UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis and reactivity of 14-membered keto lactones Maurice, Joseph Roger Luc 1991

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

Item Metadata

Download

Media
831-UBC_1991_A1 M42.pdf [ 5.78MB ]
Metadata
JSON: 831-1.0059773.json
JSON-LD: 831-1.0059773-ld.json
RDF/XML (Pretty): 831-1.0059773-rdf.xml
RDF/JSON: 831-1.0059773-rdf.json
Turtle: 831-1.0059773-turtle.txt
N-Triples: 831-1.0059773-rdf-ntriples.txt
Original Record: 831-1.0059773-source.json
Full Text
831-1.0059773-fulltext.txt
Citation
831-1.0059773.ris

Full Text

SYNTHESIS AND REACTIVITY OF 14-MEMBERED KETO LACTONES by Joseph Roger Luc Maurice B.Sc. (Honours), University of Ottawa, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLEMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1991 ©Joseph Roger Luc Maurice, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ii Abstract A convergent synthesis using dithiane alkylations to produce a long chain hydroxy acid, and subsequent macrolactonization of the latter, yielded the 14-membered macrolides 47 and 58. Alkylation of 47 with methyl iodide gave a diastereomeric product mixture of oc-methylated lactones with the major product having the 2R*,13S* stereochemistry (71). The product ratio was explained by using low energy starting material conformations combined with MM2 calculations. Alkylation of 58 in the same manner also yielded a mixture with the major product having the 2R*,13S* stereochemistry (77a). 7 1 2R*,13S* 77a 2R*,13S* Hydrolysis of the dithiane ring on 47 and 58 gave the keto lactones 28 and 29 respectively. Reduction of the ketone in 28 yielded a mixture of diastereomeric hydroxy lactones of which the major product had the 5S*,13S* stereochemistry (61); this stereochemistry was confirmed by the X-ray crystallographic structure of the p-bromobenzene sulfonate derivative. Reduction of the ketone in 29 also yielded a diastereomeric mixture of hydroxy lactones; the 7S*,13S* stereochemistry of the major product 67 was suggested by MM2 calculations. 0. 28 29 HO/, 61 5S*,13S* 67 7S*,13S* Treatment of both 28 and 29 with trimethylsilyl trifluoromethanesulfonate and tert-butyldimethylsilyl trifluoromethanesulfonate in the presence of triethylamine gave rise to mixtures of isomeric enol ethers; the product distribution could be partly explained by MM2 calculations based on product conformations. iv Tahle of contents Abstract ii Table of contents iv List of figures viii List of tables x List of abbreviations xi Acknowledgements xiii CHAPTER I Introduction 1.1 Structure and chemistry of large rings 1 1.2 Conformational analysis of large rings 3 1.3 Conformational control in synthesis of macrocyclic natural products 9 1.4 Conformational analysis of 14-membered macrolides 17 1.4.1 Conformational analysis of cyclotetradecane 17 1.4.2 Polar maps 18 1.4.3 The corner and pseudo-corner positions 22 1.4.4 The lactone linkage 22 CHAPTER II Results and discussion 2.1 Synthesis of 5-oxo-13-tetradecanolide (28) 26 2.1.1 Synthesis of 9-bromo-2-(2-tetrahydropyranyloxy)-nonane (37) 28 2.1.2 Synthesis and cyclization of hydroxy acid 46 30 2.2 Synthesis of 7-oxo-13-tetradecanolide (29) 33 2.2.1 Synthesis of 7-bromo-2-(2-tetrahydropyranyloxy)-heptane (48) 35 2.2.2 Synthesis and cyclization of hydroxy acid 57 35 2.3 Low energy conformations of macrolides 28 and 29 38 2.4 Diastereoselective reduction of 28 and 29 44 2.4.1 Diastereoselective reduction of 28 45 2.4.2 Diastereoselective reduction of 29 51 2.5 Alkylation of 5-(r,3'-dithian-2'-yl)-13-tetradecanolide (47) and 7-(l',3'-dithian-2'-yl)-13-tetradecanolide (58) 55 2.5.1 Alkylation of S^l'^'-dithian-r-yl)-^-tetradecanolide (47) 55 2.5.2 Alkylation of 7-(l',3'-dithian-2'-yl)-13-tetradecanolide (58) 66 2.6 Preparation of enol ethers from 28 and 29 72 2.6.1 Enol ether generation from 28 73 2.6.2 Enol ether generation from 29 76 2.7 Conclusion 79 CHAPTER IU Experimental 3.1 General 80 3.2 6-Bromo-l-hexanol (31) 82 3.3 6-Bromo-l-(2-tetrahydropyranyloxy)hexane (32) 83 3.4 9-Hydroxy-2-nonanone (34) 84 3.5 9-Hydroxy-2-nonanone ethylene acetal (36) 85 3.6 9-Bromo-2-nonanone ethylene acetal (37) 86 3.7 5-(r,3'-Dithian-2'-yl)-l-(2-tetrahydropyranyloxy)pentane (41) 87 3.8 10-(l\3'-Dithian-2'-yl)-14-(2-tetrahydropyranyloxy)-2-tetradecanone ethylene acetal (42) 88 3.9 10-(l',3'-Dithian-2'-yl)-14-(2-tetrahydropyranyloxy)-2-tetradecanone (43) 89 3.10 5-(l\3'-Dithian-2'-yl)-13-oxotetradecanal (44) 90 v i 3.11 5-(l\3'-Dithian-2'-yl)-13-oxotetradecanoic acid (45) 91 3.12 5-(r,3'-Dithian-2'-yl)-13-hydroxytetradecanoic acid (46) 92 3.13 5-(l',3'-Dithian-2'-yl)-13-tetradecanolide (47) 93 3.14 5-Oxo-13-tetradecanolide (28) 94 3.15 6-Bromohexanal (48) 95 3.16 6-Bromohexanal ethylene acetal (51) 96 3.17 7-Bromo-2-heptanol (49) 97 3.18 7-Bromo-2-(2-tetrahydropyranyloxy)heptane (50) 98 3.19 7-(l',3'-Dithian-2'-yl)heptanal ethylene acetal (53) 99 3.20 7-(l\3'-Dithian-2'-yl)-13-(2-tetrahydropyranyloxy)-tetradecanal ethylene acetal (54) 100 3.21 7-(l',3'-Dithian-2,-yl)-13-hydroxytetradecanal (55) 101 3.22 7-(l',3'-Dithian-2'-yl)-13-hydroxytetradecanoic acid (57) 103 3.23 7-(l',3'-Dithian-2'-yl)-13-tetradecanolide (58) 104 3.24 7-Oxo-13-tetradecanolide (29) 105 3.25 5-Hydroxy-13-tetradecanolide (61,62) 106 3.26 (5S*,13S*)-5-(p-Bromobenzenesulfonyloxy)-13-tetradecanolide (63) 109 3.27 7-Hydroxy-13-tetradecanolide (66,67) 110 3.28 2-Methyl-5-(l',3'-dithian-2'-yl)-13-tetradecanolide (68) 112 3.29 2-Methyl-13-tetradecanolide (69,70) 113 3.30 2,2-Dimethyl-5-(l',3'-dithian-2,-yl)-13-tetradecanolide (74) 114 3.31 2,2-Dimethyl-5-oxo-13-tetradecanolide (75) 116 3.32 2-Methyl-7-(r,3'-dithian-2'-yl)-13-tetradecanolide (77) 117 3.33 2-Methyl-13-terradecanolide (69,70) from 2-methyl-7-(l',3'-dithian-2'-yl)-13-tetradecanolide (77) 118 3.34 2,2-Dimethyl-7-(l',3,-Dithian-2'-yl)-13-tetradecanolide (79) 118 vii 3.35 2,2-Dimethyl-7-oxo-13-tetradecanolide (80) 119 3.36 5-(Trimethylsilyloxy)-4-tetradecen-13-olide (81) and 5-(trimethylsilyloxy)-5-tetradecen-13-olide (82) 120 3.37 7-(Trimethylsilyloxy)-6-tetradecen-l 3-olide (85) and 7-(trimethylsilyloxy)-7-tetradecen-l3-olide (86) 122 3.38 5-(rerr-Butyldimethylsilyloxy)-4-tetradecen-l 3-olide (83) and 5-(rm-butyldimethylsilyloxy)-5-tetradecen-l 3-olide (84) 123 3.39 7-(/erf-Butyldimethylsilyloxy)-6-tetradecen-l3-olide (87) and 7-(rm-butyldimethylsilyloxy)-7-tetradecen-13-olide (88) 124 References 125 Spectral appendix 130 vi i i List of figures Page Figure 1. The diamond lattice conformation of cyclodecane 3 Figure 2. Diamond lattice conformation of cyclotetradecane 4 Figure 3. Intramolecular hydrogen interactions in cyclotetradecane 5 Figure 4. The corner and pseudo-corner positions 6 Figure 5. Celmer and Perun conformations for aglycone of erythromycin B 8 Figure 6. Site selectivity in enolization of 18 13 Figure 7. Conformations of cyclotetradecane and relative strain energies in kcal/mol (MM2) 18 Figure 8. Construction of a polar map 19 Figure 9. Rule for determining the sign of an anti dihedral angle 20 Figure 10. Polar maps for the five lowest energy conformations of cyclotetradecane 21 Figure 11. Comparison of the two corner positions in the C conformation of cyclotetradecane 22 Figure 12. The three possible substitution patterns for 13-tetradecanolide with an s-trans lactone geometry and a C-O-C-H dihedral angle of less than 60° in a [3434] conformation 23 Figure 13. Retrosynthetic analysis of 5-oxo-13-tetradecanolide (28) 27 Figure 14. Retrosynthetic analysis of 7-oxo-13-tetradecanolide (29) 34 Figure 15. Polar maps of the three lowest energy conformations of 28 40 Figure 16. Polar maps of the three lowest energy conformations of 29 43 Figure 17. Stereodiagram for macrolide 63 46 Figure 18. Investigation of peripheral attack on the three lowest energy conformations of 28 48 ix Figure 19. Possible mechanism for the reduction of 28 with MAD and tert-butylmagnesium chloride 50 Figure 20. Investigation of peripheral attack on the three lowest energy conformations of 29 54 Figure 21. Proton abstraction of the three lowest energy conformations of 47 59 Figure 22. Stereodiagram for macrolide 74 63 Figure 23. Polar maps for macrolide 74 from crystal structure (A) and from MM2 calculation (B) 64 Figure 24. Stereodiagram for macrolide 77a 68 Figure 25. Polar maps for macrolide 77a from crystal structure (A) and from MM2 calculation (B) 68 Figure 26. Proton abstraction of the three lowest energy conformations of 58 70 Figure 27. Lowest energy conformations (MM2) of 81a and 83a 76 Figure 28. Lowest energy conformations (MM2) of 86a and 88a 79 X List of tables Page Table I. The three lowest energy conformations of cyclotetradecane 7 Table II. The three lowest energy conformations of 28 39 Table III. The three lowest energy conformations of 29 42 Table IV. Reduction of 5-oxo-13-tetradecanolide (28) with various reducing agents 47 Table V. Reduction of 7-oxo-13-tetradecanolide (29) with various reducing agents 52 xi List of abbreviations A angstrom A N A L elemental analysis calcd calculated CD circular dichroism COSY correlation spectroscopy d doublet D B U l,8-diazabicyclo[5.4.0]undec-7-ene DCC 1,3-dicyclohexylcarbodiimide DMF A^N-dimethylformamide DMSO dimethylsulfoxide Et ethyl eV electron volts FT Fourier transform G C gas chromatography HMPA hexamethylphosphoramide H P L C high-performance liquid chromatography HRMS high resolution mass spectroscopy IR infrared J coupling constant LDA lithium di-isopropylamide LRMS low resolution mass spectroscopy m meta m multiplet MAD methylaluminium bis(2,6-di-terr-butyl-4-methylphenoxide) Me methyl MM mp n NBS NMR NOEDS P Ra-Ni Rf Rt s t TBDMS tert Tf THF TMS molecular mechanics melting point normal N-bromosuccinimide nuclear magnetic resonance nuclear Overhauser effect difference spectroscopy para Raney nickel in thin layer chromatography, ratio of distance traveled by the compound to the distance traveled by the solvent retention time singlet triplet rm-butyldimethylsilyl tertiary trifluoromethanesulfonyl tetrahydrofuran trimethylsilyl Acknowledgements First, I would like to thank Dr. Weiler for his encouragement and his patience during my studies in his group. The help of the NMR, mass spectroscopy, and micro-analysis departments was greatly appreciated. I am also very grateful to Dr. Steve Rettig for the X-ray crystallographic analyses and for some helpful advice. Finally, I would like to thank my colleagues in the Weiler group for making the laboratory an enjoyable, if not downright looney, working environment. CHAPTER I 1 Introduction In the last 50 years, the laboratory synthesis of complex natural products has presented a formidable challenge to the organic chemist. The complexity and the cyclic nature of many of these natural products have led chemists to a better understanding of the importance of conformation to chemical reactivity.1'2 Today, the concepts and ideas of conformational analysis have a prevalent place in the rationalization of reaction mechanisms and chemical transformations.3 1.1 Structure and chemistry of large rings Muscone (1) and civetone (2), components of musk oil, were the first macrocyclic natural products ever isolated, by Ruzicka in 1926.4 With no modern spectroscopic methods available, he was able to ascertain the macrocyclic nature of these two compounds