Ketone 114 was obtained in 90% yield (0.41 g). The physical properties of 114 are as follows: . 1 JR (film) vmax.: 3040 (C-H stretching), 1710 (C=O stretching) cm ‘H-NMR (400 MHz, CDC1 ) 6: 0.23 (1H, dd, 3=4.8 and 8.0 Hz), 0.45 (1H, 3  t,  1=4.8 Hz),  0.80-0.99 {7H, m, including 0.86 (3H, d, J=6.4 Hz), 0.93 (3H, d, J=6.4 Hz)), 1.20 (3H, s), 1.25-1.50 (2H, m), 1.60-1.95 (4H, m), 2.10-2.25 (2H, m), 2.30-2.55 (2H, m). MS m/z: 206 (M, 23.4%), 191 (6.3%), 188 (10.6%), 173 (16.7%), 163 (38.6%), 93 (100.0%). 2.3.7.  AIdol Condensation:  hydroxydione 106 to hydroxyenones 117 and 118  [1 aR-( 1 aa, 1b13,5a13,6ac)] 1 a, 1 b,2,3,6,6a-Hexahydro-6a-( 1 -hydroxyl- 1 -methylethyl)- 1 b methyl-cycloprop[a]inden-4( 1H)-one (117) or [6R,7R,9R] 9-( 1 -Hydroxyl- 1 -methylethyl-6methyl)tricyclo [4.4.0.0 .]dec-1(10)-en-3-one (117) 9 ’ 7  89  [1 aR-( 1 ax, 1 b13,5a13,6aa)J 1 a, lb,2,3,5 ,6a-Hexahydro-6a-( 1 -hydroxyl- 1 -methylethyl)- ib methyl--cycloprop[a]inden-4( 1H)-one (118) or [6R,7S,9R] 9-( 1 -Hydroxyl- 1 -methylethyl) 6-methyltricyclo 7 [4.4.0.0 . 9 ’ ]dec- 1 (10)-en-3-one (118)  117  118  Compound 106 (1.94 g, 8.15 mmol) in benzene (50 ml) was treated with pyrrolidine (0.82 ml, 9.8 mmol) and refluxed for 5 hours with a dean-stark trap. Concentration in vacuo gave a brown viscous oil which was chromatographed with ethyl acetate:hexanes mixture (1:1, v/v) to provide 117 (0.54 g, 30%) and 118 (0.79 g, 44%) in a total yield 74%. The physical properties of 117 are as follows: [aJ=+119 (c=1.OO, CHC1 ). 3 UV (MeOH, c=20.0 mg/i) max.: 234 nm (log e=4.21 1). JR (film) vmax.: 3420 (0-H stretching), 3050 (C-H stretchings of cyclopropyl group), 1655 (conjugated C=0 stretching) cm . 1 ‘H-NMR (400 MHz, CDC1 ) & 0.80 (1H, J=4.4 Hz), 1.12 (3H, s), 1.20 (6H, s), 1.32 (1H, 3 dd, J=4.4 and 8.8 Hz), 2.00-2.85 (6H, m), 5.60 (1H, bs). MS m/z: 220 (M, 0.8%), 202 (9.1%), 57 (34.7%), 43 (100.0%) High resolution mass measurement: calculated for 0 10 C 2 H : 2 4 220.1463; found: 220.1464. The physical properties of 118 are as follows: m.p.=70-7 1°C. [c]=+6l (c=0.58, CHC1 ). 3 JR (CHC1 ) vmax.: 3450 (0-H stretching), 1702 (C=0 stretching), 1642 (C=C stretching) 3 . 1 cm  90  CDC1 6: 0.34 (1H, t, J=4.4 Hz), 1.06 (1H, dd, J=4.4 and 8.2 Hz), ) H4MR (400 MHz, 3 1  1.19 (3H, s), 1.25 (3H, s), 1.27 (3H, s), 1.50-1.70 (2H, m), 1.87 (1H, m), 2.35-2.65 (2H, m), 2.80-3.10 (211, m), 5.53 (1H, d, J=1.2 Hz). MS m/z: 220 (M, 3.5%), 202 (23.3%), 187 (15.0%), 43 (100.0%). High resolution mass measurement: calculated for C 4H2002: 220.1463; found: 220.1471. 1 Catalytic Hydrogenation:  2.3.8.  hydroxyenones 117 and 118 to ketol 120  [1 aR-( 1 acz,213,5a13,6aa)1 1 a, 1 b,2,3,5 ,5a6,6a-Octahydro-6a-( 1 -hydroxyl- 1 -methylethyl)- 1 b methylcycloprop [a] inden-4( 1H)-one (120) or [1 R,6S ,7R,9S] 9-( 1 -Hydroxyl- 1.]decan-3-one (120) 9 ’ 7 methylethyl)-9-methyhricyclo [4.4.0.0  0tt”,’ 120  To enones 117 and 118 (536 mg, 2.44 mmol) in ethanol (20 ml) solution was added 10% palladium on charcoal catalyst (130.3 mg). The mixture was then stirred under 1 atm hydrogen (1 atm) for 1.2 hours. After the mixture was filtered through a layer of Celite and washed with additional ethanol (20 ml), the solution was concentrated in vacuo. Column chromatography of the crude oil with hexanes:ethyl acetate (1:1) gave ketol 120 (519 mg, 96.0%). The physical properties of 120 are as follows: . CHC1 ) [aJ=+61.8 (c=1.00, 3 . 1 JR (film) Vmax.: 3100-3700 (0-H stretching), 1710 (C=O stretching) cm ) 6: 0.44 (1H, dd, J=4.0 and 5.4 Hz), 0.63 (111, ddd, J=1.2, 3 ‘H-NMR (400 MHz, CDC1 5.4, and 8.6 Hz), 1.06 (1H, bs), 1.10-1.30 {(1OH, m, including 1.14 (311, s), 1.21 (3H, s)  91  and 1.25 (3H, s)}, 1.63 (1H, m), 1.72-1.92 (4H, m), 2.12-2.25 (2H, m), 2.35-2.52 (2H, m). MS m/z: 222 (M, 1.4%), 204 (16.9%), 189 (13.5%), 133 (74.9%), 59 (100.0%). High resolution mass measurement: calculated for 0 12 C 2 H : 2 4 222.1620; found:. 222.1618. 2.3.9.  Methylation:  ketol 97 and 120 to ketol 121  [1 aR-( 1 aa, 1 bf3,5af,6aa)] 1 a, 1 b,2,3,5,5a,6,6a-Octahydro-6a-(1 -hydroxyl- 1 -methylethyl) 1 b,5 ,5,trimethyl-cycloprop [a]inden-4( IH)-one (121) or [1 S,6R,7R,9R] 9-( 1 -Hydroxyl- 1methylethyl)-2,2,6-trimethyltricyclo [4.4.0.0 .]decan-3-one (121) 9 ’ 7  121 Method A: To the solution of ketol 120 (50 mg, 0.23 mmol) in anhydrous t-butanol (2.0 ml) was added potassium t-butoxide (174 mg, 1.42 mmol) and iodomethane (85 .tl, 1.4 mmol). The mixture was then refluxed under nitrogen for 2 hours, cooled, and quenched with water (10 ml). Extraction with diethyl ether (2x10 ml), drying with magnesium sulfate, and evaporation of solvent in vacuo gave an oil which was chromatographed to afford 121 (33 mg, 62%). Method B: Ketol 97 (48 mg, 0.20 mmol) in t-butanol (2.0 ml) was treated with potassium t butoxide (124 mg, 1.01 mmol) and iodomethane (65 p.1, 1.0 mmol) under nitrogen. The mixture was refluxed for 1.5 hours, cooled down, and quenched with water (10 ml). Extraction with diethyl ether (2x10 ml), drying with magnesium sulfate, and evaporation of  92  ether in vacuo provided an oil which was chromatographed with ethyl acetate:hexanes mixture (3:7, v/v) to give 121 (41 mg, 8 1%). The physical properties of 121 are as follows: [x]=-19.2 (c=0.0832, dioxane). JR Vmax. (film): 3 100-3650 (0-H stretching), 1705 (C=0 stretching). ‘H-NMR (400 MHz, CDC1 ) 8: 0.41 (1H, dd, J=4.8 Hz), 0.58 (1H, dd, J=4.8 and 7.8 Hz), 3 1.05-1.40 { 18H, m, including 0.96 (3H, s), 1.12 (3H, s), 1.22 (3H, s), 1.24 (3H, s) and 1.34 (3H, s)}, 1.46-1.62 (2H, m), 1.75-1.90 (2H, m), 2.17 (1H, m), 2.70 (1H, m). MS m/z: 250 (M, 1.7%), 235 (9.4%), 232 (3.7%), 217 (6.4%), 192 (30.5%), 177 (18.1%), 133 (47.1%), 59 (100.0%). High resolution mass measurement: calculated for 0 16 C 2 H : 2 250.1934; found: 250.1936. 2.3.10.  Robinson Annulation:  thujonol (94) to hydroxyenone 122  [1 aR-( 1 ao, 1 b3,6act)] 1 a, 1 b,2,3,6,6a-Hexahydro-6a-( 1 -hydroxyl- 1 -methylethyl)- 1 b,5dimethyl-cycloprop[a]inden-4( 1H)-one (122) or [6R,7R,9R1 9-( 1 -hydroxyl- 1 -methylethyl) 2,6-dimethyltricyclo [4.4.0.0 .]dec- 1 (2)-en-3-one (122) 9 ’ 7  °‘IH 122  To 1-dimethylaminopentan-2-one--iodomethane salt (2.84 g, 9.44 mmol) in ethanol (80 ml) was added the solution of ketol 94 (1.43 g, 8.51 mmol) in ethanol (20 ml). After potassium hydroxide (0.92 g, —80% pure, 13 mmol) was added, the mixture was refluxed under nitrogen for 3 hours. Concentration of the reaction mixture in vacuo gave a yellow oil  93  which was chromatographed using ethyl acetate:hexanes mixture (1:1, vlv) to provide compound 122 as a colorless oil (636 mg, 32%). The physical properties of 122 are as follows: ). 3 [x]=+90.3 (c2.03, CHC1 UV (MeOH, c=40.6 mg/l) max.: 248 nm (loge=4.04). JR (film) vmax.: 3200-3600 (0-H stretching), 1645 (C=0 stretching) cm. 1 ‘H-NMR (400 MHz, CDC1 ) & 0.76 (1H, t, J=4.7 Hz), 1.11 (3H, s), 1.20 (611, s), 1.67 3 (3H, s) ppm. MS m/z: 234 (M, 1.2%), 216 (3 1.5%), 201 (48.0%), 173 (34.7%), 59 (100.0%). High resolution mass measurement: calculated for 0 12 C 2 H : 2 5 234.1619; found: 234.1613. 2.3.11. Cyclopropane Ring Opening Reaction:  ketol 97 to chloroketone 123  [1 R-( 1 a,3ax,413,7act)] 3,3a,4,6,7,7a-Hexahydro- 1 -chloromethyl-4,7a-dimethyl-2( 1I1)-( 1methylethylidene)-5H-inden-5-one (123)  123 Ketol 97 (78 mg, 0.33 mmol) in methylene chloride (5.0 ml) was stirred with concentrated hydrochloric acid (5.0 ml) at room temperature for 30 minutes. Water (20 ml) was added to quench the reaction. After methylene chloride extraction (2x10 ml), drying over magnesium sulfate, and evaporation of solvent in vacuo, the crude product was chromatographed with ethyl acetate:hexanes mixture (1:8, v/v) to afford the starting ketol 97 (15 mg, 19%) and chloride 123 (51 mg, 74%). 94  The physical properties of 123 are as follows: [czJ=+1.4x102 (c=0.50, CHC1 ). 3 JR Vmax. (film): 1700, 1641 cm . 1 ‘H-NMR (400 MHz, CDC13) & 1.00 (3H, d, J=7.0 Hz), 1.05 (1H, m), 1.24 (1H, m), 1.44 (3H, s), 1.58-1.80 {7H, including 1.60 (3H, s) and 1.71 (3H, s)}, 2.10 (6H,m), 3.45-3.65 (2H, m). MS m/z: 256/254 (M, 0.6%/2.2%), 239 (0.5%), 218 (34.6%), 203 (18.4%), 133 (85.0%), 41 (100.0%). High resolution mass measurement: calculated for C1: 15 254.1437, C 3 0 3 H2 found: 254.1437; calculated for C1: 17 C 2 H 3 0 5 256.1408, found: 256.1412. 3 Radical-mediated Rearrangement:  2.3.12.  chloroketone 123 to enones 125  and 126 [1R-( 1 ct,3 aa,43,7ax)] 3,3a,4,6,7 ,7a-Hexahydro- 1 ,4,7a-trimethyl-2( 1H)-( 1-methylethylidene)-5H-inden-5-one (125) [1 S-( 1 x,4af3,8a)] 4,4a,5,6,8 ,8a-Hexahydro- 1 ,4a-dimethyl-7(3H)-( 1 -methylethylidene) naphthalen-2( 111)-one (126)  125  126  Chloride 123 (50.4 mg, 0.198 mmol) in benzene (20 ml) was treated with tributyltin hydride (82 p.1, 0.30 mmol, 1.5 eqv.) and AIBN (3.2 mg, 0.019 mmol, 0.10 eqv.) under nitrogen. This mixture was then refluxed for 2 days. Concentration in vacuo gave the crude product which was chromatographed with ethyl acetate:hexanes mixture (1:8) to afford 126  95  (19.9 mg) and 125 (7.1 mg) in a total yield 80%, based on the recovery of chloride 123 (12.0 mg). The physical properties of 126 are as follows: m. p.=85°C. ). 3 [c]=-23.9 (c=1.00, CHC1 ): 1700 cm 3 . 1 JR Vm (CHC1 CDC1 & 0.90-2.00 { 17H, m, including 1.02 (3H, d, J=), 1.26 (3H, ) ‘H-NMR (400 MHz, 3 s), 1.64 (3H, s) and 1.66 (3H, s)), 2.20-2.65 (6H, m), 2.93 (JH, m). MS m/z: 220 (M, 44.4%), 203 (8.2%), 187 (8.7%), 148 (47.3%), 135 (100.0%). High resolution mass measurement: calculated for 0: 14 C 2 H 5 220.1827; found:. 220.1822. The physical properties of 125 are as follows: [aj=+35 (c=0.94, CHC1 ). 3 IRVm, (film): 1710 cm-’. HNMR (400 MHz, CDC1 1 ) 6: 0.80-2.00{ 18H, m, including 0.92 (3H, d, J=), 0.99 (3H, 3 d, J=), 1.22 (3H, s), 1.54 (3H, s), 1.64 (3H, s)}, 2.10 (4H, m), 2.50 (1H, m), 2.57 (1H, m). MS m/z: 220 (M, 23.5%), 205 (8.2%), 187 (4.9%), 175 (11.4%), 163 (50.3%), 135 (100.0%). High resolution mass measurement: calculated for 0: 14 C 2 H 5 220.1827; found: 220.1822. 2.3.13.  Methylation:  ketone 96 to ketone 119  [1 aS-( lacc, 1bf3,5af3,6ac)] 1 a, lb,2,3,5,5a,6,6a-Octahydro- lb,5,5-trimethyl-6a-( 1methylethyl)-cycloprop [a]inden-4( 1H)-one (119) or [iS ,6R,7S,9S1 2,2,6-triimethyl-9-( 1methylethyl)tricyclo [4.4.0.0 .]decan-3-one (119) 9 ’ 7  96  119 To the solution of ketone 96 (62.0 g, 0.282 mol) in anhydrous t-butanol (700 ml) was added potassium t-butoxide (130.5 g, 1.07 mol) slowly under nitrogen. lodomethane (66.6 ml, 1.07 mol) was added in a dropwise manner with stirring to ensure a gentle reflux. Upon finishing the addition, refluxing continued for 30 minutes. The mixture was cooled down, quenched with water (700 ml), extracted with petroleum ether (3x500 ml). evaporation of solvent gave an oil which was chromatographed to provide the methylated ketone 119 (55.1 g, 84%). The physical properties of 119 are as follows: ); [aj=0.00 (c=0.993, CHC1 3 ). 3 =+ 14.0 (c=0.993, CHC1 5 [x] JR Vmax (film): 3060, 1700 cm . 1 H4MR (400 MHz, CDC1 1 ) : 0.18 (1H, dd, J=4.0 and 8.0 Hz), 0.40 (1H, 3  t,  J=4.0 Hz),  0.80-0.88 {4H, m, including 0.85 (3H, d, J=7.2 Hz)}, 0.90 (3H, d, J=7.2 Hz), 0.97 (3H, s), 1.22 (3H, s), 1.25-1.50 (6H, m, including 1.32 (3H, s)}, 1.65-1.90 (3H, m), 215 (1H, td, J=4.4 and 15.2 Hz), 2.70 (1H, m). MS m/z: 234 (M, 55.8%), 219 (14.3%), 201 (22.0%), 191 (29.9%), 173 (51.1%), 43 (100.0%). High resolution mass measurement calculated for 0: 16 234.1983; found: C 2 H 234.1986. 2.3.14.  WoIf-Kishner-Huang Minion Reaction:  ketone 119 to alkane 128  [1 aS-( 1 aa, 1 b3,5a3,6act)J Decahydro- 1 b,5,5-trimethyl-6a- (1 -methylethyl) cycloprop[a]indene (128) or [1 R,6S ,7S,9R] 9-( 1 -Methylethyl)-2,2,6-thmethyltricyclo .]decane (128) 9 ’ 7 [4.4.0.0  97  128  Ketone 119 (42.0 g, 180 mmol) in diethylene glycol (300 ml) was treated with potassium hydroxide (37.0 g, -80% pure, 528 mmol) and hydrazine monohydrate (26.8 ml, 552 mmol). The mixture was heated at 100°C for 1.5 hours under nitrogen. The temperature was then raised to 220°C to distill away water and excess hydrazine. Refluxing continued at 2 10°C for 4 hours. The mixture was cooled down, diluted with water (11), and extracted with petroleum ether (3X600 ml). Evaporation of the solvent gave a brown oil which was chromatographed with petroleum ether through a short column gave 128 as a colorless oil (24.50 g, 62%). The physical properties of 128 are as follows: [cx ]=+42.5 (c=1.00, CHC1 ). 3 JR Vm (film): 3060 cm . 1 ‘H-NMR (400 Mhz, CDC1 ) & 0.04 (1H, dd, 3=4.6 and 8.4 Hz), 0.40 (1H, 3  t,  J=4.6 Hz),  0.72-0.82 (4H, including 0.78 (3H, s)}, 0.82-1.65 {22H, including 0.88 (3H, d, J=7.2 Hz), 0.95 (3H, d, J=7.2 Hz), 0.98 (3H, s) and 1.07 (3H, s)}. MS m/z: 220 (M, 3.4%), 205 (10.3%), 177 (45.5%), 109 (100.0%). High resolution mass measurement: calculated for : 18 C 2 H 6 220.2191; found: 220.2198. 2.3.15.  Ozonation:  alkane 128 to alcohol 130 and ketone 131  [1 aR-( 1 aa,2a13,5af3,6acc)] Decahydro-cc-hythoxy-cx,cc, 1 b,5 ,5-pentamethylcycloprop[a] inden -6a-methanol (130) or [1 R,6S ,7R,9S] 9-( 1 -Hydroxyl- 1 -methylethyl)-2,2,6-trimethyltricyclo .jdecane (130) 9 ’ 7 [4.4.0.0  98  [1 aR-(1 acc2af3,5a13,6acx)] 6a-Acetyl-decahydro- 1 b,5,5-trimethylcycloprop[ajindene (131) or [1 R,6S ,7R,9S] 9-Acetyl-2,2,6-trimethyltricyclo  decan-3-one (131)  131  130  Compound 128 (4.50 g, 20.4 mmol) in ethyl acetate (500 ml) was cooled to -40°C. A stream of ozone in oxygen (90 volts, flow rate 9.1 mi/sec) was passed for 6.5 hours. The oxygen flow continued to pass the solution till the blue color disappeared. Dimethyl sulfide (1.0 ml) was added and the mixture was warmed slowly to room temperature with stirring. After washed with water and saturated sodium bicarbonate solution, the mixture was dried with magnesium sulfate. Solvent evaporation gave an oil which was chromatographed to afford compounds 130 (2.01 g, 42%) and 131 (1.23 g, 27%) in a total yield 69%. The physical properties of 130 are as follows: [cc]=+49.2 (c=0.995, CHC1 ). 3 . 1 JR Vm. (film): 3 400 (0-H stretching), 3060 (C-H stretching) cm HJ4MR (400 MHz, CDC1 1 ) 6: 0.40-0.55 (2H, m), 0.72 (3H, s), 0.80-1.90 (23H, 3 including 0.92 (3H, s), 1.05 (3H, s), 1.10 (3H, s) and 1.19 (3H, s)}. MS m/z: 236 (M, 1.0%), 218 (35.8%), 203 (26.9%), 178 (41.3%), 163 (59.5%), 59 18 C 2 H 6 236.2140; found: (100.0%). High resolution mass measurement: calculated for 0: 236.2140. Elemental analysis: caic. for 0: 18 C 2 H 6 C 81.29, H 11.94; found: C 81.23, H 12.00. The physical properties of 131 are as follows: [cc]=+82 (c=0.24, CHC1 ). 3 IR Vm. (film): 1675 cm 1 (C=0 stretching).  99  CDC1 & 0.83 (3H, s), 1.00 (3H, s), 1.05 (1H, m), 1.10-1.70 {12H, ) ‘H-NMR (400 MHz, 3 m, including 1.15 (3H, s)}, 1.80 (1H, dd, 3=4.0 and 12.0 Hz), 2.00 (3H, s), 2.20 (1H,  t,  J=12.0 Hz). MS m/z: 220 (M, 33.6%), 205 (16.5%), 177 (17.0%), 109 (33.9%), 43 (100.0%). High resolution mass measurement: calculated for C 5H240: 220.1827; found: 220.1825. 1  2.3.16.  Dehydration:  alcohol 130 to alkene 138  [laR-( 1 acç 1bI3,5aI3,6ac)] Decahydro- lb,5,5-trimethyl-6a-( 1 -methylethenyl)-cycloprop[a] indene (138) or [1R,6S ,7R,9Sj 9-( 1 -methylethenyl)-2,2,6-trimethyltricyclo decane (138)  1_—_ 138  To the alcohol 130 (171 mg, 0.724 mmol) in benzene (15.0 ml) solution was added pyridinium tosylate (28 mg, 0.11 mmol, 0.15 eqv.). The mixture was refluxed with a DeanStark trap on for 15 minutes. After the reaction mixture was washed with saturated sodium bicarbonate solution (10 ml), the organic layer was separated and concentrated in vacuo. Column chromatograghy by ethyl acetate:hexanes mixture (8:1, vlv) gave the vinyl cyclopropane 138 (127 mg, 91%) and the starting alcohol 130 (20.0 mg, 11.7%). The physical properties of 138 are as follows: =+87.9 (c=1.09, CHC1 5 [x] ). 3 JR Vm (film): 3075, 1650 cm. 1  100  ‘H-NMR (400 MFIz, CDC1 ) ö: 0.52 (1H, dd, 1=4.8 and 8.8 Hz), 0.68 (1H, t, J=4.8 Hz), 3 0.81 (3H, s), 1.00 (3H, s), 1.02-1.58 (11H, m, including 1.13 (3H, s)}, 1.65 (3H, s), 1.701.92 (2H, m), 4.65-4.85 (2H, two broad singlets). MS m/z: 218 (M, 26.5%), 203 (23.2%), 189 (4.2%), 175 (20.9%), 147 (3 1.3%), 147 (5 1.3%), 109 (100.0%). High resolution mass measurement calculated for 2 6H 1 C 6 : 218.2035; found: 218.2030.  2.3.17. [1R-( 1  Cyclopropane Ring Opening Reaction:  alcohol 130 to chloride 132  3ac,7a)] 3a,4,5 ,6,7,7a-Hexahydro- 1 -chloromethyl-2(3H)-( 1 -methylethylidene)  4,4,7a-trimethyl- 1H-indene (132)  132  Alcohol 130 (100 mg, 0.420 mmol) in methylene chloride (5.0 ml) was stirred with concentrated hydrochloric acid (5.0 ml) at room temperature for 30 minutes. Separation and concentration of the methylene layer gave the crude product which was chromatographed with ethyl acetate:hexanes (1:8, v/v) to afford 132 as a colorless oil (92 mg, 85%). The physical properties of 132 are as follows: ). 3 [aJ=+33.5 (c=1.00, CHC1 IR Vmax (film): 2910 (C-H stretching). H-NMR (400 MHz, CDC13) 8: 0.75-1.85 {22H, m, including 0.84 (3H, s), 1.04 (3H, s), 1  1.22 (3H, s), 1.63 (3H, s) and 1.70 (3H, s), 2.04-2.55 (3H, m), 3.40-3.75 (2H, m).  101  MS m/z: 256/254 (M, 4.8/14.8%), 241 (12.8%), 239 (37.6%), 203 (96.4%), 109 C1: 256.1772, found: 37 (100.0%). High resolution mass measurement: calculated for C16H27 256.1763; calculated for C161127 C1: 254.1801, found: 254.1801. 35  2.3.18.  Ozonolysis:  alkene 138 to ketone 131  131 Method A: To a solution of vinylcyclopropane 138 (200 mg, 0.9 17 mmol) in a mixture solvent  t  BuOH:water (9.0 ml, 2:1, v/v) was added potassium permanganate (436 mg, 2.76 mmol) at room temperature; the dark purple solution was stirred first at room temperature for 40 minutes.and then at 40°C for 10 minutes. Afterwards, the mixture was diluted with water (20.0 ml) and extracted with ethyl acetate (2x25m1). The combined extract was washed with brine (10 ml) and concentrated in vacuo. The oil obtained above was then dissolved in methanol (10 ml) and treated with 4 (313 mg, 0.706 mmol) for 1 hour at room temperature. After concentration in Pb(OAc) vacuo, the crude product was column chromatographed using ethyl acetate:hexanes mixture (1:8, v/v) to give the starting vinylcyclopropane 138 (4.1 mg, 2.0%) and ketone 131 (180 mg, 91% based on recovery). Method B: A stream of ozone was passed through a solution of vinylcyclopropane 138 (117 mg, 0.536 mmol) in methylene chloride (5.0 ml) at -40°C for 30 minutes. After the addition of dimethyl sulfide (2.0 ml), the mixture was warmed up slowly and then stirred at room  102  temperature for two days. Concentration of the reaction mixture in vacuo gave a crude product which was chromatograghed with ethyl acetate:hexanes mixture (1:8, v/v) to provide ketone 131 (73 mg, 62%). 2.3.19.  Nucleophilic Addition by MeLi:  ketone 131 to alcohol 130  To compound 131 (91 mg, 0.41 mmol) in anhydrous THF (2.0 ml) was added methyl lithium (1.40 M, THF) in a dropwise manner at -40°C with bipyridyl as the indicator till an orange color was observed persistently. The mixture was warmed to room temperature, stirred for an additional 60 minutes, quenched with water (15 ml), and extracted with diethyl ether (2x15 ml). The ether solution was dried over magnesium sulfate. Solvent evaporation gave an oil which was chromatographed with ethyl acetate:hexanes mixture (1:8, v/v) to provide alcohol 130 as a colorless oil.(65.5 mg, 75% based on recovery of starting material) and the starting compound 131 (9.1 mg, 10%). 2.3.20.  Conversion of 138 to 133 via 139  [1R-( 1 cc,3acç7ac)j 3a,4,5 ,6,7,7a-Hexahydro- 1 ,4,4,7a-tetramethyl-2(3H)-( 1-methyl ethylidene)- 1H-indene (133) [1 R-( 1 ct,3acx,7ac.t)j 3a,4,5 ,6,7 ,7a-Hexahydro- 1 ,4,4,7a-tetramethyl-2-( 1(phenylthiomethyl)ethylidene- 1H-indene (139)  139  133  The mixture of vinylcyclopropane 138 (50.5 mg, 0.23 mmol) and thiophenol (50 .tl, 0.49 mmol, 2.0 eqv.) in benzene (2.0 ml) was refluxed for 24 hours under nitrogen. This mixture was concentrated and chromatographed to give the starting vinyl cyclopropane 103  mixture was concentrated and chromatographed to give the starting vinyl cyclopropane 138 (10.1 mg, 20%) and and a polar fraction containing 139 (mg). The concentrated polar fraction was dissolved in THF (2.0 ml). To this solution was distilled ammonia (-3 ml) under nitrogen. Small pieces of lithium were added with stirring till a dark blue color persisted. The reaction mixture was then treated with ammonium chloride, filtered, and concentrated to provide a crude product. The crude product was purified by column chromatography to afford a mixture (28 mg, 69%) containing 133 and 128 (2.3:1) as indicated by GC. 2.3.21.  Reduction by Bu SnH: 3  chloride 132 to alkene 133  133 Homoallylic chloride 132 (50.2 mg, 0.197) in benzene (19 ml) was treated with tributyltin hydride (66 p1, 0.24 mmol, 1.2 eqv.) and AIBN (3.2 mg, 0.19 mmol, 0.15 eqv.) under nitrogen. The mixture was refluxed for 2 days. Evaporation of the solvent gave an oil which was then chromatographed with hexanes to give hydrocarbon 133 as a colorless oil (30 mg, 69%). The physical properties of 133 are as follows: JR (film) vmax.: 2950 cm . 1 H-NMR (400 MHz, CDC1 1 ) & 0.84 (3H, s), 0.87 (3H, s), 1.02 (3H, s), 1.05 (3H, d), 3 1.58 (3H, s), 1.62 (3H, s), 2.04-2.35 (3H, m). MS m/z: 220 (M, 23.1%), 205 (77.6%), 177 (30.6%), 41(100.0). 2.3.22.  Cyclopropane Sliding Reaction:  104  alcohol 130 to alcohol 144  [3aS-(3ax,7acc)] 3a,4,5,6,7,7a-Hexahydro-a,x,3a,7,7-pentamethyl- JH-indene-2-ethanol (144)  b 144 To the solution of alcohol 130 (80 mg, 0.34 mmol) in a dioxane:water mixture solvent (4.00 ml, 1:1, v/v) was addedp-toluentsulfonic acid hydrate (20 mg, 0.10 mmol, 0.30 eqv.). The mixture was heated at 85°C for 1 hour and cooled to room temperature. Water (10 ml) was added and methylene chloride (2x10 ml) was used to extract the aqueous solution. The methylene solution was washed with brine (10 ml), dried over magnesium sulfate, and Column chromatography of the crude product with ethyl  concentrated in vacuo.  acetate:hexanes mixture (1:8, v/v) gave homoallylic alcohol 144 (70 mg, 87%). The physical properties of 144 are as follows: [cxJ=+45.2 (c=l.00, CHC1 ). 3 JR Vmax. (film): 3 100-3650 (OH stretching). H-NMR (400 MHz, CDC1 1 ) & 0.88(3H, s), 1.02(3H, s), 1.07-1.70{ 18H, m, including 3 1.18 (3H, s), 1.21 (3H, s) and 1.22 (3H, s)}, 2.05-2.45 (4H, m), 5.33 (1H, bs) MS m/z: 236 (M, 0.1%), 218 (1.6%), 203 (5.0%), 178 (7.1%), 163 (100.0%), 135 18 C 2 H 6 236.2140; found: (21.1%). High resolution mass measurement: calculated for 0: 236.2145. 2.3.23.  Epoxidation:  alcohol 144 to epoxyalcohol 147  [2R-(2a,3c,3ax,7ac)] 2,3,3a,4,5,6,7,7a-Octahydro-a,x,3a,7,7-pentamethyl- 1H-2,3epoxyindene-2-ethanol (147)  105  147 To a solution of alcohol 144 (172 mg, 0.729 mmol) in chloroform (5.0 ml) was added m-CPBA (243 mg, --80% pure, 1.1 mmol, 1.5 eqv.). The mixture was stirred at room temperature for 1 hour. After addition of methylene chloride (5.0 ml) and washing with sodium bicarbonate solution (10 ml, 10%), the mixture was dried over magnesium sulfate and concentrated in vacuo.  Column chromatography of the crude product with ethyl  acetate:hexanes mixture (2:8, v/v) gave epoxide 147 (159 mg, 87%). The physical properties of 147 are as follows: m.p.: 82-84°C. ). 3 [cxJ=+56.7 (c=1.00, CHC1 JR Vmax. (film):3700 (0-H stretching). ‘H-NMR (400 MHz, CDC1 ) ö: 0.70-1.70 (24H, m, including 0.80 (3H, s), 0.98 (3H, s), 3 1.20 (3H, s), 1.24 (3H, s) and 1.31 (3H, s)}, 1.75-2.02 (2H, m), 2.04-2.15 (1H, dd, J=7.2 and 13.6 Hz), 2.85 (1H, s). MS m/z: 252 (M, 0.2%), 234 (4.1%), 219 (6.9%), 194 (17.9%), 179 (19.8%), 161 (19.3%), 123 (100.0%), 109 (90.4%). High resolution mass measurement: calculated for 18 C 2 H : 2 0 6 252.2089; found: 252.2088. Elemental Analysis: calculated for H2 12: C 0 8 6 C 76.14, H 11.18; found: C 76.14, H 1.05. 2.3.24.  Reductive Fragmentation by LAH:  epoxyalcohol 147 to allylic  Alcohol 151 [1 S-( 1 c,3ax,7ac)] 3a,4,5,6,7 ,7a-Hexahydro-4,4,7-trimethyl-2(3H)-methylene- 1H-inden- 1-  ol (151) 106  OH  151  Epoxide 147 (30.3 mg, 0.583 mmol) in anhydrous THF (1.0 ml) was added in a dropwise manner to a slurry of LAH (18.4 mg) in THF (1.0 ml) under nitrogen. The mixture was then heated at about 70°C (bath temperature) for 2 hours. After cooling to room temperature, ethanol (5.0 ml) was added and stirring continued for 10 minutes. Subsequently, water (15 ml) was added and the resulting mixture was extracted with ethyl acetate (2X10 ml). The ethyl acetate solution was dried over magnesium sulfate and concentrated in vacuo.. Column chromatography of the crude product with ethyl acetate:hexanes mixture (1:8, v/v) gave allylic alcohol 151 (20 mg, 87%) The physical properties of 151 are as follows: ). 3 [c]=+5.4 (c=1.00, CHC1 JR vmax. (film): 3 100-3650 (0-H stretching), 3060 (C-H stretching, olefinic), 1650 (C=C stretching). ‘H-NMR (400 MHz, CDC13) ö: 0.82 (3H, s), 1.02 (3H, s), 1.05-1.72 { 13H, m, including 1.14 (3H, s)}, 1.78 (1H, t, J=8.8  ), 2.20-2.60 (2H, m).  MS m/z: 194 (Mt13.l %), 179 (21.6%), 161 (13.0%), 123 (100.0%), 109 (85.5%). High resolution mass measurement: calculated for Ci 2 H2 3 0 : 194.1670; found: 194.1661. 2.3.25.  Allylic Oxidation by Mn0 : 2  homoallylic alcohol 151 to enone 152  [3aR-(3act,7act)] 3a,4,5,6,7,7a-Hexahydro-4,4,7a-trimethyl-2(3H)-methylene-1H-inden- 1one (152)  107  152  Allylic alcohol 151 (29 mg, 0.15 mmol) in methylene chloride (2.0 ml) was treated with manganese dioxide (65 mg, 0.75 mmol). The slurry was stirred at room temperature for 72 hours. After Filtering of the slurry and washing with methylene chloride (10 ml), the methylene chloride solution was concentrated in vacuo. Column chromatography of the crude product gave enone 152 (8.0 mg, 67% based on recovery)and starting allylic alcohol 151 (17 mg, 59% recovery). The physical properties of 152 are as follows:  [a]=÷57 (c=0.58, CHC1 ). 3 UV (MeOH, c=23 mg/l) max.: 235 nm (log e=4.0), 278 (log E=2.5). JR Vm. (film): 1710 (C=O stretching), 1635 (C=C stretching) cm . 1 11NMR (400 MHz, 3 1 CDC1 6: 0.75-1.70 { 1611, m, including 0.85 (311, s), 1.07 (3H, s) ) and 1.22 (3H, s)), 2.35-2.65 (2H, m), 5.37 (3H, bs), 6.07 (311, bs). MS m/z: 192 (M, 49.9%), 177 (20.3%), 149 (28.9%), 123 (80.7%), 68 (100.0%). High resolution mass measurement: calculated for 0: 10 C 2 H 3 192.15 14; found: 192.15 15. 2.3.26.  Cyclopropane Sliding Reaction:  alcohol 130 to acetates 153 and 154  [1 aS-(3ac,7acz)] 3a,4,5 ,6,7,7a-Hexahydro-c,a,3 a,7 ,7-pentamethyl- 1H-indene-2-ethyl acetate (153) [1 R-( 1 x,3ax,7ac)] 1 ,3,3a,4,5,6,7,7a-Octahydro-4,4,7a-trimethyl-2H-indene- 1-methyl acetate (154)  108  OAC<  153  154  A solution of alcohol 130 (60 mg, 0.26 mmol) in acetic acid (2.5 ml) was heated at 65°C for 2 hours. After cooling to room temperature, methylene chloride (10 ml) was added and the mixture was extracted with 10% sodium bicarbonate solution (10 ml). The methylene chloride solution was dried over magnesium sulfate and concentrated in vacuo. Column chromatography of the crude product with ethyl acetate:hexanes mixture (1:25, v/v) yielded acetate 153 (41 mg, 60% based on recovery), acetate 154 (4.0 mg, 6% based on recovery), starting alcohol 130 (2.9 mg, 5%) and vinyl cyclopropane 138 (3.1 mg, 6% based on recovery). The physical properties of 153 are as follows: ). 3 [x]=+41.7 (c=1.00, CHC1 IR Vm. (film): 1735 (C=O stretching), 1650 (C=C stretching). H-NMR (400 MHz, CDC1 1 ) & 0.85 (3H, s), 1.00 (311, s), 1.03-1.60 { 16H, including 1.15 3 (3H, s), 1.38 (3H, s) and 1.45 (3H, s)}, 1.97 (3H, s), 2.02-2.35 (2H, m), 2.39-2.62 (2H, AB type, J=7.2 Hz), 5.2 6(111, s). MS ni/z: 218 (M HOAc, 37.0%), 203 (100.0%), 175 (16.7%), 147 (21.5%). High -  16 0 C 30 HOAc): 218.2034; found: H 18 (C 2 resolution mass measurement: calculated for H -  3 as carrier gas): 279 (M+Hj, 219, 203. 218.2030. Chemical ionization (NH The physical properties of 154 are as follows: ). 3 [cz]=+63 (c=0.20, CHC1 IR Vm (film): 1730 cm 1 (C=O stretching).  109  H-NMR (400 MHz, CDC1 1 ) & 0.75-1.80 (22H, m, including 0.85(3H, s), 1.03 (3H, s), 3 1.14 (3H, s), 1.61 (3H, s) and 1.69 (3H, s)}, 2.01 (3H, s), 2.10-2.32 (2H, m), 2.39 (1H, t, J=5.6 Hz) MS m/z: 278 (M, 0.3%), 218 (26.0%), 203 (100.0%). High resolution mass measurement: 8 278.2246; found: 278.2248. 10 C 3 H : 2 calculated for 0 2.3.27.  Cyclopropane Sliding Reaction:  ketol 117 to ketol 155  [3aR-(3ax,7acx)] 3,3a,4,6,7,7a-Hexahydro-2-(2-hydroxyl-2-methylpropyl)-7a-methyl-5Hinden-5-one (155)  To the solution of ketol 117 (82 mg, 0.37 mmol) in a dioxane :water mixture solvent (4.00 ml, 1:1, v/v) was added p-toluenesulfonic acid hydrate (22 mg, 0.11 mmol, 0.30 eqv.).  The mixture was heated at 85°C for 3.8 hours. After cooling to room temperature, the mixture was diluted with water (10 ml) and extracted with methylene chloride (2x10.0 ml). The methylene solution was extracted with brine (10 ml), dried over magnesium sulfate and concentrated  in vacuo.  Column chromatography of the crude mixture with ethyl  acetate:hexanes mixture (2:8,  v/v) gave product 155 (72 mg, 87%).  The physical properties of 155 are as follows: [a]=+111 (c=1.00, CHC1 ). 3 JR Vm. (film): 3050-3650 (0-H stretching), 1700 (C=0 stretching), 1650 (C=C stretching). ‘H-NMR (400 MHz, CDC1 ) & 1.10-1.90 { 13H, m, including 1.20 (3H, s) and 1.23 (6H, 3 two singlets)}, 1.95-2.60 (7H, m), 2.75 (1H, dd, 3=8.8 and 17 Hz), 5.20 (lH, bs).  110  MS m/z: 222 (M+, 2.8%), 204 (13.6%), 189 (10.0%), 147 (100.0%), 133 (34.6%), 106 (47.4%). High resolution mass measurement: calculated for 2 1O C 2 H : 4 222.1620; found: 222.1618. 2.3.28. HOAc Promoted Ring Opening:  ketol 117 to ketoacetates 156 and 157  [1 R-( 1 x,3aa,7a)] 1 -Acetoxymethyl-3,3a,4,6,7,7a-hexahydro-7a-methyl-2(1H)-( 1methylethylidene)-5H-inden-5-one (156) [4aS-(4ax,5a,8ax)] 5-Acetoxyl-3,4,4a,5,8,8a-hexahydro-4a-methyl-7(6H)-(1-methyl ethylidene)-naphthalen-2( 1H)-one (157) cc  A solution of alcohol 117 (65.2 mg, 0.294 mmol) in acetic acid (2.5 ml) was heated at 85°C for 2 hours. After cooling to room temperature, methylene chloride (10 ml) was added and the mixture was extracted with 10% sodium bicarbonate solution (10 ml). The methylene chloride solution was dried over magnesium sulfate and concentrated in vacuo. Column chromatography of the crude product with hexanes : ethyl acetate (1:8, vlv) yielded acetate 156 (44 mg, 56% ) and acetate 157 (11 mg, 14%). The physical properties of 156 are as follows: ). 3 [cz]=+63.0 (c=1.00, CHC1 . 