by chemical degradation (not an easy task as muscone, for example, had but one functional group). The actual conformations of muscone and civetone were not known at the time; the general belief was that large rings existed in a multitude of conformations and that their chemistry should not differ much from that of cyclohexane. After Ruzicka's work, more macrocyclic compounds were isolated, identified, and 1 2 synthesized. An interesting class of macrocyclic compounds, due to their diverse biological properties, is the one known as "poly-oxo" macrolides. Poly-oxo macrolides consist of a lactone ring, normally 12-, 14-, or 16-membered. Other features include: (1) an array of substituents uniquely and systematically distributed around the ring and (2) the linkage of one or more sugar residues, very often containing nitrogen. Representative examples are pikromycin (3) (first one ever isolated, by Brockmann and Henkel in 1950),5 methymycin (4), erythromycin A (5), and carbomycin A (6). HQ 0 0 ^ 0 OR 0 3, R= desosaminyl HQ 0 V 0 OR 4, R= desosaminyl 0 HO OH 0 ^ OR 1 OR 2 5, Rl= desosaminyl R2= cladinosyl 0 CHO < CH3Q 0 OR '*O0CCH3 6, R= (isovaleryl)-mycarosyl-mycaminosyl Because of their interesting properties, chemists have tried for many years to synthesize these macrolides. However, it was not until the late 1970's that major synthetic accomplishments 3 were made in this field. There were two reasons for this: one was the difficulty in constructing a medium- or large-ring lactone and the other was the challenge of introducing chiral centers on a straight chain aliphatic acid. Today, numerous methods for macrolactonization are available and have been extensively reviewed.6 Also, introduction of chiral centers with good stereoselectivity has been achieved for a variety of transformations such as carbonyl reductions, additions to carbon-carbon double bonds, alkylations, and aldol condensations. However, in addition to the above transformations (achieved, in the great majority of cases, with the help of chiral auxiliaries), another methodology has been gaining popularity, and this factor is the effect of a molecule's conformation on the stereoselectivity of its reactions. 1.2 Conformational analysis of large rings Following up on his work on large ring compounds, Ruzicka and co-workers, along with Prelog, noted that the plot of melting point versus ring size for cyclic aliphatic hydrocarbons showed an unexpected dependency on ring size.7-8 This phenomenon was left unexplained until Huber-Buser et al., in 1961, showed that large ring molecules could exist in one of many possible conformations; X-ray crystallographic structures of many substituted cyclodecanes showed that these compounds all crystallized in the same conformation, implying that these compounds all existed in the same potential energy minimum well.9 A few years later, Dale pointed out that the conformations of the above substituted cyclodecanes were easily superimposed on a diamond lattice (Figure l) . 1 0 Figure 1. The diamond lattice conformation of cyclodecane. The diamond lattice is an array of carbon atoms with bond lengths of 1.54 A, C-C-C bond angles of 109.5°, and gauche or trans torsional angles; any conformation which can be superimposed on this lattice must represent a potential energy minimum. Dale asserted two important facts: 1) that a ring with an odd number of atoms cannot be superimposed on the diamond lattice and hence cannot exist in a strain free conformation, and 2) that 7- to 13-membered ring compounds cannot exist in a diamond lattice conformation because of the severe transannular hydrogen interactions.10 That the cyclodecane derivatives exist in a diamond lattice conformation clearly shows a compromise between angle strain, torsional energy, and transannular hydrogen repulsions in these compounds. Dale then proposed low energy conformations for all even-numbered rings from C6 to Ci6- All of these predictions have proved fairly accurate by X-ray crystallography and theoretical calculations of model compounds. An example of this is the conformation of cyclotetradecane (Figure 2).11 Side view Top view Figure 2. Diamond lattice conformation of cyclotetradecane (* corner position). If one views the strain free conformation of cyclotetradecane from the "top", an interesting feature can be seen: some carbon atoms have both hydrogens pointing outside of the ring, whereas the other carbon atoms have at least one hydrogen pointing inside the ring. The hydrogens pointing outside the ring don't experience any transannular repulsions (Figure 3) and their parent carbons are said to occupy a "corner" position. Dale concluded that these corner atoms should be the only carbons capable of accomodating geminal disubstitution without transannular interactions. Number of A: 1 internal B: 2 hydrogen C: 3 interactions Figure 3. Intramolecular hydrogen interactions in cyclotetradecane (* corner position). Examination of molecular models led to the definition of a corner position as a sequence of two contiguous gauche dihedral angles of the same sign flanked on either side by an anti dihedral angle.12 We noticed two different types of corner positions. One, the classical corner position as defined by Dale, is flanked by two gauche angles of the same sign (e.g. 180°, +60°, +60°, 180°) . 1 0 Another corner position, which we called a pseudo-corner, has two contiguous gauche angles of opposite signs (e.g. 180°, -60°, +60°, 180°). This pseudo-corner was also recognized by Borgen et al. in crystal structures of 1,4,8,11-tetraoxacyclotetradecane.10 We formulated the following definitions for the corner and pseudo-corner positions (Figure 4): 1. A corner position has two contiguous gauche angles of the same sign flanked on either side by an anti dihedral angle of opposite sign (e.g. -180°, +60°, +60°, -180°). 2. A pseudo-corner position has two contiguous gauche angles flanked on either side by an anti dihedral angle, with alternating signs (e.g. -180°, +60°, -60°, +180°). 6 Figure 4. The corner (above) and pseudo-corner (below) position. Dale used the corner positions to establish a simple nomenclature for macrocycles.12 This nomenclature consists of a series of numbers in brackets, each number representing the number of bonds between consecutive corner atoms. Thus the lowest energy conformation for cyclotetradecane is a [3434] conformation (see Table I). Dale then proceeded to calculate the strain energy of different diamond lattice conformations, using the potential energy curve of butane. The calculated strain energies were crude but the energy trends proved very reliable.12 Later, with the advent of molecular mechanics calculations, Anet and Cheng obtained more accurate strain energies which showed that cyclotetradecane should exist predominantly in the [3434] conformation.13 In addition, two non-diamond lattice conformations were discovered: the [3344] (with a strain energy of 1.1 kcal/mol relative to the [3434] conformation) and the [3335] (relative strain energy of 2.4 kcal/mol). T a b l e I. T h e three lowest energy conformations of cyclotetradecane Conformation Top view Side view Strain energy [3434] [3344] [3335] 0 kcal/mol 1.1 kcal/mol 2.4 kcal/mol * corner position The [3344] and [3335] conformations were found to be rotamers in the pseudo-rotation of the [3434] conformation. Calculated pseudo-rotational barriers of 7-8 kcal/mol agreed with the low temperature *H NMR and 1 3 C NMR studies by Anet et al., who found a pseudo-rotational barrier of 7 kcal/mol by using line shape techniques.13 Following the results of these conformational analyses, researchers began to investigate the conformations of the natural macrolides. The first model for the three-dimensional structure of a macrolide was advanced by Celmer in 1965.14 He proposed a structure for the preferred conformer of erythronolide B (8), the aglycone of erythromycin B. Patterned after Dale's [3434] conformation, the Celmer model was in reasonable agreement with the  lH NMR data (Figure 5).15 Figure 5. Celmer and Perun conformation for aglycone of erythromycin B. However, some discrepancies with other data led to the belief that this was not the most favored, low-energy conformation: disagreement with the CD data, unfavorable steric interactions between the 4- and 6-methyl groups, and the 12-methyl group and lactone functionality.16 Close re-examination of all data led to the Perun conformation for erythronolide B (Figure 5), which agreed with all experimental data.16 The Perun conformation could best explain the experimental results on several erythromycin derivatives.17 Similar studies on different macrolide antibiotics confirmed the Perun conformation and showed that oleandomycin (9) and lankamycin (10) occurred in the same low energy conformation as the erythromycins.1** 9, R1= oleandrosyl 10, R*= aracnosyl R2= desosaminyl R2= chalcosyl There has been some controversial evidence regarding the conformational equilibrium of these macrocyclic rings in the solid and liquid states. In recent studies, Shannon et al. showed that cyclotetradecane existed in the [3434] conformation to an extent of 60% at 340 K . 1 9 This is at odds with calculations by Dale which indicated that cyclotetradecane existed almost entirely in the [3434] conformation in the liquid state.20 Clearly, the conformational behavior of large rings is not clear cut and caution needs to be exerted in the conformational analysis of these compounds. 1.3 Conformational control in synthesis of macrocyclic natural products The stereoselective synthesis of complex macrocyclic natural products has been one of the major goals of organic chemistry. The majority of researchers prefer to attack such a synthesis by synthesizing parts of the molecule with stereocontrolled reactions on acyclic chains, giving fragments with the correct relative stereochemistry, then joining these parts together to give the final product with the correct absolute stereochemistry. However, some groups have shown a preference for the more elegant method of conformational control to introduce new chiral centers.21 Following are three examples on the application of this method. Still, one of the pioneers in the application of conformational control in synthesis, and co-workers used this approach in the final stages of his synthesis of baccharin B5 (11), a highly oxygenated trichothecanoid with anti-leukemic properties.22 11 Still's synthetic plan toward baccharin B5 involved the use of D-xylose as a source of the C6' and C13' asymmetric centers and conformational control for the selectivity in the macrocyclization and for the introduction of the CT and C4' chiral centers. The a,P-unsaturated aldehyde 12, the macrolide precursor, was obtained from the corresponding phosphonoester in 65% yield as a 4:1 E:Z mixture, of which the E-precursor was easily separated by column chromatography. This aldehyde was then cyclized using K2CO3 and 18-crown-6 in toluene (-20 to 0 °C) to yield a 1.5:1 mixture of (E,Z)- and (E,E)-macrolides in 75% yield. Flash column chromatography afforded the pure (E,Z)-macrolide (13) in 45% yield. Still was now ready to use his conformational control approach to complete the synthesis of baccharin B5. The hydroxyl group at C13' was protected as its ferr-butyldimethylsilyl ether to give compound 14. Next came the key reaction: treatment of 14 with /rz-chloroperbenzoic acid in benzene afforded a single triepoxide 15 in 70% yield. 12 13 The stereoselectivity of the reaction, as measured by response calibrated HPLC, appeared to be >15:1 for both the cyclohexane epoxide and the macrocyclic epoxide at C2'-C3', with the latter stereochemistry controlled by the conformation of the ring. This triepoxide was then carried through to the final product, baccharin B5 (11). Later, Schreiber and Hawley used the conformational control of enolate geometry in a synthesis of germacrene-D (16), a terpenoid compound which mimics the action of a sex pheromone of the American cockroach Periplaneta Americana?2 16 The intermediate 18 was obtained by thermal isomerization of the cyclobutene 17 in 92% yield. 17 18 The plan was to generate the enolate derivative 19 and submit it to a cuprate coupling to obtain the final product 16. Results of enolization with lithium diisopropylamide to give the required enol triflate were unsatisfactory. However, lithium hexamethyldisilazide in a mixture of THF and HMPA provided the high regioselectivity required for the enolization, and afforded a 10:1 mixture of enol triflates with the major product having structure 19. 18 19 The stereochemical assignment of the enol triflate 19 was based on its conversion to germacrene-D (16). Reaction of 19 with lithium dimethylcuprate gave a 10:1 mixture of germacrene-D and an inseparable isomer. Schreiber and Hawley rationalized the site- and stereoselectivity of the enolization reaction by considering the conformational properties of the ketone 18. The boat-chair conformation 18' is consistent with the results of COSY 2D NMR and NOEDS experiments.24 Enolization could occur in two possible directions. However, one needs only to compare the dihedral angles of the carbons involved in the enolization to see the selectivity that would occur (Figure 6). Figure 6. Site selectivity in enolization of 18. The four-carbon fragment C2-C1-C10-C9 has a dihedral angle of 180° (anti) in both 18' and 20', thus negating the need for conformational reorganization after proton abstraction from C l . On the other hand, the four-carbon fragment C1-C10-C9-C8, having a dihedral angle of 120° in 20', would require, upon proton abstraction from C9, major conformational reorganization to avoid the formation of a non-planar enolate and is clearly disfavored. Trapping of the enolate 20' would give the observed enol triflate 19. The results of the enolization shown above to give the final product corroborated and showed the usefulness of conformational analysis of medium rings. Our final example of macrocyclic control comes from a synthesis of (+)-(9S)-dihydroerythronolide A (21) performed by Paterson's group in Cambridge.25 OH HO OH 12 U M 6 OH 0 ^ ""'OH 21 The substituted macrolide 22 was obtained from its seco-acid precursor via a macrolactonization by the Yamaguchi method26 in 91-96% yield. OTBS O T B S 22 H 2 , 5% Rh/A12Q3 THF, 20 °C, 4 hrs 0' OTBS '"'OTBS 23 The enone 22 was selectively hydrogenated with 5% Rh/Al2C>3 to give the ketone 23 in 92% yield. Deprotonation of 23 by LDA in THF occurred exclusively at C6 to give a single enolate stereoisomer as shown by quenching with trimethylsilyl chloride. Reaction of the enolate with formaldehyde gave a 57:43 mixture of hydroxymethylated isomers which underwent mesylation and in situ elimination with DBU to give the a-methylene ketone 24. MM2 calculations on the enone 24 showed a preference for the s-cis conformation over the s-trans by greater than 4 kcal/mol and therefore was predicted to give the required E-enolate upon 1,4-addition of a suitable reagent27 Indeed, reaction of 24 with Li-Selectride in THF at -78 °C gave, after quenching with trimethylsilyl chloride, 93% of a single enol ether 25 with the E geometry (the Z-isomer is more strained by greater than 12 kcal/mol by MM2 calculations). OTBS 0 '"''OTBS 24 0 L-Selectride, THF -78 °C, 2 hrs ii) Me3SiCl, Et 3N OTBS **Y T)SiMe3 "'''OTBS 25 En route to the final product, the enol ether 25 was osmylated to cleanly give a single cc-hydroxyketone 26 in 85% yield. Reduction of 26 (with macrocyclic stereocontrol and chelation by the hydroxyl group at C6) using zinc borohydride gave a single diol product 27 in 70% yield. Deprotection of the hydroxyl groups and osmylation at C11-C12 gave the final product (+)-(9S)-dihydroerythronolide A (21). The usefulness of the conformational control approach is best described by Paterson himself: "By postponing the introduction of the stereochemistry and hydroxylation patterns at C5, C6, C l l , and C12 until after macrocycle formation, the synthesis has the following advantages: (i) flexibility in the manipulation of the silyl enol ether and (less reactive) C11-C12 double bond in 24; (ii) brevity due to the limited use of protecting groups (ca. 20 steps = 2 steps/chiral center); (iii) a high yielding macrolactonization (91-96%), as there are fewer transannular and vicinal interactions and the sp2 carbon skeleton imparts extra conformational rigidity, thus favoring cyclization; (iv) highly stereoselective reactions are obtained due to the conformational preferences of the macrolide intermediates."25 Everything said by Paterson regarding the latter synthesis can be applied to the synthesis of many macrocyclic compounds and the elegance of the conformational approach for generation of chiral centers will soon take its rightful place in the arsenal of the synthetic chemist. 1.4 Conformational analysis of 14-membered macrolides The computer program MACROMODEL, developed by Still and co-workers 2 8 was used to determine the low energy conformations of the macrolides studied in this work. The program generates all possible conformations of a medium or large ring by random variation of torsional angles in an acyclic molecule created by breaking a bond in the ring. The conformations with chain termini within a ring closure distance are then submitted to MM2 calculations. Previously, our group could not afford the extensive computer time required for such an operation and used a different approach; a brief synopsis of that approach is presented here.29  1.4.1 Conformational analysis of cyclotetradecane After Dale's conformational analysis of cyclotetradecane30 (in which he showed that the [3434], [3344], and [3335] were the three lowest energy conformations), Saunders31 calculated that there were thirteen possible diamond lattice conformations for cyclotetradecane. Our group constructed a model of each conformation and submitted them to MM2 calculations.32 The conformations exhibited a relative strain energy in the range of 0-12 kcal/mol, consistent with severe transannular interactions within the carbon framework of the cyclotetradecane (Figure 7). As Dale's nomenclature could not be used for the majority of these conformations, an arbitrary naming system was used (Ogura and co-workers had previously proposed this for conformations B, C, and D) . 3 3 ^ C^h PijrAr ^ L > [3434] 0.0 [3344] 1.1 [3335] 2.3 Conf.D 3.0 Conf. C 5.0 Conf. E 6.8 Conf. F 7.1 Conf. G 7.5 Conf. H 7.6 Conf. I 7.7 Conf. J 7.9 Conf. B 8.5 Conf.K 11.0 Conf.L 11.1 Conf. M 12.0 Figure 7. Conformations of cyclotetradecane and relative strain energies in kcal/mol (MM2). 1.4.2 Polar maps When looking at the conformational similarities between two different macrolides, one can be faced with a tremendous problem (particularly if the macrolides are highly substituted). Conformational similarities are not always apparent and the use of molecular models is tedious and prone to error. 19 A method used by our group,34 and introduced by Ogura et al. in his oleandomycin studies,35 is the use of polar maps. A polar map is a circular graph showing the magnitude and sign of the endocyclic dihedral angle of a cyclic molecule in relation to the bond number (Figure 8). Torsional bond Figure 8. Construction of a polar map. The dihedral angles uniquely describe the conformation of a molecule and a polar map will exhibit a characteristic pattern for each conformation. The concentric circles represent the magnitude and sign of the dihedral angle. Ogura used the smallest endocyclic dihedral angle within the ring to determine the sign of the angle and as the angle approaches 180°, the determination becomes quite arbitrary as we can choose a 180° angle to be either positive or negative. This can lead to a situation where two compounds crystallizing in the same conformation may have different polar maps because of an erroneous determination in the sign of an anti dihedral angle. To solve this problem, we proposed that only dihedral angles within the ring be used for the generation of polar maps.34 If the direction of rotation for a dihedral angle within the ring is clockwise from C l to C4 in C1-C2-C3-C4, then the dihedral angle sign is positive. Otherwise, the sign is negative (Figure 9: drawing A is used to determine the sign of the angle, not drawing B). 20 Figure 9. Rule for determining the sign for an anti dihedral angle. The polar maps for the five lowest energy conformations of cyclotetradecane are shown in Figure 10. Polar maps have proven themselves to be a very useful tool for comparison of conformations. [3434]* 0.0 kcal/mol [3344] 1.1 kcal/mol [3335] 2.3 kcal/mol [D]* 3.0 kcal/mol [C]* 5.0 kcal/mol Figure 10. Polar maps for the five lowest energy conformations of cyclotetradecane (* diamond lattice conformation; strain energy relative to lowest energy conformation). 1.4.3 The corner and pseudo-corner positions A polar map enables us to quickly determine where the corner and pseudo-corner positions can be found in a conformation by recognizing the characteristic sequence of dihedral angles at a corner and pseudo-corner (see Figure 4 for the definition of a corner and a pseudo-corner position). In observing the polar map for the C conformation (Figure 11), we can see the two different types of corner positions. Figure 11. Comparison of the two corner positions in the C conformation of cyclotetradecane. Geminal substitution at a corner or pseudo-corner position, according to MM2 calculations, would occur in the following sequence: 1. Geminal substitution at a corner atom. 2. Geminal substitution at a pseudo-corner position (introduction of an additional 1,3-diaxial interaction, raising the energy by 0.6 kcal/mol relative to a corner position). 