1 IR Vm (film): 1735 (C=O stretching of the acetate group), 1705 (C=O stretching) cm  ) 6: 1.28 (3H, s), 1.50-1.85 (8H, m, including 1.59 (3H, s) and 3 ‘H-NMR (400 MHz, CDC1 1.70 (3H, s)}, 1.92 (1H, m), 2.06 (3H, s), 2.10-2.60 (7H, m), 3.95-4.20 (2H, m).  111  MS m/z: 264 (M, 0.1%), 204 (23.6%), 189 (13.2%), 147 (100.0%), 134 (85.1%), 119 (44.4%). High resolution mass measurement: calculated for H240 16 264.1725; found: C : 3 264.1720. The physical properties of 157 are as follows: ). 3 [z]=+30 (c=0.66, CHC1 JR Vm. (film): 1710 (C=O stretching) cm. 1 H-NMR (400 MHz, CDC1 1 ) ö: 1.17 (3H, s), 1.40-1.80 {8H, m, including 1.65 (3H, s), and 3 1.72 (3H, s)), 1.90-2.80 (12H, m, including 2.10(3H, s)}, 5.19 (1H, dd, 3= 4.2 and 10.2  Hz). MS m/z: 204 (M-HOAc, 43.5 %), 189 (19.2%), 147 (91.9%), 133 (100.0%), 119 (54.4%), 105 (5 1.2%). High resolution mass measurement calculated for 0 10 C 2 H 4 (M-HOAc): 204.1514; found: .204.1508. Chemical ionization (NH ): 282 (M+NH 3 ), 265 (M+Hj, 222 4 (M-HOAc+NT14), 205 (M-HOAc+Hj. 2.2.29.  Baeyer-Villiger Reaction:  ketone 131 to acetate 160  [1 aR-(1 ax,ibf,5a13,6ac)] Decahydro- 1 b,5,5-thmethylcycloprop[a]inden-6a-yl acetate (160) or [1 R,6S ,7R,9S] 9-Acetoxyl-2,2,6-trimethyhricyclo [4.4.0.0 .Jdecane (160) 9 ’ 7  160 To the solution of ketone 131 (1.80 g, 8.18 mmol) in methylene chloride (10.0 ml) was added m-CPBA (4.45 g, 80-85% pure, 2.1 mmol, 2.5 eqv.). The above mixture was refluxed for 12 hours during which a milky thick slurry was observed. After cooling to room temperature, methylene chloride (50 ml) was added and the mixture was washed with 10% rapidly solution (50 ml). The organic layer was separated, washed with brine (20 ml), dried  112  with magnesium sulfate, and concentrated in vacuo. Column chromatography of the crude product gave acetate 160 (1.50 g, 82% based on starting material recovery) and starting ketone 131 (0.09 g). The physical properties of 160 are as follows: . CHC1 ) []=÷36.7 (c=0.995, 3 . 1 IR Vmax. (film): 3050 (C-H stretching), 1735 (C=O stretching) cm ) ö: 0.70 (1H, m), 0.80 (3H, s), 0.90-1.02 (4H, m, including 3 H-NMR (400 MHz, CDC1 1 0.97 (3H, s)), 1.05 (3H, s), 1.10-1.70 (8H, m), 1.90-2.10 {4H, including 2.10 (3H, s)}, 2.23 (1H, dd, J=8.0 and 12.0 Hz). MS m/z: 236 (M, 1.1%), 221 (19.0%), 194 (21.5%), 179 (22.2%), 109 (100,0%). High 5 236.1776; found:. 236.1774. 14 C 2 H : 2 resolution mass measurement: calculated for 0 2.3.30.  Saponification:  acetate 160 to cyclopropanol 161  [1 aR-( 1 aa, 1 bf3,5af3,6aa)] Decahydro- 1 b,5,5-trimethylcycloprop [a]inden-6a-ol (161) or decan-9-ol (161)  [1 R ,6S ,7R,9S1 2,2,6-Trimethyltricyclo  161 Acetate 160 (589 mg, 2.50 mmol) was dissolved in ethanol (20 ml) at room temperature. To this solution was added grounded potassium hydroxide (230 mg, ‘—80% pure, 3.28 mmol) under nitrogen. The resulting mixture was stirred for 30 minutes, diluted with water (20 ml), and extracted with methylene chloride (2x20 ml). The methylene chloride solution was dried over magnesium sulfate, concentrated to provide alcohol 161 as an oil (490 mg, 100%). The physical properties of 161 are as follows:  113  ). 3 [a]=+34.5 (c=1.00, CHC1 JR Vm (film): 3050-3650 cm’. H.NMR (400 MHz, CDC1 1 ) 6: 0.80 (3H, s), 0.96 (3H, s), 1.01 (3H, s), 1.98 (2H, m). 3 MS m/z: 194 (M, 3.2%), 179 (4.6%), 124 (28.8%), 109 (100.0%), 81(22.6%). High resolution mass measurement: calculated for C 22 194.1672; found:.194.1665. H 3 1 0: 2.3.31.  : 3 Cyclopropane Ring Opening Reaction by FeCI  cyclopropanol 157  to -chtoroketone 162 [4aS-(4acL,8acL)] 4-Chloro-3,4,4a,5,6,7,8,8a-octahydro-4a,8 ,8-trimethylnaphthalen-2( 1H)one  (162)  [1 R-( 1 a,3 aa,7act)] 1,3 ,3a,4,5,6,7,7a-Octahydro- 1 ,4,4,7a-tetramethyl-2H-inden-2-one (164)  162  164  The alcohol 161 (490 mg, 2.53 mmol) obtained from above was dissolved in anhydrous DMF (12.5 ml) under nitrogen and cooled to 0°C. Dry ferric chloride (1.03 g, 6.35 mmol) was added to this solution. After stirring for 1 hour, the resulting brown mixture was warmed up to room temperature and remained stirred for 24 hours. Addition of 1 M hydrochloric acid (20 ml), extraction with diethyl ether (2x20 ml), and drying over magnesium sulfate was followed by concentration to give the crude product containing 162 and 164 which was subject to elimination in the next step without separation. The physical properties of 162 are as follows: JR Vmax. (film): 1720 cm . 1  114  H-J4),4R (400 MHz, CDC13) 6: 0.76 (3H, s), 0.90 (3H, s), 1.25 (3H, s), 2.10-3.00 (4H, 1 m), 4.70 (1H, dd, J=6.0 and 12.0 Hz). MS m/z: 228/230 (M, 0.6%/0.2%), 206 (1.0%), 193 (7.5%), 43 (100.0%). High C1: 230.1251, found: 230.1223; 37 resolution mass measurement: calculated for C13H210 calculated C13H210 C1: 228.1281, found: 228.1276. 35 The physical properties of 164 are as follows: . 1 JR (film) vmax.: 1725 cm H-NMR (400 MHz, CDC1 1 ) 6: 0.84 (3H, s), 0.91 (3H, s), 1.17 (3H, s), 1.24 (3H, s), 1.74 3 (1H, t, J=10.6 Hz), 2.00 (1H, q, J=7.2 Hz), 2.04-2.35 (2H, m). MS m/z: 194 (M, 3.2%), 124 (28.8%), 109 (100.0%), 81(22.6%). 2.3.32.  Dehydrochlorination:  -chIoroketone 162 to enone 163  [4aR-(4acç8x)] 4a,5,6,7 ,8,8a-Hexahydro-4a,8 ,8 -trimethylnaphthalen-2( 1H)-one (163)  163 The above crude product containing f3-chloro-ketone 162 was dissolved in a saturated sodium acetate methanol solution (10 ml). This mixture was refluxed for 3 hours and concentrated in vacuo. Purification by column chromatography with ethyl acetate: hexanes (2:8) gave enone 163 (384 mg, 80% from acetate 160) and ketone 164 (24 mg, 5%). The physical properties of 163 are as follows: m.p.: 64-66°C. ). 3 [x]=+47.6 (c=1.00 CHC1 ,  UV (MeOH, c=20.0 mg/i) Amax.: 235 nm (log e3.842).  115  IR nmax. (film): 1664 cm-’ (C=O stretching). H4s4MR (400 MHz, CDC1 1 ) 6: 0.77 (3H, s), 0.96 (3H, s), 1.22 (3H, s), 1.22 (3H, s), 3 1.27-1.75 (7H, m), 2.50-2.80 (2H, m), 5.95 (1H, d, 3=9.6 Hz), 6.27 (1H, d, J=9.6 Hz). MS m/z: 192 (M, 13.3%), 150 (45.1%), 69 (100.0%). High resolution mass measurement: calculated for H2 13 C 0 0 : 192.1514; found: 192.1518. 10 C 2 H 3 C 81.20, H 10.50; found: C 81.13, H 10.48. Elemental Analysis: calculated for 0: 2.3.33.  Ring Opening Reaction by NBS:  cyclopropanol 161 to  13-  bromoketone 167 [1 R-( 1 x,3a,7aa)] 1 -Bromomethyl- 1,3 ,3a,4,5,6,7 ,7a-octahydro-4,4,7a-trimethyl-2Hinden-2-one (167)  167 Cyclopropanol 161 (12.8 mg, 0.066 mmol) in dimethy1su1foxide:chlorofom (4.0 ml, 1:1, v/v) mixture solvent was stirred with NBS (23.5 mg, 0.132 mmol, 2.0 eqv.) at room temperature for 3 hours. Water (5 ml) was added and methylene chloride (10 ml) was used to extract the aqueous solution. Magnesium sulfate drying and concentration in vacuo resulted in an oil which was chromatographed with ethyl acetate:hexanes mixture (2:8, v/v) to afford  13-  ketobromide 167 (6.3 mg, 60%) and cyclopropanol 161 (5.3 mg). The physical properties of 167 are as follows: JR (film) vmax.: 1730 cm 1 (C=O stretching). HNMR (400 MHz, CDC1 1 ) 6: 0.82 (3H, s), 1.00 (3H, s), 1.14 (3H, s), 1.91 (1H, t, 3=9.0 3  Hz), 2.32 (2H, m), 2.55 (1H, t, J=5.4 Hz), 3.35-3.65 (2H, m).  116  MS m/z: 274/272 (M, 2.4% /2.5%), 193 (7 1.2%), 175 (16.1%), 109 (100.0%). 2.3.34.  Dehydrogenation:  enone 163 to dienone 168  [4aR] 5,6,7,8-Tetrahydro-4a,8,8-trimethylnaphthalen-2(4aH)-one (168)  o 1 c 168 The solution of 0.42 M LDA (1.84 ml) in n-pentane was concentrated to a viscous mixture and cooled to -78°C. To this mixture was added THF (1.0 ml) and introduced the solution of 163 (135 mg, 0.703 mmol) in THF (1.5 ml) in a dropwise manner under nitrogen protection. After stirring for 1 hour, phenylselenenyl chloride (183 mg, 0.844 mmol, 1.2 eqv.) in anhydrous THF (0.50 ml) was added rapidly. The reaction mixture was warmed to room temperature, stirred for another 1 hour, and treated with 30% hydrogen peroxide (0.72 ml). After stirring for 5 hours, saturated sodium carbonate (aq., 5 ml) and diethyl ether (5 ml) were added. The organic layer was separated, washed with brine, dried over magnesium sulfate, concentrated in vacuo to afford the crude product. The crude product was purified by column chromatography using ethyl acetate:hexanes mixture (2:8, v/v) to provide dienone 168 (111 mg, 92% based on recovery of starting material) and the starting enone 163 (13 mg). The physical properties of 168 are as follows: [ct]=+57.3 (c=1.00, CHC1 ). 3 UV (MeOH, c=20 mg/I)  max.:  241 nm (log E=4.00).  . 1 JR Vmax. (film): 1660 (C=O stretching), 1620 (C=C stretching) cm  117  ‘H-NMR (400 MHz, CDC1 )& 3  1.15-2.10 { 15H, m, including 1.22 (3H, s), 1.30 (3H, s),  1.35 (3H, s)}, 6.14 (1H, dd, J=0.2 and 9.9 Hz), 6.25 (1H, d, J=0.2 Hz), 6.70 (1H, d, J=9.9 Hz). MS m/z: 190 (M+, 6.1%), 175 (12.0%), 147 (9.9%), 41(21.2%). High resolution mass measurement: calculated for 0: C 1 H 3 190.1357; found: 190.1358. 8 2.3.35.  Birch Reduction:  dienone 168 to enone 64  [4aR] 4,4a,5 ,6,7,8-Hexahydro-4a,8,8-trimethylnaphthalen-2(3H)-one (64)  64  To the solution of dienone 168 (85.9 mg, 0.452 mmol) in anhydrous THF (2.0 ml) was distilled ammonia (4 ml) from sodium under nitrogen. Small pieces of lithium were added for 30 minutes until a dark blue color persisted. After stirring was continued for 30 minutes, ammonium chloride powder was added to remove excess lithium. Evaporation of ammonia and THF gave a yellowish oil which upon column chromatography produced the desired enone 64 (58.0 mg, 74.2%) and the starting dienone 168 (8.5 mg). The physical properties of 64 are as follows: [cc]=-100 (c=1.00, CHC1 ). 3 UV (MeOH, c=20.0 mg/I) 2 max: 242 nm (log e=4.10). JR Vmax (film):  1665 cm 1 (C=O stretching).  ‘H-NMR (400 MHz, CDC1 ) ö: 1.14 (3H, s), 1.19 (3H, s), 1.34 (3H, s), 1.40-2.00 (8H, 3  m), 2.38 (1H, m), 2.59 (1H, m), 5.96 (1H, s).  118  MS m/z: 192 (M, 100.0%), 177 (30.7%), 164 (12.8%), 149 (37.0%). High resolution 10 C 2 H 3 192.15 14; found: 192.15 12. mass measurement: calculated for 0: 2.3.36.  ketone 171 to dienone 172  Dehydrogenation:  [4aR-( 1 cz,8)] 5,6,7,8-Tetrahydro-4a,8-dimethylnaphthalen-2(4aH)-one (172)  0 b 172 The mixture of ketone 171 (2.64 g, 14.7 mmol) and DDQ (7.41 g, 32.2 mmol, 2.2 eqv.) in dioxane (50 ml) was refluxed under nitrogen for 24 hours. Evaporation of the solvent in vacuo gave a brown oil which was purified by column chromatography with ethyl acetate:hexanes mixture (2:8, v/v) to provide dienone 172 in 80% yield (1.65 g) and the starting material 0.53 g. The physical properties of 172 are as follows: . 1 JR (film) vmax.: 1650 (C=O stretching), 1620 (C=C stretching) cmH-NMR (400 MHz, CDC1 1 ) : 0.95-1.45 (8H, m, including 1.14 (3H, d, 3=6.0 Hz) 1.27 3 (3H, s)}, 1.65-2.10 (4H, m), 2.51 (1H, septet, J=6.0 Hz), 6.11 (1H, s), 6.21 (1H, d, J=9.0 Hz), 6.78 (1H, d, J=9.0 Hz). MS m/z: 176 (M, 5.1%), 161 (3.1%), 149 (16.2%), 43 (100.0%). High resolution mass measurement: calculated for 0: C 1 H 2 176.1201; found: 176.1198. 6 2.3.37.  dienone 172 to enone 173 and ketone 174  Birch Reduction:  [4aR-( 1 a,8 [3)] 4,4a,5,6,7, 8-Hexahydro-4a,8-dimethylnaphthalen-2(3H)-one (173) [4aR-( 1 x,8 [3)] 3 ,4,4a,5,6,7,8 8a-Octahydro-4a,8-dimethylnaphthalen-2( 1H)-one (174) ,  119  da dc 0 :174  173  To a solution of dienone 172 (200 mg, 1.14 mmol) in anhydrous diethyl ether (3.0 ml) was distilled anhydrous ammonia (—4 ml) from sodium. Small pieces of lithium were added under nitrogen for 30 minutes until a steady dark blue was observed. This solution was stirred at -33°C for another 30 minutes, quenched with ammonium chloride powder, warmed to room temperature. Concentration of the mixture in vacuo gave the crude product which was chromatographed with ethyl acetate:hexanes mixture (2:8, v/v) to provide enone 173 (75 mg, 42%), ketone 174 (45 mg, 25%), and the starting material 171 (21 mg). The physical properties of 173 are as follows: [a]=-193 (c=1.03, CHC1 ). 3 UV (EtOH, c=10.3 mg/i) max.: 240 nm (log e=4.025) JR (film) vmax.: 3052 (olefinic C-H stretching), 1660 (C=O stretching), 1610 (C=C stretching) cm . 1 ‘H-NMR (400 MHz, CDC1 ) ö: 1.06 (3H, s), 1.15 (1H, m), 1.25 (3H, s), 1.38 (1H, m), 3 1.55-2.00 (6H, m), 2.25 (3H, m), 5.79 (1H, s). MS m/z: 178 (M, 76.0%), 162 (25.8%), 150 (54.7%), 79 (100.0%). High resolution mass measurement: calculated for 0: C 1 H 2 178.1357 ; found: 178.1354. 8 The physical properties of 174 are as follows: [x]=-39.7 (c=0.985, CHC1 ). 3 JR (film) vmax.: 1702 cm 1 (C=O stretching). ‘H-NMR (400 MHz, CDC1 ) 6: 0.81 (3H, d, J=6.0 Hz), 0.93 (1H, m), 1.05 (3H, s), 1.12 3 (1H, m), 1.30-1.80 (8H, m), 2.00 (1H, t, J=14 Hz), 2.25-2.55 (3H, m).  120  MS m/z: 180 (M, 50.1%), 165 (9.3%), 109 (100.0%). High resolution mass measurement: calculated for C 2H200: 180.1514; found: 180.1517. 1  121  Chapter 3. 3.1.  The Synthesis of Ambergris Fragrances  Introduction  3.1.1.  Ambergris Fragrances  Ambergris is one of the most valuable animal perfumes, like civet, musk, and . Its outstanding fragrance and mysterious effects of its odor account for man’s 81 castoreum familiarity with this material since long before the Christian era in all great civilizations. It is a metabolic product of the sperm whale (Physeter macrocephalus L.), which accumulates as concretions in the gut of the animal. After the concretion leaves the animal body, an aging process takes place over time, as a result of the action of sunlight and oxygen when floating in waves. During this process, the strong stecoraceous indole of fecal note and the waxy constituency disappear. At the same time, a complex yet balanced fragrance that is composed of a series of notes and subnotes, develops gradually to give a harmonious character . 82 The major constituent of ambergris is an odorless triterpene alcohol (-)-ambrein (176)83 which is responsible for the generation of odoriferous compounds 17718484 found in the steam volatile fraction (Figure 20). It can be presumed that the tricyclic compounds 177, 178, and ()Ambrox®* (179) are derived from the bicyclic part of (-)-ambrein (176), while the smaller fragments 180-184 are from the monocyclic part of the molecule. Ambrinol (181) and (+)-dihydro-’—ionone (180) are structurally related and in fact a racemate of 181 can be formed stereoselectively from the racemate of 180 in 70% yield by an intramolecular Prins reaction with Bronsted or Lewis acids as catalysts . The facile formation of (+)-ambreinolide 85 (185) and (+)-dihydro-y-ionone (180) during oxidation of 176 with permanganate supports this structural correlation . 86  *Ambrox® is a registered trade name of Firmenich SA. Systematic name of ()Ambrox® (179): [3aR(3acz,5a,9aa,9bI3)]-dodacahydro-3a,6,6,9a-tetramethy1naphtho[2,1-b]furan.  122  (-)-ambrein (176)  177 181  180  CH(  178  183  182  2 CICH  179 184 Figure 20 The Constituents of Ambergris  lb, (-)-ambrein (176) is degraded by autooxidation 8 According to one hypothesis during the aging process. Singlet oxygen may be considered as an active agent while copper ions from haemocyanin may function as a catalyst in this degradation. Porphyrins, known to a. This theory is supported by 7 be efficient photosensitizers, have been identified in ambergrisS  123  a photooxygenation experiment in which (-)-ambrein (176) was converted to compounds 178, b. 87 180, 181, and 182 by the cleavage of its allylic hydroperoxide  185  186  187  Ambergris is disappearing from the world market due to excessive whale hunting. In addition, the continued increase in the pollution of coasts makes it more difficult to find primequality material which is more and more rarely washed ashore. In the future, the perfume industry must meet its needs for the natural product with a synthetic equivalent. The racemic form of cx-ambrinol (181), possessing an exceptionally strong odor of damp earth with a crude civet subnote, is the only naturally occurring amber odorant for which a fully synthetic equivalent is used commercially. In 1950, it was established that the amber-like odor (woody nature) of the enol ether 178 was retained in its hydrogenation product, ambraoxide (186)88. The following search for adequate odorants resulted in the discovery of ()Ambrox®*, a degradation product of easily accessible sclareol (187); a breakthrough was then achieved in the commercial production of tricyclic amber odorants of woody nature in the late 1950’s89. The mixture of ()Ambrox® (179) and (+)isoAmbrox® (189) in the form of the base Fixateur 404 (trade name of Firmenich) has been available in perfumery for more than 30 years.  *The identification of ()Ambrox® (179) from ambergris is a much later event b. 84  124  3.1.2.  Structure and Activity Relationship of Ambergris Fragrances With ()Ambrox® (179) as a model compound, a large number of compounds have  been prepared for the correlation of structure and odor relationship. For example, the stable Oa 9 A/B transfused  and cis-fused 91 diastereomers of ()Ambrox® (179) have been prepared  and their odor quality and strength have been evaluated* (Figure 21). The difference in the  odors of ()Ambrox® (179) and (+)Ambrox® (188) is rather small. (+)Ambrox® (188) with its higher threshold value (2.4 ppb) and accentuated woody note lacks the strong and warm animal note of its enantiomer 179 (threshold value 0.3 ppb). Therefore, (+)Ambrox®  has been called poor man’s ambrox” by perfumers. The exotic, spicy undertone in (+)-. Ambrox® (188) disappears in its racemate, for which a threshold concentration of 0.5 ppb was measured. (+)IsoAmbrox® (189) has a threshold value of 34 ppb which is more than a hundred times weaker than its model compound 179, showing the importance of an axial methyl at C8 for the receptor event. Surprisingly, ()9epiAmbrox® (190) possesses the strongest odor and the lowest threshold concentration of 0.15 ppb; it lacks slightly the rich and complex bouquet of 179. The diastereomer 191 is unlikely to exist because the trans fusion would force the B ring into a highly strained boat-like conformation. Among the A/B cis-fused series, only racemic diastereomers were evaluated*. Only diastereomer 192 has an odor quality comparable to the prototype ()Ambrox® (179); it has a threshold value of 11 ppb which is 20 times higher than that of racemic Ambrox®. Racemic diastereomers 193, 194, and 195 are very weak odorants and almost devoid of any ambergris odor.  *A comparison of these racemic diastereomers with 179 seems permissible since there is only a small difference in the odor of ().ambrox® (179) and (+)ambrox® (188). The organoleptic evaluation was carried out using a threshold concentration method k 90  125  12 13  ç;E  188  189  190  191  192  193  194  195  Figure 21 Stereoisomers of ()Ambrox® The significance of the gem-dimethyl groups at C4 for the ambergris odor sensation was assessed (Figure 22)90. Both nor-methyl Ambrox® 196 and 197 have the Ambrox® note although 196 with an axial methyl at C4 (threshold value 1.4 ppm) has a greater strength than 197 with an equatorial methyl at C4 (threshold value 3 ppm). (±)DinorAmbrox® 198 without gem-dimethyl group possesses the same woody character of Ambrox® and a dominant earthy odor reminiscent of a freshly plowed earth; it has a threshold value 2.4 ppm. Therefore, the gem-dimethyl group at ring A has considerable influence on the quality and strength although their presence is not an absolute necessity for the ambergris sensation.  196  197  198  Figure 22 The Effect of the gem-Dimethyl Groups on the Ambergris Odor Activity  126  92 proposed a qualitative Based on a large number of analogues assessed, Ohloff “triaxial rule of odor sensation” to summarize the minimal structural requirements for a compound to have ambergris odor activity: 5, 8, lO-triaxial arrangement of the substituents R’, R”, and Ra in the trans-fused decalin ring system is the geometric requirement for a molecule in order to exhibit an ambergris type odor (Figure 23). The compound must possess an oxygen-containing group, the incorporation of which into the R’, R”, or Ra substituents is advantageous but not indispensable. Based on this rule, it is speculated that the specific site of the human olfactory receptor system reacts with the stimulating substance by an intermolecular three-point interaction in three dimensional space. Therefore, the related cis-fused decalyl derivatives, for obvious conformational reasons, do not in general fulfill the stereochemical requirements for odorants with ambergris-like properties.  z#1 Figure 23 Triaxial Rule of Ambergris Odor Sensation More recently, a so-called “ambergris triangle” rule was established by analyzing both . 93 electronic structures and stereochemical features of substituted decalin compounds According to this rule, an odorous compound should contain an “ambergris triangle” of certain dimensions formed by a carbon-attached oxygen atom (0) and two carbon-attached hydrogen 3.2A H  2.90±0.40  A 0  2.8A  an ambergris odorant Figure 24 The Ambergris Triangle Rule  127  atoms (H 1 and H) making major contribution to the LUMO of this compound (Figure 24). Typically, H and Hj are allylic, tertiary, or axial. A specific example is given in Figure 24. According to this group of authors, the interaction between the active odorous molecule and the receptor is molecular orbital controlled. a, many inactive compounds also fulfil the general 94 However, as indicated by Winter structural conditions postulated as being necessary for ambergris-type activity. He explored an approach using the concepts of oriented profile and steric accessibility of the functional group, focussing on a quantitative estimation of the degree of interaction between the poiar (hydrogen bond acceptor, e.g., oxygen) part of an odorant molecule and the hypothetical hydrogen bond donor group (e.g., hydroxyl) on the receptor site ’. The accessible polar surface area, a 941 measure of the steric accessibility, was calculated for each structure after optimization by molecular mechanics calculations. A lower limit of accessibility necessary for activity was found to 6 A. So far, only a limited range of molecules have been tested by this approach. The precise nature of the ambergris odorant and the receptor interaction is essentially a speculation. Each model reveals certain truth since each can make certain successful predictions. More precise models of greater power in the quantitative prediction are likely to evolve in the future as chemists get more acquainted with ever more sophisticated computational technology. In addition, very recent exciting progress has been made in the isolation of human olfactory receptors . Studies of receptor structures and the nature of active 95 sites will enhance our understanding of the sense of smell as a whole as well as the ambergris olfaction. 3.1.3.  Synthesis of Ambrox® Many synthetic sequences leading to ()Ambrox® (179) and its racemate have  appeared in the past few years, which reflects the reduction in available natural sources and the increasing market demand for ambergris fragrances. Most enantioselective syntheses involve  128  the use of naturally derived diterpenes or sesquiterpenes as starting material. A brief summary of the more typical synthetic sequences is presented below. The commercial production 96 of ()Ambrox® (179) is based on procedures developed by Hinder and Stoll in l95O96 (Scheme 29).  These procedures involved  degradation of natural sciareol (187), the principal source of which is clary sage (salvia sciarea L.). Direct treatment of sciareol (187) with chromium trioxide gave lactone 20196c. An d of obtaining 201 consisted of a sequence of reactions: the conversion of 96 alternative way 187 into sclareol oxide (199) by potassium permanganate, ozonolysis of 199 to yield the acetoxy acid 200, and the cyclization of 200 to lactone 201. LAH reduction of lactone 201 generated diol 202 which was then cyclized to ()Ambrox® (179) employing a catalytic amount of of f3-naphthalene sulfonic acid. Usually, (+)isoAmbrox® (189) was generated as a minor by-product. OH  187  202  tc’d  199  179  200  ; b) 03, heating; c) KOH, then HC1; d) 150°C, vacuum; e) Cr0 4 a) KMnO , AcOH; 3 f) LAH, Et 0; g) f3-naphthalenesulfonic acid 2  Scheme 29 Stoll and Hinde?s Synthesis of ()Ambrox® from Sciareol (187)  129  A short sequence using sciareol (187) as starting material was reported by Naf et al. 97 (Scheme 30) Catalytic hydrogenation of sciareol (187) gave dihydrosclareol (203) in good yield. The reduction of this double bond was necessary to ensure a regioselective cleavage in the next step. The diol 203 in carbon tetrachloride was then treated with aqueous sodium hypochiorite to provide hypochlorite 204 which was then decomposed to chloride 205 via an alkoxyl radical fragmentation mechanism. The cycization of 205 by means of sodium hydride in THF afforded ()Ambrox® (179). The overall yield from sclareol (187) was 11-12%.  cc’ Ilgg  a  sclareol 187  b  c  204  203  d  205  179  , EtOH; b) aq. NaOC1, CCL 2 a) 5% Pd-C, H ; c) 30-35°C, 3h; d) NaH, THF, 3h, reflux 1  Scheme 30 Nafs Synthesis of ()Ambrox® from Sclareol A similar sequence from sclareol (187) based on the fragmentation of an alkoxyl radical Christenson (Scheme 31). Sclareol oxide (199), previously prepared a was also reported by 98 from sclareol (187)98, was treated with hydrogen peroxide to produce a diastereomeric hydroperoxide mixture (206). Reaction of 206 with ferrous chloride and a catalytic amount of cupric chloride provided a bifunctional compound 207 which was then hydrolyzed to  Ambrox® (179). The overall yield from sclareol (187) was 34%.  130  (-)-  sclareol  a  b 1 X 206  199  OAc  d  207  179  0 2 ; b) H 4 CuC1 (cat.); d) KOH, ‘PrOH, H , HOAc; c) FeC1 0 2 ,2 2 a) KMnO  Scheme 31 Christenson’s Synthesis of ()Ambrox® from Sciareol  The fourth sequence towards ()Ambrox® (179) using sciareol (187) as starting material was reported by I. C. Coste-Manere et al. 99 (Scheme 32). Sciareol was acetylated to afford 208 which was then converted, in quantitative yield, to diene 209 by treatment with a catalytic amount of palladium acetate in quantitative yield. Reaction of 209 with potassium permanganate generated a mixture of ambreinolide (185) and sciareolide (201) (3:2) in an overall yield of 80%. LAH reduction of 185 and subsequent cyclization by p-toluenesulfonyl chloride provided ambraoxide (186). Similar treatment of 201 furnished ()Ambrox® (179).  131  QAc  a  b  sciareol  208  185  +  ,  d  209  2O1  179  186  , 24hr, 4 O/H, 2h, 96%; d) KMnO 2 Ac 0 ; b) Pd(Ac) 2 /dioxane, 100°C/l5min, 100%; c) LAH, Et a) 2 LAH, O 2 Et /H; TsC1, 2hr, 90% LAH/THF, 25°C/3hr, 1) C TsCI/CH 2 1 25°C, 90%; g) e) 98%; 80%;  Scheme 32 Coste-Manere’s Synthesis of ()Ambrox® from Sciareol Several other diterpenes (Figure 25), abietic acid (82)100, manoyl oxide (211)101, and methyl labdanolate (212) 102, were also degraded into ()Ambrox® (179).  82  211  212  Figure 25 Several Other Diterpene Starting Materials for ()Ambrox® Synthesis  132  103 (Scheme 33) reported the conversion of the sesquiterpene M. 3. Cortes et al.  (-)-  drimenol (33) into ()Ambrox® (179). Oxidation of 33 by pyridinium chlorochromate resulted in aldehyde 213 which was homologated to enol ether 214. Hydrolysis of 214 and the following LAH reduction provided alcohol 215. Protection of the hydroxyl group in 215 as acetate and subsequent dihydroxylation afforded 216 which was then cyclized into furan 217. Oxidation of 217 led to ketone 218 which was further reduced into ()Ambrox® (179). The overall yield of ()Ambrox® (179) from (-)-drimenol (33) was 19%. Notably, the ether linkage cx to the carbonyl group in 218 survived during the Woif-Kishner reduction. H OH 2 CH  b  a  33  OMe  CHO  c, d  214  213  g, h  215  216  217  çj%%Qi ç±° 218  179  0, Pyr.;f) 0s0 2 C1 b) Ph CH ; O; d) LAH; e) Ac 3 P=CH(OMe); c) H 3 a) PCC, 2 ; g) NaOH, 4 C1, Pyr.; i) KOH, DEG, 2 2 0; h) MeSO 2 H NH NH  Scheme 33 Cortes’ Synthesis of ()Ambrox® from (-)-Drimenol (33)  133  1 (Scheme 34) developed an enantionselective synthesis of ()Ambrox® Mon et al.  (179) from geranylacetone (219). Enantiomerically pure tosylate 220 was previously ’. The substitution of the tosyl group in 220 gave 41 prepared from 219 by the same group° nitrile 221 which was treated with a Wittig reagent to give the methylene nitrile 222. This nitrile was reduced with DIBAL to provide 223 and further reduction with sodium borohydride yielded alcohol 224. Stereoselective epoxidation of 224 resulted in 225. Reduction of 225 with LAH generated diol 202 which was then cyclized to ()Ambrox® (179). The overall yield of 179 from geranylacetone (219) was 2.2% in 15 steps. OTs  b  225  e  d  C  222  221  220  219  223  202  224  179  , MeOH; e) m-CPBA, 4 =CH DME; c) DIBAL; d) NaBH 3 Ph , a) NaCN, DMSO; b) 2 N H 5 C1 LiA1H, TUF; g) TsC1, C CH ; 2  Scheme 34 Mon’s Synthesis of ()Ambrox® from Geranylacetone (219)  134  The first synthesis of racemic Ambrox® was reported by Matsui et aL 105 (Scheme 35) Darzen’s condensation of dihydro-3-ionone (226) with ethyl chloroacetate and decarboxylation of the resulting glycidic acid with a catalytic amount of sodium acetate gave an aldehyde 227. Treatment of 227 with malonic acid and subsequent ethylation by titanium tetrachioride in ethanol afforded ethyl trans- 3-monocyclohomofarnesate (228). The cyclization of 228 by means of trifluoroacetic acid yielded tricyclic sclareolide (201). Reduction of this lactone and subsequent ring closure furnished (±)Ambrox®. The overall yield of (±)Ambrox® from dihytho--ionone (226) was 4.9%. c,d  226  228  227  201  179  , EtOH; 4 (COOH) Et CH N; d) TiCI 3 Et, NaOEt b) NaOAc, 200°C; c) 2 2 a) C1CHCO CH H g) TsC1, Pyr. COOH; f) ) 3 3 ( NaA1H C O ; 2 OCH e) CF  Scheme 35 Matsui’s Synthesis of (±)Ambrox® from Dihytho--ionone (226) 106 also started with The second racemic synthesis of Ambrox® by Buchi and Wuest dihydro-3-ionone (226) (Scheme 36). Condensation of 226 with dimethyl carbonate gave the monocyclic f3-ketoester 229 which was cyclized to the bicyclic 13-ketoester 230 using stannic chloride as catalyst. The 0-allylation of 230 provided an allyl ether which was heated in xylene to afford 231. Demethoxycarbonylation led to a mixture of 232 and 233 (6:1). The addition of MeMgI to 232, the ozonolysis of the resulting alcohol 234, and the subsequent treatment with sodium borohydride afforded diol 235. The cyclization of this diol with a  135  catalytic amount of p-toluenesulfonic acid in nitromethane furnished (±)Ambrox® as major product. The hydroboration of 234 yielded a diol 236 which was cyclized to (±)-ambraoxide  (186). The overall yields of (±)Ambrox® and (±)-ambraoxide from dihytho--ionone (226) were 9.0% and 5.4% respectively. CH 2 CO 3 a 3 -ThrcooCH  229  226  230 CH / 2 CO 3  —  -  c  d +  232  f  233  g  235  h,  179  g  236  186  O 3 (CH C 0, ) NaH, DMF, 20°C; b) SnCI a) 2 ,2 4 C1 520°C; c) allyl bromide, Nail, DMF; CH , , DMSO; e) MeMgI; 1)03, MeOH/NaBH 2 d) CaC1 ; g)p-TsOH, 2 4 NO h) BH 3 CH ; -THF/ 3 OH, H 0 2 Scheme 36 Buchi and Wuest’s Synthesis of (±)Ambrox® from Dihydro-3-ionone (226)  136  3.2.  