3. Geminal substitution at a non-corner atom. These results agreed with Dale's qualitative prediction.10 1.4.4 The lactone linkage A further reduction in the number of conformations which need to be considered (Figure 7) is possible as a result of the lactone functionality in our macrolides. Deslongchamps36 has estimated the energy difference between the s-trans and s-cis ester conformations to be of the order of 3.0 kcal/mol, thus explaining why open chain esters occur in the s-trans form. 23 0 0 R R' R s-trans s-cis Because a 14-membered ring is large enough to accomodate an s-trans lactone, conformations with an s-cis lactone linkage can be ignored in our conformational analysis. In addition, studies by Dunitz and Schweizer37 on over 1750 esters of secondary alcohols showed that these contain a C-O-C-H dihedral angle in the range 0-60°, with 85% of the cases in the 0-40° range. Applying this trend to lactones of secondary alcohols, and using these restrictions on the [3434] conformation, we are left with only three possible conformations of 13-tetradecanolide to consider (Figure 12). R 0 Figure 12. The three possible substitution patterns for 13-tetradecanolide with an s-trans lactone geometry and a C-O-C-H dihedral angle of less than 60° in a [3434] conformation. CHAPTER II 24 Results and discussion Still, Schreiber, and Paterson's syntheses, described in the introduction, show that conformational control of large rings can be a powerful tool in organic synthesis. With the increasing interest by pharmaceutical companies toward the laboratory manufacture of complex drugs based on macrocyclic rings (one example is the synthesis of FK-506, a potent immunosuppressant),38 their work now comes under a new light. We have chosen to study the conformational control of 14-membered keto lactones. There are various reasons for this decision. Many macrolide antibiotics contain a 14-membered lactone ring, and the effect of the conformation of the ring on their chemical properties needed closer scrutiny. Knowledge of the effect of the conformation of the ring on its chemical reactivity might lead us to a better understanding of their biological mode of action. Also, the knowledge gained about 14-membered lactones might be applied to other macrolides. Because many poly-oxo macrolides contain oxygen functionalities at C-5 and C-7, we chose to concentrate our studies on 5-oxo-13-tetradecanolide (28) and 7-oxo-13-tetradecanolide (29). The usefulness of the ketone group in introducing various functionalities in a molecule was also a powerful motivator in the choice of these two macrolides. 28 29 In addition, the presence of an asymmetric center at C-13 in both macrolides would enable us to study the effect of a remote asymmetric center on the stereoselectivity of the reactions performed on the ring. Still and Galynker,39 during their work on 10-membered lactones, found that the stereoselectivity of kinetic alkylations decreased with increasing distance from the asymmetric center. In both our macrolides, the ketone group is five atoms removed from the asymmetric center. We can expect this difference in the position of the ketone group, relative to both the asymmetric center and the lactone functionality, to have an effect on the conformation of the macrolides, which will translate into an effect on the stereoselectivity of the various reactions performed on these two macrolides. 2.1 Synthesis of 5-oxo-13-tetradecanolide (28) The retrosynthetic scheme for our synthesis of 5-oxo-13-tetradecanolide is shown in Figure 13. As should be the case in all syntheses, starting materials were inexpensive and commercially available, as large amounts of the macrolide would be required. The plan consisted of synthesizing the molecule in two parts, coupling these and cyclizing the resulting long-chain hydroxy acid to give the lactone ring. Of all the reactions involved in the synthesis, the lactonisation was thought to be the most troublesome step. However, Corey et al., 4 0 as well as Masamune et al. 4 1 have perfected many high-yielding cyclization methods, called "double activation" methods, by converting the carboxylic acid terminus of the hydroxy acid precursor to 2-pyridinethiol esters and t-butylthiol esters respectively. A modification of these methods was used in our synthesis.42 27 Figure 13. Retrosynthetic analysis of 5-oxo- 13-tetradecanolide (28). 28 2.1.1 Synthesis of 9-bromo-2-(2-tetrahydropyranyloxy)nonane (371 1,6-Hexanediol (30) was treated with aqueous HBr under continuous liquid-liquid extraction with n-heptane to give 6-bromo-l-hexanol (31) in yields ranging from 70-85%.43 The bromo alcohol was then protected as its tetrahydropyranyl ether 32 by reacting it with dihydropyran in presence of a catalytic amount of p-toluenesulfonic acid.44 32 We now needed to add three carbons to 32 to obtain the nine-carbon chain required for the first part of our molecule. One method of choice for such a transformation is the acetoacetic acid synthesis, where the alkyl halide reacts with the anion of methyl acetoacetate to give, through nucleophilic substitution, the long-chain [3-keto ester. Decarboxylation, in basic or acidic conditions, gives the long-chain ketone. Thus, the anion of methyl acetoacetate (33), generated with sodium methoxide in methanol, was treated with the protected bromo alcohol 32, the intermediate {3-keto ester immediately decarboxylated in acidic conditions to give 9-hydroxy-2-nonanone (34) , 4 5 0 0 0 NaOMe, MeOH, A ii) 32  iii) H + , H 20-MeOH, A 33 34 The yields, however, were poor and examination of the alkylation product, prior to decarboxylation, showed it to be mainly the ester 35, obtained by a retro Claisen reaction through attack of sodium methoxide on the intermediate (3-keto ester. The hydroxy ketone 34 was finally obtained in good yield (79%) by using NaH as a base in a mixture of THF and DMF, followed by decarboxylation in acidic conditions.46 0 0 0 NaH, THF-DMF, A ») 32  X)CH, n 0 H + ,H 2 O-MeOH,A 33 34 9-Hydroxy-2-nonanone ethylene acetal (36) was obtained in 91% yield by allowing 34 to react with 1,2-ethanediol and a catalytic amount of p-toluenesulfonic acid in refluxing benzene with azeotropic removal of water.47 The alcohol 36 was then converted into the bromide 37 with triphenylphosphine and carbon tetrabromide at -78 °C in dichloromethane in 88% yield.48 HQCH 2 CH 2 OH p-TsOH, benzene reflux (CH2)4 36 (C 6H 5) 3P, CBr 4 CH 2C1 2, -78 °C 2.1.2 Synthesis and cyclization of hydroxy acid 46 With the first part of our molecule in hand, namely the protected keto bromide 37, we were now ready to proceed with the synthesis of the second portion. We first had to synthesize the protected hydroxy dithiane 41. The unprotected hydroxy dithiane 40 had been synthesized by reaction between 1,3-propanedithiol (38) and 5-hydroxypentanal in the presence of boron trifluoride etherate.47 5-Hydroxypentanal can be obtained by hydration of dihydropyran (39) 4 9 but is difficult to isolate. It was thought that we could carry out a one-step reaction by using 1,3-propanedithiol (38) and dihydropyran (39). Thus, a mixture of 1,3-propanedithiol (38) and dihydropyran (39) was treated with boron trifluoride etherate in CH2CI2. The crude hydroxy dithiane 40 contained some protected hydroxy dithiane 41. This crude product was then treated with dihydropyran to give 41 in 72% yield 4 4 The protected tetradecane 42 was obtained in 88% yield by coupling the anion of the dithiane 41, generated with n-butyllithium in THF at -20 °C, with the protected keto bromide 37 4 7 The tetradecane derivative 42 was then deprotected in a mixture of acetone, methanol, and water at 55 °C, using p-toluenesulfonic acid as a catalyst, to give the hydroxytetradecanone 43 in 97% yield. 4 1 0 /i-BuLi, THF, -20 °C II) 37 Before proceeding with the cyclization it was necessary to oxidize the alcohol to an acid group and reduce the ketone to an alcohol to give the long-chain hydroxy acid 46. Unfortunately, this oxidation turned out to be the major challenge in our synthesis. Our initial attempts with Jones' reagent50 and Corey's pyridinium dichromate reagent51 ended with decomposition of the starting material 4 3 . We then decided to carry out the oxidation in two steps. First, we would oxidize the terminal alcohol to an aldehyde, then convert the aldehyde to the carboxylic acid. Here again, we ran into some problems. The Swern oxidation,52 as well as Corey's pyridinium chlorochromate or dichromate reagents51 and the Dess-Martin reagent,53 also gave decomposition of the starting material. Finally, we found that the Pfitzner-Moffat oxidation54 allowed us to smoothly oxidize the alcohol to the aldehyde 44 in 70% yield. The aldehyde was converted to the 32 acid 45 by a silver oxide oxidation, the silver oxide generated in situ from silver nitrate and sodium hydroxyde, in 97% yield.5 5 The hydroxy acid 46 was then obtained by sodium borohydride reduction of the ketone in 100% yield. 46 With the long-chain hydroxy acid in hand, we were now able to study the macrolactonization. A mixture of rm-butyl isocyanide and l-phenyl-i//-tetrazole-5-thiol in toluene was added to a solution of 46 in toluene and the resulting solution refluxed to yield the protected keto lactone 47 in a very good yield of 78%.42 The keto lactone 28 was obtained in 86% yield (17% overall) from the protected keto lactone 47 by oxidative hydrolysis of the dithiane using N-bromosuccinimide (NBS) in acetone.56 33 47 28 2.2 Synthesis of 7-oxo-13-tetradecanolide (29) As for the synthesis of 5-oxo-13-tetradecanolide (28), starting materials for the synthesis of 7-oxo- 13-tetradecanolide (29) were inexpensive and commercially available. The retrosynthetic analysis for 7-oxo-13-tetradecanolide (29) is shown in Figure 14. The plan here was also to synthesize the molecule in two parts, couple these, and cyclize the resulting hydroxy acid to the 14-membered lactone. Our plan called for the same 6-carbon starting material for each fragment. 2.2.1 Synthesis of 7-bromo-2-(2-tetrahydropyranyloxy)heptane (48) 6-Bromohexanal (48) was obtained in 82% yield from 6-bromo-l-hexanol (31) using a Swern oxidation.52 Conversion of 48 into 7-bromo-2-heptanal (49) was effected by the Grignard reaction of 48 with methylmagnesium iodide (79% yield). The secondary alcohol 49 was then protected as its tetrahydropyranyl ether 50 in 98% yield in the usual fashion.44 50 2.2.2 Synthesis and cyclization of hydroxy acid 57 6-Bromohexanal (48) was protected as its ethylene acetal 51 in 96% yield 4 7 The bromo acetal 51 was then coupled with the anion of 1,3-dithiane (52), generated with «-butyllithium in THF, to give the dithianylacetal 53 in 84% yield, which was subsequently coupled, in the same fashion, to the protected bromo alcohol 50 to give the tetradecane derivative 54 with a yield of 82%. The tetradecane derivative 5 4 was treated with pyridinium p-toluenesulfonate in a mixture of acetone, methanol, and water to give the hydroxy tetradecanal 5 5 in 33% yield.57 A by-product of the reaction, the dimethyl acetal 5 6 , was also converted to 5 5 with p-toluenesulfonic acid in acetone in 100% yield. Compound 5 5 was then converted to the carboxylic acid 5 7 using a silver oxide oxidation.55 54 Acetone/methanol/water p-TsOH, A HO HQ /XjCH 2 ) 3 JX^(CH 2 )3_X r + ^AjC H2)3jX^(CH 2) 3^i\ ( 55 AgN0 3, NaOH THF-H 2 0 56 p-TsOH, acetone, A 57 0CH3 X)CH3 Cyclization of the hydroxy acid 57 to the lactone 58 was accomplished in 55% yield by the method described previously using rm-butyl isocyanide and l-phenyl-i//-tetrazole-5-thiol.42 The keto lactone 29 was then obtained in 66% yield (9% overall) by oxidative cleavage of the dithiane in 58 using N-bromosuccinimide in acetone.56 2.3 Low energy conformations of macrolides 28 and 29 Minimum energy geometries found by molecular mechanics calculations often depend on the starting conformation of the molecule (local minimum problem).32 To calculate the low energy conformations of 14-membered lactones, previously our group used as starting conformations the five low energy conformations shown in Figure 10 -the [3434], [3344], [3335], D, and C conformations- along with the following criteria: 1. The lactone linkage should have the s-trans geometry. 2. The C-O-C-H dihedral angle in the lactone linkage should be in the 0-60° range. 3. Substituents should occupy exterior positions on the macrolide ring, and geminally disubstituted carbons occupy either a corner, or pseudo-corner, position. 39 However, the use of the MACROMODEL program, developed by Still and co-workers,28 greatly simplified our calculations. The latest version of MACROMODEL has a conformational search routine in it. After providing a starting structure for a given molecule, the program provided us with a list of conformers for the molecule, in order of increasing energy. The conformational search was thus expanded to include conformations other than the five low energy conformations mentioned earlier. Table II. The three lowest energy conformations of 28. Low energy conformations Strain energy3 Conformation 28a 28b 28c 0.00 kcal/mol [3434]* 0.26 kcal/mol [3434]* 0.35 kcal/mol [3344] a Strain energy is reported relative to the lowest energy conformation. * Diamond lattice conformation. The three lowest energy conformations for the macrolide 28, obtained from the MACROMODEL program, are shown in Table II. The conformational analysis of 28 was quite complex, with five conformations being within 0.4 kcal/mol of the global minimum. The two lowest energy conformations turned out to have the [3434] geometry while the third lowest had the 40 [3344] conformation. Conformations 28a and 28c had the same local conformation at the ketone group. 28a 28b 28c Figure 15. Polar maps of the three lowest energy conformations of 28. All three conformations had their C-O-C-H dihedral angle well inside the 0-60° range established as our criterion (43° for 28a, 46° for 28b, and 40° for 28c). Unfortunately, we were unable to obtain an X-ray crystallographic structure for 28 (obtained as an oil), which could have provided some insight into our findings. Examination of the *H NMR spectrum of 28 seemed to confirm the fact that the three lowest energy conformations (if not more) co-existed in solution. Egan et al . 1 6 had shown that extreme values for vicinal coupling constants (10-12 Hz and 0-2 Hz) indicated that the molecule existed largely in one conformation while intermediate coupling constants (6-7 Hz) were found in conformational^ mobile systems. The methine proton in 28 exhibited an eleven-line pattern with coupling constants of 4,6, and 8 Hz, which would indicate a conformationally mobile molecule, in agreement with the fact that five low energy conformations were found for 28 within 0.4 kcal/mol of the global minimum. Another surprise awaited us as we proceeded with the conformational analysis of the macrolide 29. As shown in Table UI, the lowest energy conformation 29a appeared to be a twist conformation, then a [3344] conformation, and finally a [3434] conformation although it should be noted that all three conformations are exceedingly close in energy. The three conformations all had a C-O-C-H dihedral angle within the 0-60° range, in accordance to Schweizer and Dunitz's findings.37 The polar maps for the three conformations are shown in Figure 16. Again, we were unable to obtain a X-ray crystallographic structure to evaluate these results as 29 was obtained as an oil which refused to crystallize. Table III. The three lowest energy conformations of 29. Low energy conformations Strain energya Conformation 29a 29b 29c 0.0 kcal/mol twist 0.05 kcal/mol [3344] 0.09 kcal/mol [3434]* a Strain energy is reported relative to the lowest energy conformation. Diamond lattice conformation. Although one might be surprised to see a [3344] conformation lower in energy than the usually more stable [3434], the occurrence is not uncommon. Our group has observed such results before; a few macrolides synthesized in our laboratory crystallized in a [3344] conformation.59 43 29a 29b 29c Figure 16. Polar maps of the three lowest energy conformations of 29. So named because of its resemblance to the twist-boat conformation of cyclohexane, the twist conformation was first reported by Shreiber and Liew 5 8 in a report on the crystal structure of (10E)-5,9-dimethoxy-6-oxo-10-tridecen-l 3-olide (59), a 14-membered macrolide. Our group also confirmed the existence of this twist conformation; the hydroxy lactone 60, obtained by reduction of the corresponding keto lactone, crystallized in a twist conformation.59 MM2 calculations on the twist conformation of cyclotetradecane showed it to be the fifth lowest energy conformation for a 14-membered ring, with a relative strain energy of 3.4 kcal/mol. 60 If one used Dale's nomenclature to name the twist and Ogura's D conformations, both would be designated as [311] conformations, i.e. there are 3 and 11 carbon-carbon bonds between successive corner positions. Because of this ambiguity, our group proposed to extend Dale's nomenclature system to include the pseudo-corners in the analysis; the number of bonds between a corner and a pseudo-corner, or between two pseudo-corners, shall be indicated by a primed number. With this new nomenclature, the D and twist conformations can be designated as [347'] and [3'4'3'4'] conformations. Examination of the *H NMR of 29 gave rise to the same finding as for macrolide 28; namely the methine proton in 29 also exhibited an eleven-line pattern, with coupling constants of 3,6, and 9.5 Hz, supporting the fact that the molecule is conformationally mobile.17 2.4 Diastereoselective reduction of 28 and 29 With both of our macrolides containing a ketone group, we were interested in evaluating the effect of the conformation of the ring on the reduction of the ketone group by various reducing agents. It was hoped that the possible diastereoselectivity of these reductions could be rationalized and/or predicted by molecular mechanics calculations. 2.4.1 Diastereoselective reduction of 28 From previous research in our laboratory, it was found that the reduction of 14-membered keto lactones with sodium borohydride proceeded with little or no selectivity.29 Similarly, reduction of 28 using sodium borohydride gave a 100% yield of a 43:57 mixture (GC: the response factor for the two isomers was assumed to be identical) of two diastereomeric alcohols 61 and 62 of undetermined stereochemistry. Changing the borohydride counter ion to tetramethylammonium gave a mixture of alcohols 61 and 62 in a ratio of 41:59 2 8 6 1 5 S M 3 S * 6 2 5R*,13S* 46 (98% yield), a very little change in the selectivity. At this stage, we were unable to separate the two alcohols by column chromatography and their relative stereochemistry was still unknown. Having had little success in achieving stereoselectivity with simple borohydride reduction, we decided to turn our attention to bulkier reducing agents, in the hope that they would provide increased selectivity. Thus reduction of 28 using L-Selectride (lithium tri-sec-butylborohydride) in THF at -78 °C proceeded smoothly to give a 73:27 mixture (by GC analysis) of the two diastereomeric alcohols 61 and 62 (97% yield). Fortunately, when 28 was treated with excess L-Selectride, alcohol 61 was isolated as pure colourless crystals in 52% yield. However these crystals of 61 were not adequate for structure determination. Therefore we prepared the p-bromobenzenesulfonate derivative 63 (obtained as colourless prisms) for an X-ray crystallographic analysis , which showed the relative stereochemistry of its substituents to be 5S*,13S* (Figure 17). Hence the major isomer of the diastereomeric alcohols had the 5S*,13S* stereochemistry. The asterisks indicate that both enantiomers of the compound are present. Figure 17. Stereodiagram for macrolide 63. Reduction of 28 with K-Selectride gave a 58% yield of the same diastereomeric alcohols 61 and 62 in an 84:16 ratio respectively. This reduction was very sluggish and the 84:16 ratio is not the initial ratio as the reaction invariably proceeded with decomposition of products, which produced an increase in the ratio of 61 to 62. Addition of excess reducing agent did not improve the yield of the reaction; it merely increased the yield of decomposition products. The reductions with L-Selectride and K-Selectride gave the alcohol 61 as the major product. We wondered if the selectivity could be reversed, i.e. if we could form alcohol 62 predominantly. A method for reducing carbonyl groups with the reverse selectivity of that achieved by the Selectride reagents has been developed by Yamamoto et al. . 6 0 Using a bulky organoaluminium reagent such as methylaluminium bis(2,6-di-fm-butyl-4-methylphenoxide) (64) (MAD) to complex the carbonyl group from the more open face, this allowed delivery of the hydride from the more hindered face, thus giving the opposite selectivity to the Selectride reagents (which reduce the carbonyl from the more open face). Thus, reduction of 28 with MAD and tert-butylmagnesium chloride gave a 93% yield of a 20:80 mixture of alcohols 61 and 62 respectively, a selectivity superior to that achieved by the Selectride reagents. The results for the various reductions performed on 28 are summarized in Table IV. Table IV. Reduction of 5-oxo-13-tetradecanolide (28) with various reducing agents. Reducing agents Temperature Stereoselectivity Yield % 61 62 (S*,S*) (S*,R*) NaBELt -78 °C 43 57 100 Me 4 NBH 4 -78 °C 41 59 98 L-Selectride -78 °C 73 27 97 K-Selectride -78 °C 84 16 58 MAD/rm-BuMgCl -78 °C 20 80 93 Predicted selectivity (MM2) -78 °C 0 100 Still and Galynker31 had demonstrated that reagents attack the Tc-system of a macrocyclic ring almost exclusively from the periphery of the ring, or the more open face. Our group has previously used this rationale to explain results obtained in our laboratory. We therefore decided to examine the results of peripheral attack by the reducing agents on the three lowest energy conformations of 28 (Figure 18). 28c 62c 0.35 kcal/mol R*,S* Figure 18. Investigation of peripheral attack on the three lowest energy conformations of 28. As one can see, the product expected from peripheral attack on the ketone group is the same for all three starting conformation, namely the 5R*,13S* alcohol. This is in contradiction with our experimental results which show a mixture of 61 and 62 in the case of all the reductions performed, regardless of the reducing agent. Indeed, the major product in the case of the Selectride reductions is the 5S*,13S* alcohol, as opposed to the predicted 5R*,13S*. Wu and Houk have shown that reduction of cyclic ketones occur so that there is minimal torsional strain introduced as the newly formed C-0 bond rotates inside or outside of the ring.6 1 Close examination of the conformations 28a, 28b, and 28c revealed that upon reduction from peripheral attack, the newly formed C-0 bond has to rotate past a C-C bond and this process is energetically disfavored. It seems therefore that reduction of 28 does not occur from the ground state conformations 28a, 28b, and 28c and that other conformations, of higher energy, must be involved. The fact that both unsubstituted borohydride reductions gave little or no selectivity simply confirmed that tetrahydridoborate is not a bulky group, and therefore can approach the carbonyl group from both faces more easily than do the substituted borohydride reagents. The difference in yield and isomer ratio between the Li-Selectride and K-Selectride reductions may be explained by the fact that the K-Selectride reduction gave some decomposition product. A higher decomposition rate for the minor isomer 62 would explain the higher 61:62 ratio. The most rewarding result in the various reductions performed was the reverse selectivity obtained with the Yamamoto reagent. The combination of MAD and rerr-butylmagnesium chloride gave us a selectivity of 20:80 for 61 and 62 respectively, a total reversal of the Selectride selectivity, as expected. A mechanism which may explain this result is shown in Figure 19. No stereochemical correlation between the product obtained in the figure and our results is implied: we are simply showing the manner of operation of the reduction process using MAD and tert-butylmagnesium chloride. It is interesting to note that the stereoselectivity in the MAD reduction was not only the reverse of that of the Selectride reagents, but also slightly better. The higher stereoselectivity for the Yamamoto reduction can possibly be explained by a greater steric requirement for the organoaluminium in its complexation of the ketone in 28. 