Discussion Retrosynthetic Analysis for Synthesis of ()Ambrox® (179) from  3.2.1.  Enone 163 After the completion of the formal sequence to (-)-polygodial (2), we turned our attention to ()Ambrox® (179), the synthesis of which from thujone was vigorously pursued in our laboratories. ()Ambrox® may be considered as a homo-drimane sesquiterpene. The synthetic sequence we perceived is shown in Scheme 37.  2 3  179 (-)-Ambrox®  163  I 246  189 (+)isoAmbrox®  i 251  255  Scheme 37 Retrosynthetic Analysis for Synthesis of ()Ambrox® The cis-fused enone 163 was a promising starting material: the convex  f face and the  steric hindrance from the axial methyl at C4 in the ring A of the major conformer (non steroidal) should ensure a favorable conjugate addition (e.g., by a vinyl anion equivalent) from the  face of the ring B segment thereby, generating a chiral center C9 of the same  configuration as that in the ()Ambrox® (179). The conjugate addition might provide a good  137  opportunity of introducing a methyl group into C8 regioselectively, by trapping the enolate produced in the addition reaction with methylating reagents like iodomethane. The cis-fused y,6-enone 246 thus obtained would undergo a stereochemical correction step at C5 to its epimer, the trans-fused y6-enone 251.  9 8  the major conformer of 163  We also envisaged that the furan ring C of ()Ambrox® (179) or (+)isoAmbrox® (189) may be formed by an acid catalyzed cyclization of the trans-fused 1,5-diol 255 which  would be prepared by stereoselectuive reduction of 251 and subsequent hydroboration. Compound 251 was expected to possess the conformation as drawn below. An axial orientation of the secondary hydroxyl group at C7 would be necessary to ensure a facile migration of the axial hydride from the vicinal tertiary carbon C8. Alternatively, an equatorial orientation of the secondary hydroxyl group would probably lead to some skeletal rearrangement as shown.  OH  255  260  138  Studies on Conjugate Addition to Enone 163 and Subsequent  3.2.2.  Methylation of 245 To complete the synthesis of ()Ambrox® (179) as planned above, a diastereoselective conjugate addition to enone 163 from the  13 face was necessary.  Conjugate addition to enones  . The 107 by organocopper reagents has been most widely used in organic synthesis stereochemistry of such conjugate addition to octalones analogous to 163 was first examined. Cuprous chloride-catalyzed addition of (2-propenyl) magnesium bromide to trans-fused octalones 237A and 237B (Scheme 38), gave exclusively the products 238A and 238B . This facial preference can be explained in the 108 respectively, resulting from the a face attack following manner : the incoming group has to be perpendicular to the enone plane in order 109 to have maximal orbital overlap during the progress of the reaction and therefore a minimal transition state energy (stereoelectronic requirement); as a result, the antiparallel attack* from the a face would go through a half-chair (chair-like) transition state while the parallel attack* from the  13 face will involve a skew-boat (boat-like) transition state.  The highly strained skew-  boat transition state would require much higher activation energy and therefore the antiparallel attack from the a face prevailed.  *  An antiparallel attack to a carbon-carbon double bond of a cyclohexene ring is defined as the attack antiparallel to the neighboring pseudoaxial group while a parallel attack as the attack parallel to the neighboring pseudoaxial group.  139  =MgBr 2 CH ) 3 CH(CH  237A, R=H 237B, R=CH 3  238A, R=H 3 238B, R=CH  parallel attack  antiparallel attack  Scheme 38 Conjugate Addition of Organocopper Reagents to trans-Fused Octalones  Even for cross-conjugated dienones 239A, 239B, and 239B (Scheme 39), the x face attack still predominated’ 10  1 :R  CuLi 2 ) 3 (CH  R,=CH R , =H 2 239A, 3 =H 2 239B, 3 =CO 1 R C 2 , HR 239C, Rl=CH , 2 3 CH 3 = R ) —C(=CH  R,=CH R , =H 2 240A, 3 =H 2 CH R 2 R,=CO , 240B, 3 3 = R ) -C(CH CH 240C, Rl=CH ,2 3  Scheme 39 Conjugate Addition of Organocopper Reagents to Cross-conjugated Dienones  140  In the case of cis-fused octalone 242 (Scheme 40), only the product 243 resulting from  face attack was obtained by lithium dimethylcuprate addition” . Since octalone 242 1  does not have a rigid conformation, a consideration of all its conformers is necessary to understand this reverse facial stereoselectivity. For the steroid-like conformer 239a, which should be more stable, the parallel attack from the a face is especially disfavored because of the highly hindered concave geometry of the a face and the skew-boat transition state involved. For the non-steroid-like conformer 239b, the antiparallel attack from the a face is effectively blocked by the concave face and the axial acetoxyl grouping which remains in the approaching path of the reagent, despite that the transition state of this antiparalell attack has a half-chair conformation. In the event, the 3 face attack was the reaction path observed.  CuLi 2 ) 3 (CH  243  242  242b  242a  Scheme 40 Conjugate Addition of Organocopper Reagents to a cis-Fused Octalone Using the same argument, we concluded that  f3 face attack of the cis-fused enone 163  would be the favored mode. It is expected that 163 may exist in a conformational equilibria between 163a and 163b. In contrast to 242, the non-steroid-like conformer (i.e., 163b) is more stable than the steroid-like conformer (i.e., 163a). The  f3 face attack at 163a is favored  over the a face attack since it represents an antiparalle which requires a half-chair transition state rather than the less favorable skew-boat essential for the a face attack. The  141  face attack  at 163b is expected as preferred because the severe steric hindrance from the axial methyl at C4 of ring A and the concave geometry of cc face would block the cc face antiparallel attack.  Nu  Nu  163a  0  163b  The use of the trans-fused enone 244 and dienone 165 (Figure 26), which were derivable from 160112, would probably produce a face addition compounds. Thus, these two compounds were very unlikely to be the backup or alternative intermediates towards the synthesis of ()Ambrox® (179).  244  168  163  Figure 26 Potential Candidate Intermediates for the Stereoselective Conjugate Addition Experimentally, conjugate addition of 163 with 2.0 equivalents of vinyl magnesium bromide and a catalytic amount of cuprous iodide in dimethyl sulfide:THF (1:5, v/v) solution  142  gave the  f  face addition product 245 in 70% yield, the stereochemistry of which was  confirmed by the ‘H-.NMR spectrum of the methylation product 246 (see Figures 27 and 28 113 or the use of cuprous and the corresponding discussion). The absence of dimethyl sulfide bromide as catalyst led to decrease in yields, likely due to the formation of competing 1,2addition by-products. Some poiar by-products were often observed in the reaction. y,6-Enone 245 had its mass spectrum showing the molecular ion peak at m/z 220 while its IR spectrum indicated absorptions at 1710 and 1635 cm, corresponding to the presence of the carbonyl 1 group and the terminal carbon-carbon double bond. Its ‘H-NMR spectrum displayed three methyl singlets at 60.90, 0.95, and 1.09 ppm, a complex five-proton multiplet at 6 2.25-3.10 ppm corresponding to the two methylene groups x to the carbonyl group and the tertiary allylic proton, and a three-proton multiplet at 6 4.95-5.80 ppm corresponding to the three olefinic protons in the terminal carbon-carbon double bond.  =CHMgBr 2 CH  1) LDA, DME  TNF, Cul DMS  163  I 3 2) CH KOH, MeOH 3)  245  246  After y,6-enone 245 was treated with LDA in DME” 4 initially at -78°C for 30 minutes, the mixture was warmed to approximately 45°C. lodomethane (5.0 eqv.) was added rapidly to the mixture. Compound 246 together with its minor epimer 247 (6:1) was isolated  247  143  in 60-70% yield.  Since these two epimers were not separable, a basic treatment with  potassium hydroxide in methanol was performed to convert the epimeric mixture into the more stable compound 246 (--97% pure by GC). When THF was used as the solvent for the methylation reaction, a very large recovery (>60%) of starting material was observed presumably because the sluggishness of the methylation reaction led to a quick equilibration of . The 5 the initially generated enolate with the methylation products through proton exchange” . The 6 use of DME as a solvent to improve the ailcylation reaction has been reported” improvement can be rationalized as follows: the bidentate chelation of DME causes the equilibration of the enolate mixture in the direction of the monomer which is more reactive and The mass spectrum of 246 revealed the  the alkylation reaction is thus  molecular ion peak at mlz 234. Its JR spectrum indicated an olefinic C-H stretching absorption , and a carbon-carbon double bond 1 at 3060 cnv’, a carbonyl stretching absorption at 1700 cm3 displayed three methyl . Its ‘H-NMR spectrum in CDC1 1 stretching absorption at 1630 cmsinglets at 60.82, 0.94, and 1.14 ppm and one methyl doublet (J=6 Hz) at 6 1.04 ppm. There were a one-proton multiplet at 62.30 ppm and a complex three-proton mukiplet at 62.40-2.70 ppm, corresponding to the allylic proton at C9 and three protons c to the carbonyl group. The olefinic signal at 6 5.01 ppm appeared as a doublet of doublets of doublets (J=17.0, 1.8, and 0.5 Hz). This signal was assigned to Rb since the resonance of Hb was expected to have a large J value (—15-20 Hz), due to coupling with Hc which was trans to Hb, and two small I values due to coupling with Ha and H9. Thus, the three J value were tentatively assigned as I (Hb,Hc)17.0 Hz, J (Ha,Hb)=1.8 Hz, and J (Hb,H9)=0.5 Hz. The olefinic proton signal at 6 5.14 ppm appeared as a doublet of doublets (J=10.2 and 1.8 Hz). This signal was assigned to Ha since the resonance of Ha should have a 3 value at 8-12 Hz, due to coupling with Hc which was cis to Ha, and a small J value of 1.8 Hz due to coupling with  Rb.  Thus, we obtained I  (Ha,Hc)10.2 Hz. The olefinic proton signal at 6 5.55 appeared as a doublet of triplets (J=17.0 and 10.2 Hz). It was assigned to Hc since the resonance of Hc was expected to appear at lower field and should show 3 values of 17.0 and 10.2 Hz. Thus, we obtained 3 (Hc,H9)=J  144  (Ha,Hc)10.2 Hz. The large coupling constant between Hc and H9 (J=10.2 Hz) indicated a near coplanarity of the C9-H9 and Cl l-Hc bonds. In the nonsteroid-like conformer 246a, the vinyl side chain is drawn as shown in order to portray this situation and to indicate minimal interactions with neighboring groups, as revealed from molecular models. Ha  246  248  The epimer 247, generated as a minor product together with 246 during the methylation of 245, was not characterized by spectroscopy since its separation from 246 was very difficult. However, a partially enriched sample (50% by GC) obtained from the methylation in THF was converted into 246 (97% pure by GC) by treatment of the mixture with a dilute KOH-methanol solution. The existence of 247 was then indirectly confirmed.  247  To confirm the stereoselectivity of the conjugate addition reaction and the regioselectivity of the methylation reaction, a detailed ‘H-NMR spectrum analysis of 246 was conducted (Figure 27). The ‘H-NMR spectrum in deuteriated benzene 118 afforded a clear resolution of the four proton signals at  2.20-2.70 ppm previously observed in the spectrum taken in  145  deuteriated chloroform. Decoupling by irradiation at 6 5.75 ppm (Hc signal) caused a oneproton triplet (J=1O.2 Hz) at 6 2.42 ppm to collapse into a doublet (J=1O.2 Hz) in addition to the collapsing of the Ha signal at 6 4.94 ppm and Hb at 6 4.82 ppm into two broad singlets. Therefore, the triplet signal at 6 2.42 ppm was clearly due to the allylic proton (H9) which apparently coupled only with Hc and one neighboring axial proton (J=10.2 Hz). Thus, the methylation of 245 must have taken place at C8 rather than at C6. The methylation at C6 would have produced 248 which should have a more complex signal (doublet of doublets or triplet) for the allylic proton (H9) if irradiation at the Hc signal had occurred. The only methyl 3 spectrum appeared at 6 1.15 (J=5.l Hz) was assigned to the doublet signal in the CDC1 methyl group at C8. A one-proton mutiplet, consisting of six lines of equal spacing (5.1 Hz) D spectrum and at 62.28 6 and an intensity ratio 1:3:4:4:3:1, appeared at 6 1.90 ppm in the C ppm CDC1 3 respectively. The splitting pattern of this signal was indeed a doublet of quartets with J=10.2 Hz for the doublet coupling and J=5.1 Hz for the quartet coupling. Thus, this signal was assigned to the ct methine proton at C8.  Hb Hc  246a 246  Hc 5 H  249a 249  146  iJj  a)  b)  ij_I  c) z  ss  Figure 27 Decoupling Experiments of 246 a) the ‘H-NMR off-resonance spectrum in CDC1 . 3 b) proton-proton homonuclear decoupling at 5.75 ppm (C D6). 6 c) the ‘H-NMR off-resonance spectrum in C6D6.  147  In fact, the structure 249a (a conformer of compound 249 which had reverse configurations at C8 and C9 with regard to 246) could also account for the result of decoupling experiments, especially the large coupling constant between Hc and Hi (J (H,H1)=10.2 Hz) due to the diaxial orientation of these two protons. Therefore, to eliminate further doubt about the stereochemistry of 246, NOE difference experiments were carried out. Unfortunately, this effort did not prove to be productive. The crowdedness of signals in the aliphatic proton region in both CDC1 3 and C D spectra caused the interpretation of NOE 6 difference experiments very difficult. However, important evidence was later obtained from compound 251, the epimer of 246 (see section 3.2.3.). To obtain 246 more effectively, we had tried to carry out an “one-pot reaction” by quenching the enolate generated in the conjugate addition with iodomethane. The one-pot operation would eliminate intermediate isolation and could be an efficient way to ensure the desired regioselective methylation” . However, the one pot operation proved to be very 9 sluggish and produced mostly by-products in addition to compound 246 (10%) and 245 (10%).  Attempts to improve the reaction by changing solvents (Et 0, THF, DME), 2  temperature, and using HMPA as additive failed. Since the conjugate addition worked well to produce 245 in good yield, the methylation step must be responsible for the sluggishness of this one-pot operation. The slowness of the methlyation reaction under the experimental conditions could lead to the proton exchange between the initially formed enolate from the conjugate addition reaction and the methylation product 246. The enolate of 246 thus generated would be then further methylated.  0-methylation of enolates might be also  responsible for some by-reactions. 3.2.3.  Conversion of cis-fused y,-enone 246 to trans-fused y,5l 2 The stereochemical conversion of the A/B cis fusion in compound 246 to the desired  A/B trans fusion in 251 was realized through a two-step sequence: the introduction of a  148  double bond to give 250 and a stereoselective reduction of 250 to produce the trans-fused compound 251.  Hb 1)LDA,TF]F  Li, NH 3  2) PhSeC1 3) 11202, Pyr.  0 2 E  246  250  251  Slow addition of 246 dissolved in THF solution to a lithium diisopropylamide-THF solution at -78°C under nitrogen was followed by a rapid injection of a phenylselenenyl chloride-THF mixture. After stirring 1 hour at room temperature, THF was evaporated in vacuo. Methylene chloride, pyridine, and hydrogen peroxide were added and the resulting mixture was stirred overnight. Dienone 250 was then isolated in 65% yield based on a 25% recovery of starting material. The mass spectrum of 250 indicated a molecular ion peak at m/z 232. The UV spectrum showed maximal absorptions at 242 nm (log E=3.96) and 383 nm (log c=3.46) corresponding to it-it’ and n1t* transition absorptions of the enone moiety. The IR , a 1 spectrum displayed an olefinic carbon-hydrogen stretching absorption at 3060 cm, and a carbon-carbon double bond 1 conjugated carbonyl stretching absorption at 1660 cm stretching absorption at 1630 cm . The 1 1 H-NMR spectrum revealed a methyl doublet (J=5.1 Hz) at 6 1.07 ppm and three methyl singlets at 6 1.17, 1.22, 1.25 ppm. A triplet (J=10.2 Hz) at 6 2.12 ppm and a multiplet at 6 2.40 ppm, consisting of six equally-spaced (5.1 Hz) lines which had an intensity ratio 1:3:4:4:3:1, were assigned to the allylic proton at C9 and the methine proton at C8 (dq, J=10.2 and 5.1 Hz). Four olefinic protons at 6 5.05 (dd, J=17.0 and 1.8 Hz), 5.18 (dd, J=10.2 and 1.8 Hz), 5.64 (dt, J=17.0 and 10.2 Hz), and 6.00 (s) ppm, were attributed to Hb, Ha, Hc, and H6.  149  To reduce the large recovery of starting material and improve the yield of . One-phase elimination of the 20 dehydrogenation, different conditions were applied’ 0 (30%) addition to the phenylselenide prepared in THF or 2 phenylselenide oxide by direct H DME (dimethoxylethane) led to a even greater recovery of starting material (50%) and a lower yield of 250 (50% based on recovery). Another one-phase elimination of phenylselenide oxide by transferring the phenylselenide into a sodium periodate solution in methanol-H20 (1:1) mixture resulted in the formation of a new compound of unknown structure (25% based on recovery) in addition to 250 (40% based on recovery) and the recovered starting material (25%). TLC monitoring of the reaction showed that little starting material 246 was left after the introduction of phenylselenenyl chloride and the newly generated phenylselenides appeared as two UV active spots of apparenly different intensities. Attempts to separate these two spots by silica gel chromatography failed as only the starting material 246 was isolated. Apparently, the selenides were very labile. In fact, even leaving the phenylselenides in THF-H20 mixture overnight regenerated the starting material 246 as exclusive product. It was expected that the phenylselenylation of 246 would produce two diastereomers (iv) and (v), resulting from the attacks on a and The  13 faces of the enolate of 246 (Figure 28).  13 face attack of phenylselenenyl chloride on the dominant conformer (i) of the enolate has  the advantage of going through a less strained half-chair transition state (ii) but suffers the steric hindrance from the angular methyl group. The a face attack has to go through a strained skew-boat transition state (iii) and suffers the hindrance from the gem-dimethyl groups in ring . Therefore, the 21 A’  13 face attack has some advantage overall. As we know, the [2,3]  . 22 sigmatropic elimination of selenoxides goes through a syn-coplanar transition state’ Therefore, only the selenoxide derived from (iv) will undergo elimination to afford 250 while the selenoxide from (iv) will likely decompose back to the starting material 246.  150  ci  (i)  SePh  (v) (iii)  CI  Figure 28 Stereochemistry of Phenylselenenylation of 246 In another attempt, the trimethylsilyl enolether 252 was prepared by reaction of 246 with LDA and trimethylsilyl chloride in THF and subsequent treatment with different oxidizing acetatel and trityl fluoroborate (ie., triphenylcarbenium d, DDQl palladium (II) 23 c, agents, 23 ’. None of these treatments gave any new product. 231 tetrafluoroborate)’  252  124 and The stereochemistry of Birch reduction on octalones has been well studied 26 For simple octalones, the trans theories to rationalize the data have been forwarded . 1 ’ 125 fused products were frequently obtained. It is assumed that the reduction goes through a dianion intermediate (i). The protonation of (i) at the 3 position produces enolate (ii) which is then hydrolyzed to give the saturated product. Thus, the stereochemistry of the final product is 125 assumed the dianion (i) has a decided by the protonation step of dianion (i). Stork et al.  151  tetrahedral  13  carbon while Robinson 126 instead proposed that the  13  carbon is trigonal. The  importance of orbital overlap in the transition state of the transformation between (i) and (ii) was recognized by both groups.  3 Li, NH  0 2 H  (i)  (ii)  Treatment of 250 in anhydrous ether:ammonia (1:2) employing a slightly excessive lithium for one hour and quenching the resulting mixture by ammonium chloride, afforded 251 in 90% yield. The GC retention times of the epimers 246 and 251 were very different from each other. The mass spectrum of 251 showed its molecular ion peak at m/z 234 while the JR spectrum indicated the carbonyl stretching absorption at 1706 cm’. The ‘H-NMR spectrum in 3 displayed three methyl singlets at 60.88, 0.90, and 1.09 ppm, a methyl doublet (J=6.0 CDC1 Hz) at 60.93 ppm, a four-proton multiplet at 62.00-2.50 ppm, and three olefinic protons at 6 H  248  152  4.98 ppm (dd, J=16.8 and 1.6 Hz), 5.12 ppm (dd, J=l0.0 and 1.6 Hz), and 5.56 ppm (dt, J=16.8 and 10.0 Hz). As in the case of 246, these three olefinic signals were assigned to Hb at C12, Ha at C12, and Hc at Cli respectively. The large coupling constant (J=12.0 Hz) between the H9 and H8 was again observed from the H9 signal  ( 1.96 ppm, t, J=12.0 Hz) in  the ‘H-NMR spectrum (C ), which proved beyond any doubt that the diequatorial D 6 orientation of vinyl group at C9 and methyl group at C8 in 251. Therefore, the diequatorial orientation of these two groups in the structure 246a were further confirmed. The structure 249a can now be firmly excluded since its corresponding trans-fused product would have these two groups diaxially oriented. 3.2.4.  Synthesis of Diol 255 from trans-Fused y,6-Enone 251 As stated earlier, an axial secondary hydroxyl group at C7 in the diol 255 was required  for the cyclization to occur in a desired manner.  • ‘  L-Selecinde  1) BH -TFIF 3 2) H 0 2  251  253  255  To this end, ‘y,S-enone 251 was treated with L-Selectride (i.e., lithium tn-sec butylborohydride) in THF at -7 8°C. The axial alcohol 253 was isolated in nearly quantitative yield. The mass spectrum indicated the molecular ion peak at m/z 236 and a fragment ion at m/z 218 due to the loss of a water molecule. The IR spectrum showed a broad intense  hydroxyl stretching absorption near 3450 cm 1 and a carbon-carbon double bond stretching absorption at 1630 cm . The ‘H-NMR spectrum revealed a quartet (J=3.0 Hz) at ö 3.92 ppm 1  corresponding to the ct proton of the axial hydroxyl group and three one-proton multiplets at 8  153  -  4.94 (dd, J=17.2 and 2.4 Hz), 5.02 (dd, J=10.4 and 2.4 Hz), and 5.52 (dt, J=17.2 and 10.4 Hz) ppm. The fact that the a proton of the newly introduced secondary hydroxyl appeared as a quartet at 6 3.92 ppm with a small coupling constant (J=3.0 Hz) confirmed its equatorial orientation. As shown in Figure 29, an equatorial proton at C7 in compound 253 is expected to couple nearly equally with the vicinal axial protons (Hax at C8 and Hax at C6) and the equatorial proton (Heq) at C6 because of the close dihedral angles, <Heq(7)-C7-C8-Hax(8), <Heq(7)-C7-C6-Hax(6), and <Heq(7)-C7-C6-Heq(6) . Therefore, a quartet with a small 127 coupling constant is expected for an equatorial proton at C7. Instead, an axial proton at C7 in compound 254 would couple nearly equally with the axial protons at C6 and C8 but differently to the equatorial proton at C6. A doublet of triplets with Jax,eq  -  3 Hz and Ja,(ax  10Hz would  be expected for this axial proton at C7.  OH  253  254  H C H 8 eq  OH  C4T’He q OH  eq  Figure 29 Structural Analysis of Stereoselective Reduction Product 253 The stereochemical outcome of the reaction between cyclic ketones and various hydrides has been frequently reviewed . For highly hindered hydride reagents like lithium 128 a and lithium tris(trans2methylcyclopentyl)borohydridel 29 thsecbutylborohydridel b, the 29 product from the less hindered face attack is usually expected. The most remarkable feature of these hindered hydride reagents lies in their ability to deliver the hydride almost exclusively in  154  an equatorial manner, even in the absence of any other nearby differentiating groups in the cyclohexanone ring, to give an axial alcohol. Therefore, we could predict with confidence that lithium tri-sec-butylborohydride reduction of 251 would produce the axial alcohol 253. This is indeed the case.  equatorial face  253  251 axial face  The hydroboration of 253 with borane in ThF, followed by basic hydrogen peroxide workup, produced mainly the 1,5-diol 255 (70%) in addition to a minor 1,4-diol 256 (10%). Diol 255 had its mass spectrum showing the molecular ion peak at m/z 254 and two fragment ions at m/z 236 and 218 corresponding to loss of one and two water molecules from the parent molecular ion. The ‘H-NMR spectrum revealed two overlapping methyl singlets at 6 0.82 ppm, one methyl singlet at 6 0.85ppm, a methyl doublet at 6 0.98 (J =6.0 Hz), a one-proton multiplet (dt, J=7.2 and 9.6 Hz) at 6 3.50, a one-proton multiplet at 6 3.62 ppm (J=5.6 and 9.6 Hz), and a quartet (J=3.0 Hz) at 6 3.85 ppm corresponding to the c hydrogen attached to the secondary hydroxyl group at C7. The two one-proton multiplets at 6 3.50 and 3.62 ppm were due to the methylene group attached to the newly created primary hydroxyl group. Therefore, the hydroboration reaction of 253 proceeded regioselectively according to the general rule that the hydroxyl group is preferentially situated at the less substituted end of a double bond in hydroboration reaction . 130 12  12  256  255  155  The minor product 256 appeared to be a mixture of two diastereomers with a ratio of 4:1, as indicated in the ‘H-NMR spectrum. These two diastereomers were difficult to separate by column chromatography. The mass spectrum of the mixture revealed a molecular ion peak . In the ‘H 1 at m/z 254. The JR spectrum showed an intense hydroxyl absorption at 3400 cm NMR spectrum, the minor diastereomer had a broad singlet at 63.77 ppm corresponding to H7 and a quartet (J=8.0 Hz) at 6 4.12 ppm corresponding to the a proton at Cli while the major isomer had a broad singlet at 3 3.87 ppm corresponding to H7 and a quartet (J=8.0 Hz) at 6 4.21 ppm corresponding to the a proton at Cii in the ‘H-NMR spectrum. Further assignment of the stereochemistry at Cii to these two diastereomers was not possible based on the above obtained data. 3.2.5.  Cyclization of Diol 255 to ()Ambrox® (179) As shown in the Introduction, most of the synthetic sequences to natural or racemic  Ambrox® involved a cyclization of a 1,4-difunctional (at C8 and C12) intermediate to form the tetrahydrofuran ring. Cyclizations of the the 1,4-diol 199 by -naphthalenesulfonic acid in 4 and the epimeric 1 ,4-diol 232 by p ”° 99 96 or p-toluenesulfonyl chloride in pyridine toluene toluenesulfonic acid in nitromethane 106 are of more direct relevance to our designed cyclization of diol 255 (Figure 30).  235  202  Figure 30 1,4-Diols Utilized for Acid Catalyzed Cyclization to ()Ambrox® It is assumed 106 that these cyclization reactions catalyzed by acids proceed through a tertiary carbocation (i) which is formed by elimination of the tertiary hydroxyl group at C2  156  (Figure 31). The x face attack by the primary hydroxyl is kinetically preferred to the  face  attack because the latter would be subjected to steric hindrance from the angular methyl group in the transition state (iii). Thus, Ambrox® (179) is preferentially produced through a lower energy transition state (ii). However, iso-Ambrox® (189) resulting from the  face attack will  become the major product under prolonged treatment and can actually be obtained from Ambrox® (179) under the same condition . This reflects that (+)isoAmbrox® (189) with a 96 cis-fused tetrahydrofuan ring is thermodynamically more stable.  4SOH  H  (ii)  /  179  slow  (i)  -H (iii)  189 Figure 31 Mechanistic Analysis of Cationic Cyclizations of 202 and 235  Under even more dramatic condition, i.e., boiling toluene with a cation-exchange resin la, ()Ambrox® (179), initially formed from diol 202, was rapidly 3 as catalystl converted to a hydrocarbon mixture of unidentified structures (60%) and a new tetrahydrofuran  *  The Chemical Abstract registry number for KU-23 is [9049-63-2]. This resin was first recorded in Chemical 13 lb Abstract in 1966 (CA 65: 15600a). Regretfully, we have no access to the corresponding original article by Soviet chemists. Later reports on the application of KU-23 contain no specific information about its preparation and structural characterization. Since KU-2, another cation exchange resin, is a sulfonated copolymer of styrene and divinyl benzene 13 (CA 55: 27959i) and KU-21, also a cation exchange resin, is a modification of KU-2 containing additional hydroxyl and carboxyl groups 13 (CA 55: 4819g), KU-23 is probably a modification of KU-2, i.e., a modified sulfonated copolymer of styrene and divinyl benzene.  157  257 of a rearranged bicyclofarnesane skeleton (32%) in addition to a small amount of epi Ambrox® (190) and isoAmbrox® (189) (Scheme 41). This resulting dehydration product of definite chemical composition was called “ionoxide”. It was claimed that the ‘ionoxide’ had a very distinct musk-ambergris odor and a very high rating as a perfume. On the whole, the smell of “ionoxide” was determined by the tricyclic compound 258, which had a strong musk At lower temperature (90°C), the major products  odor, reminiscent of the odor of muscone.  were detected to be Ambrox® (179), (+)isoAmbrox® (189) and a mixture of unsaturated alcohols 259 as shown in Scheme 41.  KU-23  Hydrocarbons  +  +  toluene 257  202  190  189  179  258  Scheme 41 The Formulation of “lonoxide” establishedl l a by a chemical The gross structure and stereochemistry of 257 was 3 correlation with the known compound 259. The configurational reversal at C5 is especially noteworthy.  259  158  The choice of diol 255 as the substrate to be cyclized to ()Ambrox® (179) has been briefly justified in Section 3.2.1. Diol 260, the epimer of 255, has an equatorial hydroxyl group at C7 and therefore it is very likely to undergo a skeletal rearrangement (ring contraction), as indicated below, when treated with acids*.  