50 Figure 19. Possible mechanism for the reduction of 28 with MAD and terf-butylmagnesium chloride. 2.4.2 Diastereoselective reduction of 29. Reductions on 29 were carried out with the same reagents as for 28. Thus, reduction of 29 with L-Selectride in THF at -78 °C gave a 100% yield of a 9:1 mixture of two diastereomeric alcohols of undetermined stereochemistry. This ratio was determined by NMR as we could not separate the diastereomers by GC. We were also unable to separate the two diastereomers by column chromatography. 66 7RM3S* 67 7S*,13S* Reduction of 29 with K-Selectride at -78 °C in THF gave an 83% yield of a mixture of alcohols in a slightly decreased selectivity of 80:20 compared to L-Selectride. This reduction proceeded much more smoothly than the same one performed on macrolide 28, with much less decomposition products. The sodium borohydride reduction of 29 gave, as for macrolide 28, little or no selectivity: a 1:1 ratio of alcohols in 91% yield. This confirmed again the inability of borohydride to differentiate between the si and re faces of the carbonyl group because of its small size. The Yamamoto reduction of 29, with MAD and rm-butylmagnesium chloride, also yielded poor results; a 74% yield of a 1:1 mixture of diastereomeric alcohols, with the same major product being predominant as was found in the Selectride reductions. The fact that the desired reverse selectivity (as compared to the Selectride reagents) was not achieved is unexplained at the moment. A summary of the results obtained is shown in Table V. Tab le V. Reduct ion of 7-oxo-13-tetradecanoIide (29) with various reducing agents . Reducing reagents Temperature Stereoselectivity Yield % 67 66 (S*,S*) (R*,S*) NaBH 4 -78 °C 50 50 91 L-Selectride -78 °C 90 10 100 K-Selectride -78 °C 80 20 83 MAD/rm-BuMgCl -78 °C 50 50 74 Predicted selectivity (MM2) -78 °C 100 We were now faced with a major problem: the determination of the relative stereochemistry of the two products in the reduction of 29. We could not obtain adequate crystals of 29, or of its p-bromobenzenesulfonate derivative, for a structure determination. We had to search for other means of determining the stereochemistry of the products. One possibility available to us was *H NMR. If the product existed largely in one major conformation, then we could possibly correlate the coupling constants of the alcohol methine proton with the coupling constants on low energy conformations obtained from a MACROMODEL search for both diastereomeric alcohols. As it turned out, the coupling constants for the alcohol methine proton were 8.5, 4, and 4 Hz, i.e. coupling constants consistent with a conformationally mobile molecule.17 This was not surprising in light of the report by Shannon et al. that cyclotetradecane itself existed in several different conformations in the liquid state.19 From these results, it was apparent we could not use lH NMR to ascertain the relative stereochemistry of the two alcohol products. With no physical means of assigning the stereochemistry of the products, we found ourselves restricted to the use of molecular mechanics calculations as a possible way of predicting the stereochemistry of the major reduction product. The results of peripheral hydride attack on macrolide 29's three lowest energy conformations are depicted in Figure 20. From our calculations, the three lowest energy conformations of 29 would all give rise to the same product upon peripheral hydride attack, namely the 7S*,13S* alcohol (67b is actually the enantiomer of 67a and 67c). It turned out that the five lowest energy conformations of 29 give the 7S*,13S* alcohol from peripheral hydride attack. MM2 calculations therefore suggest that the major product obtained from our reductions with the bulky Selectride reagents might be the 7S*,13S* alcohol 67, and the minor product the 7R*,13S* alcohol 66. Caution must be emphasized here because of the results obtained in the reduction of 28. 29a 0.0 kcal/mol 67a 0.66 kcal/mol 29b 67b 0.05 kcal/mol 0.0 kcal/mol 29c 67c 0.09 kcal/mol 3.25 kcal/mol Figure 20. Investigation of peripheral hydride attack on the three lowest energy conformations of 29. With the predicted ratio of 100:0 in favor of the 7S*,13S* over the 7R*,13S*, it was gratifying that the Selectride reagents again gave high facial selectivity toward the carbonyl's si and re faces. Once again, the sodium borohydride reduction displayed poor diastereoselectivity. The surprising result was the poor diastereoselectivity exhibited by MAD and rm-butylmagnesium chloride as opposed to the excellent selectivity in the reduction of keto lactone 28. At first it was thought that the MAD reagent had failed to complex with the ketone; however, upon addition of the MAD reagent to the keto lactone, the yellow colour indicative of successful complexation appeared instantly. Since complexation was occurring, there had to be another reason for the lack of selectivity. The rationale for this lack of selectivity is still unknown at this time. 2.5 Alkylation of 5-(r.3'-dithian-2'-vlV13-tetradecanolide (47) and 7-fl'.3'-dithian-2'- yl)-13-tetradecanolide (58) Diastereoselective carbon-carbon bond formation has figured predominantly in organic synthesis. Control of the diastereoselectivity has been achieved mostly through the use of chiral auxiliaries such as Evans' chiral amide and imide enolates,62 Masamune's optically active mandelic acid derivatives,63 and Meyers' chiral enamines and oxazolines.64 We were interested in seeing whether the conformational preferences of our two macrolides would induce diastereoselectivity in a-alkylation of the ketone and lactone groups. 2.5.1 Alkylation of 5-(l'.3'-dithian-2'-vlV13-tetradecanolide (47) Our first attempts involved alkylation of 5-oxo-13-tetradecanolide (28). Treatment of 28 with lithium di-isopropylamide, followed by addition of methyl iodide, gave only decomposition products with some recovered starting material. The result was the same using either lithium tetramethylpiperidide or lithium hexamethyldisilazide as bases. With the lack of success in alkylation of 28, we turned to the alkylation of 5-(l',3'-dithian-2'-yl)-13-tetradecanolide (47). Treatment of 47 with lithium di-isopropylamide at -78 °C, followed by addition of methyl iodide, gave a 95% yield of an 85:15 mixture (GC: the response factor is assumed to be the same for both isomers) of diastereomeric oc-methylated lactones 68. No dialkylated product was observed. We were unable to separate the diastereomers by column chromatography. 47 68 85:15 To determine the relative stereochemistry of the products, 68 was desulfurized with Ra-Ni in THF; work-up and co-injection with authentic materials for GC identification established that the major product was (2R*,13S*)-2-methyl-13-tetradecanolide (69) and the minor product (2S*,13S*)-2-methyl-13-tetradecanolide (70).65 57 68 Ra-Ni THF, A 69 2R*,13S* 70 2S*,13S* Therefore the major product from the alkylation of 47 was (2R*,13S*)-2-methyl-5-(l',3'-dithian-2'-yl)-13-tetradecanolide (71) and the minor product (2S*,13S*)-2-methyl-5-(r,3'-dithian-2'-yl)-13-tetradecanolide (72). The kinetic nature of the reaction was confirmed by base equilibration of the 85:15 mixture of 71:72 to a ratio of 49:51. 71 72 2R*,13S* 2SM3S* A MACROMODEL search showed that the three lowest energy conformations of 47 are, in order of increasing energy, a [3434] (47a), a [3434] (47b), and a [3344] (47c) conformation. Interestingly, the two [3434] conformations are the same, the only difference being in the orientation of the dithiane ring relative to the macrocycle. All three conformations have an s-trans dihedral angle at the lactone group. In accordance with Dale's prediction on gem-disubstituted 14-membered rings, the dithiane ring occupied a corner position (Figure 21).10 Deprotonation of 47 should involve loss of the hydrogen which is the closest to being parallel to the p orbitals of the TC bond of the carbonyl group (stereoelectronic control approach).66 This results in 73a, 73b, and 73c as possible conformations for the enolate (Figure 21). 59 0.0 kcal/mol 0.67 kcal/mol 47c 73c 1.28 kcal/mol 2.25 kcal/mol Figure 21. Proton abstraction of the three lowest energy conformations of 47: strain energies for enolates relative to lowest energy enolate 73e. Alkylation of the three conformations 73a, 73b, and 73c should occur from the same direction as the deprotonation. This would then lead to formation of the 2R*,13S* isomer from the three conformations as the sole product of the reaction. This is in reasonable agreement with our results, in which we saw an 85:15 ratio of products in favor of the 2R*,13S* isomer over the 2S*,13S* isomer. We were curious to see whether the gem-dimethylated compound obtained from alkylation of either 71 or 72 would adopt a conformation in which the two gem-disubstituted carbons would occupy corner positions. Accordingly, the original 85:15 mixture of 68 was treated with lithium di-isopropylamide in T H F at -78 °C. Subsequent addition of methyl iodide gave an 87% yield of an 85:15 mixture of 2,2-dimethyl-5-(l',3'-6 8 85:15 0 L D A , THF, -78 °C ii) CH 3 I , -78 °C 7 4 7 2 2 S M 3 S * dithian-2'-yl)-13-tetradecanolide (74) and (2S*,13S*)-2-methyl-5-(l',3,-dithian-2,-yl)-13-tetradecanolide (72). Again the mixture proved very difficult to separate. We decided to hydrolyze the dithiane ring in the hope that the resulting product mixture would be easier to separate. Thus, the 85:15 mixture of 74 and 72 was treated with N-bromosuccinimide in aqueous acetone at 0 °C to give an 85:15 mixture of 2,2-dimethyl-5-oxo-13-tetradecanolide (75) and (2S*,13S*)-2-methyl-5-oxo-13-tetradecanolide (76). Even this mixture proved quite difficult to separate, but very careful column chromatography enabled us to obtain a small quantity of pure 75 as an oil. Although it would have been interesting to get a structure determination for 75 to study its conformation, we were unable to crystallize the oil. 74 72 2S*,13S* NBS, aqueous acetone 0 ° C 75 76 2S*,13S* Pure 2,2-dimethyl-5-(r,3'-dithian-2'-yl)- 13-tetradecanolide (74) was obtained in 89% yield as colourless crystals by treating 75 with propanedithiol (38) in the presence of boron trifluoride-etherate in dichloromethane at 0 °C. Fortunately, crystals suitable for structure determination were obtained for 74. The stereodiagram for 74 is shown in Figure 22. Figure 22. Stereodiagram for macrolide 74. As can be seen quite clearly on the stereodiagram, 74 exists in a low energy [3434] conformation, with an s-trans dihedral angle at the lactone group, a C-O-C-H angle of 43°, and with both gem-disubstituted carbons occupying corner positions, as predicted by Dale.1 0 In addition, a MACROMODEL search of compound 74 showed that the lowest energy conformation was a [3434] conformation identical to the one shown in the stereodiagram of 74, as can be seen on the polar maps (Figure 23): a very satisfying result. 