H 2  10  6  255  OH  OH  260  It is well known that 3-hydroxy-thterpenoids, e.g., (i), undergo ring A contraction to give isopropylidene derivatives of partial formula (iii) via carbocation (ii) when treated with 133 This rearrangement is of diagnostic value, since 3c-hydroxytriterpenoids (iv), ’ 132 acids . when treated under the same conditions, yield principally products of partial structures (v) and (vi) due to a simple 1,2-elimination and a methyl migration. It is assumed that the four centers involved in the migration or elimination should adopt an anti co-planar conformation.  (i)  (iv)  (ii)  (v)  *  (iii)  (vi)  It should be noted that no precedent for the cyclization of a 1 ,5-diol into a tetrahydrofuran could be found in the literature.  159  The cyclization of 255 was effected under different conditions, as is summarized in Table 5. Treatment of 255 with p-toluenesulfonic acid (2.0 eqv.) in nitromethane at 80°C for 2 hours gave ()Ambrox® (179) (31%), (+)isoAmbrox® (189) (30%), a mixture of alcohols 261 (15%), and a mixture of hydrocarbons. All spectroscopic data of 179, 189 were consistent with those recorded in literature. The yield of 179 increased to 48% and the production of (+)isoAmbrox® (189) decreased to 15% by using toluene as the solvent. The melting point and specific rotation  of the obtained ()Ambrox® were  measured to be 74-76°C and -25.1 (c=l.00, 3 CHC1 ) . They agree well with the reported values [m.p. 77-77.5°C; [aJ=-24.7 (c=1.0, . 90 ) 3 CHC1 1 The IR, ‘H-NMR, and mass spectroscopic data are identical with those recorded in the literature. The melting point and specific rotation [z] of the obtained (+)isoAmbrox® were measured to be 57-59°C and +7.3 (c=1.00, 3 CHCI ) . They agree well with the reported values [m.p. 60-60.5°C; [a]=+7.5 (c=1 .0, . 90 ) 3 CHC1 ] The IR, ‘H-NMR, and mass spectroscopic data are identical with those recorded in the literature . 90 The mixture of alcohols 261 contained a few compounds, as shown the gas chromatogram and the complex ‘H-NMR spectrum. This mixture could not be further purified by column chromatography. It displayed a hydroxyl stretching absorption at 3450 cm 1 in the mass spectrum and a molecular ion peak at m/z 220 in the mass spectrum. Thus, this mixture was probably composed of monodehydrated compounds from diol 255. The non-polar hydrocarbon mixture was obtained from the earliest fractions from column chromatography.  It contained several compounds, as revealed from the gas  chromatogram. The JR spectrum indicated no hydroxyl stretching absorption while the mass spectrum showed a molecular ion peak at m/z 218. Thus, this mixture must be a doubly dehydrated product of diol 255. The ‘H-NMR spectrum displayed poorly resolved aliphatic proton signals at 30.60-2.65 ppm and olefinic proton signals at 3 5.00-5.60 ppm. It could not be further separated.  160  Under a more dramatic condition 2 NO 100°C, 3.0 eqv. HOTs), the cyclization 3 (CH , produced compound 257, the principal component of “ionoxide”, in 34% yield and (+)-iso Ambrox® (189) in 19% yield. The specific rotation [cL] 5 of compound 257 was +37.1 (c=1.00, CHC1 ), which is in good agreement with the reported value ([ct1 3 =+39.1, c= 6.7, 8 131 Its other spectroscopic data, including IR, ‘H-NMR, and MS spectra are ) 3 CHC1 a. reportedl l a. Similar to what was reported by Viad et consistent with those 3  a, a 3 al.l  hydrocarbon mixture was isolated in large amount (41%) from this reaction. Table 5 Cyclization of the 1,5-Diol 255 under Different Conditions Composition of dehydration products, % Conditions  179  189  HOTs (2.0 eqv.), 2 NO 3 CH 80°C, 2 hrs  31  30  HOTs (2.0 eqv.), Toluene 80°C, 2 hrs  48  HOTs (3.0 eqv.), 2 NO 3 CH 100°C, 0.5 hrs  257  --  10  19  --  34  261  hydrocarbons  15  22  15  8  --  40  No significant difference was observed when p-toluenesulfonic acid was replaced with 3-naphthalenesulfonic acid. Non-protonic, poorly ionizing solvents, i.e., nitromethane and toluene, were used for our cyclization of the 1,5-diol-255. These two solvents have been employed previously in the cyclization of 1,4-diols to ()Ambrox®  (179)96.106.131.  The yield of 1,2-elimination  (dehydration) by-products 258 (Scheme 41) was minimized by using these solvents. Presumably, mainly ion pairs rather than free carbocations are involved under these . The loss of I protons has to take place from the same side of the leaving group 34 conditions’  161  ___  Presumably, mainly ion pairs rather than free carbocations are involved under these . The loss of f3 protons has to take place from the same side of the leaving group 134 conditions (i.e., H20) which, instead of the solvent, acts as the base. Such a stereochemical requirement reduces the possibilities of 1,2-eliminations in rigid trans-fused decalone systems. If highly ionizing solvents were used, free planar carbocations would be formed and therefore the loss of j3 protons would occur from either the same or the opposite side of the leaving group. The mechanism for the formation of 257 from (+)isoAmbrox® (189)13 as well as from our 1,5-diol 255 is proposed (Figure 32). The consecutive 1,2-shifts of peripheral axial hydrogens and the angular methyl group as indicated may produce an olefin 261 which has two conformers 262 (i) and 262 (ii). Cyclization of the more stable conformer (ii) via an intramolecular anti addition produces 257, the principal component of “ionoxide”.  CH3  262 (i) OH  3 CH 257  262 (ii)  Figure 32 Mechanism for the Formation of 257  162  the triterpenoid 3-3-friedelanol (263) was transformed into 13 (18)-oleanene (264) by acid . Presumably, the carbocation (i) (Figure 33) generated from 263 undergoes six 135 catalysis stereoelectronically controlled 1,2-shifts as shown to afford the carbocation (ii). The loss of a proton results in 13 (18)-oleanene 264.  H  263  264  IL H  H  (i)  (ii)  Figure 33 The Conversion of 3-13-Friedelanol (263) into 13 (18)-Oleanene (264) In conclusion, we have succeeded in synthesizing ()Ambrox® (179) enantio selectively from the thujone-derived enone 163 in seven steps in an overall yield of 9.5%. Moreover, a novel synthesis of 257, the principal component of “ionoxide”, was discovered. The successful strategy should be applicable to the synthesis of other ambergris fragrances, which will be discussed in the next section.  163  3.3.  Future Developments The synthesis of ()Ambrox® (179) dictates the preparation of its precursor 255  (Scheme 37). During the preparation of 255 from enone 163, two steps, i.e., 246 to 250 and 250 to 251 (Section 3.2.3.) are required in order to reverse the configuration at C5. ) from t However, for the direct synthesis of 257 (i.e., the principal component of “ionoxide enone 163, it is unnecessary to have these two steps, since the configuration at C5 of compound 257 is the same as that of 163. Thus, a shorter route is perceived, as shown in Scheme 42. 0H %S  3 b) BH  257  265  246  3 CH  3 CH 3 .CH  3 CH - -  0 2 -H  3 CH  3 CH 3 CH  265  3 CH  (i)  257  Scheme 42 A Possible Shorter Route to Compound 257 cis-Fused Ketone 246, prepared in two steps from 163 (Section 3.2.3.), might be subjected to L-Selectride reduction and hydroboration to give the cis—fused diol 265. An acidcatalyzed cyclization of 265 could then provide 257. Mechanistically, a series of consecutive 1,2-shifts of peripheral axial groups in 265 would first generate the tertiary carbocation (i). The subsequent ring closure of (i) should afford the desired product 257.  164  The developed strategy (Scheme 37) to the synthesis of ()Ambrox® (179) may be further extended to the synthesis of other diastereomers possessing significant odoriferous properties, for example, the cis-fused isomer 189 and ()epiAmbrox® (187). The diol 265, if prepared as outlined in Scheme 42, would be cyclized to afford 192 under mild acid catalysis (Scheme 43). Functioning as the reactive species, the stable conformer of 265 could follow the reaction path as envisaged to yield the desired 192 stereoselectively.  192  265  3 CH  3 CH  H  265  192  Scheme 43 A Possible Synthesis of Compound 192 To synthesize ()epiAmbrox® (190), dienone 168, prepared previously from 163 (Section 2.2.9.), would be converted to 266 by a cuprous iodide-catalyzed conjugate addition and a subsequent methylation (Scheme 44). According to the argument presented in Section 3.2.2.(Scheme 39), the x face attack in the conjugate addition reaction is expected to be dominant. Birch reduction of compound 266 could generate the trans-fused ketone 267, which might undergo L-Selectride reduction and hydroboration to provide diol 268. The stereoselective cyclization of this diol by acid catalysis, following the reaction path as shown, could finally lead to the ()epiAmbrox® (190).  165  a) LDA, PhSeC1  a) Cul =CH)MgBr 2 (CH  0 2 b) H  I 3 b) LDA, CH 266  168  163  Li, NH 3  a) L-Selectride 3 b)BH  190  268  267  H  0 2 -H  c 190  268  Scheme 44 A Possible Synthesis of (-)-epi-Ambrox (190) In replacing vinylmagnesium bromide with allylmagnesium bromide in the conjugate addition step, it should be possible to obtain the 1,6-diol 269 from enone 160, by using the same strategy (Scheme 37). The acid-catalyzed cyclization of 269 would then furnish another ambergris odorant: ambraoxide (186), the homologue of (.)Ambrox® (179) (Scheme 45). 0H  H 0 2 -H  163  269 Scheme 45 A Possible Synthesis of Ambraoxide (186)  166  186  3.4.  Experimental See Section 2.3.1. for General experimental.  3.4.1.  Conjugate Addition:  c-enone 163 to cis-fused y,ö-enone 245  [4R-(4cc,4ax,8act)] 4-Ethenyl-3,4,4a,5,6,7,8,8a-octahydro-4a,8,8-trimethylnaphthalen2(1H)-one (245)  245 To a solution of enone 163 (718 mg, 3.74 mmol) in anhydrous THF (20.0 ml), cuprous iodide (112 mg, 0.59 mmol, 0.15 eqv.) and dimethyl sulfide (5.0 ml) were introduced under a nitrogen atmosphere. This mixture was cooled to 0°C and 0.66 M vinylmagnesium bromide in TFIF solution (8.2 ml, 5.41 mmol, 1.4 eqv.) was added in a dropwise manner over a period of 1 hour. After the mixture was warmed to room temperature and stirred for another 1 hour, saturated sodium chloride (20 ml) was introduced to quench the excess vinylmagnesium bromide. The organic layer was separated; the aqueous layer was extracted with diethyl ether (10 ml). The combined organic solution was dried over magnesium sulfate and concentrated in vacuo to give a crude oil which was then chromatographed to provide enone 245 in 70% yield (576 mg). The physical properties of 245 are as follows: [(x]=+22.2 (c=1.00, CHC1 ). 3 IR Vm (film): 3065(C-H stretch olefinic), 1710(C=O stretch), 1635(C=C stretch). ,  H-NMR (400 MHz, CDC1 1 ) 6: 0.80-1.70 { 14H, including 0.90 (3H, s), 0.95 (3H, s) and 3 1.09 (3H, s)}, 1.90 (1H, m), 2.25-3.10 (5H, m), 4.95-5.15 (2H, m), 5.70 (111, m). 167  MS m/z: 220 (M, 15.6%), 205 (4.0%), 123 (67.8%), 43 (100.0%). High resolution mass 14 C 2 H 5 220.1828; found:. 220.1831. measurement: calculated for 0: 3.4.2.  Methylation by EDA and lodomethane:  cis-fused y,&enone 245 to cis  fused y,-enone 246 [3S-(3cc,4I3,4a3,8af3)J 4-Ethenyl-3,4,4a,5,6,7,8,8a-octahydro-3,4a,8,8tetramethylnaphthalen-2( 1H)-one (246)  246 A LDA / n-pentane solution (0.68 M, 2.6 ml, 1.77 mmol) was concentrated to remove n-pentane. The resulting white viscous mixture was cooled to -40°C, to which anhydrous dimethoxyethane (1.0 ml) was then added under nitrogen. Enone 245 (350 mg, 1.59 mmol) in dimethoxyethane (3.5 ml) was introduced to the LDA solution in a dropwise manner over a period of 1 hour. This enolate solution was warmed rapidly to 50°C and freshly distilled iodomethane (0.40 ml, 6.42 mmol) was added rapidly. The resulting turbid yellowish mixture was stirred at 50°C for 30 minutes and quenched with a solution of potassium hydroxide (100 mg) in methanol (10 ml). After stirring for 30 minutes, the reaction mixture was concentrated in vacuo to give the crude product which was chromatographed with ethyl acetate:hexanes (1:8, v/v) to afford 246 (207 mg, 65% based on recovery of 245) and the starting enone 245 (53 mg, 15%). The physical properties of 246 are as follows: [aJ=+29.3 (c=1.00, CHC1 ). 3 JR Vm. (film): 3060 (C-H stretching, olefinic), 1700 (C=O stretching), 1630 (C=C 168  stretching). ‘H-NMR (400 MHz, CDC1 ) 8: 0.70-1.60 (18H, m, including 0.82 (3H, s), 0.94 (311, s) 3 and 1.04 (3H, d, J=6.4 Hz), 1.14 (3H, s)}, 1.90 (1H, m), 2.30 (1H, m), 2.45-2.70 (3H, m), 4.95-5.20 (2H, m), 5.55(111, m). MS m/z: 234 (M, 1.7%), 219 (0.5%), 167 (35.7%), 149 (100.0%). 3.4.3.  Dehydrogenation by PhSeCI/H : cis-fused y,6-enone 246 to dienone 0 2  250 [3S-(3a,4f.,4a3,8a)] 4-Ethenyl-4,4a,5 ,6,7 ,8-hexahydro-3 ,4a,8,8-tetramethylnaphthalen2(3H)-one (250)  250 A LDA solution in n-pentane (0.50 M, 2.30 ml, 1.15 mmol) was concentrated in vacuo to remove n-pentane. The viscous mixture was cooled to -40°C and THF (1.0 ml) was then added under nitrogen. The solution of 246 (250 mg, 1.07 mmol) in THF (3.0 ml) was added in a dropwise manner with stirring over a 45 minute period. The resulting mixture was warmed to room temperature and phenylselenyl chloride (212 mg, 1.10 mmol) was introduced. Stirring at room temperature was continued for 1.5 hours before addition of pyridine (0.50 ml), methylene chloride (5.0 ml), and hydrogen peroxide (0.50 ml of 30% H 0 in 3.0 ml of 2 water). The two-phase mixture was stirred at room temperature for 5 hours and separated. The aqueous layer was extracted with methylene chloride (5.0 ml). The combined organic layers were dried over magnesium sulfate and concentrated in vacuo to yield a crude product. Purification by column chromatography with ethyl acetate:hexanes (1:8, v/v) afforded dienone 169  250 (134 mg, 62% based on recovery of starting material) and the starting material 246 (31 mg). The physical properties of 250 are as follows: [(x]=-4.2 (c=1.0O, CHC13). UV (MeOH, c=40.0 m g/l)  242 nm (log c=3.96).  JR Vm. (film): 3060 (olefinic C-H stretching), 1665 (C=O stretching). H-NMR (400 MHz, CDC13) 6: 0.80-1.90 { 18H, m, including 1.07 (3H, d, J=5.1 Hz), 1.17 1 (3H, s), 1.22 (3H, s) and 1.25 (3H, s}, 2.12 (1H, t, J=10.2 Hz), 2.40 (1H, m), 5.00-5.25 (2H, m), 5.69 (1H, m), 6.00 (1H, s). MS m/z: 232 (M, 21.2 %), 217 (17.2%), 189 (2.7%), 178 (7.7%), 164 (100.0%), 149 (48.7%), 121 (14.9%). High resolution mass measurement calculated for 0: 14 C 2 H 6 232.1827; found: .232.1829. 3.4.4.  Birch Reduction:  dienone 250 to trans-fused ‘y,6-enone 251  [3S-(3a,413,4a13,8ax)] 4-Ethenyl-3,4,4a,5,6,7,8,8a-octahydro-3,4a,8,8tetramethylnaphthalen-2(1H)-one (251)  251 To a solution of dienone 250 (300 mg, 1.29 mmol) in anhydrous THF (2.0 ml), anhydrous ammonia (4 ml) was distilled from sodium under a nitrogen atmosphere. Small pieces of lithium were added slowly over a 30 minute period until a persistent dark blue color remained. After stirring for 1 hour at -33°C, ammonium chloride powder was introduced to  170  quench excess lithium. Evaporation of ammonia and THF gave a yellowish oil which was chromatographed to afford the trans-fused decalone 251 (268 mg, 90%). The physical properties of 251 are as follows: . CHC1 ) [a}=-8.38 (c=2.40, 3 JR Vm. (film): 3060 (C-H stretching, olefinic), 1706 (C=O stretching). ) ö: 0.75-1.80 { 19H, m, including 0.88 (3H, s), 0.90 (3H, s), 3 H-NMR (400 MHz, CDC1 1 0.97 (3H, d 3=6.0 Hz), and 1.09 (3H, s)}, 2.00-2.50 (4H, m), 4.98 (1H, dd, J=1.6 and 16.8 Hz), 5.12 (1H, dd, J=1.6 and 10.0 Hz), 5.56 (1H, dt, 3=10.0 and 16.8 Hz) MS m/z: 234 (M, 33.8%), 219 (8.8%), 203 (0.3%), 137 (10.8%), 123 (100.0%), 109 16 236.2140; found: C 2 H (19.2%). High resolution mass measurement calculated for 0: 236.2097. 3.4.5.  Reduction by L-Selectride:  trans-fused 7,8-enone 251 to alcohol 253  [2R-(2a,3a,4f3,4a,8aa)j 4-Ethenyl-decahydro-3 ,4a,8 ,8-tetramethylnaphthalen-2-ol (250)  253 The trans-fused ketone 251 (250 mg, 1.07 mmol) in anhydrous THF (2.0 ml) was added in a dropwise manner to L-Selectride (0.72 M, 3.0 ml, THF) at -78°C for 30 minutes. The solution was stirred for 1.5 hour, warmed to 0°C, and stirred for an additional 1 hour. Aqueous sodium hydroxide solution (3.0 ml, 3 M) and aqueous 30% hydrogen peroxide (3.0 ml) were then introduced. The resulting mixture was stirred 30 minutes, saturated with potassium carbonate, and separated. The aqueous layer was further extracted with diethyl ether (2x10 ml). The organic solutions were combined and concentrated in vacuo. Purification by 171  column chromatography with ethyl acetate:hexanes (2:8, v/v) gave alcohol 253 (240 mg, 95%). The physical properties of 253 are as follows: ). 3 [cL]=-40.9 (c=1.00, CHC1 JR Vm. (film): 3450 (0-H stretching), 3060 (C-H stretching, olefinic), 1630 (C=C stretching). ‘H-NMR (400 MHz, CDC1 ) & 0.70-1.85 {24H, m, including 3  },  3.92 (1H, q, J=3.0 Hz),  4.94 (1H, dd, 3=2.4 and 17.2 Hz), 5.02 (1H, dd, J=2.4 and 10.4 Hz), 5.52 (1H, td, J=10.4 and 17.2 Hz). MS m/z: 236 (M, 2.4%), 218 (2.1%), 203 (4.5%), 123 (100.0%). High resolution mass measurement: calculated for 0: 18 C 2 H 6 236.1240; found: 236.2136. Elemental Analysis: calculated for 0: 18 C 2 H 6 C 81.29, H 11.09; found: C 81.22, H 11.11  3.4.6. [1 S-( 1  Hydroboration:  alcohol 253 to 1,5-diol 255  ,4,4a,8acc)] Decahydro-3-hydroxyl- 2,5,5,8a--tetramethylnaphthalene- 1-  ethanol (255)  çbE:H 255 To a cooled solution (0°C) of alcohol 253 (300 mg, 1.27 mmol) in THF (2.0 ml) was added borane in THF solution (7.0 ml, 0.56 M) in a dropwise manner under nitrogen over a period of 30 minutes. The solution was warmed to room temperature and then stirred for 1.5 hours. After water (1.0 ml), aqueous sodium hydroxide (3.0 ml, 3M), and aqueous hydrogen  172  peroxide (3.0 ml, 30%) were introduced, the resulting mixture was stirred overnight, saturated with sodium chloride, and separated. The aqueous layer was further extracted with diethyl ether (10 ml). The organic solutions were combined, dried over magnesium sulfate, and concetrated in vacuo to give the crude product. The crude product was chromatographed with ethyl acetate: methanol:hexanes (1:1:2, v/v/v) to afford diol 255 in 71% yield (229 mg). The physical properties of 255 are as follows: m.p.: 128-130°C. [c]=-16.9 (c=1.00, CHC1 ). 3 JR Vmax. (film): 3400 (0-H stretching). ‘H-NMR (400 MHz, CDC1 ) & 0.70-1.80 {27H, m, including 0.82 (6H, two overlapped 3 singlets), 0.86 (3H, s), and 0.98 (3H, d, 3=6.0 Hz)), 3.50 (1H, dt, J=7.2 and 9.6 Hz), 3.62 (1H, dt, J=5.6 and 9.6), 3.85 (1H, q, 3=3.0 Hz). MS m/z: 236 (M-H20, 6.6%), 221 (6.5%), 191 (16.9%), 177 (6.0%), 167 (18.9%), 138 (62.2%), 123 (100.0%). High resolution mass measurement: calculated for 11 16 C 02: 30 254.2236; found: .254.2241. Elemental Analysis: calculated for 0 10 C 3 H : 2 6 C 75.53, H 11.89; found: C 75.75, H 12.00. 3.4.7.  Cyclization:  1,5-Diol 255 to 179, 189, and 257  [3aR-(3aa,5aI3,9acc9b)] Dodecahydro-3a,6,6,9a-tetramethyl- 1H-naphtho[2, 1 -blfuran (179) [3aS-(3acc,5 acx,9a,9bct)] Dodecahydro-3a,6,6,9a-tetramethyl- 1H-naphtho[2, 1-b] furan (189) *)j Dodecahydro-3a,4,7,7-tetramethyl-2H-naphtho[8a, 1 -b]furan (257) [3aR-(3arx,4cç6ax, lOaS  179  189  173  257  Procedure #1: Diol 255 (20 mg, 0.079 mmol) in anhydrous toluene (2.0 ml) was treated with p toluenesulfonic acid (27 mg, 0.16 mmol, 2.0 eqv.) under a nitrogen atmosphere. This solution was then heated at 80°C for 2 hours. The resulting mixture was transferred by diethyl ether (10 ml) to a separatory funnel, washed with saturated sodium carbonate solution, dried over magnesium sulfate, and concentrated in vacuo. Column chromatography with hexanes, ethyl acetate :hexanes (1:50, v/v), ethyl acetate:hexanes (1:20, v/v), and ethyl acetate:hexanes (1:8, v/v) consecutively gave a mixture of hydrocarbons (1.4 mg, 8%), (+)isoAmbrox® (189) (1.9 mg, 10%), ()Ambrox® (179) (8.9 mg, 48%), and a mixture of alcohols 258 (2.8 mg, 15%). Procedure #2: Diol 255 (20 mg, 0.079 mmol) in nitromethane (2.0 ml) was treated with p toluenesulfonic acid (40 mg, 0.23 mmol, 3.0 eqv.) under a nitrogen atmosphere. The solution was then heated at 100°C for 30 minutes. After a workup similar to that in the procedure #1, the crude product was chromatographed with hexanes and ethyl acetate:hexanes (1:50, v/v) to give a mixture of hydrocarbons (7.0 mg, 40%), the ionoxide principal 257 (6.4 mg, 34%), and (+)isoAmbrox® (189) (3.5 mg, 19%). The physical properties of 179 are as follows: m.p.=74-76°C. [z]=-25.1 (c=1.00, CHC1 ). 3 JR (CHC1 ) vmax.: 1455, 1380, 1000, 975 cm 3 . 1 ‘H-NMR (400 MHz, CDC1 ) 6: 0.83 (3H, s), 0.84 (311, s), 0.88 (3H, s), 1.09 (3H, s), 3.83 3 (1H, q, J=8.0 Hz), 3.92 (1H, m). MS m/z: 236 (M, 3.4%), 221 (100.0%), 205 (6.8%), 177 (3.8%), 137 (40.2%), 97 (37.5%), 84 (23.8%), 81(20.3%), 69 (20.4%), 59 (22.8%), 55 (18.8%), 43 (20.6%). High resolution mass measurement: calculated for 0: 18 C 2 H 6 236.2140; found: 236.2139.  174  The physical properties of 189 are as follows: m.p.=57-59°C. ). 3 [cc]=+7.4 (c=1.00, CHC1 . 1 JR (CHC1 ) vmax.: 1450, 1375, 1070, 1035 cm3 ) 6: 0.86 (3H, s), 0.89 (3H, s), 0.90 (3H, s), 1.06 (3H, s), 3.70 3 H-N4R (400 MHz, CDC1 1 (q, J=8.0 Hz), 3.80 (1H, dt, J=3 and 8.0 Hz). MS m/z: 236 (M, 0.0%), 221 (M-CH , 100.0%), 177 (1.7%), 137 (21.3%), 109 (7.9%), 3 97 (33.0%), 84 (3 1.6%), 69 (21.5%), 55 (34.0%), 47 (7.5%), 43 (49.8%). High resolution 15 (M-CH C 2 H ): 221.1905; found: 221.1906. 3 mass measurement: calculated for 0 ), 237 (M-i-Hj. 5 Chemical ionization MS using methane as carrier gas: 251 (M÷CH The physical properties of 257 are as follows: [a]=÷37.1 (c=1.00, CHC1 ). 3 . 1 JR (film) vmax.: 1455, 1370, 1040, 1025 cm ‘H-NMR (400 MHz, CDC1 ) 6: 0.81 (3H, s), 0.83 (3H, d, 3=6.6 Hz), 0.87 (3H, s), 0.94 3 (3H, s), 3.70 (1H, dt, J=2 and 8.0 Hz), 3.81 (1H, q, J=9 Hz). ‘H-NMR (400 MHz, CC14) 6: 0.80 (3H, s), 0.82 (3H, d, 3=6.6 Hz), 0.87 (3H, s), 0.92 (3H, s), 3.67 (111, dt, 3=2 and 8.0 Hz), 3.76 (1H, q, 3=8.0 Hz). MS m/z: 236 (M, 7.7%), 221 (7.0%), 194 (13.7%), 193 (100.0%). High resolution mass measurement: calculated for 0: 18 C 2 H 6 236.2140; found: 236.2146.  175  Chapter 4  Exploratory Studies of Different Strategies to Develop Thujone as a Chiral Building Block  The synthetic strategy described in the previous two chapters focussed primarily on the cleavage of the isopropyl side chain of thujone as an important operation, to afford eventually target molecules like (-)-polygodial (2) and ()Ambrox® (179). A direct result of such a strategy is that the synthesized target molecule always incorporates seven of the ten carbon atoms in the starting thujone molecule. A question is then raised: is it possible to develop other strategies which incorporate different number of carbon atoms into these target molecules? If developed, each of such new strategies would be characteristic of its own carbon incorporation, providing novel entries into various natural products. A closely related issue is that, in principle, for a given strategy which incorporates a certain number of carbons, there can be various methods which integrate the same number of carbon atoms into target molecules, depending on how the starting structure is incorporated or how and where some parts of the starting structure are removed during the incorporation. With these general considerations in mind, we decided to integrate the isopropyl side chain of thujone into target molecules as much as possible, rather than to cleave the isopropyl side chain completely as in the previous studies. Thus, strategies of different degrees of carbon incorporation can be developed. This chapter summarizes some exploratory studies in this direction. For the purpose of presentation, strategies incorporating seven, nine, and ten carbons are called  2, and £1Q strategies respectively.  4.1. Studies on “Homothujone” and Its Derivatives: a new .7. strategy As shown in Section 2.2.4. and 2.2.7., previously synthesized thujone-derived cyclopropylcarbinols of the general skeleton (i) (Scheme 46) usually undergo acid-promoted ring cleavage reactions through exo-type 1 and exo-type 2 cleavage pathways, rather than the endo-type cleavage pathway to provide the desired cyclohexane ring (Figure 11). It was  176  hypothesized that the preferred exo-type 1 cleavage was due to the exposed nature of the methylene in the cyclopropyl ring, towards the incipient nucleophiles (Figure 12). If the bicylo[3.1.O]hexane system (i) could be expanded to the bicyclo[4.1.O]heptane system (ii) in a regioselective manner shown in Scheme 46, the homoallylic halide (iii) with a desired cyclohexane ring would become the “logical product” due to the preferred exo-type 1 cleavage. With a versatile homoallylic halide group, (iii) may be readily elaborated into  (-)-  polygodial (2) and ()Ambrox® (179).  ring expansion -- -  *  (i)  (ii)  *  (iii)  Scheme 46 The Potential of a Regioselective Ring Expansion Reaction This ring expansion reaction had not been considered in our earlier studies nor in other laboratories in which other avenues of thujone chemistry had been developed. Therefore, its evaluation would also make a fundamental contribution to thujone chemistry. Scheme 47 shows the overall plan in which this strategy may afford alternative syntheses of (-)-polygodial (2) and ()Ambrox® (179). Ring expansion of thujone may be expected to generate a “homothujone” (272) which could be then converted to enone 274. Birch reduction followed by enolate trapping should produce a trans-fused ketone 281 which would be reduced to hydrocarbon 284. Ozonation should then form both alcohol 294 and ketone 295. Exo-type 1 cleavage of alcohol 294 would result in homoallylic chloride 296 and the latter could be ozonized to a 3-chloro-ketone 298 while the ketone 295 could be converted to f3-bromoketone 297 using m-CPBA and NBS as described in Section 2.2.8..Versatile functional groups in both 297 and 298 would allow them to be readily converted to either (-)-polygodial (2) or ()Ambrox® (179).  177  272  274  +  281  284  x -  295  297: X=Br 298: X=Cl  296  Scheme 47 “Homothujone” Strategy for Syntheses of Various Natural Products The apparent advantage of the homothujone strategy is that the trans A/B ring fusion  could be possibly realized by Birch reduction directly, rather than through a tedious stereochemical correction sequence from the A/B cis-fused systems obtained earlier. As a new  2 strategy, the homothujone strategy incorporates seven of the original ten carbon atoms present in thujone into potential target molecules in a novel way.  4.1.1.  Regioselective Ring Expansion of Thujone The desired regioselective ring expansion of thujone was accomplished by treating  thujone with ethyl diazoacetate and boron trifluoride etherate under nitrogen at room . The 13-ketoester 27O, which existed mainly in its enol form, was isolated in 37 temperature’ $ Because thujone in use was a mixture of a-thujone and f-thujone (10:1), the product 270 was a mixture of two diastereomers in a similar ratio, as revealed by GC. No attempt was made to separate these two diastereomers. The 1 H-NMR spectral data presented here represent the characterization of the major c  178  70% yield. The mass spectrum of 270 showed a molecular ion at mlz 238. The UV spectrum indicated an absorption band at 258 nm (log e=3.980) while the JR spectrum displayed a broad hydroxyl absorption at 3370 cm’, an intense conjugated ester carbonyl stretching absorption at 1 1655 cm, and a weak carbon-carbon double bond stretching absorption at 1615 cnr’. The 1  CHCO N E 2 t -Et 3 BF 0 2 3  presence of an intense UV absorption and the lack of any non-conjugated carbonyl absorption demonstrated domination of the enol form in compound 270138. The ‘H-NMR spectrum revealed three separate one-proton signals at high field 6 0.30 (dd, J=4.4 and 8.8 Hz), 0.39 (t, J=4.4 Hz), and 0.68 (dd, J=4.4 and 8.8 Hz), corresponding to the three protons in the cyclopropane ring. Two methyl doublets (J=5.6 and 4.4 Hz), corresponding to the two methyl groups at the isopropyl side chain, overlapped at 6 0.98 ppm. There were a one-proton multiplet at 8 1.03 ppm corresponding to the methine proton at C8, a methyl doublet (J=7.2 Hz) at 6 1.24 ppm corresponding to the methyl at C2, and a methyl triplet (J=6.8 Hz) at 6 1.31 ppm corresponding to the methyl of the ethyl ester group. A two-proton signal of AB type at 6 2.25-2.57 ppm (J=16 Hz) was assigned to the methylene at C5 while a quartet (J=7.2 Hz) at 6 2.64 ppm was due to the methine at C2. The 3 coupling constant between the methine protons at Cl and C2 was zero!. A two-proton multiplet at 64.21 ppm corresponded to the methylene in the ethyl ester group and a very low field singlet signal at 6 12.24 ppm was due to the hydroxyl proton in the enol form of 270. The spectroscopic data presented above could not differentiate the enol form of 270  diastereomer while other spectroscopic data are the gross properties of the diastereomeric mixture. This situation remains the same for homothujone 272.  179  from that of 271, which would be the product of carbon insertion from the more substituted side of carbonyl function in thuj one. Crucial evidence was obtained, however, from the next step, i.e., the decarboxylation of the 3-keto ester.  271 T  r  e  a  t  m  e  n  produced 272# in 95% yield’ . The mass spectrum of 272 showed its molecular ion at m/z 39 166 while the IR spectrum indicated a carbonyl absorption at 1700 cm. It is expected that 1 272 would have its three a protons (to the carbonyl group) in the region between 62.00 ppm and 6 3.00 ppm in the ‘H-NMR spectrum whereas 273 available from 271 would reveal four protons in this region. To our surprise, 272 contained four protons in this region: two at 6 2.10 ppm (m), one at 6 2.35 ppm (m), and one at 6 2.47 ppm (dt, J=3.0 and 8.0 Hz). A series of decoupling experiments were performed to clarify the situation (Figure 34). frradiation of the one-proton signal at high field (60.72 ppm), which was assigned to one of the three cyclopropane protons, caused the multiplet at 6 2.47 ppm to collapse into a quartet (J=8.0 Hz) in addition to the simplification of the complex two-proton signal at high field (6 0.50 ppm), which was assigned to the two remaining cyclopropane protons. frradiation of the signal at 60.50 ppm resulted in only the collapse of the signal at 60.72 ppm. Thus, the signal at 60.72 ppm was clearly due to the Cl proton while the signal at 62.47 ppm was assigned to the methine proton at C2. Irradiation of the methyl doublet resonance (J=8.0 Hz) at 6 1.22  # The product 272 was a mixture of a and diastereomers (10:1), as indicated by GC. The 1 H-NMR data I described here represent the characterization of the major a diastereomer while other spectroscopic data are the gross properties of the diastereomeric mixture. See also the footnote at p. 178.  180  t  e)  C)  b)  a) 3.0  2.0  1.0  0.0  6 (ppm) Figure 34 Decoupling Expenments of 272 a) off-resonance spectrum. b) proton-proton homonuclear decoupling at 0.50 ppm. c) proton-proton homonuclear decoupling at 0.72 ppm. d) proton-proton homonuclear decoupling at 1.22 ppm. e) proton-proton homonuclear decoupling at 2.47 ppm. 181  ppm led to the collapse of the C2 proton signal into a doublet (3=3.0 Hz) and irradiation at 6 2.47 ppm transformed the methyl doublet signal at 6 1.22 ppm into a singlet and the Cl proton signal at 6 0.72 ppm into a doublet of doublets (J=4.8 and 8.8 Hz), further confirming the assignment. The fact that the C2 proton was coupled only to the Cl proton and and the methyl protons at 6 1.22 ppm suggested the correct structural assignment to 272 and thus 270. The proton at C2 of 273 would have coupled to the C3 protons in addition to the methyl protons and the Cl proton. Seemingly, one of the methylene protons at C5 of 272 had an unusually high chemical shift between 62.00 to 2.70 ppm. 7 8  DMSO  270  0, NaC1 2 H 272:  aJI  273:  a/f3  DMSO 271  0, NaC1 2 H  It is noteworthy that the coupling between the Cl proton and the C2 proton in compound 272 (J=3.0 Hz) was rather different from that in 270 (3=0 Hz). This can be explained when one considers the possible conformations of these two compounds (Figure 35). The enol form of compound 270ct* can have two boat-like conformers 270a and 270b. Conformer 261b is less stable because of the repulsion between the axial methyl at C2 and the axial hydrogen at C5. Inspection of models reveals a dihedral angel <H1-C1-C2-H2 close to 900  in conformer 261a which possesses an equatorial methyl group at C2. Therefore, the  coupling constant between Hi and H2 is expected to be small. Among the two half-chair *  As indicated in the footnotes at p. 178 and p. 180, the 1 H-NIvIR data thus far described for 270 and 272 represent their a diastereomers only.  182  conformers of 272cx*, 272b with an axial methyl group is considered more stable because it is devoid of the C2-methyl bond and the Cl-Hi bond eclipsing interaction present in 272a and the flat nature of the plane involving C2-C1-C6-C5 also greatly reduces the repulsion between the axial methyl at C2 and the axial proton at C4 in 272b. The dihedral angle <Hi-C i-C2-H2 is approximately 300 and therefore a larger coupling constant between Hi and H2 is expected. 2  5  L 2  270b  270a  2  2  4  272b  272a  Figure 35 Conformational Analysis of 27Oc and 272x; The insertion reaction of a ketone by ethyl diazoacetate usually take place from the less  substituted or less bulky side. The formation of the reactive conformer shown in Figure 36 is . 140 presumably faster than other possible conformers due to minimal gauche steric repulsions Assuming that the subsequent migration is a faster process than the internal rotation about the carbon-carbon bond, the insertion from the less substituted side becomes the dominant product.  0  3 BF  CHCO N E t 2 2 R(’ R  3 OBF + (M) 2 N2•yCO E t (L) (L) 2 R(’R (S or M) H (S)  COCHR 1 R C E 2 O t  Figure 36 Explanation for Regioselectivity of the Carbon Insertion Reaction  183  4.1.2.  Stereoselective Robinson Annulation of Homothujone (272) The Robinson annulation of homothujone (272) was carried out by refluxing the  starting material with potassium hydroxide and the salt of 1-diethylaniino-3-pentanone and one equivalent iodomethane in ethanol. Enone 274 as shown was isolated in 70% yield.  EVK, KOH  EtOH, reflux 272  274  The mass spectrum of 274 indicated the molecular ion at mlz 232 corresponding to the formula H2 16 C 0 4 . The UV spectrum showed an intense absorption band at 250 nm (log c=4. 133) corresponding to the it to  it  transition in the enone chromophore. The JR spectrum  displayed a conjugated carbonyl absorption at 1660 cm. The ‘H-NMR spectrum was fairly 1 well resolved. Three low field one-proton signals at 5 0.30 (dd, J=4.8 and 9.6 Hz), 0.50 (dd, 3=4.8 and 9.6 Hz), and 0.66 ppm  (t,  3=4.8 Hz) were assigned to the cyclopropane protons.  Two methyl doublets (both J=7.2 Hz) at 5 0.90 and 0.93 ppm were due to the two methyl groups of the isopropyl side chain while a neighboring multiplet at 5 0.95 was assigned to the methine proton of the side chain. Two methyl singlets at 5 1.16 and 1.74 ppm corresponded to the angular methyl (at ClO) and the vinylic methyl (at C4) protons respectively. Three multiplets at 3 1.58 (2H), 1.82 (1H), and 1.93 ppm (1H) were assigned to the methylene protons at Cl and C7 while two other lower field multiplets at 5 2.12 ppm (1H, dt, J=5.2 and 14.0 Hz) and at 32.35-2.70 ppm (3H) were due to the four methylene protons at C2 and C6. The structure of 274 was further confirmed by a series of NMR experiments. The structure 275, which might possibly be formed by the EVK Robinson annulation from the less substituted side of the carbonyl group, is inconsistent with the fact that only two methyl doublets were observed in the spectrum of the isolated product 274. However, the structure  184  276, which was possibly generated from the  face attack of the more substituted side, could  accommodate all the spectroscopic data so far obtained. More evidence was needed to differentiate 274 and 276.  276  274  Inspection of molecular models reveals that the angular methyl groups at ClO have different spatial relationships with the three cyclopropane protons in the diastereomers 274 and 276. In the case of 274, the angular methyl is relatively close to the cyclopropane methylene ) but distant 11 proton directed into the concave face of the bicyclo[4. 1 .O]heptane moiety (i.e., Hi from the cyclopropane methine proton (i.e., H9) and the other methylene proton which is directed away from the concave face of the bicyclo[4.1.O]heptane moiety (i.e., H). For 11 276, the angular methyl is relatively close to H9 but distant from both methylene protons Hj and HQUI. Thus, if the the angular methyl is irradiated, a positive NOE enhancement for H will indicate the presence of 274 while a positive enhancement for 119 will suggest the existence of 276.  Hin Hout  H  274  276  185  Fortunately, the ‘H-NMR spectrum was fairly well resolved. The methyl singlet signal at 6 1.22 ppm, previously assigned to the angular methyl, was well separated from nearby signals and the three cyclopropane proton signals at high field were also well separated from each other. From a large number of recorded spectra of substituted cyclopropanes, it is generally observed that, in any designated cyclopropane, the magnitude of the vicinal coupling constant for cis protons (protons on the same side of a cyclyopropane plane, e.g., H9 and H) is always larger than that for trans protons (e.g., H9 and Hj) . Since each of the three 141 coupling constants in the AMX system, composed by the three cyclopropane protons of 274 or 276, had to be either 4.8 Hz or 9.6 Hz  ,  the coupling constant between H9 and Huit 1)  (H9,H)] and the coupling constant between H9 and Hin [J (H9,Hin)] should have values 9.6 Hz and 4.8 Hz respectively, in order to satisfy the relationship: J (H9,Hjjj> J (H9,Hm). 3 (Ht,Hin) had to be 4.8 Hz to produce a thplet of J=4.8 Hz observed in the spectrum and this triplet signal was due to Hin. Otherwise, if J (Hi,Hjn) were 9.6 Hz, a triplet of 3=9.6 Hz would have been observed and this triplet would have been due to H. Thus, the consideration of magnitude for coupling constants enabled us to assign the triplet (J=4.8 Hz) at 6 0.66 ppm to Hj but the two doublet of doublets signals at 6 0.30 and 0.50 ppm cannot be assigned further. H-’ heteronuclear correlation spectrum (2D-HETCOR, Figure 37) 1 C A two dimensional 3 further confirmed the assignment. The proton (doublet of doublets) at 60.50 ppm correlated intensely with a tertiary carbon at 6 33.00 ppm but weakly with a secondary carbon at 6 12.60 ppm. Both the proton (triplet) at 60.66 pm and the proton (doublet of doublets) at 60.30 ppm correlated intensely with the secondary carbon at 6 12.60 ppm but not with the tertiary carbon at 6 33.00 ppm. This suggested that the proton (doublet of doublets) at 60.50 ppm was due to H9 and the quartet proton at 60.30 ppm was due to HQUI. The determination of substitution of the above mentioned carbons was facilitated by an APT (Attached Proton Test) experiment (Figure 38). The carbon at 6 12.60 ppm was assigned as secondary since it was very intense in the off-resonance spectrum and did not invert its 186  F2  (PPM)  —D  4.0  15  3.02826242.22.0181.61.4121.00.8060.40.2  Figure 37 2D-HETCOR spectrum of 274 a) H (0.30 ppm) ---C (12.60 ppm). b) H (0.66 ppm) ---C (12.60 ppm). c) H (0.50 ppm) ---C (33.00 ppm).  187  00 00  220  220  —.  —.  200  —  (APT)  200  (BB)  160  160  140  140  .-..  120  —-  120  100  t.  100  I  80  80  60  .-w-  60  I  Figure 38 The ‘ C Broad Band Decoupling (BB) and APT Spectra of 274 3  180  180  III I  I  I  40  I 40  I  I  Lii  1  f  i  Fill  r  I  .1  I  111111  2(  E’l  hurT  20 PP.4  ii  L  I  C  II  0  phase in the APT spectrum. Among the six carbons of inverse phase (which can be either primary or tertiary carbons) in the APT spectrum, four of them were sorted out as primary carbons since they had low chemical shifts in the ‘ C spectrum (6: 10.35, 18.55, and 19.20 3 ppm). As shown from the HETCOR spectrum, these four carbons also correlated well with H-NMR spectrum. Thus, the other two carbons at 6 33.00 and four methyl singlets in the 1 36.65 ppm must be tertiary carbons. NOE experiments’ 43 were then carried out on compound 274 (Figure 39). Irradiation at 111 at 60.66 ppm the angular methyl signal at 6 1.22 ppm resulted in a 4.0% enhancement of H but no enhancement of either 119 or H. Therefore, the stereochemistry of 274 was finally confirmed. Irradiation of Hm at 6 0.66 ppm did not give a clear enhancement of the angular methyl signal but did cause a 10% enhancement of Hui and a negative enhancement of 119. 4.1.3.  Attempted Generation of the trans-Fused Hydrocarbon 284  Having obtained the desired intermediate 274 in good overall yield from thujone, it was appropriate to evaluate some chemistry with this compound. Birch reduction of 274, followed , gave the gem-dimethylated ketone 1 by iodomethane addition to trap the generated enolate 277 in low yield (15%). Attempts to improve this reaction by addition of proton donors (i.e., water and t-butanol) during the Birch reduction step, quenching of excess lithium with isoprene, and removal of ammonia prior to iodomethane addition proved to be infertile. The by-products were relatively non-polar and difficult to separate from each other. Simple reduction of 274 and polymethylation of 274 and 277 might be responsible for their generation.  1) Li, NH , THF 3 2) CH I 3  274  277  189  c)  b)  a)  3.0  1.0  2.0  Figure 39 The NOE Experiments of 274 a) off-resonance spectrum. b) irradiation at 0.62 ppm. c) irradiation at 1.22 ppm.  190  0.0  8 (ppm)  The mass spectrum of 277 revealed the molecular ion peak at m/z 248 while the JR H-NMR . The relatively complex 1 1 spectrum indicated a carbonyl absorption at 1703 cm spectrum could be analyzed in support of structure 277. Two multiplets appearing at 6 0.08 ppm (1H) and at 6 0.35 (2H), corresponded to the three cyclopropane protons. A triplet (J=2.4 Hz) at 6 0.85 ppm consisted of six protons, probably due to the overlapping of two doublets of the methyl groups of the isopropyl group. Three methyl singlets at 8 1.03, 1.21, and 1.22 ppm and two multiplets at 8 2.30 ppm (2H) and 2.62 ppm (1H) were also observed. The A/B ring junction was assumed to be trans, in accord with the expected stereochemistry of Birch reduction ( see Chapter 3, Section 3.1.3.) although insufficient evidence is available to be certain. The poorly resolved ‘H-NMR spectrum discouraged attempts to use NOE experiments to elucidate the nature of A/B ring junction at this stage. An alternative route to 277 was perceived. Reduction of 274 by lithium and ammonia produced a mixture of two compounds 278 and 279 in a ratio of 4:1. Because these two compounds were not convertible by reaction with potassium hydroxide in methanol, they were assumed to be two diastereomers of opposite A/B ring fusion with the major isomer 278 presumed to possess the trans ring fusion as in compound 277. These two compounds were difficult to separate by column chromatography. The mass spectrum of the mixture (278 and 279) indicated a molecular ion at m/z 236.while the JR spectrum showed a carbonyl absorption . Catalytic hydrogenation of 274 with 5% Pd-C at room temperature in ethanol 1 at 1700 cm generated 278 and 279 in a ratio of 6:1. Thus, the ‘H-NMR spectrum of this mixture could reveal some characteristic signals of 278. Two multiplets appeared at 8 0.09 ppm (1H,  t,  J=5.2 Hz) and 0.40 ppm (2H, m), corresponding to the three cyclopropane protons. Three methyl doublets at 6 0.85 ppm (J=6.0 Hz), 0.88 ppm (J=6.0 Hz), and 0.93 ppm (J=8.0 Hz) corresponded to the two methyl groups of the isopropyl side chain and the methyl group at C4. The angular methyl appeared at 8 1.35 ppm as a singlet.  191  Li, NH 3  I 3 CH HOtBu, KOtBu  or H , Pd, EtOH 2 H  274  278/279  Refluxing the reduction mixture containing 278 and 279 with iodomethane and potassium t-butoxide in anhydrous t-butanol under nitrogen resulted in a mixture which did not contain 277, as indicated by GC. No further attempt was made to elucidate this mixture. The reaction carried out at room temperature gave only recovered starting material.  I’”ø  H  280 Treatment of the mixture of 278 and 279 (6:1) with sodium methoxide and iodomethane produced 280 in 54% yield. This compound was characterized by a molecular ion peak at m/z 248 in its mass spectrum, a carbon-carbon double bond stretching absorption at 1680 cm 1 in its JR spectrum, and two methyl singlets at ö 1.57 and 3.50 ppm, corresponding to the vinylic methyl and the methoxyl methyl in the ‘H-NMR spectrum. The A/B ring  junction of 280 was assumed to be trans, the same as that of 278 and 277. At this point, an alternative sequence to the A/B trans-fused hydrocarbon 273, which was based on the rearrangement of the original sequence as shown in Scheme 47, was considered. The order of steps involved in this new sequence (Scheme 48) would be methylation, decarbonylation, hydrogenation; whereas the original sequence would have steps  in a different order: hydrogenation, methylation, decarbonylation.  192  21*12  NaOMe, DMSO 274  KOH,DEG  I 3 CH 282  283  -THF 3 1)BH  2) HOAc, heating 284 Scheme 48 An Alternative Sequence to Hydrocarbon 284  Enone 274 was first methylated to 282 in 60% yield using sodium methoxide in DMS 0145. The mass spectrum of 282 showed its molecular ion at m/z 246 while the JR . The ‘H-NMR spectrum revealed three 1 spectrum displayed a carbonyl absorption at 1700 cmmethyl singlets at 6 1.01, 1.18, and 1.20 ppm, corresponding to the angular methyl group and the two geminal methyl groups, and a one-proton triplet (J=4.0 Hz) at 6 5.42 ppm corresponding to the olefinic proton. Decarbonylation of 282 utilizing the Woif-Kishner-Huang Minion conditions proceeded smoothly to give 283 in 67% yield. The mass spectrum of 283 revealed the molecular ion peak at m/z 232 while the JR spectrum indicated the absence of carbonyl H-NMR spectrum showed three one-proton multiplets at 6 0.14, 0.36, and absorption. The 1 0.50 ppm, corresponding to the three cyclopropane protons, and a one-proton triplet (J=4.0 Hz) at 65.30 ppm corresponding to the olefinic proton. Treatment of carbon-carbon double bonds by borane to form organoboranes which are then decomposed with acetic acid to produce saturated C-C bonds is a useful indirect method of . However, such a treatment of 283 generated a 146 carbon-carbon double bond reduction complex mixture which was composed of several compounds as detected by GC and the ‘H NMR spectrum.  193  11  1) BH -THF 3  13  2) H 0 2  71  51  OH  283  285  286  To understand the complication, an oxidative treatment of the intermediate organoboranes by basic hydrogen peroxide was carried out. Diol 285 and alcohol 286 were isolated in 39% and 29% yield respectively. The mass spectrum of 285 had its molecular ion peak at m/z 268 while the JR spectrum indicated an intense hydroxyl absorption near 3500 cm 1  The 1 H-NMR spectrum showed three methyl singlets at 6 0.96, 0.98, 1.00 ppm and two  methyl doublets at 6 1.01 ppm (J=7.0 Hz) and 1.15 ppm (J=7.0 Hz). Two multiplets appearing at 6 3.72 ppm (211) and 4.04 ppm (111) corresponded to the the methylene and methine protons attached to Cli and C7. An X-ray structure of 285 (crystalized from methylene chloride) is shown in Figure 40 (see also Appendix 2). The cis A/B ring fusion is clearly indicated.  Alcohol 286 had its molecular ion peak at m/z 250 in the mass spectrum and an hydroxyl absorption at 3450 cm 1 in the JR spectrum. The ‘H-NMR spectrum indicated three multiplets at 6 0.14, 0.45, and 0.64 ppm, corresponding to the three cyclopropane protons. Two methyl doublets appeared at 60.85 ppm (J=6.0 Hz) and 0.90 ppm (J=6.0 Hz) while three methyl singlets were observed at 60.98, 1.10, and 1.16 ppm. A doublet of doublets at 6 2.14 ppm (111, J=5.2 and 7.4 Hz) was probably due to one of the methylene protons at C7 which was neighboring to the cyclopropane ring). A one-proton complex multiplet at 6 3.87 ppm was assigned to the proton at C6. By analogy to structure 285 and the following mechanistic explanation, the ring fusion of 286 was presumed to be cis and the hydroxyl should have orientation.  194  285  Figure 40 Single Crystal X-ray Structure of 285 (ORTEP Drawing) The oxidation of 286 by Jones reagent produced ketone 287 in 80% yield. Compound 287 was characterized by its molecular ion peak at m/z 248 in the mass spectrum, a t in the IR spectrum, and a two-proton signal non-conjugated carbonyl absorption at 1700 cm  195  of strongly coupled AB type at 62.29 ppm corresponding to the two methylene protons at C7 in the ‘H-NMR spectrum. 11  Jones reagent OH  286  287  The formation of 285 and 286 can be rationalized as follows (Figure 41). The hydroboration of the carbon-carbon double bond in 283 probably takes place from the  f face  to generate the cis-fused organoborane (i) which undergoes a direct oxidation to yield alcohol 286. Isomerization of (i) may afford another organoborane (ii) which rearranges to the third  alcohol 286 [011 2 NBR  283  (i)  (ii) 2 BR  [0]  2 BBR  diol 285 2 BR H  (iii)  (iv)  Figure 41 Novel Cyclopropane Ring Cleavage in the Hyciroboration of 283  196  organoborane (iii) via a novel cyclopropane ring cleavage. A stereoselective hydoboration of (iii) provides the fourth organoborane intermediate (iv) and a two-fold oxidation of the latter results in the isolated diol 285.  Cleavage of vinyl cyclopropanes has been previously  147 but usually a drastic condition is required. Notably, the cleavage reaction of 283 observed took place at room temperature. After all, the complication of this indirect reduction was due to the unexpected conversions occurring during the hydroboration step.  1 2 CH  1) Li, NH 3  274  2) TMSC1 0 2 Et  288  H  289/290  KOH, MeOH  0 277 Scheme 49 An Alternative Route to Ketone 277 In a last attempt to improve the yield of 277, the sequence in Scheme 49 was considered and put into experimental test. Trimethylsilyl enol ether 288 was prepared by . 148 trapping the enolate generated in the Birch reduction of 274 with trimethylsilyl chloride The crude product thus obtained was then converted into a mixture of trimethylsilyl cyclopropyl ethers (289/290) using the Simmons-Smith . 149 reaction 5 ” 0 This mixture 1 probably contained two diastereomers 289 and 290 which had the newly created cyclopropyl ring  and  f  oriented since ThC indicated more than two spots. The crude product from  Simmons-Smith reaction was hydrolyzed in warm potassium hydroxide-methanol 151 A major compound isolated was identified as 277 by comparing its MS, IR, ’ 150 solution .  197  NMR data with 277 previously obtained in the Stork enolate trapping reaction. The nature of  A/B ring junction in 288, 289, and 299 was uncertain although tentatively assumed to be trans as for 277. The overall yield of 277 from 274 was 45%.  291 The reduction of 277 by the Woif-Kishner-Huang Minion method gave hydrocarbon 291 in 70% yield. The mass spectrum of 291 indicated the molecular ion at ni/z 234 while the IR spectrum showed the absence of carbonyl absorption. The ‘H-NMR spectrum revealed two muitiplets at 6 0.07 ppm (1H) and 0.40 ppm (2H), corresponding to the three cyclopropane protons. A triplet (6H, J=2.5 Hz) and three methyl singlets appeared at 6 0.84 ppm, 1.10, 1.20, and 1.22 pm respectively. 4.1.4.  Ozonation of 291 Ozonation of hydrocarbon 291 in ethyl acetate, as before, resulted in the isolation of  ketone 292 (35%) and alcohol 293 (5%) instead of the expected compounds 294 and 295.  7  çOH  cHO  292  293  +  291  294  198  295  Ketone 292 in its mass spectrum revealed a molecular ion at m/z 248 and its JR spectrum displayed a conjugated carbonyl absorption at 1665 cm. Its ‘H-NMR spectrum 1 indicated two methyl doublets at 6 0.84 ppm (J=6.6 Hz) and 0.94 ppm (J=6.6 Hz), corresponding to the two methyl groups of the isopropyl side chain, and three methyl singlets at 6 0.78, 1.11, and 1.30 ppm. A one-proton septet (J=6.6 Hz) corresponding to the methine proton in the isopropyl side chain appeared at 6 1.84 ppm. A multiplet containing two protons at 6 2.00-2.30 ppm corresponded to the two methylene protons at C6. An attempt to prepare suitable crystals for X-ray diffraction analysis of the solid 292 was not successful. The mass spectrum of alcohol 281 showed the molecular ion peak at m/z 250 and its JR spectrum 1 which corresponded to hydroxyl stretching revealed a broad absorption band near 3405 cmabsorption. Its ‘H-NMR spectrum indicated two multiplets at 60.13 ppm (1H) and 0.45 ppm (2H), corresponding to the three cyclopropane protons, and two methyl doublets at 60.89 and 0.98 ppm, corresponding to the two methyl groups of the isopropyl side chain. Three methyl singlets appeared at 6 0.86, 1.10, 1.20 while a multiplet (1H) at 64.16 ppm corresponded to the proton at C8, the hydroxyl bearing carbon. The orientation of the hydroxyl group was not determined. It is surprising that the previously noted selective ozonation of thuj one derivatives could not be applied to the homothujone derivative 291. The reasons for this reactivity change are unknown. Generally, a cyclohexane ring is more puckering than a cyclopentane. This may allow one of carbon-hydrogen bonds at C7 properly oriented towards the cyclopropane ring in 291. This orientation may then facilitate the participation of the cyclopropyl group in the ozone insertion into this particular carbon-hydrogen bond*. The unusual reactivity of the carbon-hydrogen bonds of the methylene neighboring to the cyclopropane ring in homothujone derivatives was assumed to be general, which discouraged further pursuit of the homothujone strategy at that time. Since the oxidation of cyclopropylmethylene to cyclopropylketone has  *  For discussion on the mechanism of ozone insertion into carbon-hydrogen bonds, see Section 2.2.2. 199  been observed by other oxidizing reagents’ , the oxidation of homothujone derivatives may 52 find application in a way complementary to the ozonation of thujone derivatives in the future. 4.2.  Studies on Utilizing the C2-C3 Bond Cleavage Products:  a j strategy  Cyclopropylcarbinol of the general structure (ii) was considered as potentially useful intermediate in the thujone chemistry (Scheme 50). They might be available from thujone derived cyclopropylcarbinol (i) by cleavage of the C2-C3 bond.  Because the relief of the  cyclopentane ring constraint, this seco-(C2-C3) cyclopropylcarbinol could possibly undergo acid-promoted ring opening via the cleavage of C1-C5 bond (endo type cleavage), rather than the cleavages of C1-C6 and C5-C-6 bonds (exo-type 1 and exo-type 2) usually observed for (i).  HX  (i)  (ii)  (iii)  Scheme 50 Ring Cleavage of seco-(C2-C3) Cyclopropylcarbinols A more attractive sequence leading to syntheses of (-)-polygodial (7) and its analogues involved the utilization of a seco-(C2-C3) intermediate (Scheme 51). Trione 107, which could not find a ready application like its congener 106*, might undergo aldol condensation to afford enone 299 which would subsequently be methylated and selectively reduced to 300. An oxidative cleavage of the C2-C3 bond should produce trione 301 which could be then recyclized to 303. The seco-(C2-C3) compound 301 was considered equivalent to 302. Conjugate addition of geminally diactivated cyclopropane 3o353 would then generate  *  Both 106 and 107 were derived from ozonation (Section 2.2.2., Scheme 17) Compound 106 was used in the studies on synthesis of steroid analogues from thujone, which is not described in this thesis.  200  compound 304 and the latter could then be reduced to the trans-fused decalone 305 by Birch reduction. Application of 305 in the syntheses of (-)-polygodial (7) and its analogues can be readily perceived. 10  10  10  10  10  2 0:  seco  6  2  CHO  corro  CCHO -...  -  9  302  303  CHO •  304  305  (-)-polygodial (7)  Scheme 51 A Novel Sequence to (-)-Polygodial (7); This novel sequence belongs to  strategy in which nine of the ten carbons in thujone  is incorporated into the target molecule (-)-polygodial (7). The cleavage of the C2-C3 bond and the following cyclization are interesting from the structural point of view and they are termed seco (from seco -thujone) and corro (from corre lation or connection of two seemingly distant carbons C3 and C9#) operations. These two operations reveal an inherent topology or connectivity of the thujone carbon skeleton. The direct creation of a trans A/B ring fusion and the use of a electrophilic cyclopropane are quite appealing from the chemical point of view.  The numbering for the structural segment derived from thujone is kept the same as that for thujone to facilitate analysis.  201  Experimentally, the cyclization of 107 turned out to be a difficult reaction to perform. Treatment of 107 with pyrrolidine in refluxing benzene produced a rather complex mixture. Thus, no further attempt was made to carry out the above sequence. Fortunately, Dr. Dominik Guggisberg obtained ketoacid 308 as a by-product in the preparation of diol 307 from olefin 306*. A similar sequence to that in Scheme 51 was perceived starting with 308 (Scheme 52).  4 KMnO  N 2 CH  0:tBuOH 2 H  0 2 Et  306  307  308  03, EtOAc 0°C  311  310  309  Scheme 52 The Utilization of a seco-(C2-C3) Intermediate 308 Thus, methylation of ketocarboxylic acid 308 with diazomethane in diethyl ether gave ketoester 309 in 95% yield’ . had The mass spectrum of compound 309 showed the 54 molecular ion at m/z 280 corresponding to the molecular formula 0 18 C 2 H 3 7 while Its JR spectrum revealed absorptions at 1710 cm 1 and 1685 cm 1 corresponding to the stretching absorptions of the conjugated ester carbonyl and the carbonyl in the cyclohexane ring. Two  methyl doublets at  0.67 ppm (J=7.2 Hz) and 0.92 ppm (J=7.2 Hz) corresponding to the two  methyl groups of the isopropyl side chain and three methyl singlets at ppm were observed in the ‘H-NMR spectrum. A methyl singlet at  3.65 corresponded to the  methyl of the methoxycarbonyl group.  *  I am grateful to Dr. Dominik Guggisberg for providing a sample of compound 308.  202  1.11, 1.13, and 1.20  The ozonation of 309 in ethyl acetate at 0°C generated 310 in 45% yield. The tertiary alcohol 312 (Scheme 53) was not isolated. Probably it was rapidly dehydrated to a terminal olefin at 0°C; the latter was then ozonized to 310 (see Section 2.2.2.). Diketoester 310 had its mass spectrum showing the molecular ion at m/z 280 ppm and the IR spectrum showing carbonyl stretching absorptions at 1710 cm 1 and 1690 cm-’. The ‘H-NMR spectrum indicated a methyl singlet at 6 1.08 ppm, an overlap of two methyl singlets as a broad singlet at 6 1.10 ppm, a methyl singlet at 6 2.28 ppm corresponding to the methyl of the acetyl group, and a methyl singlet at 63.76 ppm corresponding to the methyl of the methoxycarbonyl group. LDA treatment of 310 in THF resulted in only recovery of the starting material. Instead, pyrrolidine treatment in refluxing benzene resulted in a messy mixture which was not analyzed further. At this stage, we realized that a mistake had been made. Compound 301 is not equivalent to 302 at all (Scheme 51) but actually identical to 313 (Figure 42). Therefore, the cyclization of 301 and 310 would not produce 303 and 311 as drawn in Scheme 51 and 52 but highly stained compounds 314 and 315 (Figure 42) which have trans-fused bicyclo[4. 1.0] heptane moieties . 16  313  312  301  314 Figure 42 A Structural Misperception for 301  203  302  315  With this consideration in mind, a new sequence was devised as shown in Scheme 53. A selective conjugate addition of the geminally diactivated cyclopropane 308 from the less substituted carbon 155 would generate 316 which could be cyclized to the highly functionalized octalone 317. The further elaboration of 317 to (-)-polygodial (7) can be readily envisaged. This new sequence has the advantages stated for that in Scheme 51 and is no doubt a worthwhile undertaking in the future.  Me 2 .CO  0 316  310  317  (-)-polygodial (7)  Scheme 53 The Final “seco/corro”  Strategy to the Synthesis of (-)-Polygodial (7);  4.3.  A Formal Synthesis of (+)-3-Cyperone:  a ClO strategy  Thujonol (94)* as prepared earlier (Section 2.2.2.) was treated with concentrated hydrobromic acid in methylene chloride at room temperature for two hours. Enone 318 and phenol 319 were isolated in 85% and 10% yield respectively.  *  “Thujonol” was a mixture of x and f3 diastereomers in a ratio of 10:1. See the footnote at p. 28. 