8 8 A B Figure 23. Polar maps for macrolide 74 from crystal structure (A), and from MM2 calculation (B). The polar map of the lowest energy conformation from MM2 calculation (B) is actually the mirror image of the one obtained by plotting the torsional angles read from the crystal structure (A); for purpose of comparison, the signs of the torsional angles in (B) have been reversed.. Something puzzled us in the alkylation of 68 to the dimethyl compound 74, and that was that compound 72 did not react to give compound 74. One possible explanation for this was that the lactone's remaining oc-hydrogen in 72 might not be parallel to the p orbitals of the TC bond of the carbonyl group, thus rendering its abstraction by base more difficult. It was immediately apparent that if 72 was the trans isomer and adopted a conformation similar to one of the starting low energy conformations 47a, 47b, and 47c, the methyl group would be sitting in a corner position; this would then give us a product in which the lactone's remaining oc-hydrogen is in a good position for abstraction by base and compound 72 should then react to give 74. However, competitive protonation of such an enolate (the protonated base being the proton source) may give back compound 72. A MACROMODEL conformational search performed on the intermediate enolate 73 revealed that the three lowest energy conformations, 73d, 73e, and 73f all had the methyl group in a pseudo-axial position, presumably to relieve interaction between the methyl group and the oxygen anion in the enolate. Alkylation from the more open face would then give as a product the (2R*,13R*)-dimethylated lactone 72. Thus the 85:15 ratio of 71:72 might be explained by alkylation of the enolates 73a-73f in unequal amounts. 73f 0.77 kcal/mol If compound 72 exists in conformation 72a, then the lactone's remaining cc-hydrogen would be in a position where its abstraction by base under stereoelectronic 66 control is much less likely, thus preventing 72 from reacting to form 74. This is another possible explanation for the lack of reaction of 72. 2.5.2 Alkylation of 7-fl'.3'-dithian-2'-vlV13-tetradecanolide (58) Treatment of 58 with lithium di-isopropylamide in THF at -78 °C, followed by addition of methyl iodide, gave a 100% yield of a 75:25 mixture (GC: the response factor is assumed to be the same for both isomers) of diastereomeric oc-methylated lactones 77. No dialkylated product was observed. To determine the relative stereochemistry of the products, 77 was desulfurized with Ra-Ni in THF; work-up and co-injection with authentic materials for GC identification established that the major product was (2R*,13S*)-2-methyl-13-tetradecanolide (69) and the minor product (2S*,13S*)-2-methyl-13-tetradecanolide (70).65 Therefore the major product from the alkylation of 7-(l',3'-dithian-2'-yl)-13-tetradecanolide (58) was (2R*,13S*)-2-methyl-7-(r,3'-dithian-2'-yl)-13-tetradecanolide (77a) and the minor product (2S*,13S*)-2-methyl-7-(l*,3'-dithian-2,-yl)-13-tetradecanolide (77b). 77a 77b 2R*,13S* 2S*,13S* major minor Fortunately, we obtained further confirmation on the stereochemistry of the products as the major isomer 77a crystallized out of the oily product mixture in crystals suitable for a structure determination. As one can see in the stereodiagram (Figure 24), the stereochemistry is clearly 2R*,13S*. In addition, the lactone group is in the s-trans arrangement and the C-O-C-H dihedral angle has a value of 22°. Both facts agree with the restrictions previously used for 14-membered lactone rings. Comparison of the polar map of the crystal structure of 77a with that of the lowest energy conformation obtained in a MACROMODEL search showed the two to be quite similar. The conformation is the same but for a difference of a few degrees in some of the torsional angles (Figure 25): the only noticeable difference is for torsional angle 12. Once again, the polar maps are mirror images of each other, the sign of the torsional angles in (B) have been reversed for ease of comparison. Figure 24. Stereodiagram for macrolide 77a. Figure 25. Polar maps for 77a from crystal structure MM2 calculation (B). (A) and The stereoelectronic control approach was used again to rationalize the results of alkylation of 58. However, these results are not as satisfactory as for the alkylation of 47. The three lowest energy conformations for 58, obtained in a MACROMODEL search (Figure 26), were all [3344] conformations; two of them, 58b and 58c, are epimers. The Boltzmann distribution for these three conformations at -78 °C was calculated to be 37:37:26 for 58a:58b:58c. Abstraction of the proton closest to being parallel to the p orbital of the TC bond of the carbonyl group would give rise to the enolates 78a, 78b, and 78c.66 Alkylation from the same direction as proton abstraction should then give (2S*,13S*)-2-methyl-7-(l',3'-dithian-2'-yl)-13-tetradecanolide (77b) from 78a and 78b, and (2S*,13R*)-2-methyl-7-(l',3'-dithian-2'-yl)- 13-tetradecanolide (77a) from 78c. Using the Boltzmann distribution obtained for the starting conformations, the calculated product ratio was 74:26 in favor of 77b over 77a; this was at odds with our experimental ratio of 80:20 in favor of 77a over 77b. H 58a 78a 0.0 kcal/mol Base Base 58b 78b 0.02 kcal/mol 58c 78c 0.83 kcal/mol Figure 26. Proton abstraction of the three lowest energy conformations of 58. Thus we considered the possibility that the intermediate enolate 78 might undergo a conformational change. A MACROMODEL search on the intermediate enolate 78 showed that the three lowest energy conformations would yield 77b as final product. After this 71 negative result, we were at a loss as to finding an adequate explanation for the stereochemical outcome of the alkylation of 58, and this stereoselectivity remains unexplained. Once again, we decided to verify whether gem-disubstitution would occur at a corner position, as predicted by Dale.1 0 Thus, the mixture of 77a and 77b was treated with lithium di-isopropylamide in THF at -78 °C to give a 33% yield of 2,2-dimethyl-7-(l',3'-dithian-2'-yl)-13-tetradecanolide (79) as an oil. We were unable to obtain suitable crystals of 79 for a structure determination, which would have enabled us to further corroborate Dale's prediction. i) LDA, THF -78 °C ii) CH3I, -78 °C 77a,b 79 We also prepared 2,2-dimethyl-7-oxo-13-tetradecanolide (80), which was obtained by oxidative hydrolysis of the dithiane group in 79 by N-bromosuccinimide. Although 80 was obtained as white crystals, they were inadequate for an X-ray structure determination. 72 k ^ 5 NBS Acetone/water 0 ° C 0' 79 80 2.6 Preparation of enol ethers from 28 and 29 Enol ethers are very useful synthetic intermediates. We decided to see whether we could selectively form one of the four possible enol ethers obtainable from 28 and 29. As selective generation of enol ethers is quite difficult from ketones having a secondary carbon atom on each side, success in this matter on our part would be a step forward for conformational control. 0 R R, ORi ss . ^ R + R 2 \ ^ Many methods for formation of enol ethers are available in the literature.67 Most involve vigorous conditions; however, Emde et al. have reported that the use of silyl trifluoromethanesulfonates in the presence of triethylamihe is mild and convenient for the preparation of enol ethers 6 7 c ' d They also noted that the use of only 0.8 equivalents of triethylamine led to an equilibrium controlled mixture of enol ethers. We used this method on both our macrolides, 28 and 29. 2.6.1 Enol ether generation from 28 We decided to generate two different kinds of enol ethers from macrolide 28, namely the trimethylsilyl and rm-butyldimethylsilyl enol ethers. Initial treatment of 28 with 1 equivalent of trimethylsilyl trifluoromethanesulfonate and 0.8 equivalent of triethylamine gave us no product. We had to increase the amount of reagents to 1.5 equivalents of trimethylsilyl trifluoromethanesulfonate and 2 equivalents of triethylamine before we were able to isolate a mixture of three isomeric enol ethers in a ratio of 14:72:14 (GC). 82a 82b The three isomeric enol ethers were unseparable, and we could not determine their structure. Still and co-workers had previously used the results of MM2 calculations to rationalize the results of equilibrium controlled reactions in cyclic compounds.29 We therefore conducted a MACROMODEL search in order to find the lowest energy conformations for each of the possible products. The Boltzmann distribution at 0 °C for the lowest energy conformations of all four possible enol ethers gave a 98:1.5:0.5 ratio for 81a:82b:81b respectively; that ratio was quite different from the experimental ratio, which could imply that the reaction did not proceed under strict equilibrium control. If we compare the calculated ratio to our experimental ratio of 14:72:14, one might conclude that the enol ether 81a, the 4Z isomer, was the major product of the reaction, with the remainder composed of 82b (the 5Z isomer) and 81b (the 4E isomer). Regardless of the true structure of the product, an experimental ratio of 14:72:14 represents an acceptable example of conformational control, especially in such a mobile system. It was thought that the use of a bulkier silylating reagent might result in a greater selectivity in enol ether formation from 28. Thus, we turned toward rm-butyldimethylsilyl trifluoromethanesulfonate as a silylating agent. Treatment of 28 with tert-butyldimethylsilyl trifluoromethanesulfonate and triethylamine yielded a mixture of three isomeric enol ethers, again unseparable, in a ratio of 19:70:11. This was a rather disappointing result. One would have thought that the size of the ferr-butyldimethylsilyl group would have induced a higher selectivity than that obtained. Once again, a MACROMODEL search was carried out on all four possible enol ethers, and a Boltzmann distribution calculated at 0 °C. The resulting calculated ratio of 80:19.5:0.5 for 83a:83b:84b respectively was closer to the experimental result of 19:70:11 than in the TMS enol ether formation. From a comparison of the calculated and experimental ratios, one might conclude that the major product in the formation of the TBDMS enol ethers from 28 was compound 83a. An interesting point was the fact that in both reactions, the product with the lowest energy conformation was the 4Z isomer for both the TMS and the TBDMS enol ethers. In addition, the lowest energy conformations for 83a (the TMS enol ether) and 85a (the TBDMS enol ether) were the same; both had the twist conformation (Figure 27). 