204  6  HBr  6  +  C1 CH 2  318  94  319  . The UV CHC1 ) The specific rotation [a] of 318 was measured to be +42 (c=0.29, 3 spectrum displayed a broad absorption peak maximal at  c=20 mg/i), corresponding to the  it  OH, 3 234.3 nm (loge=3.95, CH  to iu transition of the enone chromophore. The mass  spectrum indicated the molecular ion peaks at m/z 232 and 230 (intensity ratio  =  1:1),  0 The 5 9 Br. C 1 H 7 0 and O 5 1 Br C 1 H 8 corresponding to two isotopic parent ions of formulas O 1 and a weak JR spectrum showed an intense conjugated carbonyl absorption at 1670 cm . The ‘H-NMR spectrum was well 1 carbon-carbon double bond absorption at 1630 cmresolved. An apparent doublet at 6 1.11 ppm (J=6.8 Hz), corresponding to the two methyl groups of the isopropyl side chain. A methyl doublet at 6 1.34 ppm (J=7.1 Hz) corresponded to the methyl at C4. A septet at 6 2.43 ppm (1H), a multiplet at 6 2.55 ppm (1H), and another multiplet at 62.92 ppm (2H) were assigned to the methine proton of the isopropyl side chain, the methine proton at C4, and the two methylene protons at C6. A doublet of triplets signal at 6 4.19 ppm (J=4.4 and 10.2 Hz) was due to the methine proton at C5 (i.e., the bromine bearing carbon) while a broad singlet at 6 5.97 ppm was clearly due to the olefinic proton at C2.  c diastereomer of 94  3 diastereomer of 94  Since the x diastereomer of thujonol (94) was the predominant component (—90%) of  205  the starting material*, it was reasonable to assume that the major ring cleavage product 318 (85%) had the configuration at C4 as shown. The configuration at C5 was assigned as shown by analogy with the observed nucleophilic attack on C5 from the back side of the cleaving Cl C5 bond during the acid promoted ring cleavage of an analogous cyclopropylcarbinol.  94 These two configurational assignments were supported by the ‘H-NMR spectral data. As indicated above, the methine proton at C5 appeared as doublet of triplets at J=4.4 for doublet and J=1O.2 Hz for triplet.  4.19 ppm with  This can be well understood from the  conformational analysis of 318. As shown below, compound 318 have two half-chair-like conformer 318a and 318b. Conformer 318a is the predominant one since it has both the methyl group at C4 and the bromo group at C5 equatorially oriented. The gross ‘H-NMR spectrum can be approximately represented by conformer 318a. The axial proton at C5 of 318a should couple with two axial protons at C4 and C6 nearly equally (J8-13 Hz) and with the equatorial proton at C6 relatively weakly (J3-5 Hz). We may predict with confidence that the methine proton at CS will appear as a triplet splitting into three doublets with J values in  318a  318b 3 )2 R=-CH(CH  *  See footnotes at page 28 and 204.  206  ranges just indicated. This is indeed the case. In fact, except the enantiomer of 318, no other diastereomer of 318 can explain the particular splitting pattern of the signal at 84.19 ppm. Phenol 319 is known as carvacrol . The mass spectrum of 319 indicated the 164 C 1 H 0 The JR spectrum 4 molecular ion peak at m/z 150, consistent with the formula 0. showed an intense hydroxyl absorption at 3300 cm. The ‘H-NMR spectrum was 1 exceedingly simple. An apparent doublet at 6 1.22 ppm (6H, J=7.2 Hz) was assigned to the two methyl groups of the isopropyl side chain. A methyl singlet at 6 2.20 was due to the methyl group at C4 while a one-proton septet was assigned to the methine proton of the isopropyl side chain. A broad one-proton singlet at 63.96 ppm corresponded to the hydroxyl proton. Three olefinic proton signals at 6 6.65 (1H, d, J=1.8 Hz), 6.73 (1H, dd, J=7.5 and 1.8 Hz), and 7.04 (1H, d, J=7.5 Hz) corresponded to protons at C2, C6 and CS respectively. When thujonol (94) was treated with concentrated hydrochloric acid at room temperature, chioro-enone 320 and carvacrol (319) were isolated in 45% and 40% yield respectively.  HC1  6 +  CI CH 2  94  320  319  The mass spectrum of compound 320 revealed molecular ion peaks at m/z 188 and 186 (intensity ratio 1 :3), corresponding to two isotopic parent ions of formulas C1 C 1 H 3 0 0 and 5 7 C 1 H 3 0 C1. 0 The IR spectrum indicated an intense conjugated carbonyl stretching absorption 5 at 1675 cm-’ and a weak carbon-carbon double bond absorption at 1630 cm-’. This chioro enone was rather unstable and the obtained ‘H-NMR spectrum always contained extra signals due to the presence of carvacrol (319). However, a “difference spectrum” between the “contaminated spectrum” and the spectrum of 319 revealed all signals of 320 clearly. In fact, this “difference spectrum” of 320 was very similar to the spectrum of 318. An apparent  207  doublet at 6 1.09 ppm (6H, J=7.2 Hz) was due to the two methyl groups of the isopropyl side chain while a methyl doublet at 6 1.30 ppm was assigned to the methyl group at C4. A septet at 6 2.43 ppm (1H), a multiplet at 62.54 ppm (1H), and another multiplet at 6 2.78 ppm (2H) were further assigned to the methine proton of the isopropyl group, the methine proton at C4, and the two methylene protons at C6. A doublet of triplet signal at 64.06 ppm (1H, J=4.4 and 9.8 Hz) corresponded to the methine proton at C5 (i.e., the chlorine bearing carbon) while a broad singlet at 6 5.95 ppm (1H) was due to the olefinic proton at C2. Based on the analysis of the splitting pattern of the CS methine proton signal in a way similar to that for 318, the stereochemistry of 320 was determined to be as shown. Br  Br  HBr 2 2  bOH  94  (i)  (ii): 318/321  Br  i 4 Ni  -HBr  319  (iii)  Figure 43 The Endo-type Cleavage Mechanism for the Formation of 318 and 319 The mechanism in Figure 43 was proposed to explain the formation of 318 and 319. The HBr promoted ring opening through the C1-C5 bond cleavage (i.e., the endo-type cleavage) produces (i) which undergoes a double bond migration to give the more stable isomer (ii), i.e., 318 and its C4 epimer 321. The acid catalyzed enolization of (ii) generates dienol (iii) and the latter may lose a HBr molecule either through a 1,2-elimination to yield 319 directly or through a 1,4-elimination to afford dienone (iv) first and then 319 later.  208  It is noted above that chioroenone 320, although structurally similar to bromoenone  318, was much less stable. It decomposed into carvacrol (319) in deuteriated chloroform at room temperature. This instability may account for the fact that more carvacrol (319) was isolated from the HC1 promoted ring cleavage of thujonol (94). Both 318a and 320a, the major half-chair-like conformers of 318 and 320, are suitable for acid catalyzed enolization since they all have axial protons at C4.  X=Br, 318a X=Cl, 320a  X=Br, 318b X=Cl, 320b 2 ) 3 R=-CH(CH  However, the formation of enol from 318a is likely to be more difficult because of the greater steric interaction (allylic strain) between the more bulky equatorial bromine at C5 and the methyl at C4 during the enolization. The relative ease of enolization for 320a allows the following dehydrobromination to take place (Figure 43) and carvacrol (319) is thus more readily converted from 320.  [  X=Br, Cl 2 ) 3 R=-CH(CH  I  The endo-type cleavage pathway during the acid promoted ring opening of another thujone-derived cyclopropylcarbinol was again observed (Scheme 54). Hydroxyenone 122,  209  previously obtained from Robinson annulation of thujonol (94) with EVK in 35% yield (Section 2.2.3.), was treated with hydrobromic acid in methylene chloride. Bromo-dienone 322 was isolated in 91% yield. Compound 322 has been previously reduced to (+)-a. 3 cyperone (8) by tributyltin hydride in an earlier synthesis of (+)-3-cyperone from thujonel Thus, a new sequence to (+)-3-cyperone was completed in four steps using ozonation of thujone, Robinson annulation of thujonol (94), ring opening of 122, and radical-mediated reduction of 322. The new sequence incorporates all the ten carbons of thujone into the target molecule (+)--cyperone (8). This synthesis provides an example ofj.Q strategy.  EVK  rN  KOH, EtOH  HBr CI CH 2  TrH  fH  122 Br 3 HSnBu MEN  322  (+)--cyperone (8)  Scheme 54 A Formal Synthesis of (+)--Cyperone (8)  The specific rotation [aj 5 of 322 was measured to be +420 (c=1.00, CHC1 ), which 3 13 The UV spectrum ) 3 CHC1 is in good agreement with the reported value +430 (c=1 .0, a. displayed an intense absorption peak maximal at 2. 293 nm (log =4.40, MeOH). Thus, a conjugation among the carbonyl group, the C4-C5 double bond, and the C6-C7 double bond was suggested. The mass spectrum indicated molecular ion peaks at m/z 298 and 296 (intensity ratio  1:1), corresponding to two isotopic parent ions of formulas O Br 1 C 2 H 8 5 1  and O Br. 1 C 2 H 7 5 The JR spectrum revealed a intense conjugated carbonyl absorption at 1 9  210  1660 cm1 and a weak carbon-carbon double bond absorption at 1620 cm . The 1 1 H-NMR spectrum showed an apparent doublet at 6 1.12 ppm (6H, J=6.0 Hz), corresponding to the two methyl groups of the isopropyl side chain, and two methyl singlets at 6 1.17 and 1.86 ppm, corresponding to the angular methyl group at ClO and the vinylic methyl group at C4. A oneproton doublet of doublets signal at 6 4.14 (J=6.0 and 10.0) was assigned to the methine proton at the bromine bearing carbon (C9) while a one-proton singlet at 66.31 ppm was clearly due to the olefinic proton at C6. The particular splitting pattern of the signal at 6 4.14 ppm allowed the assignment of the configuration at C9. The molecular model of 322 revealed a rather rigid conformation in order to accommodate the full conjugation of the three double bonds. The axial proton at C9 would couple with the axial proton at C8 (usually, J=8-13 Hz) and the equatorial proton at C8 (usually, J=3-5 Hz) quite differently and a doublet of doublets with 3 values in regions just indicated should be expected for the C9 proton signal. This prediction is quite close to what was observed. Compound 323, the epimer of 322 with the bromine at C9 x oriented, would show the C9 proton signal as a triplet with 3=3-5 Hz or a doublet of doublets with both 3=3-5 Hz. The slightly larger 3 value for the coupling between the C9 axial proton and the C8 equatorial proton in 322 (3=6.0 Hz), than normally observed for the coupling between an axial proton and an equatorial proton, is probably due to some geometric distortion. H  323  322  The mechanism shown in Figure 44 was proposed to rationalize the formation of 322 from 122, which is similar to what was proposed for the formation of 318 from thujonol (94) (Figure 43). The nucleophilic ring opening generates an unstable intermediate (i) which then  211  rearranges to the more stable dienone 322 with a fully conjugated dienone system. The ring opening reaction proceeds through the cleavage of C1-C5 bond* , i.e., the endo-type cleavage. The bromide anion attacks on the C5 from the backside of the cleaving C1-C5 bond, leaving the  13 orientation of the bromo group in (i) and thus 322. Br  I  Br  Br  H —  (i)  122  322  Figure 44 The Ring Opening reaction of 122 via the Endo-type Cleavage Pathway It is speculated that the interaction of the double bond exo to the bicyclo[3. 1 .O]hexane and the C1-C5 bond in hydroxyenone 122 and thujonol (94) leads to the weakening of C1-C5 and eventually its facile cleavage under acidic conditions, A possible new way of incorporating all the ten carbons into target molecules is shown in Scheme 55. Rearrangement of vinylcyclopropanes of general structure 324 available from ozonation of thujone derivatives may provide useful intermediates of general structure 325 to synthesis of polyquinanes which possess a bicylo[3.3.O]octane unit. It also serves as one way to correlate two “distant carbons”: C6 and C8. 10  10  324  325  Scheme 55 A Potential New £J Strategy *  In order to facilitate the comparison with the ring opening of thujonol 94, the numbering of those carbons in the bicyclo[3. 1 .Olhexane moiety in 120 is, at this point, kept the same as that in thujonol (94).  212  A few polyquinanes containing such a dimethylated bicyclo[3.3.O]octane unit are known, for example, (-)-retigeranic acid 304. In a recent total synthesis of 304, a chiral starting material 305 of the dimethylated bicylo[3.3.O]octane unit was incorporated into the target molecule 157 (Figure 45). 0 H  H  +  I’ll’’,  C 2 HO  Et 2 CO  326  327  328  Figure 45 Incorporation of a Dimethylated Bicylo[3.3.O]octane unit 4.4. Concluding Remarks:  prospect of thujone chemistry  The abstract of this thesis summarizes the highlights on applying the ozonation methodology into specific directions of investigation. Solutions to some remaining problems in these directions are suggested along the presentation while some possible extensions of the present work are also discussed. It remains to present some reflections on the subject of thujone chemistry as a whole. The enrichment of thujone chemistry and the enhancement of its versatility as a chiral starting material for the synthesis of biologically active natural products depend largely on the accumulation of fundamental knowledge about this unique entity in structurally diverse environments.  213  The ozonation of thujone and its derivatives allowed a novel functionalization of these molecules and opened the door to apply the cyclopropane chemistry on a different level in the last few years. This kind of carbon-hydrogen bond functionalization may be realized through a and should be 59 other more recently developed reagents’ ’, for example, dioxyranel 591 explored in the future. Functionalization of other positions in the thujone framework should be considered too. Ring expansion or contraction of thujone may provide interesting new avenues of thujone chemistry with regard to cyclopropane ring opening control and carbocyclic ring incorporation. The Robinson annulation of thujone is regioselective and stereoselective due to the substitution pattern of the thujone frame work and the particular geometry of the bicyclo[3.1.O]hexane unit. Annulations of opposite or complementary regioselectivity and stereoselectivity will enhance the versatility of thujone as a chiral building block. The 6membered ring annulation may be changed to annulations forming other ring sizes, for example, 5-membered and 4-membered rings when suitable new target molecules are chosen. Bridged and spiral annulations should be subjected to similar studies when needed. The degree of carbon incorporation may guide the planning in a more thorough and systematic manner. The seco and corro operations reveal an inherent connectivity of the thujone skeleton and allow novel chemistry to unfold. This may provide some novel solutions to difficult problems, for example, the direct creation of 6,6-A/B trans ring fusion. Abstraction of such formal operations from synthetic studies is intellectually inspiring and may find applications somewhere else*. As stated in Section 1.1. of Chapter 1 (General Introduction), the often tedious and lengthy process to prepare an intermediate, the racemate of which could be synthesized in a simple manner, is a serious drawback of using chiral building block. To avoid this problem,  *  For relevant discussion on general problem solving techniques, see ref. 156, under “Can you use the result?”.  214  such simple intermediates possibly derived from chiral building blocks should not be considered favorably. Highly functionalized intermediates, like functionalized cyclopropanes, cyclopentanes, cyclohexanes, and bicyclo systems like decalones, indenone, pentalenones are to be chosen as sub-goals early in the planning stage. There are other versatile chiral starting materials in use, for example, camphor and D glucose. Applications of D-glucose and other simple sugars have been numerous and provide the major basis for a systematic analysis of some synthetic problems, the so-called “chiron” . Since sugars are highly functionalized molecules, the application of them 60 approach’ frequently requires the removal of functional groups, a feature contrasting very much to  OH HO  D-glucose (pyranose)  camphor  the application of terpenes as starting materials. The camphor has been established as a very material* Comparison of thujone and the camphor chemistry will . versatile chiral starting 161 reveal some important elements responsible for their own effectiveness as chiral building block. The cross fertilization from the chemistry of camphor and other monoterpenes will certainly stimulate the chemistry of thujone. Different from camphor, thujone has not been employed as a chiral auxiliary so far. The diastereomeric impurity of thujone and unavailability of its enantiomer can attenuate its usefulness in this regard. However, derivatization of the diastereomeric mixture by converting the C4 chiral center into a trigonal center or into a quartery center may provide diastereomerically pure thujone derivatives useful as chiral auxiliaries.  *  We would like to thank Dr. T. Money for providing his newest review on this subject for our reference. 215  4.5.  Experimental See Section 2.3.1. for General experimental.  4.5.1.  Ring Expansion:  thujone (3) to ketoester 270  [1 R-( 1 x,2a/I3,6c)] 4-Ethoxycarbonyl-2-methyl-6-( 1 -methylethyl)bicyclo[4. 1 .0]heptan- 3-one (270, the ketoester form) [1 R-( 1 cc,2a/f3,6x)] 4-Ethoxycarbonyl-2-methyl-6-( 1 -methylethyl)bicyclo[4. 1 .0]hept-3-ene3-ol (270, the enolester form)  To a cooled solution (0°C) of thujone (3) (3.04 g, 20.0 mmol) and boron trifluoride etherate (4.26 g, 30.0 mmol) in anhydrous diethyl ether (25 ml), ethyl diazoacetate (3.42 g, 30.0 mmol) in anhydrous diethyl ether (5 ml) was added dropwise over a period of 30 minutes. The resulting solution was stirred under nitrogen at room temperature overnight, made basic with saturated aqueous sodium carbonate solution, and extracted with diethyl ether. The diethyl ether solution was washed with brine, dried over magnesium sulfate, and concentrated in vacuo.  Column chromatography of the crude product with ethyl  acetate:hexanes (1:30, v/v) mixture produced 13-ketoester 270 in 70% yield (3.34 g). The physical properties of 270 are as follows*: UV (MeOH, c=20.4 mg/I) max.: 258 nm (log e=3.980).  *  All data were taken from spectra of the mixture of a and 1 diastereomers (9:1 from GC). The 1 H-NMR spectral signals should be those of the predominant a diastereomer since these signals can be easily selected by comparing the integrations. The 1 H-NMR spectral signals of the minor f3 diastereomer were hardly observable from the spectrum. See foots at p. 178 and 180.  216  JR (film) vmax.: 3370 (0-H stretching), 1655(C=0 stretching), 1615 (C=C stretching) cm. 1 HNMR (400 MHz, CDC1 1 ) & 0.30 (1H, dd, J=4.4 and 8.8 Hz), 0.39 (1H, t, J=4.4 Hz), 3 0.68 (1H, dd, J=4.4 and 8.8 Hz), 0.98 (6H, two overlapped doublets, J=5.6 and 4.4 Hz), 1.03 (1H, m), 1.24 (3H, d, J=7.2 Hz), 1.31 (3H, t, J=6.8 Hz), 2.25-2.57 (2H, AB type, J=16 Hz), 2.64 (1H, q, J=7.2 Hz), 4.21 (2H, m), 12.24 (1H, s). MS m/z: 238 (M, 35.0%), 192 (79.7%), 177 (66.4%), 149 (100.0%). High resolution mass measurement calculated for 22 4H 238.1569; found: 238.1570. 1 C : 3 0 4.5.2.  Decarboxylation:  ketoester 270 to homothujone (272)  [1 R-( 1  2a/f3,6c)] 2-Methyl-6-( 1 -methylethyl)bicyclo[4. 1 .0]heptan-3-one (272)  I. —. . 272 To ketoester 270 (2.70 g, 11.3 mmol) in DMSO (20 ml) was added sodium chloride (1.20 g, 20.9 mmol) and water (1.0 ml). The resulting mixture was refluxed at 140°C for 4 hours, cooled down, diluted with water (40 ml), and extracted with diethyl ether (3X25 ml). The ether solution was dried over magnesium sulfate and concentrated in vacuo to give a crude product which was chromatographed with ethyl acetate:hexanes (1:8, v/v) mixture. Homothujone 272 was obtained in 96% yield (1.80 g). The physical properties of 272 are as follows*: JR (film) vmax.: 3060, 1700 cm. 1 ‘H-NMR (400 MHz, CDC13) 6: 0.50 (2H, m), 0.72 (1H, m), 0.95 (3H, d, J=6.4 Hz), 0.98  *  All data were taken from spectra of the mixture of a and f diastereomers (9:1 from GC). The 1 H-NMR spectral signals should be those of the predominant a diastereomer since these signals can be easily selected by comparing the integrations. The 1 H-NMR spectral signals of the minor f3 diastereomer were hardly observable from the spectrum. See also foots at p. 178 and 180.  217  (3H, d, J=6.4 Hz), 1.06 (1H, m), 1.22 (3H, d, J=8.0 Hz), 1.84 (1H, m), 2.10 (2H, m), 2.35 (1H, m), 2.47 (1H, m). MS m/z: 166 (M, 18.3%), 123 (29.7%), 109 (58.0%), 96 (91.2%), 41 (100.0%). High resolution mass measurement calculated for H 1 C 8 1 O: 1 166.1358; found: 166.1360. 4.5.3.  Robinson Annutation:  homothujone (272) to enone 274  [1 aR-( 1 ax,7a13,7bx)J 1,1 a,2,3 ,6,7,7a,7b-Octahydro-4,7a-dimethyl- 1 a-( 1 -methylethyl)-5Hcyclopropa[ajnaphthalen-5-one (274)  274 Homothujone 272 (341 mg, 2.05 mmol) was mixed with 1-diethylamino-3-pentanoneioclomethane salt (675 mg, 2.26 mmol) in anhydrous ethanol (20 ml) under an atmosphere of nitrogen. After the addition of potassium hydroxide (184 mg, —80% putre, 2.57 mmol), the reaction mixture was heated to reflux for 1 hour, cooled down, and diluted with water (30 ml). Petroleum ether (2x20 ml) was used to extract the above aqueous mixture. Concentration of the combined petroleum ether solution in vacuo furnished an oil which was chromatographed to provide 274 in 70% yield (332 mg). The physical properties of 274 are as follows: [a]t=+1.94X102 (c=1.00, 3 CHC1 ) . UV (MeOH, c=20.0 mg/I) max.: 250 nm ((log e=4.133). JR (film) vmax.: 3060, 1660, 1620 cm-’. ‘H-NMR (400 MHz, CDC1 ) & 0.30 (1H, dd, J=4.8 and 9.6 Hz), 0.50 (1H, dd, 3=4.8 and 3  9.6 Hz), 0.66 (1H, t, J=4.8 Hz), 0.90 (3F1, d, J=7.2 Hz), 0.93 (3H, d, J=7.2 Hz), 1.01 (1H,  218  m), 1.16 (3H, s), 1.58 (2H, m), 1.74 (3H, s), 1.82 (1H, m), 1.93 (1H, m), 2.12 (1H, dt, J=5.2 and 14.0 Hz), 2.35-2.70 (3H, m). MS m/z: 232 (M, 57.2%), 217 (18.3%), 189 (60.1%), 161 (100.0%). High resolution mass measurement calculated for H24O: 16 232.1827; found: 232.2819. C Elemental analysis: calc. for F1240: 16 C 82.70, H 10.41; found: 82.58, H 10.44. C 4.5.4. Birch Reduction-CH I Trapping and Birch Reduction-TMSCI Trapping3 Simmons-Smith Reaction-Hydrolysis Sequences: [1 aR-( 1  enone 274 to ketone 277  7af,7bc)] Decahydro-4,4,7a-trimethyl- 1 a-( 1 -methylethyl)-5H-cyclopropa[a}  naphthalen-5-one (277)  cyt 277 MethodA: Ammonia was distilled from sodium to a flask charged with enone 274 (419 mg, 1.81 mmol) under nitrogen. Pieces of lithium metal (13.8 mg, 1.99 mmol, 1.1 eqv.) were added and the resulting dark purple solution was stirred at -33°C for 1 hour before iodomethane (1.3 ml) and anhydrous diethyl ether (5.0 ml) were introduced. The dry ice-acetone condenser was removed to allow ammonia to evaporate. The reaction mixture was stirred overnight and transferred to a separatory funnel containing water (15 ml) and ether (20 ml). The ether layer was separated, washed with brine (10 ml), dried over magnesium sulfate. Evaporation of diethyl ether in vacuo resulted in a yellowish oil which was chromatographed first with ethyl acetate:hexanes (1:15, v/v) and then benzene to furnish ketone 277 in 15% yield (63 mg).  219  Method B: Ammonia (—20 ml) was distilled from sodium to a solution of enone 274 (1.10 g, 4.74 mmol) in anhydrous ether (10 ml) under nitrogen. Lithium (35 mg, 4.98 mmol,1.05 eqv.) was added. The dark purple mixture was stirred for 1.5 hours at -33°C before freshly distilled trimethylsilyl chloride (1.20 ml, 2.0 eqv.) was injected. The resulting yellowish solution was warmed to room temperature and stirred for 1 hour. Evaporation of ammonia and ether gave a yellowish crude oil. Anhydrous ether (10.0 ml) was introduced to the above crude product. Half of the solution (—5.0 ml) thus prepared was transferred to a new dry flask. Zinc-copper couple (powder, 314 mg) and distilled diiodomethane (0.80 ml) were added and the greyish mixture was refluxed overnight. Filtration through a layer of Celite afforded an ether solution which was condensed to a colorless oil. This oil was then dissolved in methanol (10 ml). After introduction of potassium hydroxide (100 mg, —80% pure, 1.78 mmol), the solution was refluxed 1 hour and cooled down. Evaporation of solvent in vacuo and repeated column chromatography with ethyl acetate:hexanes (1:8, v/v) mixture yielded 277 in 45% (262 mg). The physical properties of 277 are as follows: ). 3 [a]=-8.3 (c=0.42, CHC1 JR (film) Vmax.: 1703 1 cm (C=O stretching). ‘H-NMR (400 MHz, CDC13) 6: 0.08 (1H, m), 0.35 (2H, m), 0.70-1.55 {20H, including 0.85 (6H,  t,  J=2.4 Hz), 1.04 (3H, s), 1.21 (3H, s), 1.22 (3H, s)}, 1.70 (1H, m), 1.85 (1H,  m), 2.30 (2H, m), 2.62 (1H, m). MS m/z: 248 (M, 18.6%), 230 (12.0%), 205 (27.2%), 41(100.0%). High resolution mass measurement calculated for 0: 18 C 2 H 7 248.2140; found: 248.2135.  220  4.5.5. Catalytic Hydrogenation: [1 aR-( 1  enone 274 to ketone 278  7a,7bc)] Decahydro-4,7a-dimethyl- 1 a-( 1 -methylethyl)-5H-cyclopropa[aJ  naphthalen-5-one (278)  278 Method A: Ammonia (5 ml) was distilled from sodium to a flask containing 274 (151 mg, 0.500 mmol) in anhydrous ether (3.0 ml) under an atmosphere of nitrogen. While the flask was kept at -33°C, small pieces of lithium were added slowly for about 1 hour until a blue color persisted. After further stirring for 30 minutes, ammonium chloride was added to destroy excess lithium and ammonia was evaporated during warming up to room temperature. Concentration of the reaction mixture gave an oil which was chromatographed with ethyl acetate:hexanes (1:8, v/v) mixture to give a mixture of 278 and 279 (124 mg, 82%) of at a ratio 4.3:1 as indicated by GC. Method B: The solution of enone 274 (368 mg, 1.59 mmol) in ethanol (15.9 ml) was mixed with 10% palladium-charcoal catalyst (85 mg). The mixture was charged with 1 atm hydrogen at room temperature and stirred for 2 hours. Filtration through a layer of Celite gave a colorless solution which was then concentrated in vacuo. A mixture of 278 and 279 at a ratio 6:1 as shown from GC were thus obtained (350 mg, 95% yield).  221  The physical properties of 278 are as follows:* JR (film) vmax.: 1705 cm’ (C=O stretching). ‘H-NMR (400 MHz, CDC13) 6: 0.09 (1H, t, J=5.2 Hz), 0.40 (2H, m), 0.85 (3H, d, J=6.0 Hz), 0.88 (3H, d, J=6.0 Hz), 0.93 (3H, J=8.0 Hz), 1.45 (3H, s), 1.73 (1H, m), 1.87 (1H, m), 2.24 (2H, m), 2.50 (1H, m), 2.89 (1H, m). MS m/z: 234 (M, 30.6%), 219 (16.3%), 191 (20.2%), 41 (100.0%). High resolution mass measurement: calculated for 0: 16 234.1984; found: 234.1980. C 2 H 4.5.6.  Methylation:  c,I3-enone 274 to 13,y-enone 282  [1 aS-( 1 x,7a3,7ba)] 1,1 a,2,4,6,7 ,7a,7b-Octahydro-4,4,7a-trimethyl- 1 a- (1 -methylethyl)-5Hcyclopropa[ajnaphthalen-5-one (282)  282 To the solution of enone 274 (109 mg, 0.470 mmol) in anhydrous DMSO (5.0 ml) was added sodium methoxide (55 mg, 1.0 mmol, 2.1 eqv.) under nitrogen. After the mixture was stirred for 5 hours, iodomethane (100 p1, 1.61 mmol, 4.0 eqv.) was injected. Stirring continued for another 3 hours. The reaction mixture was poured to a funnel containing 20 ml water. The aqueous mixture was extracted with hexanes (2X15ml). After drying over magnesium sulfate, evaporation of solvent in vacuo, and chromatography with ethyl  *  All data were taken for the spectra of the mixture. The 1 H-NMR spectral signals were those of the predominant diastereomer 278 since they can be easily selected by comparing the integrations while the 1 H. NMR spectral signals of 279 were difficult to observe from the spectrum. See also footnotes at p. 178 and 180.  222  acetate:hexanes (1:8, v/v) mixture, ketone 282 was obtained (62 mg, 60 % yield based on 10% recovery of starting material). The physical properties of 282 are as follows: 1 JR (film) vmax.: 1700 cm‘H-NMR (400 MHz, CDC13) 8: 0.20 (1H,  t,  3=4.0 Hz), 0.39 (1H, dd, J=4.0 and 10.0 Hz),  0.58 (1H, dd, J=4.0 and 10.0 Hz), 0.80-1.40 { 16 H, m, including 0.94 (6H, d, J=6.0 Hz), 1.01 (3H, s), 1.18 (3H, s), 1.20 (3H, s)}, 1.84 (1H, m), 1.98 (1H, m), 2.14-2.35 (2H, m), 2.38-2.65 (2H, m), 5.41 (1H,  t,  3=4.0 Hz).  MS m/z: 246 (M, 36.4%), 231 (42.1%), 218 (5.9%), 203 (50.2%), 105 (100.0%). 4.5.7.  WoIf-Kishner-Huang Minion Reaction:  f3,y-enone 282 to alkene 283  [1 aS-( 1 cc7aI3,7bct)] 1 a,2,4,5 ,6,7 ,7a,7b-Octahydro-4,4,7a-trimethyl- 1 a-( 1 -methylethyl)- 1Hcyclopropa[ajnaphthalene (283)  283 To the mixture of ketone 282 (500 mg, 2.03 mmol) in diethylene glycol (10 ml) was added potassium hydroxide (422 mg, —80% pure, 6.02 mmol) and hydrazine hydrate (300 tl, 6.18 mmol) under nitrogen. After refluxing at 100-150°C for 1 hour, water and excess hydrazine hydrate were distilled away through a Dean-Stark trap until the temperature reached 250°C. Further refluxing at 200°C continued for 4 hours. The reaction mixture was then cooled to room temperature and and diluted with water (20 ml). The aqueous mixture was extracted with petroleum ether  (3x 10 ml). Evaporation of solvent in vacuo and column  chromatograghy with petroleum ether afforded 283 (316 mg, 67%).  223  The physical properties of 283 are as follows: JR (film) vmax.: 3050 cm1 (C-H stretching). H-NMR (400 MHz, CDC13) 6: 0.15 (1H, m), 0.40 (2H, m), 0.87 (6H, d, J=6.0 Hz), 1.05 1 (3H, s), 1.09 (3H, s), 1.16 (3H, S), 5.30 (1H, t, 3=4.0 H). MS m/z: 232 (Mt 53.4%), 217 (39.3%), 204 (6.7 %), 189 (62.1%), 105 (100.0%). 4.5.8.  Hydroboration:  alkene 283 to diol 285 and alcohol 286  [1 S-( 1 a,213,3cx,4aa,8aa)J Decahydro-3-hydroxy-5,5,8a-trimethyl-2-( 1 -methylethyl) naphthalenemethanol (285) [1 aS-( 1  3f,3a3,7af3,7bc)} Decahydro-4,4,7a-Trimethyl- 1 a-( 1 -methylethyl)-3H-  cyclopropa[a]naphthalen-3-ol (286)  OH  286  285  To the solution of 283 (100 mg, 0.43 mmol) in THF (5.0 ml) at 0°C under nitrogen was added borane (0.35 M in THE, 1.0 ml) in a dropwise manner. The resulting mixture was stirred for 5 hours at room temperature and cooled to 0°C again. Aqueous sodium hydroxide solution (3.0 M, 1.0 ml) and hydrogen peroxide solution (aq., 30%, 1.0 ml) were added slowly. The resulting two-phased mixture was warmed to room temperature, stirred for 2 hours, and saturated with sodium chloride. The THF layer was separated and the aqueous layer was extracted with ether (5 ml). The organic layers were combined and concentrated in vacuo. Column chromatography of the crude product with ethyl acetate:hexanes mixture (1:8 first and then 3:7, v/v) generated 285 (45 mg, 39%) and 286 (31 mg, 29%). The physical properties of 285 are as follows: 224  m.p.=136-138°C. . CHC1 ) [a]=+23 (c=0.84, 3 . 1 JR (film) vmax.: 3500 (0-H stretching) cm ‘H-NMR (400 MHz, CDC13) & 0.90-1.80 {26H, 0.96 (3H, s), 0.98 (3H, s), 1.00 (3H, s), 1.01 (3H, d, J=7.0), 1.15 (3H, d, J=7.0)}, 1.97 (2H, m), 2.18 (1H, m), 3.72 (2H, m), 4.04 (1H, m). , 1.2%), 235 (3.1%), 232 (1.7%), 123 (100%). High resolution mass (M-H 0 MS m/z: 250 2 7 268.2402; found: 268.2215. Chemical ionization measurement calculated for 0 12 C 3 H 2 3 as carrier gas) mlz: 286 (M+NH), 269 (M+Hj. MS (using NH 7 C 76.06, H 12.02; found: C 76.26, H 12.02. 2 0 1 C 3 H : Elemental Analysis: calculated for 2 The physical properties of 286 are as follows: . CHC1 ) [x]=+13 (c=0.50, 3 . 1 JR (film) vmax.: 3400(0-H stretching), 3060 (cyclopropane C-H stretching) cm H-NMR (400 MHz, CDC13) & 0.14 (1H, dd, J=4.4 and 8.8 Hz), 0.45 (lH, dd, J=4.4 and 1 8.8 Hz), 0.64 (1H,  t,  J=4.4 Hz), 0.85 (3H, d, J=6.0 Hz), 0.90 (3H, d, 3=6.0 Hz), 0.98 (3H,  s), 1.10 (3H, s), 1.16 (3H, s), 2.14 (JH, dd, J=7.5 and 15.0), 3.87 (1H, m). MS m/z: 250 (M, 2.1%), 232 (10.5%), 217 (12.8%), 207 (10.6%), 109 (100.0%). High 7 250.2297; found: 250.2307. 10 C 3 H resolution mass measurement: calculated for 0: 4.5.9. Oxidation by Jones Reagent:  alcohol 286 to ketone 287  [1 aS-( I a,3a13,7aI3,7bc)] Decahydro-4,4,7a-trimethyl- 1 a- (1 -methylethyl)-311-cyclopropa[a] naphthalen-3-one (287)  287  225  To the solution of alcohol 286 (20 mg, 0.080 mmol) in acetone (2.5 ml) was added Jones reagent (12M Cr03 in concentrated sulfuric acid) in a dropwise manner until the mixture changed to a steady orange color. Water (10 ml) was added and the aqueous mixture was extracted with hexanes (2x5 ml). Evaporation of solvent in vacuo and column chromatography with ethyl acetate:hexanes (1:8, v/v) mixture afforded 287 (16 mg, 80%). The physical properties of 287 are as follows: JR (film) vmax.: 1700 cm. 1 ‘H-NMR (400 MHz, C ) 8: 0.18 (2H, m), 0.53 (1H, m), 0.65 (3H, d, J=7.2 Hz), 0.90 D 6 (3H, d, J=7.2 Hz), 0.97 (3H, s), 1.01 (3H, s), 1.19 (3H, s), 2.29 (2H, AB type, J=16.0 Hz). MS m/z: 248 (M, 10.6 %), 233 (2.8%), 205 (5.2%), 177 (9.6%), 109 (100.0%). 4.5.10.  0-Methylation:  ketone 278/279 to methyl enol ether 280  [laR-( 1 a,7aI3,7bc)J 1 a,2,3,3a,6,7,7a,7b-Octahydro-4,7a-dimethyl- 1 a-( 1 -methylethyl)-5methoxyl- 1H-cyclopropra[a]naphthalene (280)  280 Ketone 278/279 (6:1, 200 mg, 0.855 mmol), obtained from palladium-charcoal catalyzed hydrogenation of 274, was treated with sodium hydride (70 mg, 2.0 eqv., 60% in mineral oil) in anhydrous DMSO (5.0 ml) under nitrogen at room temperature for 1 hour. Freshly distilled iodomethane (106 p.1, 1.71 mmol, 2.0 eqv.) was added rapidly and the resulting mixture was stirred for another 1 hour. The reaction mixture was then poured to water (20 ml) and the aqueous mixture was extracted with hexanes (2X15ml). Evaporation of  226  hexanes in vacuo and column chromatography with ethyl acetate:hexanes (1:8, v/v) mixture gave 280 (91 mg, 54% based on recovery of starting material) and starting material 278/279 (42 mg). The physical properties of 280 are as follows: JR (film) vmax.: 3050, 1680 (C=C stretching) cm . 1 lH4sMR (400 MHz, CDC1 ) & 0.08 (1H, m), 0.30 (2H, m), 0.70-1.70 {(22H, m, 3 including 0.87 (6H,  t,  J=6.0 Hz), 0.95 (3H, s), 1.60 (3H, s)}, 3.47 (3H, s).  MS m/z: 248 (M, 40.2%), 233 (8.4%), 216 (2.9%), 137 (90.2%), 41(100.0%). 4.5.11. WoIf-Kishner-Huang Minion Reaction:  ketone 277 to Alkane 291  [1 aR-( 1 a,7a3,7bcL)] Decadydro-4,4,7a-thmethyl- la-( 1 -methylethyl)- 1H-cyclopropa[a] naphthalene (291)  291 Ketone 277 (250 mg, 1.01 mmol) in diethylene glycol (20 ml) was treated with potassium hydroxide (370 mg, 5.28 mmol) and hydrazine monohydrate (270 p.1, 5.56 mmol). The mixture was heated at 100-150°C for 1.5 hours under nitrogen. The temperature was then gradually raised up to 220°C to distill away water and excess hydrazine over a period of 1.5 hours. Refluxing continued at 210°C for 4 hours. The mixture was cooled down, diluted with water, and extracted with petroleum ether (3X20 ml). Evaporation of the solvent in vacuo gave a brown oil which was chromatographed with petroleum ether through a short column to yield 291 as a colorless oil (175 mg, 75%). The physical properties of 291 are as follows: JR (film) vmax.: 3050 (cyclopropane C-H stretching) cm . 1  227  H-NMR (400 MHz, CDC1 1 ) 6: 0.07 (1H, m), 0.40 (2H, m), 0.84 (6H, t, J=3.0 Hz), 1.10 3 (3H, s), 1.20 (3H, s), 1.22 (3H, s). MS m/z: 234 (M, 2.7%), 219 (4.5%), 191 (11.0%), 43 (100.0%). High resolution mass measurement: calculated for : 10 C 3 H 7 234.2348; found: 234.2358. 4.5.12. Ozonation:  alkane 291 to ketone 292 and alcohol 293  [1 aS-( 1 a,7a13,7bx)] 1,1 a,3,3a,4,5,6,7,7a,7b-Decahydro-4,4,7a-trimethyl-1 a-(1 -methylethyl) -2H-cyclopropra[a]naphthalen-2-one (292) [1 aS-( 1 cz,7a13,7ba)] 1,1 a,3,3a,4,5,6,7,7a,7b-Decahydro-4,4,7a-trimethyl- 1 a-( 1 -methylethyl) -2H-cyclopropra[a]naphthalen-2-ol (293)  ccI 292  293  A stream of ozone-oxygen gas was passed through the solution of 291 (200 mg, 0.855 mmol) in ethyl acetate (10.0 ml) at -40°C for 7 hours. Oxygen was passed through the solution for 15 minutes to remove the residual ozone in the solution. The reaction mixture was then treated with dimethyl sulfide (0.5 ml), extracted with water (10 ml), and 10% aqueous sodium bicarbonate solution (10 ml). Removal of solvent in vacuo and chromatography of the crude product with ethyl acetate:hexanes (2:8, v/v) mixture provided ketone 292 (74 mg, 35%) and alcohol 293 (10 mg, 5%). The physical properties of 292 are as follows: IR (film) vmax.: 1665 (C=O stretching) cm . 1 H4JMR (400 MHz, CDC1 1 ) 6: 0.75-1.35 (23H, m, including 0.78 (3H, s), 0.84 (3H, d, 3 J=6.6 Hz), 0.97 (3H, d, J=6.6 Hz), 1.11 (3H, s), 1.30 (3H, s)}, 1.47 (1H, m), 1.63 (1H,  228  m),  1.84 (1H, septet, J=6.6 Hz), 2.00-2.30 (2H, m).  MS m/z: 248 (M, 18.4%), 233 (15.2%), 205 (23.2%), 177 (42.3%), 41(100.0%). High resolution mass measurement: calculated for 0: 18 C 2 H 7 248.2140; found: 248.2135. The physical properties of 293 are as follows: 1R (film) Vmax.: 3405 cm’ (0-H stretching). ) 6: 0.13 (1H, m), 0.45 (2H, m), 0.86 (3H, s), 0.89 (3H, d, 3 ‘H-NMR (400 MHz, CDC1 J=6.0 Hz), 0.98 (3H, d, J=6.0 Hz), 1.10 (3H, s), 1.20 (3H, s), 4.16 (1H, m). MS m/z: 250(M, 0.8%), 232 (5.3%), 217 (4.5%), 43 (100.0%). High resolution mass 10 C 3 H 7 250.2297; found: 250.2301. measurement: calculated for 0: 4.5.13.  Ketoacid 308  [1 S,2R, 1 ‘(2)R] 2- (2’-oxo- 1 ‘,3’,3’-trimethylcyclohexyl)cyclopropaneformic acid (308)  oco( 308 The physical properties of 308, which was provided by Dr. Dominik Guggisberg are as follows: m.p.: 88-90°C. ). 3 [x]=-4.9 (c=1.00, CHC1 JR (film) vmax.: 2300-3650 (0-H stretching), 1685 (C=0 stretching), 1645 (carboxylic acid . 1 group’s C=0 stretching) cm ) 6: 0.62 (1H, dd, J=5.6 and 8.8 Hz), 0.85-1.30 (19H, 3 ‘H-NMR (400 MHz, CDC1 including 0.93 (1H,  t,  J=8.8)}, 1.60-1.95 (6H, m).  MS m/z: 266 (M, 21.0%), 251 (4.9%), 238 (3.0%), 220 (15.0%), 205 (10.8%), 195  229  (3.3%), 109 (100.0%). High resolution mass measurement: calculated for 0 16 C 2 H : 3 266.1881; found: 266.1874. 4.5.14.  Methylation by Diazomethane:  ketoacid 308 to ketoester 309  [2R, 1 ‘(2)R,2’(2)S] 2- [2-( 1-Methylethyl))-2-(methoxycarbonyl)Jcyclopropyl-2,6,6,trimethylcyclohexanone (309)  X 2 oco 309 To the solution of 308 (500 mg, 1.88 mmol) in anhydrous diethyl ether (10.0 ml) at 0°C was added 0.35M diazomethane-diethyl ether solution (6.0 ml, 2.1 mmol) in a dropwise manner. The resulting mixture was stirred at room temperature for 2 hours. Solvent removal in vacuo and column chromatography with diethyl ether:hexanes (2:8, v/v) yielded ketoester 309 (501 mg, 95% yield). The physical properties of 309 are as follows: IR (film) vmax. 2960 (C-H stretching), 1710 (C=O stretching), 1685 (CO stretching) cm . 1 ‘H-NMR (400 MHz, CDC1 ) & 0.67 (1H, dd, J=4.8 and 9.6 Hz), 0.92 (3H, d, J=7.2 Hz), 3 1.07 (3H, d, J=7.2 Hz), 1.11 (3H, s), 1.13 (3H, s), 1.17-1.82 {12H, m, including 1.20 (3H, s)}, 3.65 (3H, s). MS m/z: 280 (M, 13.1%), 265 (2.3%), 248 (3.7%), 233 (2.4%), 220 (14.9%), 205 (10.0%), 177 (11.8%), 69 (100.0%). High resolution mass measurement: calculated for 16 280.2038; found: 280.2035. C : 3 0 26 11 4.5.15.  Ozonation:  ketoester 309 to compound 310  [2R, 1 ‘(2)R,2’(2)R] 2- [2-Acetyl-2-(methoxycarbonyl)]cyclopropyl-2,6,6,trimethylcyclohexanone (310)  230  310  The solution of ketoester 309 (241 mg, 0.861 mmol) in ethyl acetate (20 ml) was cooled to 0°C and passed with ozone-oxygen stream through a gas dispersion tube for 6 hours. The stream of oxygen was passed for 15 minutes to remove excess ozone. Dimethyl sulfide (0.5 ml) was added and the resulting mixture was stirred for 10 minutes at room temperature, extracted with water (10 ml), 10% aqueous sodium bicarbonate solution (2x10 ml), dried over magnesium sulfate, and concentrated in vacuo. The crude product was purified by column chromatography (petroleum ether:ether 8:2, v/v) to give 310 (63 mg, 45% based on recovery of starting material) in addition to the starting material 309 (101 mg, 42%). The physical properties of 310 are as follows: IR (film) vmax.: 1725, 1690 cm . 1 H-NMR (400 MHz, CDC13) & 1.08 (3H, s), 1.10 (6H, bs), 1.34 (1H, dd, 3=4.7 and 9.0 1 Hz), 1.50-2.05 (9H, m), 2.10 (1H, t, J=9.0 Hz), 2.28 (3H, s), 3.76 (3H, s). MS m/z: 280 (M, 0.5%), 262 (2.9%), 252 (4.8%), 43 (100.0%). High resolution mass 14 C 2 H : 4 6 280.1675; found: 280.1675. measurement: calculated for 0 4.5.16.  Cyclopropane Ring Opening Reaction:  thujonol (94) to bromoenone  318 and carvacrol (319) [2R,3S] 3-Bromo-2-methyl-5-( 1 -methylethyl)cyclohex-5-en- 1-one (318) 2-Methyl-5-( 1 -methylethyl)phenol (319)  231  x5  HJCy  318  319  Thujonol (94) (600 mg, 3.57 mmol) in methylene chloride (25 ml) was stirred with concentrated 48% hydrobromic acid (25 ml) for 1.5 hours at room temperature. The organic layer was separated, dried over magnesium sulfate, and concentrated in vacuo. The crude product was purified by column chromatography using ethyl acetate:hexanes (1:15, v/v) mixture to provide bromoenone 318 (700 mg, 85%) and carvacrol (319) (51 mg, 10%). The physical properties of 318 are as follows: [a]=+42 (c=0.29, CHC1 ). 3 UV (MeOH, c=20 mg/I) ?max.: 234 nm (log E=3.95) IR (film) Vmax.: 1670 (C=0 stretching), 1630 (C=C stretching). HNMR (400 MHz, CDC1 1 ) ö: 1.11 (6H, d, J=6.8 Hz), 1.34 (3H, d, J=7.1 Hz), 2.43 (1H, 3 septet, J=6.8 Hz), 2.55 (1H, m), 4.19 (1H, dt, J=4.4 and 10.2 Hz), 5.97 (1H, bs). MS m/z: 232/230 (M, 1.1%/1.3%), 151 (100.0%), 135 (33.6%), 123 (60.2%). High resolution mass measurement: calculated for Br C 1 H 7 O 5 232.0287 and 9 C 1 H 8 O 0 and Br: 5 1 230.0130; found: 232.0280 and 230.0116. The physical properties of 319 are as follows: IR (film) vmax.: 3400 (0-H stretching) cm . 1 1144MR (400 MHz, CDC13) ö: 1.22 (6H, d, J=6.6 Hz), 2.21 (3H, s), 2.82 (1H, septet, 1 3=6.6 Hz), 3.96 (1H, bs), 6.66 (1H, d, 3=1.8 Hz), 6.72 (1H, dd, J=1.8 and 7.1), 7.04 (1H, d, J=7.1 Hz). MS m/z: 150 (M, 35.5%), 135 (100.0%), 107 (15.6%). High resolution mass measurement: calculated for 0: C 1 H 0 150.1045; found: 150.105 1. 4  232  Cyclopropane Ring Opening Reaction:  4.5.17.  thujonol (94) to chioroenone  320 and carvacrol (319) [2R,3S1 3-Chloro-2-methyl-5-( 1 -methylethyl)cyclohex-5-en- 1-one (320)  Thujonol (94) (500 mg, 2.98 mmol) was treated with concentrated 35—36% hydrochloric acid (25 ml) in methylene chloride (25 ml) at room temperature for 1.5 hours. The methylene chloride solution was separated, dried over magnesium sulfate, and concentrated in vacuo. Column chromatography with ethyl acetate:hexanes (1:20, v/v) provide chloro-enone 320 (252 mg, 45%) and carvacrol (319) (177 mg, 40%). The physical properties of 320 are as follows: IR (film) vmax.: 1675 (C=O stretching), 1630 (C=C stretching) cm . 1 )& 3 ‘H-NMR (400 M}lz, CDC1  1.09 (6H, d, J=7.2 Hz), 1.30 (3H, d, J=7.8 Hz), 2.43 (1H,  septet, J=7.2 Hz), 2.54 (1H, m), 2.78 (2H, m), 4.06 (1H, dt, 1=4.4 and 9.8 Hz), 5.95 (1H, bs) ppm. MS m/z: 188/186 (M, 5.8%/19.1%), 151 (100.0%), 135 (15.9%). 4.5.18.  Cyclopropane Ring Opening Reaction:  hydroxyenone 122 to  bromodienone 322 [4aS-(4acz,5a)] 5-Bromo-2,3,3a,4,5,6-hexahydro- 1 ,4a-dimethyl-( 1 -methylethyl) naphthalen 2(3H)-one (322)  233  322 Hydroxyl-enone 122 (39 mg, 0.17 mmol) in methylene chloride (5 ml) was stirred with concentrated 48% hydrobromic acid (5 ml) at room temperature for 3 hours. The methylene chloride layer was separated and the aqueous layer was extracted with methylene chloride (5 ml). The combined methylene chloride solution was dried over magnesium sulfate and concentrated in vacuo. Column chromatography of the crude product afforded bromo dienone 322 (45 mg, 9 1%). The physical properties of 322 are as follows: [a]=+42O (c=1.00, CHC1 ). 3 UV (MeOH, c=20 mg/i) max.: 293 nm (log E4.40) JR (film) vmax.: 1660 (C=O stretching), 1620 (C=C stretching) cm . 1 ‘H-NMR (400 MHz, CDC1 ) ö: 1.12 (6H, d, J=6.0 Hz), 1.17 (3H, s), 1.86 (3H, s), 2.003 2.90 (7H, m), 4.14 (JH, dd, J=6.0 and 10.0 Hz), 6.31 (1H, s). MS m/z: 298/296 (M, 80.0%/88.5%), 217 (100.0%), 175 (56.3%). High resolution mass measurement: calculated for Br: 19 C 2 H 7 O 5 296.0775; found: 296.0768. 1  234  Bibliography Chapter 1 1.  E. 3. Ariens, W. Soudijn, and P. B.M. W. M. Timmermans; Stereochemistry and Biological Activity ofDrugs; Blackwell Scientific: Palo Alto, CA, 1983.  2.  E. J. Ariens, J. S. van Rensen, and W. Welling, eds.; Stereochemistry of Pesticides: Biological and Chemical Problems; Elsevier Science Publishers B. V., 1988.  mt. Ed. Engl. 19,  13(1980).  3.  G. Blaschke; Angew. Chem.  4.  J. A. Pickett; Phil. Trans. R. Soc. Lon. B 310: 235-239(1985).  5.  J. W. Scott; Asymmetric Synthesis ; J. D. Morrison and J. W. Scott, eds.; Academic Press, vol. 4, 1984; p. 1-226.  6.  D. Whittaker and D. V. Banthorpe; Chem. Rev. 72 (4), 305-3 13 (1972). A review on chemistry of thujane monoterpenes.  7.  J. Werner and M. T. Bogert; J. Org. Chem. 3, 578 (1939).  8.  T. Norm and M. S. Bergquist; Arkiv. Kemi. 22, 137 (1964).  9.  D. V. Banthorpe and H. S. Davies; J. Chem. Soc. B, 1356 (1968).  10.  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, and B. R. Worth; Can. J. Chem. 57, 3145 (1979).  11.  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. Becaiski, W. R. Cullen, M. D. Fryzuk, G. Herb, BR. James, 3. P. Kutney, K. Piotrowska, and D. Tapiolas; Can. J. Chem. 66, 3108 (1988).  12.  C. Carvaiho, 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).  13.  a) b) c) d) e)  14.  a) J. P. Kutney, P. Grice, K. 0. Pascoe, K. Piotrowska, S. 3. Rettig, J. Szykula, and J. Trotter Helv. Chim. Acta 65, 2345 (1982). b) J. P. Kutney, P. Grice, K. Piotrowska, S. J. Retttig, J. Szykula, J. Trotter, and L. V. Chu; Helv. Chim. Acta 66, 1806 (1983). c) J. P. Kutney, P. Grice, K. 0. Pascoe, K. Piotrowska, S. J. Rettig, J. Szykula, and J. Trotter; Helv. Chim. Acta 66, 1820 (1983).  15.  J. P. Kutney, K. Piorowska, Y-H. Chen, K-P. N. Cheng, Z. Gao, and S. J. Rettig;  J. P. J. P. I. P. J. P. J. P.  Kutney, J. Balsevich, and P. Grice; Can. J. Chem. 58, 2641(1980). Kutney and A. K. Singh; Can. J. Chem. 60, 1842(1982). Kutney and A. K. Singh; Can. J. Chem. 61, 1111(1983). Kutney and A. K. Singh; Can. J. Chem. 62, 1407 (1984). Kutney and A. K. Singh; Can. J. Chem. 62, 2813 (1984).  235  Can. J. Chem. 68, 1698 (1990). 16.  A. Greenberg and 3. F. Liebman; Strained Organic Molecules; Academic Press, Inc., 1978; P. 81-83.  17.  a) H. N. C. Wong, M-Y Hon, T. Hudlicky, C-W Tse, and Y-C Yip; Chem. Rev. 89, 165-198 (1989). b) Z. Rappoport, ed.; The Chemistry of Cyclopropyl Group; John Wiley & Sons, New York, 1987.  18.  D. A. Oare and C. H. Heathcock; Topics in Stereochemistry ; E. L. Eliel and S. H. Wilen, eds.; John Wiley and Sons, Vol. 19, 1989; p. 227-407.  Chapter 2 Introduction 19.  a) T. A. van Beek and Ae. de Groot; Red. Tray. Chim. Pays-Bas 105, 5 13-527 (1986). b) ibid.; Red. Tray. Chim. Pays-Bas 106, 1-18 (1987). c) B. 3. M. Jansen and Ae. de Groot; Nat. Prod. Rep. 8, 309-3 18 (1991). d) B. J. M. Jansen and Ae. de Groot; Nat. Prod. Rep. 8, 3 19-337 (1991).  20.  G. W. Ware; Pesticide, Theory and Application ; W. H. Freeman and Company, San Francisco, 1983.  21.  W. S. Bowers; Natural Productsfor Innovative Pest Managementt ; D.L. Whitehead and W. S. Bowers, eds.; Pergman Press, Oxford, 1983; P. 4758.  22.  E. A. Bell, L. E. Fellows, and M. S. J. Simmonds; Safer Insecticides, Development and Use ; E. Hodgson and R. 3. Kuhr, eds.; Marcel Dekker, New York and Basel, 1990; P. 337-350.  23.  a) K. Munakata; Pure Appl. Chem. 42, 57 (1975). b) J. R. Miller and K. L. Strickler; Chemical Ecology ofInsects; W. J. Bell and R. T. Carde, eds.; Chapman and Hall London, 1984; P. 127. c) L.M. Schoonhoven; Entomol. Exp. Appl. 31, 57 (1982).  24.  D. H. Williams, M. J. Stone, P. R. Hauck, and ans S. K. Rahman; J. Natural Products 1189-1208 (1990).  25.  T. Jermy; Natural Productsfor Innovative Pest Managementt ; D. L. Whitehead and W. S. Bowers, eds.; Pergamon Press, Oxford, 1983; P. 223-236.  26.  a) I. Kubo, Y-W. Lee, M. Pettei, F. Pilkiewicz, and K. Nakanishi; J. Chem. Soc., Chem. Comm. 1013 (1976). b) I. Kubo and K. Nakanishi; Advances in Pesticide Science ; H. Geissbuehler, ed.; Pergman Press, Oxford, Vol. 2, 1979; p. 284. c) K. Nakanishi and I. Kubo; Israel J. Chem. 16, 18 (1977). d) K. Nakanishi; Insect Biology ofFuture; M. Locke and D. S. Smith, eds.; Academic Press, New York, 1980; p. 603.  27.  a) D. C. Griffiths, J. A. Pickett, L. E. Smart, and C. M. Woodlock; Pestic. Sci. 27, 269-276 (1989). b) 3. A. Pickett; J. Chem. Ecol. 10, 1845 (1988).  236  28.  I. Kubo and I. Ganjian; Experientia 37 (1981).  29.  a) C. S. Barnes and J. W. Loder Austr. J. Chem. 15, 322 (1962). b) Y. Fukuyama, T, Sato, I. Miura, and Y. Asakawa; Phytochemistry 24, 1621 (1985) c) J. W. Loder; Austr. J. Chem. 15, 389 (1962). d) F. S. El-Feraly, A. T. McPhail and K. D. Onan; J. Chem. Soc.,Chem. Comm. 75 (1978). e) C. J. W. Brooks and G.H. Draffan; Tetrahedron 25, 2887 (1969). f) L. Canonica, A. Corbella, G. Jommi, J. Krepinsky, G. Ferrari, and C. Casagrande; Tetrahedron Lett. 2137 (1967).  30.  a) G. Sodano, M. D’Ischia, and G. Prota; Tetrahedron Lett. 23, 3295-3298 (1982). b) G. Sodano, G. Cimno, A. Spella; Tetrahedron Lett. 25, 4151-4152 (1984). c) G. Sodano, V. Caprioli, C. Cimino, R. Colle, M. Gavagnin, and A. Spinella; .1. Nat. Prod. 50, 146-15 1 (1987).  31.  a) W-C. Ma; Physiol. Entom. 2, 199-207 (1977). b) I. Kubo; Agric. Biol. Chem. 48, 73-78 (1984). c) G. L. Feritz, G. D. Mills, J. D. Warthen, and R. M. Waters; J. Chem. Ecol. 15 (12), 2607-2623 (1989).  Chapter 2 Total Synthesis of Polygodial and Warburganal 32. (±) Polygodial a) T. Kubo, T. Suzuki, M. Tanemura, A. S. Kumanireng, N. Ototani, and Y. Kitahara; Tetrahedron Lett. 1961 (1971). b) S. P. Tanis, and K. Nakanishi; J. Am. Chem. Soc. 101, 4398 (1979). c) S. C. Howell, S. V. Ley, and M. Mahon; J. Chem. Soc., Chem. Comm. 507 (1981). d) M. Jallali-Maini, G. Boussac, P. Lemaitre, M. Larcheveque, D. Guillerm, andl. Y. Lallemand; Tetrahedron Lett. 22, 2995m (1981). e) R. L. Snowden; Tetrahedron Lett. 25, 3835 (1984). f) Ae de Groot and B. J. M. Jansen; J. Org. Chem. 855 (1988). 33.  (-)-Polygodial a) K. Mori and H. Watanabe; Tetrahedron 42, 273 (1986). b) M. J. Cortes, I. Razmilic, J. R. Sierra, and J. Lopes; Chem. and md. 735 (1985) c) ibid.; Syn. Comm. 17, 95-103 (1987). d) J-F. He and Y-L. Wu; Tetrahedron 44, 1933-1940 (1988). e) B. J. Jansen, J. A. Kreuger, and Ae. de Groot; Tetrahedron 45, 1447 (1989)  34.  (±)-Warburganal a) A. Ohsuka, and Matsukawa; Chem. Lett. 635 (1979). b) H. Okawara, H. Nakai, and M. Ohno; Tetrahedron Lett. 23, 1987 (1982). c) T. Nakata, H. Akita, T. Naito, T. Oishi; J. Am. Chem. Soc. 101, 400 (1979). d) S. V. Ley and M. Mahon; Tetrahedron Lett. 22, 3909 (1981). e) D. M. Hollinshead, S. C. Howell, S. V. Ley, M. Mahon, and N. M. Ratcliffe; J. Chem. Soc., Perkin I 1579 (1983). f) P. A. Wender and S. L. Eck; Tetrahedron Lett. 23, 1871 (1982). 237  g) D. J. Goldsmith and H. S. Hezar; Tetrahedron Lett. 21, 3543 (1980). h) A. S. Kende and T. J. Blacklock; Tetrahedron lett. 21, 3119 (1980). 35.  (-)-Warburganal a) M. L. Oyarzum, M. Cones, and J. Sierra; Synthtic Comm. 12, 951 (1982). b) H. Okawara, N. Nakai, and M. Ohno; Tetrahedron Lett. 23, 1087 (1982). c) I. Razmilic, J. Sierra, J. Lopes, M. Cones; Chem. Lett. 1113 (1985). d) S. Manna, P. Yadagiri, and J. R. Faick; J. Chem. Soc., Chem. Comm. 1324 (1987) e) T. Nakano, M. A. Mailo, andA. Rojas; J. Chem. Soc., Perkin Trans I 2137 (1987). f) I. A. Hueso-Rodriguez and B. Rodriguaz; Tetrahedron 45, 1567 (1989).  36. Cinnamidial a) T. Nakai, T. Nakata, and T. Oishi; Chem. Lett. 445 (1980). b) L. P. J. Bueton, and J. White; J. Am. Chem. Soc. 103, 3226 (1981); c) ibid.; J. Org. Chem. 50, 357 (1985). 37. Muzigodial a) M. P. Bosch, F. Camps, J. Coil, A. Guerrero, T. Tatsuoka, and J. Meinwald; J. Org. Chem. 51, 773 (1986). b) Ae. de Groot, M. P. Broekhuysen, L. Doddema, M. C. Voliering, and J. M. M. Westerbeek; Tetrahedron Lett. 23, 4831 (1982). c) Ae. de Groot and B. J. M. Jansen; J. Org. Chem. 49, 2034 (1984). 37. Muzigodial a) M. P. Bosch, F. Camps, J. Coil, A. Guerrero, T. Tatsuoka, and J. Meinwald; J. Org. Chem. 51, 773 (1986). b) Ae. de Groot, M. P. Broekhuysen, L. Doddema, M. C. Vollering, and 3. M. M. Westerbeek; Tetrahedron Lett. 23, 4831 (1982). c) Ae. de Groot and B. J. M. Jansen; J. Org. Chem. 49, 2034 (1984). 38.  P. S. Bailey; Ozonation in Organic Chemistry ; Academic Press, Vol. 2, 1982.  39.  3. R. Durland and H. Adkins; J. Am. Chem. Soc. 61, 429(1939).  40.  H. Varkony, S. Pass, and Y. Mazur; Chem. Commun. 437(1974)  41.  Z. Cohen, E. Keinan, Y. Mazur,and T. H. Varkony; J. Org. Chem. 40, 2141 (1975)  42.  G. A. Cook, A. D. Kiffer, C. V. Klumpp, A. H. Malik, and L. A. Spence; Adv. Chem. Ser. 21, 44 (1959).  43.  D. Tal, E. Keinan, and Y. Mazur; J. Am. Chem. Soc. 101, 502 (1979).  44.  5. Sarel, J. Yovell, and M. Sarel-Imber; Angew. Chem. inter. Edit., 7, 577 (1968).  45.  Z-Y Gao, Master thesis, University of British Columbia, 1989. 238  46. P. Gunning, Ph. D. thesis, University of British Columbia, 1991. 47.  Y. M. Hellman and G. A. Hamilton; .1. Am. Chem. Soc. 96, 1530 (1974).  48.  a) H. M. White and P. S. Bailey; J. Org. Chem. 30, 3037 (1965). b) C. C. Price and A. L. Tumolo; J. Am. Chem. Soc. 86, 4691 (1964). c) S. W. Benson; Adv. Chem. Ser. 77, 74 (1968). d) R. W. Murray; Singlet Oxygen; H. H. Wasserman and R. W. Murray, eds.; Academic Press, New York, 1979; p. 93-114. e) C. C. Schubert and R. N. Pease; J. Am. Chem. Soc. 78, 5553 (1956). f) A. A. Popov, S. K. Rakovski, D. M. Shopov, and L. V. Ruban; Izc. Akad. Nauk SSSR, Ser. Khim. 1950 (1976); Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transi.) 958 (1976).  49.  A. J. Beckwith, C. L. Bodkin, and T. Duong; Aust. J. Chem. 30, 2177 (1977).  50.  a) E. Proksch and A. de Meijere; Angew. Chem. mt. Ed. Engl. 15, 761 (1976). b) A. Pross; Adv. Phys. Org. Chem.; V. Gold and D. Bethell, eds.; Academic Press, Vol. 14, 1977. c) C. D. Johnson; Tetrahedron 36, 3461 (1980).  51.  A. A. Akhrem and Y. A. Titov; Total Steroid Synthesis; Plenum Press, New York, 1970; p. 64-72.  52.  a) R. Fraisse-Jullien, C. Frajaville, V. Toure, and M. Derieux; Bull. Soc. Chim. Fr. 3725 (1966). b) J. P. Kutney, J. Cable, W. A. F. Gladstine, H. W. Hanssen, E. 3. Torupka, and W. D. C. Warnock; J. Am. Chem. Soc. 90, 5332 (1968).  53.  G. Stork and D. E. Kanhne; J. Am. Chem. Soc. 105, 1072 (1983).  54.  M. Bartok, et al.; Stereochemistiy of Heterogeneous Metal Catalysis; John Wiely & Sons, 1985.  55.  3. P. Mecomick and D. L. Barton; .1. Org. Chem. 45, 2566 (1980).  56.  a) H. G. Kuivila; Synthesis 499 (1970). b) W. P. Neumann; Synthesis 665 (1987).  57.  a) L. H. Slaugh; J.Am. Chem. Soc. 87, 1522 (1965). b) Monygomery and Matt; J. Am. Chem. Soc. 89, 6553 (1967). c) T. A. Holgren, ME. H. Howden, ME. Medof; 3. 0. Roberts; J. Am. Chem. Soc. 89, 3051 (1967). d) S. J. Cristol and R. V. Barbour; J. Am. Chem. Soc. 90, 2832 (1968). A. L. J. Beckwith and V. W. Bowry; J. Org. Chem. 54, 2681 (1989).  58.  H. Laurant and R. Wiechert; Organic Reactions in Steroid Chemistry; J. Fried and A. Edwards, eds.; Van Norstrand Reinhold Company, New York, 1972; p. 87.  59.  W. Reusch; Reduction: techniques and applications in organic synthesis; R. L. Augustine, ed.; Marcel Dekker, Inc., New York, 1968; p. 17 1-211.  60.  W. G. Dauben and G. H. Berezin; J. Am. Chem. Soc. 89. 3449 (1967). 239  61.  a) P. A. Wender and J. J. Howbert; Tetrahedron Lett. 24, 5325 (1983). b) R.A. Roberts, V. Schull, and L. Paquette; J. Org. Chem. 48, 2076 (1983).  62.  B. Fuchs; Topics Stereochemistry ; E. L. Eliel and N. L. Alinger, eds.; Wiley, New York; Volume 10, 1978; p. 1-94.  63.  R. Bucourt; Conformational Analysis: scope and present pimitations ; G. Chiurdoglu, ed.; Academic Press, New York, 1971; p. 59-62.  64.  a) E. Toromanoff; Tetrahedron 36, 2809-2931(1980). b) M. L. Sinnott; Adv. Phys. Org. Chem.; Academic Press, Vol. 24, 1988.  65.  a) H. E. Paren, W. S. Mellon, H.K. Schnoes, and H. F. DeLuca; J. Bioorg. Chem. 13, 62-75 (1985). b) H. E. Paaren, H. F. DeLuca, and H. K. Schnoes; J. Org. Chem. 45, 3253 (1980)  66.  G. Frater and U. Muller; Helv. Chim. Acta 72, 653 (1989).  67.  I. M. Goldman; J. Org. Chem. 34, 1979 (1969).  68.  a) K. Hayano, H. Shirahama, and T. Matsumoto; Bull. Chem. Soc. Jpn. 63, 628 (1990). b) K. Hayasaka, T. Ohshikazu, H. Shiorahama, and T. Matsumoto;Tetrahedron Lett. 26, 873 (1985). c) T. Fujita, T. Ohtsuka, H. Shirahama, and T. Matsumoto; Tetrahedron Lett. 23, 4091 (1982).  69.  a) W. D. Closson and G. T. Kwiatkowski; Tetrahedron 21, 2779 (1965). b) W. G. Bauben, L. E. Friderich, P. Obserhanli, E. I. Aoyagi; J. Org. Chem. 37, 9 (1972). c) R. W. Thies and J. E. Billigmeier; J. Org. Chem. 38, 1758 (1973). d) R. W. Thies and H. J. Shih; J. Org. Chem. 42, 280 (1977).  70.  a) W. D. Emmons and G. B. Lucas; J. Am. Chem. Soc. 77, 2287 (1955). b) R. R. Sauers and R. W. Ubersax; J. Am. Chem. Soc. 3939 (1965).  71.  a) P. A. S. Smith; Molecular Rearangements ; P. de Mayo, ed.; Interscience, New York, Vol. 1, 1963; p. 577-589. b) H. 0. House; Modern Synthetic Reactions; Benjamin, New York, 1972; p. 327. c) G. R. Krow; Tetrahedron 37, 2697-2724 (1981).  72.  B. Plesnicar; Oxidation in Organic Chemistry; W. S. Trahanovsky, ed.; Academic Press, New York, Part C, 1978; p. 254-262.  73.  C. H. DePuy and R. J. Van Lanen; J. Org. Chem. 39, 3360 (1974).  74.  a) Y. Ito, S. Fujii, and T. Saegusa; J. Org. Chem. 41, 2073 (1976). b) Y. Ito, S. Fujii, M. Nakatsuka, F. Kawamoto, and T. Saegusa; Org. Synth. 59, 113-122 (1979).  75.  R. M. Moriarty, R. K. Vaid, T. E. Hopkins, B. K. Vaid, and 0. Prakash; Tetrahedron Lett. 31, 197 (1990). 240  76.  a) M. Miyashita, H. Uda, and A. Yoshikoshi; J. Chem. Soc. Chem. Comm. 1396, (1969). b) M. Miyashita and A. Yoshikoshi; J. Am. Chem. Soc. 96, 1917 (1974).  77.  P. S. Mariano and E. Bay; J. Org. Chem. 45, 1763 (1980).  78.  I. Fleming; Frontier Obital s and Organic Chemical Reactions; John Wiley & Sons, New York, 1976.  79.  D. Caine; Organic Reactions; John Wiley & Sons, Inc., Vol. 23, 1976; P. 20-39.  80.  von B. Maurer, M. Frachebvoud, A. Grieder, G. Ohloff; Helv. Chim. Acta 65, 2371 (1972).  Chapter 3 81.  a) G. Ohloff; Fragrance Chemistry---The Science of the Sense of Smell; E.T. Theimer, ed.; Academic Press, Inc., 1983; Chapter 15, The Fragrance of Ambergris, p. 535 573. b) E. Lederer; Prog.Chem. Org. Nat. Prod. 6, 87 (1950). -  82.  P. A. Dubois; Parfums. Cosmet. Aromes. 19, 35 (1978).  83.  a) E. Lederer and D. Mercier; Experientia 3, 188 (1947). b) L. Ruzika, H. Gutman, 0. Jeger, E. Lederer; Helv. Chim. Acta 31, 1746 (1948).  84.  a) E. Jegou, J. Polonsky, E. Lederer, K. -H. Schulte-Elte, B. Egger, and G. Ohloff; Nouv. J. Chim. 1, 529 (1977). b) B. D. Mookherjee and R. R. Patel; mt. Congr. Essent. Oils, 7th, Kyoto, Oct. 711, 1977; p 479.  85.  A, G. Armor, G. Buchi, A. Eschenmoser, A. Storni; Helv. Chim. Acta 42, 2233 (1959).  86.  a) M. Stoll, C. F. Seidel, B. Wilihaim, M. Hinder; Helv. Chim. Acta 39, 183 (1956). b) G. Ohioff and C. Vial; Helv. Chim. Acta 60, 2767 (1977).  87.  a) E. Lederer and R. Tixier; C. R. Acad. Sci. 225, 531 (1947). b) G. Ohloff, K. H. Schultz, and B. L. Muller; Helv. Chim. Acta 60, 2763 (1977).  88.  L. Ruzika and C. F. Seidal; Helv. Chim. Acta 33, 1285 (1950).  89.  M. Stoll and M. Hinder; Helv. Chim. Acta 33, 1251 (1950).  90.  a) G. Ohloff, W. Giersch, W. Pickenhagen, A. Furrer, and B. Frei; Helv. Chim. Acta 68, 2022 (1985). b) S. Schwimmer and D. G. Guadagni; J. Food Sci. 27, 94 (1979).  91.  S. Escher, W. Giersch, Y. Niclass, G. Bernardinelli, and G. Ohloff; Helv. Chim. Acta 73, 1935 (1990). 241  92.  a) G. Ohloff; Gustation and Olfaction; G. Ohioff and A. Thomas, eds; Academic Press, London & New York, 1971; p 168. b) G. Ohioff, F. Naf, R. Decorzant, W. Thommen, and E. Sundt; Helv. Chim. Acta 56, 1414 (1973). c) ref. 81a, p 563.  93.  I. B. Bersuker, A. S. Dimoglo, M.Yu. Gorbachav, M. N. Koitsa, and P. F. Vald; Nouv. J. Chim. 9, 211 (1985).  94.  a) B. Winter; Pure & App!. Chem. 62, 1377 (1990). b) M. G. J. Beets; Structure -activity Relationships in Human Chemoreception; Applied science publishers, London, 1978; p. 149-172.  95.  R. Axel and L. Buck; Cell 65, 75 (1991). This is a recent paper describing the discovery of 18 genes responsible for coding in the odor receptor proteins. It is also summarized in Time, April 15th, 1991.  96.  a) J. N. Schumacker, W. M. Henley, E. Bernasek, and C. E. Teague; U. S. Patent 3, 050, 532 (1962). b) M. Stoll; U. S. Patent 3, 029, 255 (1962). c) M. Stoll and M. Hinder; Helv. Chim. Acta 33, 1251 (1950). d) M. Hinder and M. Stoll; Helv. Chim. Acta 33, 1308(1950).  97.  R. Decorzant, C. Vial, F. Naf, and G. M. Whitesdes; Tetrahedron 43, 1871 (1987).  98.  a) P. A. Christenson; Tetrahedron 44, 1925 (1988). b) D. B. Bigley, N. A. J. Rogers, and J.A. Barltrop; J. Chem. Soc. 4613 (1960).  99.  I. C. Coste-Maniera, J. P. Zahra, and B. Waegell; Tetrahedron Lett. 29, 1017 (1988).  100. H. Koyama, Y. Kaku, and M. Ohno; Tetrahedron Lett. 28, 2863 (1987). 101. R. C. Cambie, K. N. Joblin, and A. F. Preston; Aust. J. Chem. 24, 583 (1971). 102. J. de P. Terasa, J. G. Urones, A. M. Pedraro, P. B. Barcals; Tetrahedron Lett. 26, 5717 (1985). 103. M. Gonzalez-Sierra, E. A. Ruveda, J. T. Lopez, and M. J. Cones; Heterocycles 26, 2801 (1987). 104. a) K. Mon and H. Tamura; Liebigs Ann. Chem. 361 (1990). b) K. Mori and Komatsu; Bull. Soc. Chim. Belge 95, 771 (1986). 105. T. Kawanobe, K. Kogami, and M. Matsui; Agric. Biol. Chem. 50, 1475 (1986). 106. G. Buchi and H. Wuest; Helv. Chim. Acta 72, 996 (1989). 107. a) G. H. Posner; An Introduction to Synthesis Using Organocopper Reagents ; John Wiley & Sons, Inc.; 1980. b) G. H. Posner, Organic Reaction; John Wiley, New York; Vol. 19, 1972; p. 1.  242  108. a) E. Piers and W. de Waal, and R. W. Britton; J. Chem.Soc., Chem. Comm. 188 (1968). b) ibid.; Can. J. Chem. 47, 4299 (1969). c) ibid.; Can. J. Chem. 47, 4307 (1969). 109. E. Toromanoff; Bull. Soc. Chim. Fr. 708 (1960). 110. a) J. A. Marshall and S. F. Brady; Tetrahedron Lett. 1387 (1969). b) ibid.; J. Org. Chem. 35, 4068 (1970). c) J. A. Marshall and R. A. Ruden; Tetrahedron Lett. 2875 (1971). 111. J. A. Marshall and G. M. Cohen; J. Org. Chem. 36, 877 (1971). 112. P. Beak and B. M. Monroe; J. Org. Chem. 32, 2778 (1963). 113. S. H. Benz and G. Dabbagh; Tetrahedron 45, 425 (1989). 114. M. Gall and H. House; Org. Synth. 52, 39 (1972). 115. H. 0. House; Modern Synthetic Reactions; W. A. Benjamin, Inc., 1972; p. 502. 116. R. M. Coates and L. 0. Sandefur; J. Org. Chem. 39, 275 (1974). 117. a) H. 0. House, R. A. Auerbach, M. Gall, N. P. Peet; J. Org. Chem. 514 (1973). b) ibid; J. Org. Chem. 1209 (1976). 118. T. J. Flautt and W. F. Erman; J. Am. Chem. Soc. 85, 3212 (1963) 119. a) R. J. K. Taylor; Synthesis 364-392 (1985). b) M. J. Chapdelaine and M. Hulce; Organic Reaction ; John Wiley, New York; Vol.38, 1990; p. 225-654. 120. a) H. J. Reich, J. M. Renga, and I. L. Reich; J. Am. Chem. Soc. 97, 5434 (1975). b) H. J. Reich, S. Wollowitz, J. E. Trend, F. Chow, D. F. Wendelborn; J. Org. Chem. 43, 1697 (1978). 121. J. Valls and E. Toromanoff; Bull. Soc. Chim. France; 758 (1961). 122. a) D. L. J. Clive; Tetrahedron 34, 1049 (1980). 123. a) I. Fleming and I. Paterson; Synthesis 736 (1979). b) M. E. Jung, Y-G. Pan, M. W. Rathke, D. F. Sullivan, R. P. Woodbury; J. Org. Chem. 42, 3961 (1977). c) I. Ryn, S. Murai, Y. Hatayama, N. Sonoda; Tetrahedron Lett. 3455 (1978). d) Y. Ito, T. Hirao, and T. Saegusa; J. Org. Chem. 43, 1011 (1978). 124. D. Caine; Organic Reaction 23, 1 (1976). 125. G. Stork and S. D. Darling; J. Am. Chem. Soc. 86, 1761 (1964). 126. M. J. T. Robinson; Tetrahedron 21, 2475 (1965).  243  127. a) M. Karplus; .1. Chem. Phys. 30, 11(1959). b) M. Karplus; J. Am. Chem. Soc. 85, 2870 (1963). 128. E.C. Ashby and S. A. Noding; J. Am. Chem. Soc. 98, 8616 (1976). 129. a) H. C. Brown and S. Krishnamurthy; J. Am. Chem. Soc. 94, 7159 (1972). b) ibid.; J. Am. Chem. Soc. 98, 3383 (1976). 130. H. C. Brown; Organic Synthesis via Boranes; John Wiley & Sons, New York, 1975. 131. a) P. F. Viad, M. N. Koltsa, I. P. Dragalin, L. A. Zadorozhnaya, V. E. Sibirseva, and L. M. Sitnova; Russ.Chem. Rev. 58, 2037 (1988). b) N. M. Kruglikove and A. B. Pashkov; Plastichekie Massy 27-9 (1966). c) K. M. Saldadze; Mezhdunarod. Simpozium P0 Makromol. Khim. Sektsiya 3, 107-14 (1960); Dokiady, Moscow. d) F. T. Shostak, M. V. Vihikh, A. K. Sharova, and Z. G. Chufarova; Izvest. Sibir. Otdel. Akad. Nauk. S.S.S.R. No. 8, 69-74 (1960). 132. E. L. Eliel, N. L. Allinger, S. J. Angyal, and G. A. Morrison; Conformational Analysis; John Wiley & Sons, Inc., 1965; p. 301. 133. J. F. Biellman and G. Qurisson; Bull. Soc. Chim. France 331 and 341 (1962). 134. T. H. Lowry and K. S. Richardson; Mechanism and Theory in Organic Chemistry; Harper & Row, Publishers, New York, 1981; p. 531-534. 135. R. M. Coats; Tetrahedron Lett. 4143 (1967).  Chapter  4  136. J. M. Bobbitt and K. -P. Segebarth; Cyclopentanoid Terpene Derivatives ; W. I. Taylor and A. R. Battersby, eds; Marcel Dekker, Inc., New York, 1969; p. 1-146. 137. H. J. Liu and S. P. Majumdar; Synth. Comm. 5(2), 125 (1975). 138. C. R. Noller; The Chemistry of Organic Compounds; W. S. Saunders Company, Pilidelphia & London, 1965; p. 902-904. 139. a) A. P. Krapcho and A. J. Lovey; Tetrahedron Lett. 957 (1973). b) A. P. Kracho, J. F. Weimaster, J. M. Eldridge, E. G. E. Jahngen, Jr., A. J. Lovey, and W. P. Stephens; J. Org. Chem. 43, 138 (1978). 140. a) W. L. Mock and M. E. Hartman; J. Org. Chem. 42, 459 (1977). b) ibid.; J. Org. Chem. 42, 466 (1977). 141. a) L. M. Jackman and S. Sternhall; Applications ofNuclear magnetic Resonance spectrometry in Organic Chemistry; 2nd ed.; Pergman Press, Toronto, 1969; p. 228-229 and 286. b) M. Tandon, L. I. Wiebe, and E. E. Knaus; Can. J. Chem. 67, 1484 (1989). 142. a) J. K. M. Sanders and B. K. Hunter; Modern NMR Spectroscopy: a guide for chemists; Oxford University Press, 1987. 244  b) A. Derome; Modern NMR Techniques in Chemistry Research; Pergaman Press, 1987. 143. a) J. H. Noggle and R. E. Schirmer; The Nuclear Overhauser Effect: Chemical Applications; Academic Press, Inc., New York, 1971. b) I. H. Sadler; Nat. Prod. Rep. 5, 101-27 (1988). 144. G. Stork, P. Rosen, and N. L. Goldman; J. Am. Chem. Soc. 83, 2965 (1961). 145. D. Caine; Carbon-Carbon Bond Formation; R. L. Augustine, ed.; Marcell Dekker, Vol. 4, New York & Basel, 1979; P. 85-352. 146. H. C. Brown and K. Murray; J. Am. Chem. Soc. 81, 4108 (1959). 147. a) J. B. Pierce and H. M. Walborsky; J. Org. Chem. 33, 1962 (1968). b) B. Rickbom and S. E. Wood; J. Am. Chem. Soc. 93, 3940 (1971). c) W. A. G. Graham and F. G. A. Stone; Chem. and md. (London) 1096 (1957). 148. G. Stork and P. F. Hudrlik; J. Am. Chem. Soc. 90, 4462 & 4464 (1968). 149. R. D. Smith and H. E. Simmons; Org. Synth. 41, 72 (1961). 150. a) G. M. Rubottom and M. I. Lopez; J. Org. Chem. 38, 2097 (1973). b) S. Murai, T. Aya, and N. Sonoda; J. Org. Chem. 38, 4354 (1973). 151. J. M. Conia and G. Girard; Tetrahedron Lett. 2771 (1973). 152. T. Hasegawa, H. Niwa, and K. Yamada; Chem. Lett. 1385 (1985). 153. S. Danishefshy; Accounts of Chemical Research; 12, 66-72 (1979). 154 TH. J. DeBoer and H. J. Backer; Org. Synth. 36, 16 (1956). 155. C. Mioskowski, S. Manna, 3. R. Faick; 24, 5521 (1983). 156. George Polya; How to Solve It: a new aspect ofmathematical method; Princeton University Press, 1971. 157. J. P. Kutney, G. M. Hewitt, T. Kurihara, P. J. Salisbury, R. D. Sindelar, K. Stuart, P. M. Townsley, W. T. Chalmers, G. G. Jacoli; Can. J. Chem. 29, 2677-1683 (1981). 158. T. Hudlicky, L. Radesca-Kwart, L. Li, and T. Bryant; Tetrahedron Lett. 29, 3283 (1988). 159. a) W. Adams, Ruggerro, and J. 0. Edwards; Acc. Chem. Res. 22, 205-211 (1989). b) L. Hill, ed.; Nouv. J. Chim. 13, 645-793 (1990). 160. S. Hanessian; Total Synthesis of Natural Products: ‘The Chiron Approach’; Pergman Press, 1983. 161. a) T. Money; Natural Product Reports; 253-289 (1985). b) T. Money; Studies in Natural Products Chemistry; Elsevier, vol. 4, 1989; p. 625697. 162. D. D. Perrin, W. L. Armarego, D. R. Perrin; Purication of laboratory Chemicals; 245  2nd Edition; Pergamon Press, 1980. 163. W. C. Still, M. Kahn, and A. Mitra; J. Org. Chem. 43, 2923 (1978). 164. a) M. S. Carpenter and W. M. Easter; J. Chem. Soc. 20, 401 (1955). b) E. Zavarin and A. B. Anderson; J. Chem. Soc. 20, 83 (1955). 165. S. W. C. Watson and J. F. Eastham; J. Organometal. Chem. 9,165-168 (1967). 166. a) S. Motherwell and W. Clegg; PLUTO: programfor plotting molecular and crystal structures; University of Cambridge, England, 1978. b) C. K. Johnson; ORTEP II: Report ORNL-5138; Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1976.  246  Appendix 1.  X-ray Structure Report on Epoxide 147 A.  Crystal Data  Empirical Formula  0 1 C 2 H 2 6 8  Formula Weight  252.40  Crystal Color, Habit  colorless, prism  Crystal Dimensions (mm)  0.300 X 0.400  Crystal System  monoclinic  No. Reflections Used for Unit Cell Determination (28 range)  25 (100.7  Omega Scan Peak Width at Half—height  0.37  —  x  0.500  109.00)  Lattice Parameters: a—  6.767 (1)A 9.616 (1)A 12.205 (l)A 98.84 (1)° 784.7 (2)A 3  Space Group  1 P2  Z value  2  Dcalc  1.068 g/cm 3  F 0 00  280  ( CuK)  4.97 cm B.  (44)  Intensity Measur e men t S  Di ffractometer  Rigaku AFC6S  Radiation  CuK (X  Tempe rature  21°C  Take—off Angle  6.0°  Detector Aperture  6.0 mm horizontal 6.0 mm vertical  247  —  1.54178  A)  Crystal to Detector Distance  285 mm  Scan Type  >— 2  Scan Rate  32.0°/mm (in omega) (8 rescans)  e -  Scan Width  (1.15  +  0.20 tane)°  154.6° No. of Reflections Measured  Total: 1929 Unique: 1776 (Rjt  Corrections  —  .035)  Lorentz—polarization Absorption (trans. factors: 0.82 — 1.00) Secondary Extinction (coefficient: 0.70718E—04) C.  Structure Solution and Refinement  Structure Solution  Direct Methods  Refinement  Full—matrix least—squares  Function Minimized  Z w (IFol  Least—squares Weights  /a Fe 2 4Fo ( ) 2  p—factor  0.04  Anomalous Dispersion  All non—hydrogen atoms  No. Observations (I>4.OOu(I)) No. Variables Reflection/Parameter Ratio  1656 167 9.92  Residuals:  0.045; 0.067  R;  Goodness of Fit Indicator  2.83  Max Shift/Error in Final Cycle  0.01  Maximum Peak in Final Diff. Map Minimum Peak in Final Diff. Map  248  0.24 3 e/A —0.12 e/A 3  —  2 IFCI)  H22  H23  H26  H14 Hi:  Figure 46 Single Crystal X-Ray Structure of Epoxide 147 (PLUTO Drawing)*  *  The numbering of carbon atoms here is different from that used in the Discussion (Section 2.2.7.). 249  Figure 47 The Unit Cell Structure of Epoxide 147 (Packing Diagram)  250  Table 6 Final Atomic Coordinates (fractional) and B (A ) of Epoxide 147 2  atom  x  y  z  B  0(1)  0.1042(3)  0.3562  0.3502(2)  5.46(7)  0(2)  0.1848(3)  0.0318(3)  0.5658(1)  5.04(7)  C(1)  0.2368(3)  0.1703(3)  0.2154(1)  4.00(7)  C(2)  0.3048(4)  0.0589(4)  0.1368(2)  5.5(1)  C(3)  0.5313(5)  0.0506(5)  0.1542(3)  7.7(2)  C(4)  0.6307(4)  0.1929(6)  0.1443(3)  7.9(2)  C(S)  0.5753(3)  0.2917(5)  0.2316(2)  6.6(1)  C(6)  0.3497(3)  0.3117(3)  0.2235(2)  4.56(8)  C(7)  0.3116(3)  0.3663(3)  0.3346(2)  4.81(8)  C(S)  0.2487(3)  0.2547(3)  0.4026(2)  3.87(6)  C(9)  0.2407(3)  0.1229(3)  0.3363(2)  4.23(7)  C(10)  0.2170(8)  —0.0818(5)  0.1588(3)  9.3(2)  C(11)  0.2282(5)  0.0955(6)  0.0149(2)  7.6(2)  C(12)  0.2684(6)  0.4159(5)  0.1330(3)  7.7(2)  C(13)  0.2878(4)  0.2592(3)  0.5272(2)  4.67(8)  C(14)  0.1451(4)  0.1742(3)  0.5876(2)  4.41(8)  C(15)  —0.0724(4)  0.2082(4)  0.5470(2)  5.7(1)  C(16)  0.1953(5)  0.1997(4)  0.7123(2)  6.7(1)  251  Table 7 Hydrogen Atom Coordinates (fractional) and Bj 0 atom  x  H(1)  0.100(4)  H(2)  0.0969  H(3)  ) of Epoxide 147 2 (A B so  z 0.599(3)  5.3(6)  0.1913  0.1864  4.8  0.5773  0.0131  0.2284  9.2  H(4)  0.5719  —0.0120  0.0984  9.2  H(5)  0.7762  0.1813  0.1549  9.5  H(6)  0.5850  0.2316  0.0705  9.5  H(7)  0.6273  0.2543  0.3050  7.9  H(8)  0.6370  0.3823  0.2223  7.9  H(9)  0.3944  0.4422  0.3702  5.8  M(10)  0.1198  0.0699  0.3437  5.1  H(11)  0.3589  0.0655  0.3608  5.1  11(12)  0.0707  —0.0758  0.1462  11.2  H(13)  0.2601  —0.1512  0.1085  11.2  11(14)  0.2635  —0.1094  0.2357  11.2  11(15)  0.2920  0.1816  —0.0047  9.1  11(16)  0.2610  0.0198  —0.0330  9.1  11(17)  0.0829  0.1084  0.0048  9.1  11(18)  0.3456  0.5024  0.1438  9.3  11(19)  0.2798  0.3767  0.0601  9.3  H(20)  0.1276  0.4355  0.1371  9.3  H(21)  0.4236  0.2244  0.5513  5.6  H(22)  0.2794  0.3565  0.5499  5.6  11(23)  —0.0979  0.3058  0.5631  11(24)  —0.1022  0.1924  0.4669  6.9  11(25)  —0.1578  0.1483  0.5848  6.9  11(26)  0.1112  0.1403  0.7513  8.1  11(27)  0.3364  0.1778  0.7374  8.1  11(28)  0.1705  0.2975  0.7282  8.1  —0.016(4)  252  -  -  -  6.9  Table 8 Bond Lengths (A) of Epoxide 147 with Estimated Standard Deviations in Parentheses  atom  atom  distance  atom  atom  distance  0(1)  C(7)  1.448(3)  C(4)  C(S)  1.517(5)  0(1)  C(8)  1.459(3)  C(5)  C(6)  1.527(3)  0(2)  C(14)  1.428(3)  C(6)  C(7)  1.513(3)  C(1)  C(2)  1.554(3)  C(6)  C(12)  1.530(4)  C(1)  C(6)  1.55S(3)  C(7)  C(8)  1.460(3)  C(1)  C(9)  1.541(3)  C(8)  C(9)  1.500(3)  C(2)  C(3)  1.517(4)  C(8)  C(13)  1.503(3)  C(2)  C(10)  1.519(5)  C(13)  C(14)  1.537(3)  C(2)  C(11)  1.539(4)  C(14)  C(15)  1.515(4)  C(3)  C(4)  1.537(7)  C(14)  C(16)  1.527(3)  253  Table 9 Bond Angles (deg) of Epoxide 147 with Estimated Standard Deviations in Parentheses  atom  atom  atom  angle  atom  atom  atom  angle  C(7)  0(1)  C(8)  60.3(1)  C(7)  C(6)  C(12)  109.1(2)  C(2)  C(1)  C(6)  116.9(2)  0(1)  C(7)  C(6)  113.4(2)  C(2)  C(1)  C(9)  115.1(2)  0(1)  C(7)  C(8)  60.2(1)  C(6)  C(1)  C(9)  105.2(2)  C(6)  C(7)  C(8)  111.0(2)  CCI)  C(2)  C(3)  109.8(2)  0(1)  C(8)  C(7)  59.5(1)  C(1)  C(2)  C(10)  110.1(2)  0(1)  C(8)  C(9)  111.3(2)  C(1)  C(2)  C(11)  110.6(3)  0(1)  C(8)  C(13)  115.2(2)  C(3)  C(2)  C(10)  110.0(3)  C(7)  C(8)  C(9)  107.7(2)  C(3)  C(2)  C(11)  109.3(2)  C(7)  C(8)  C(13)  122.2(2)  C(10)  C(2)  C(11)  107.0(3)  C(9)  C(8)  C(13)  123.8(2)  C(2)  C(3)  C(4)  112.8(3)  C(1)  C(9)  C(8)  105.1(2)  C(3)  C(4)  C(5)  110.1(2)  C(8)  C(13)  C(14)  116.4(2)  C(4)  C(S)  C(6)  112.7(2)  0(2)  C(14)  C(13)  105.7(2)  C(1)  C(6)  C(S)  111.8(2)  0(2)  C(14)  C(15)  110.3(2)  CU)  C(6)  C(7)  102.2(2)  0(2)  C(14)  C(16)  108.7(2)  C(1)  C(6)  C(12)  114.1(2)  C(13)  C(14)  C(15)  112.3(2)  C(5)  C(6)  C(7)  107.0(2)  C(13)  C(14)  C(16)  109.4(2)  C(S)  C(6)  C(12)  112.0(2)  C(15)  C(14)  C(16)  110.3(2)  254  Table 10 Torsional or Conformational Angles (deg) of Epoxide 147 (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  angle  0(1) C(7) C(6) C(1)  —47.1(2)  C(1) C(2) C(10)H(13)  180  0(1) C(7) C(6) C(S)  —164.7(2)  C(1) C(2) C(10 )H(14)  —60  0(1) C(7) C(6) C(12)  74.0(3)  C(1) C(2) C ( 11 ) H C 15)  66  0(1) C(7) C(8) C(9)  104.7(2)  C(1) C(2) C( 11 )H( 16)  —174  0(1) C(7) C(8) C(13) —102.3(2)  C(1) C(2) C(11)H(17)  —54  48.9(3)  0(1) C(8) C(7) C(6)  —105.6(2)  C(1) C(6) C(5) C(4)  0(1) C(8) C(7) H(9)  109  CU) C(6) C(S) H(7)  —72  C(1) C(6) C(S) H(8)  169  0(1) C(8) C(9) C(1)  46.1(2)  18.4(2)  0(1) C(8) C(9) H(1O)  —73  C(1) C(6) C(7) C(8)  0(1) C(8) C(9) H(11)  165  C(1) C(6) C(7) H(9)  164  0(1) C(8) C(13)C(14)  86.4(2)  C(1) C(6) C( 12 )H( 18)  179  0(1) C(8) C(13)H(21)  —153  C(1) C(6) C(12)H(19)  —61  0(1) C(8) C(13 )H(22)  —35  CU) C(6) C(12)H(20)  59  0(2) C(14) C(13)C(8)  C(1) C(9) C(8) C(7)  67.7(3)  —17.3(2)  C(1) C(9) C(8) C(13) —169.8(2)  0(2) C(14)C(13)H(21)  —53  0(2) C(14)C(13)H(22)  —171  C(2) C(1) C(6) C(S)  —43.3(3)  0(2) C(14)C(15)H(23)  179  C(2) C(1) C(6) C(7)  —157.4(2)  0(2) C(14)C(15)H(24)  —61  C(2) C(1) C(6) C(12)  0(2) C(14)C(15)H(25)  59  0(2) C(14)C(16)H(26)  —61  C(2) C(1) C(9) H(10)  —82  0(2) C(14)C(16)H(27)  59  C(2) C(1) C(9) H(11)  39  0(2) C(14)C(16)H(28)  179  C(2) C(3) C(4) C(S)  60.9(4)  C(1) C(2) C(3) C(4)  —53.1(3)  C(2) C(3) C(4) H(S)  —179  C(1) C(2) C(3) H(3)  67  C(2) C(3) C(4) H(6)  —59  C(1) C(2) C(3) H(4)  —174  C(3) C(2) C(1) C(6)  45.2(3)  60  C(3) C(2) C(1) C(9)  —79.1(3)  C(1) C(2) C(10)H(12)  C(2) C(1) C(9) C(8)  85.0(3) 158.8(2)  The sign is positive if when looking from atom 2 to atom 3 a clock— wise motion of atom 1 would superimpose it on atom 4.  255  Table 10 Torsional or Conformational Angles (deg) of Epoxide 147 (4)  (cont) angle  angle  (1)  C(9) C(1) C(2) C(1O)  42.2(3)  C(13)C(14)O(2) H(1)  C(9) C(1) C(2) C(l1)  160.2(2)  C(13)C(14)C(lS)H(23)  —63  C(9) C(1) C(6) C(12) —145.9(2)  C(13)C(14)C(15)H(24)  57  C(13)C(14)C(15)H(25)  177  (1)  (2)  (3)  (4)  C(9) C(8) C(7) H(9)  —146  (2)  (3)  —178(2)  C(9) C(8) C(13)C(14)  —56.5(3)  C(13)C(14)C(16)H(26)  —176  C(9) C(S) C(13)H(21)  64  C(13)C(14)C(16)H(27)  —56  C(9) C(S) C(13)H(22)  —178  C(13)C(14)C(16)H(28)  64  C(10)C(2) C(1) H(2)  —75  C(15)C(14)O(2) B(1)  C(10)C(2) C(3) H(3)  —54  C(15)C(14)C(13)H(21)  —174  C(10)C(2) C(3) H(4)  65  C(15)C(14)C(13)H(22)  68  C(10)C(2) C(11)H(15)  —175  C(15)C(14)C(16)H(26)  60  C(10)C(2) C(11)H(.16)  —55  C(15)C(14)C(16)H(27)  —180  C(10)C(2) C(11)H(17)  65  C(15)C(14)C(16)H(28)  —60  C(11)C(2) C(1) H(2)  43  C(16)C(14)O(2) H(1)  64(2)  C(11)C(2) C(3) 11(3)  —171  C(16)C(14)C(.13)H(21)  64  C(11)C(2) C(3) 11(4)  —52  C(16)C(14)C(13)H(22)  —54  C(11)C(2) C(10)H(12)  —61  C(16)C(14)C(15)H(23)  59  C(11)C(2) C(10)H(13)  59  C(16)C(14)C(15)H(24)  179  C(11)C(2) C(10)H(14)  179  C(16)C(14)C(15)H(25)  —61  C(12)C(6) C(1) 11(2)  —33  11(2) C(1) C(9) H(10)  35  C(12)C(6) C(S) 11(7)  159  11(2) C(1) C(9) H(11)  157  C(12)C(6) C(S) H(S)  40  11(3) C(3) C(4) 11(5)  61  C(12)C(6) C(7) 11(9)  —75  11(3) C(3) C(4) 11(6)  —180  C(13)C(8) C(7) 11(9)  7  11(4) C(3) C(4) H(S)  —58  C(13)C(8) C(9) 11(10)  71  11(4) C(3) C(4) 11(6)  61  C(13)C(8) C(9) H(11)  —50  H(S) C(4) C(S) H(7)  —57  —57(2)  The sign is positive if when looking from atom 2 to atom 3 a clock— wise motion of atom 1 would superimpose it on atom 4.  256  Table 10 Torsional or Conformational Angles (deg) of Epoxide 147 (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  (cont) angle  164  C(6) C(1) C(9) 11(10)  148  C(3) C(2) C(10)H(12)  —179  C(6) C(1) C(9) 11(11)  —91  C(3) C(2) C(i0)H(13)  —59  C(6) C(5) C(4) H(S)  —178  C(3) C(2) C(10)H(14)  61  C(6) C(S) C(4) H(6)  62  C(3) C(2) C(11)H(15)  —55  C(6) C(7) 0(1) C(8)  101.7(2)  C(3) C(2) C(11)H(16)  65  C(6) C(7) C(S) C(9)  —0.9(2)  C(3) C(2) C(11)H(17)  —175  C(3) C(2) C(1) H(2)  C(S) C(7) C(8) C(13)  152.0(2)  C(7) 0(1) C(S) C(9)  —98.6(2)  C(7) 0(1) C(8) C(13)  113.9(2)  C(7) C(S) C(1) C(9)  —28.3(2)  C(4) C(3) C(2) C(10) —174.5(3)  C(7) C(6) C(1) 11(2)  84  C(4) C(3) C(2) C(11)  68.3(4)  C(7) C(6) C(5) 11(7)  39  C(4) C(5) C(6) C(7)  160.0(3)  C(7) C(6) C(S) 11(8)  —80  C(4) C(S) C(6) C(12)  —80.5(4)  C(7) C(6) C(12)H(18)  66  C(S) C(4) C(3) 11(3)  —60  C(7) C(6) C(.12)H(19)  —174  C(5) C(4) C(3) 11(4)  —179  C(7) C(6) C(12)H(20)  —54  C(S) C(6) C(1) C(9)  85.8(2)  C(7) C(8) C(9) 11(10)  —137  C(5) C(S) C(1) 11(2)  —162  C(7) C(S) C(9) 11(11)  102  C(3) C(4) C(S) C(6)  —57.9(4)  C(3) C(4) C(S) 11(7)  63  C(3) C(4) C(S) 11(8)  —178  C(S) C(6) C(7) C(8)  —99.1(2)  C(7) C(8) C(13)C(14)  154.9(2)  C(S) C(6) C(7) 11(9)  46  C(7) C(8) C(13)H(21)  —84  C(S) C(6) C(12)H(18)  —52  C(7) C(8) C(13)H(22)  34  C(S) C(6) C(12)H(19)  68  C(S) C(6) C(12)H(20)  —172  C(8) 0(1) C(7) 11(9) C(S) C(7) C(6) C(12)  C(6) C(1) C(2) C(10)  166.5(3)  C(S) C(9) C(1) 11(2)  C(6) C(1) C(2) C(11)  —75.S(2)  C(8) C(13)C(14)C(15)  C(6) C(1) C(9) C(S)  —109 139.S(3) —84 —52.6(3)  C(8) C(13)C(14)C(16) —175.4(2)  28.6(2)  The sign is positive if when looking from atom 2 to atom 3 a clock— wise motion of atom 1 would superimpose it on atom 4.  257  Table 10 Torsional or Conformational Angles (deg) of Epoxide 147 (1)  (2)  (3)  (4)  angle  H(5) C(4) C(5) R(8)  62  H(6) C(4) C(5) R(7)  —177  H(6) C(4) C(5) 14(8)  —58  (1)  (2)  (3)  (4)  (cont) angle  The sign is positive if when looking from atom 2 to atom 3 a clock wise motion of atom 1 would superimpose it on atom 4.  258  Appendix 2.  X-ray Structure Report on Diol 285 A.  Crystal Data  Empirical Formula  0 1 C 3 H 2 7 2  Formula Weight  268.44  Crystal Color, Habit  colorless, needle  Crystal Dimensions  0.120  (mm)  x  0.180 X 0.480  Crystal System  orthorhombi c  No. Reflections Used for Unit Cell Determination (2 range)  25  Omega Scan Peak Width at Half—height  0.37  ( 57.4  81.8°)  —  Lattice Parameters: a b c  —  V  —  10.730 (2)A 20.411 (2)A 7.484 (l)A  = —  1639.0  Space Group  2 1 P2  Z value  4  Dcalc  1.088 g/cm 3  F 0 00  600  1 ’ (CuK)  4.97 cm 1 B.  3 (3)A  (*19)  Intensity Measurements  Diffractometer  Rigaku AFC6S  Radiation  CuK (X  Temperature  21°C  Take—off Angle  6.0°  Detector Aperture  6.0 mm horizontal 6.0 mm vertical  Crystal to Detector Distance  285 mm  259  =  1.54178 A)  Scan Type  -  2e  Scan Rate  16.0°/mm (in omega) (8 rescans)  Scan Width  (0.94  0.20 tan8)°  155.00  max 2 e No.  +  Total:  of Reflections Measured  1921  Lorentz—polari zation Absorption 1.07) 0.92 (trans. factors: Secondary Extinction 0.26140E—04) (coefficient:  Corrections  —  C.  Structure Solution and Refinement  Structure Solution  Direct Methods  Refinement  Full—matrix least—squares  Function Minimized  E w (IFol  Least—squares Weights  ) 2 ( Fo / 2 4Fo  p—factor  0. 025  Anomalous Dispersion  All non—hydrogen atoms  No. Observations (I>3.00(I)) No. Variables Reflection/Parameter Ratio  1570 181 8.67  Residuals:  0.036;  R;  Goodness of Fit Indicator  1.93  Max Shift/Error in Final Cycle  0.002  Maximum Peak in Final Diff. Map Minimum Peak in Final Diff. Map  260  —  0.047  3 0.18 e/A 3 —0.11 e/A  2 (Fc()  H2  02 HiS H27  -415  H18 H29  C12  H13  C16 Cli co  H14  H17  H28  HiS C2 H32 C13  Hil H12  ci  H3  31  C3 C17  CS  H20 Cs CS  H6 HG  C4  H30  Hi  23 CS  C14  HID  HS  01  C7  H24  CiS  H26  H21  H22  H7  H25  Figure 48 Single Crystal X-ray Structure of Diol 285 (PLUTO Drawing)* *  The numbering of carbon atoms here is different from that used in the Discussion (Section 4.1.3.). 261  L’J  :1  0  0  C.  C)  C,)  CD  (-)  C.  CD  H  ‘71  ) of Diol 285 2 Table 11 Final Atomic Coordinates (fractional) and B (A  x  atom  Y  z  Beg  0(1)  0.3431(2)  0.47048(9)  0.7356(2)  4.39(7)  0(2)  0.3951(1)  0.51200(9)  0.0739(2)  4.28(7)  C(1)  0.4487(2)  0.3707(1)  0.2595(3)  3.66(9)  C(2)  0.5054(2)  0.4402(1)  0.2899(3)  3.27(8)  C(3)  0.4447(2)  0.4791(1)  0.4452(3)  3.35(8)  C(4)  0.4255(2)  0.4375(1)  0.6131(3)  3.49(8)  C(S)  0.3754(2)  0.3697(1)  0.5802(3)  3.82(9)  C(6)  0.4429(2)  0.3302(1)  0.4364(3)  3.8(1)  C(7)  0.5628(2)  0.2940(1)  0.5021(4)  4.5(1)  C(8)  0.6244(3)  0.2597(1)  0.3415(5)  6.0(1)  C(9)  0.6465(3)  0.3047(2)  0.1864(4)  6.4(2)  C(10)  0.5229(3)  0.3310(1)  0.1208(4)  5.6(1)  C(11)  0.3155(2)  0.3759(1)  0.1845(4)  4.8(1)  C(12)  0.5111(2)  0.4819(1)  0.1214(3)  4.1(1)  C(13)  0.5125(2)  0.5433(1)  0.5005(4)  4.9(1)  C(14)  0.6601(2)  0.3372(1)  0.5965(4)  5.4(1)  C(15)  0.5242(3)  0.2405(1)  0.6344(5)  6.7(2)  C(16)  0.4755(3)  0.6043(1)  0.3994(5)  6.5(1)  C(17)  0.6530(3)  0.5358(2)  0.5186(51  6.6(2)  eq  —  263  ) of Diol 285 2 (A  Table 12 Hydrogen Atom Coordinates (fractional) and atom  x  y  z  H(1)  0.269(3)  0.472(1)  0.694(4)  6.2(7)  H(2)  0.372(3)  0.500(2)  —0.029(4)  5.9(8)  H(3)  0.5922  0.4329  0.3256  3.9  H(4)  0.3616  0.4920  0.4041  4.0  H(5)  0.5066  0.4329  0.6720  4.2  H(6)  0.2879  0.3738  0.5445  4.6  H(7)  0.3808  0.3452  0.6925  4.6  H(8)  0.3848  0.2946  0.4087  4.6  H(9)  0.5700  0.2240  0.3022  7.2  H(10)  0.7047  0.2418  0.3800  7.2  H(11)  0.6876  0.2806  0.0898  7.7  H(12)  0. 6996  0.3413  0.2242  7.7  H(13)  0.5390  0.3594  0.0178  6.7  H(14)  0.4716  0.2938  0.0834  6.7  H(15)  0.3172  0.3999  0.0711  5.8  H(16)  0.2626  0.3993  0.2701  5.8  H(17)  0.2820  0.3319  0.1646  5.8  H(18)  0.5729  0.5166  0.1401  4.9  H(19)  0.5376  0.4540  0.0221  H(20)  0.4838  0.5514  0.6228  5.9  H(21)  0.7295  0.3098  0.6375  6.5  H(22)  0.6215  0.3589  0.6993  6.5  H(23)  0.6913  0.3703  0.5128  6.5  H(24)  0.4663  0.2102  0.5759  8.0  264  Bso  -  4.9  Table 12 Hydrogen Atom Coordinates (fractional) and B 0  ) of Diol 285 2 (A 2  atom  x  y  H(25)  0.4832  0.2606  0.7379  8.0  H(26)  0.5982  0.2165  0.6741  8.0  H(27)  0.5116  0.6029  0.2791  7.8  H(28)  0.5064  0.6430  0.4628  7.8  H(29)  0.3844  0.6064  0.3905  7.8  H(30)  0.6718  0.4953  0.5840  8.0  H(31)  0.6869  0.5733  0.5840  8.0  H(32)  0.6908  0.5337  0.3995  8.0  Biso  Table 13 Bond Lengths (A) of Diol 285 with Estimated Standard Deviations in Parentheses  atom  atom  distance  atom  at ozn  distance  0(1)  C(4)  1.440(2)  C(4)  C (5 )  1.505(3)  0(2)  C(12)  1.432(3)  C(5)  C(6)  1.527(3)  C(1)  C(2)  1.560(3)  C(6)  C(7)  1.563(3)  C(1)  C(6)  1.562(3)  C(7)  C(8)  1.540(4)  CU)  C(10)  1.539(3)  C(7)  C(14)  1.538(4)  C(1)  C(11  1.539(3)  C(7)  C(15)  1.532(4)  C(2)  C(3)  1.552(3)  C(8)  C(9)  1.500(4)  C(2)  C(12)  1.523(3)  C(9)  C(10)  1.512(5)  C(3)  C(4)  1.531(3)  C(13)  C(16)  1.510(4)  C(3)  C(13)  1.553(3)  C(13)  C(17)  1.522(4)  265  (cont.)  Table 14 Bond Angles (deg) of Diol 285 with Estimated Standard Deviations in Parentheses  atom  atom  atom  angle  atom  atom  atom  angle  C(2)  C(1)  C(6)  111.9(2)  C(1)  C(6)  C(S)  109.7(2)  C(2)  C(l)  C(1O)  112.0(2)  C(1)  C(6)  C(7)  119.0(2)  C(2)  C(1)  C(11)  110.6(2)  C(5)  C(6)  C(7)  114.7(2)  C(6)  C(1)  C(10)  108.3(2)  C(6)  C(7)  C(8)  108.8(2)  C(6)  C(1)  C(11)  108.0(2)  C(6)  C(7)  C(14)  115.6(2)  C(10)  C(1)  C(11)  105.7(2)  C(6)  C(7)  C(15)  108.5(2)  C(1)  C(2)  C(3)  114.3(2)  C(8)  C(7)  C(14)  109.1(2)  C(1)  C(2)  C(12)  113.8(2)  C(8)  C(7)  C(1S)  107.2(2)  C(3)  C(2)  C(12)  110.5(2)  C(14)  C(7)  C(15)  107.2(2)  C(2)  C(3)  C(4)  112.8(2)  C(7)  C(8)  C(9)  113.2(2)  C(2)  C(3)  C(13)  115.8(2)  C(8)  C(9)  C(10)  109.3(3)  C(4)  C(3)  C(13)  108.2(2)  C(1)  C(10)  C(9)  114.9(2)  0(1)  C(4)  C(3)  110.2(2)  0(2)  C(12)  C(2)  114.2(2)  0(1)  C(4)  C(S)  108.4(2)  C(3)  C(13)  C(16)  116.0(2)  C(3)  C(4)  C(S)  115.1(2)  C(3)  C(13)  C(17)  113.8(2)  C(4)  C(S)  C(6)  115.6(2)  C(16)  C(13)  C(17)  112.8(2)  266  Table 15 Torsional or Conformational Angles (deg) of Diol 285 (1)  (2)  (3)  (4)  0(1) C(4) C(3) C(2) 0(1) C(4) C(3) C(13) 0(1) C(4) C(3) H(4)  angle  (1)  —166.2(2) 64.4(2) —SO  (2)  (3)  (4)  C(2) C(1) C(1O)C(9)  angle 74.5(3)  C(2) C(1) C(10)H(13)  —46  C(2) C(1) C(10)H(14)  —165  0(1) C(4) C(S) C(6)  173.2(2)  C(2) C(1) C(11)H(15)  59  0(1) C(4) C(S) H(6)  52  C(2) C(1) C(11)H(16)  —61  0(1) C(4) C(S) H(7)  —66  C(2) C(1) C(11)H(17)  179  0(2) C(12)C(2) C(1)  79.5(2)  C(2) C(3) C(4)C(5)  -43.2(2)  0(2) C(12)C(2) C(3)  —50.6(2)  C(2) C(3) C(4) H(S)  77  0(2) C(12)C(2) H(3)  —165  C(2) C(3) C(13)C(16)  87.6(3)  C(1) C(2) C(3) C(4)  44.7(2)  C(2) C(3) C(13)C(17)  —45.7(3)  170.1(2)  C(2) C(3) C(13)H(20)  —159  —72  C(2) C(12)0(2) H(2)  —121(2)  C(1) C(2) C(12)H(18)  —160  C(3) C(2) C(1) C(6)  —50.9(2)  C(1) C(2) C(12)H(19)  —41  C(1) C(2) C(3) C(13) C(1) C(2) C(3) H(4)  C(3) C(2) C(1) C(10) —172.7(2)  C(1) C(6) C(S) C(4)  —53.3(2)  C(3) C(2) C(1) C(11)  69.7(2)  C(1) C(6) C(S) H(6)  68  C(3) C(2) C(12)H(18)  70  C(1) C(6) C(5) H(7)  —174  C(3)  C(2) C(12)H(19)  —171  —44.2(3)  C(3)  C(4) 0(1) H(1)  69(2)  79.0(3)  C(3)  C(4) C(S) C(6)  49.3(2)  C(1) C(6) C(7) C(8) C(1) C(6) C(7) C(14)  C(1) C(6) C(7) C(15) —160.6(2)  C(3) C(4) C(5) H(6)  —72  C(1) C(10)C(9) C(8)  60.5(3)  C(3) C(4) C(5) H(7)  170  C(1) C(10)C(9) B(11)  —179  C(3) C(13)C(16)H(27)  —74  C(1) C(10)C(9) H(12)  —59  C(3) C(13)C(16)H(28)  166  C(2) C(1) C(6) C(S)  53.2(2)  C(3) C(13)C(16)H(29)  46  C(2) C(1) C(6) c(7)  —81.6(2)  C(3) C(13)C(17)H(30)  —45  C(3) C(13)C(17)H(31)  —165  C(2) C(1) C(6) H(8)  164  The sign is positive if when looking from atom 2 to atom 3 a clock— wise motion of atom 1 would superimpose it on atom 4.  267  Table 15 Torsional or Conformational Angles (deg) of Diol 285 (1)  (2)  (3)  (4)  angle  C(3) C(13)C(17)H(32)  75  C(4) C(3) C(2) C(12)  174.6(2)  (1)  (2)  (3)  (4)  (cont.) angle  C(6) C(7) C(8) H(9)  —68  C(6) C(7) C(8) H(10)  173  C(6) C(7) C(14)H(21)  177  C(4) C(3) C(13)C(16) —144.7(2)  C(S) C(7) C(14)H(22)  57  C(4) C(3) C(13)C(17)  C(S) C(7) C(14)H(23)  —63  C(6) C(7) C(15)H(24)  58  C(4) C(3) C(2) H(3)  C(4) C(3) C(13)H(20)  —71  82.0(3) —31  C(4) C(5) C(6) C(7)  83.6(2)  C(6) C(7) C(15)H(25)  —62  C(4) C(S) C(6) H(8)  —164  C(6)  C(7) C(15)H(26)  178  C(S) C(4) 0(1) H(1)  —58(2)  C(7) C(6) C(1) C(1O)  42.3(3)  C(5) C(4) C(3) C(13) —172.6(2)  C(7) C(6) C(1) C(11)  156.4(2)  C(S) C(4) C(3) H(4)  73  C(7) C(6) C(S) H(6)  —1SS  C(S) C(6) C(1) C(10)  177.2(2)  C(7) C(6) C(S) H(7)  —37  C(S) C(6) C(1) C(11)  —68.8(2)  C(7) C(8) C(9) C(10)  —61.5(3)  —176.9(2)  C(7) C(8) C(9) H(11)  179  C(S) C(6) C(7) C(14)  —53.7(3)  C(7) C(8) C(9) H(12)  58  C(S) C(6) C(7) C(15)  66.7(3)  C(S) C(6) C(7) C(8)  C(8)  C(S) C(1) C(2) C(12) —179.1(2)  C(7) C(6) H(8)  71  C(8) C(7) C(14)H(21)  —SO  C(6) C(1) C(2) H(3)  65  C(8) C(7) C(14)H(22)  —180  C(6) C(1) C(10)C(9)  —49.4(3)  C(8) C(7) C(14)H(23)  60  C(S) C(1) C(10)H(13)  —170  C(8) C(7) C(15)H(24)  —59  C(6) C(1) C(10)H(14)  71  C(8) C(7) C(15)H(2S)  —179  C(6) C(1) C(11)H(1S)  —178  C(8) C(7) C(1S)H(26)  61  C(6) C(1) C(11)H(16)  62  C(8) C(9) C(10)H(13)  —179  C(S) C(1) C(11)H(17)  —58  C(8) C(9) C(10)H(14)  —SO  C(S) C(S) C(4) H(S)  —71  C(9) C(8) C(7) C(14)  —74.5(3)  C(9) C(8) C(7) C(15)  169.7(2)  C(6) C(7) C(8) C(9)  52.5(3)  The sign is positive if when looking from atom 2 to atom 3 a clock wise motion of atom 1 would superimpose it on atom 4.  268  Table 15 Torsional or Conformational Angles (deg) of IDiol 285  (1)  (2)  (3)  (4)  angle  (1)  (2)  (3)  (4)  (cont.)  angle  C(9) C(10)C(1) C(11) —164.9(2)  C(15)C(7) C(8) H(10)  —70  C(10)C(l) C(2) C(12)  C(15)C(7) C(14)H(21)  56  59.1(2)  C(10)C(1) C(2) H(3)  —57  C(15)C(7) C(14)H(22)  —64  C(10)C(1) C(6) H(8)  —72  C(15)C(7) C(14)H(23)  176  C(10)C(1) C(11)H(15)  —62  C(16)C(13)C(3) H(4)  —31  C(10)C(1) C(11)H(16)  178  C(16)C(13)C(17)H(30)  —179  C(10)C(1) C(11)R(17)  58  C(16)C(13)C(17)H(31)  61  C(10)C(9) C(8) H(9)  59  C(16)C(13)C(17)H(32)  —59  C(10)C(9) C(8) H(10)  178  C(11)C(1) C(2) C(12)  —58.6(2)  C(17)C(13)C(3) H(4)  —164  C(17)C(13)C(16)H(27)  59  C(11)C(1) C(2) H(3)  —174  C(17)C(13)C(16)H(28)  —61  C(1l)C(1) C(6) H(8)  42  C(17)C(13)C(16)H(29)  179  C(11)C(1) C(10)H(13)  74  H(1) 0(1) C(4) H(5)  C(11)C(1) C(10)H(14)  —44  H(2) 0(2) C(12)H(18)  118  C(12)C(2) C(3) C(13)  —60.0(2)  14(2) 0(2) C(12)H(19)  0  —174  C(12)C(2) C(3) 14(4)  58  14(3) C(2) C(3) 14(4)  172  C(13)C(3) C(2) 14(3)  54  14(3) C(2) C(12)H(18)  —44  C(13)C(3) C(4) H(S)  —53  14(3) C(2) C(12)H(19)  75  C(14)C(7) C(6) 14(8)  —166  C(14)C(7) C(8) 14(9)  165  14(4) C(3) C(4) H(S) 14(4) C(3) C(13)H(20)  —167 83  C(14)C(7) C(8) 14(10)  46  14(5) C(4) C(S) 14(6)  168  C(14)C(7) C(15)H(24)  —176  H(S) C(4) C(S) H(7)  50  C(14)C(7) C(15)H(25)  64  H(6) C(5) C(6) 14(8)  —43  C(14)C(7) C(15)H(26)  —56  H(7) C(S) C(6) 14(8)  75  C(15)C(7) C(6) 14(8)  —46  14(9) C(S) C(9) H(11)  —61  C(15)C(7) C(S) H(9)  49  14(9) C(S) C(9) 14(12)  179  The sign is positive if when looking from atom 2 to atom 3 a clock wise motion of atom 1 would superimpose it on atom 4.  269  Table 15 Torsional or Conformational Angles (deg) of Diol 285 (4)  H(10)C(8)  C(9)  H(l1)  58  H(lO)C(8)  C(9)  H(12)  —62  H(11)C(9)  C(l0)FI(13)  —59  H(1l)C(9) C(1O)H(14)  60  H(12)C(9) C(10)H(13)  61  C(1O)H(14)  180  H( 20)C(13)C( 16 )H( 27)  172  H(20)C(13 )C(16)H(28)  52  H( 20)C(13 )C(16 )H( 29)  —68  H(20)C(13)C(17)H(30)  68  H(20)C(13 )C( 17 )H( 31)  —52  H(20 )C(13 )C( 17)H( 32)  —172  (2)  H(12)C(9)  (1)  angle  (3)  (1)  (2)  (3)  (4)  (cont.) angle  The sign is positive if when looking from atom 2 to atom 3 a clock wise motion of atom 1 would superimpose it on atom 4.  270  

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.831.1-0061731/manifest

Comment

Related Items