81a 83a Figure 27. Lowest energy conformations (MM2) of 81a and 83a. 2.6.2 Enol ether generation from 29 Treatment of 29 with trimethylsilyl trifluoromethanesulfonate and triethylamine yielded a mixture of four isomeric enol ethers in a ratio of 74:12:6:8. We searched, with the help of MACROMODEL, for the lowest energy conformations for the four possible products. A Boltzmann distribution at 0 °C gave a ratio of 98.2:1.7:0.1 for 86a:85b:86b respectively. This calculated ratio and the experimental one are quite different again, as for the TMS enol ether formation from 28. As for the previous enol ether formations, the difference between the calculated and the experimental ratios could possibly be explained by incomplete equilibration of the product mixture. To compare the selectivity of the TMS enol ether formation with that of a bulkier silylating agent, 29 was treated with rm-butyldimethylsilyl trifluoromethanesulfonate and triethylamine at 0 °C to give a mixture of isomeric enol ethers in a ratio of 69:6:16:9. Again, we were disappointed with the fact that selectivity did not increase with the bulkier silylating agent. A MACROMODEL search gave us the lowest energy conformations of each isomeric product and the calculated Boltzmann distribution turned out to be 83:17 for 88a and 87b respectively. This is much closer to the experimental ratio than was the case in the TMS enol ether formation. Of course, as we could not assign any structures to the product obtained because they were unseparable, we could only assume that the two major products of the reaction were 88a and 87b. TBDMSO 88b The product with the lowest energy conformation was the 7Z isomer for both the TMS and the TBDMS enol ethers of 29. Both products, 86a and 88a, had the same low energy conformation; recall that this was also the case in the formation of enol ethers from 28, where both products, the TMS and the TBDMS enol ethers, had the 4Z geometry (Figure 28). •TBDMS 86a 88a Figure 28. Lowest energy conformations (MM2) for 86a and 88a. 2.7 Conclusion We have found that the conformational behavior of 14-membered lactones can be complex; in particular, we showed, in the case of lactone 28 for example, that the lowest energy conformations obtained from a MACROMODEL search need not necessarily be involved in reduction reactions. We performed a number of reactions on 14-membered lactones with a ketone group either at C5 (compound 28) or C7 (compound 29) to investigate the conformational control due to large rings, particularly in hydride reductions, alkylations, and enol ether formations. The diastereoselectivity of some reactions could be rationalized with the aid of Boltzmann distributions of the conformations of starting materials; some results, on the other hand, were left unexplained. This rationalization based on extensive modeling had the disadvantage of requiring large amounts of computer time for the conformational searches used in MACROMODEL. The results of the conformational searches carried out, in combination with other arguments such as increase of torsional strain in the ketone reductions, showed that one cannot use only ground state conformations to explain product formation in various reactions. CHAPTER III 80 Experimental 3.1 General Unless otherwise stated, all reactions were performed under N2 in flame-dried glassware. The cold temperature baths used were: dry ice-acetone (-78 °C), dry ice-chloroform (-60 °C), dry ice-carbon tetrachloride (-20 °C) and ice-water (0 °C). Anhydrous reagents and solvents were prepared according to the procedure given in the literature.68 Alkyllithium reagents (Aldrich Chemical Co.) were standardized against diphenylacetic acid in THF, a faint yellow color being indicative of the end-point. Analytical gas-liquid chromatography (GC) was performed on a Hewlett-Packard model 5880A, equipped with a split mode capillary injector and a flame ionization detector, using either a 0.22 mm diameter DB-210 column of 12 m in length or a 0.22 mm diameter SE-30 column of 12 m in length. Helium was the carrier gas. Retention times (Rt) are reported as follows: Rt (column type, initial temperature/time at initial temperature/rate of temperature increase/final temperature, carrier gas pressure). Silica gel 60, 230-400 mesh, supplied by E. Merck Co., was used for preparative flash column chromatography. Thin layer chromatography was performed on aluminium-backed plates of silica gel 60 (2 mm thickness), also supplied by E. Merck Co. Melting points were performed on a Fisher-Johns hot stage melting point apparatus and are corrected. Infra-Red (IR) spectra were recorded on a BOMEM FT-IR Michaelson-100 connected to an IBM compatible computer. IR spectra were taken neat on NaCl plates or in chloroform solution using NaCl cells of 0.2 mm thickness. The spectrometer is internally calibrated by a HeNe laser to a resolution of 4 cm"1. Proton nuclear magnetic resonance (*H NMR) spectra were recorded in deuterochloroform solutions on a Bruker AC-200 (200 MHz), Varian XL-300 (300 MHz), or a Bruker WH-400 (400 MHz). Chemical shifts are given in parts per million (ppm) on the 8 scale versus tetramethylsilane (8 0 ppm) or chloroform (5 7.27 ppm) as internal standards. Signal multiplicity, coupling constants, and integration ratios are given in parentheses. Low resolution mass spectra (LRMS) were recorded on a Kratos-AEI model MS 50 or model MS 9 spectrometer. Only peaks with greater than 20% relative intensity or those which were analytically useful are reported. High resolution mass spectra (HRMS) were obtained from a Kratos-AEI model MS 50 spectrometer. An ionization potential of 70 eV was used in all measurements. Microanalyses were done at the microanalytical laboratory at the University of British Columbia on a Carlo Erba Elemental Analyzer 1106. 8 2 3.2 6-Bromo-l-hexanol 31 A suspension of 1,6-hexanediol (17.1 g, 0.144 mol) and 48% aqueous hydrobromic acid (8.9 M , 40 mL, 0.36 mol) was added to a liquid-liquid extractor. The mixture was heated at 80 °C under continuous extraction with n-heptane. After 44 hours, the two layers were separated. The aqueous layer was saturated with NaCl and extracted with diethyl ether. The combined organic phases were washed successively with saturated NaHCC>3 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. Evaporation of solvents afforded a golden oil which was purified by column chromatography with 3:1 petroleum ether-ethyl acetate as eluant to give 20.7 g (79%) of 31 as a colourless oil. R t (SE-30, 80 °C/2 min/20 °C min-V200 °C, 10 psi) 3.22 min. IR (neat) 3340, 2980, 2845, 1448, 1260, 1055 cm-l. *H N M R (CDCI3, 400 MHz) 5 1.33-1.53 (m, 5H), 1.58 (m, J = 7 Hz, 2H), 1.87 (m, J = 6.5 Hz, 2H), 3.40 (t, J = 7 Hz, 2H), 3.63 (t, J = 6.5 Hz, 2H). L R M S (m/z) 164 (8lBr, M+ - H 2 0 , 1), 162 (79Br, M+ - H 2 0 , 2), 136 (8), 134 (9), 83 (54), 55 (100). H R M S calcd for C 6 H n 8 l B r (M+ - H 2 0) : 164.0025; found: 164.0022. calcd for Q3H1 ^ B r (M+ - H 2 0): 162.0045; found: 162.0037. 3.3 6-Bromo-l-(2-tetrahydropyranyIoxy)hexane 32 To a solution of 6-bromo-l-hexanol (10.0 g, 55 mmol) in CH2CI2 (300 mL) was added 3,4-dihydro-2//-pyran (6.0 mL, 66 mmol) and a catalytic amount of p-toluenesulfonic acid monohydrate. The resulting mixture was stirred at room temperature for 1.5 hours, after which time it was washed successively with saturated NaHCC>3 solution and saturated NaCl solution, then dried (MgSC»4) and filtered. Removal of solvent and column chromatography (12:1 petroleum ether-ethyl acetate as eluant) yielded 13.2 g (90%) of 32 as a clear, colourless oil. RT (SE-30, 80 °C/2 min/20 °C min-V200 °C, 10 psi) 8.01 min. IR (neat) 2936, 2862, 1448, 1352, 1200, 1030, 985 cm-1. *H NMR (CDCI3, 400 MHz) 5 1.34-1.76 (m, 12H), 1.87 (m, J = 7 Hz, 2H), 3.34-3.45 (m, 3H), 3.50 (m, IH), 3.73 (dt, J = 7, 10 Hz, IH), 3.86 (m, IH), 4.56 (t, J = 4 Hz, IH). LRMS (m/z) 265 ( 8 1Br, M+ - 1, 2), 263 ( 7 9Br, M+ - 1, 2), 185 (1), 85 (100), 56 (38), 55 (48), 41 (44). HRMS calcd for C n H 2 o 0 2 8 1 B r (M+ - 1): 265.0628; found: 265.0634. calcd for CnH 2 o0 2 7 9 Br (M+ - 1): 263.0647; found: 263.0651. 84 3.4 9-Hydroxy-2-nonanone 34 Methyl acetoacetate (7.5 mL, 70 mmol) was slowly added to a suspension of NaH (2.8 g of a 60% suspension in mineral oil, 70 mmol) in THF (150 mL) and DMF (50 mL). After the effervescence had sudsided, 6-bromo-l-(2-tetrahydropyranyloxy)hexane (9.2 g, 35 mmol) in THF (50 mL) was added dropwise over 15 minutes and the resulting solution was refluxed for 44 hours. The mixture was then quenched with NH4CI, diluted with diethyl ether and washed with water, followed by saturated NaCl solution. The solvent was removed and the golden oil obtained was refluxed in 3M HCl (100 mL) and methanol (35 mL) for 1.5 hours. The mixture was then extracted with CH2CI2. The organic phase was washed successively with saturated NaHCC«3 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. Evaporation of the solvent and column chromatography (3:1 petroleum ether-ethyl acetate as eluant) gave 4.3 g (79%) of 34 as a colourless oil. R t (SE-30, 80 °C/2 min/20 °C min-1/200 °C, 10 psi) 4.30 min. IR (CHCI3) 3621, 2933, 2875, 1710, 1363, 1042 cnr*. !H NMR (CDCI3, 400 MHz) 5 1.20-1.45 (m, 6H), 1.58 (m, 4H), 2.14 (s, 3H), 2.43 (t, J = 7.5 Hz, 2H), 3.64 (t, J = 6.5 Hz, 2H). LRMS (m/z) 158 (M+ 2), 140 (3), 100 (26), 97 (35), 83 (56), 82 (100), 71 (59), 58 (66), 55 (21). HRMS calcd for C 9 H i 8 0 2 : 158.1306; found: 158.1311. 85 3.5 9-Hydroxy-2-nonanone ethylene acetal To a solution of 9-hydroxy-2-nonanone (12.4 g, 78 mmol) in benzene (200 mL) was added ethanediol (13.1 mL, 0.23 mol) and a catalytic amount of pyridinium p-toluenesulfonate. The mixture was refluxed with water separation by a Dean-Stark trap for 4 hours. The solvent was removed and the oil obtained was dissolved in diethyl ether. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC»4) and filtered. Evaporation of the solvent afforded 14.4 g (91%) of 36 as a colourless oil. R t (SE-30, 80 °C/2 min/20 °C min-1/200 °C, 10 psi) 5.67 min. IR (CHC13) 3685, 2935, 2870, 1477, 1428, 1216, 1023, 928 cm"1. tH NMR (CDCI3, 400 MHz) 8 1.31 (s, 3H), 1.32-1.42 (m, 8H), 1.55 (m, 2H), 1.63 (m, 2H), 2.33 (br s, IH), 3.61 (t, J = 6 Hz, 2H), 3.93 (m, 4H). LRMS (m/z) 187 (M + - C H 3 , 13), 87 (100), 43 (28). HRMS calcd for C10H19O3 (M+ - CH 3): 187.1334; found: 187.1334. 8 6 3.6 9-Bromo-2-nonanone ethylene acetal A solution of 9-hydroxy-2-nonanone ethylene acetal (14.4 g, 91 mmol) in CH2CI2 (350 mL) was cooled to -78 °C. Triphenylphosphine (25.2 g, 96 mmol) was added in one portion, followed by carbon tetrabromide (31.8 g, 96 mmol). The reaction mixture was stirred at -78 °C for 1.5 hours and the cooling bath removed. After stirring at room temperature for 1 hour, petroleum ether was added and the precipitated triphenylphosphine oxide was filtered off. Solvents were removed and the oil dissolved in 1:1 petroleum ether-diethyl ether to again precipitate, and filter off, the triphenylphosphine oxide. Solvent evaporation and column chromatography (9:1 petroleum ether-ethyl acetate eluant) gave 16.7 g (88%) of 37 as a colourless oil. R t (SE-30, 80 °C/2 min/20 °C min-V200 °C, 10 psi) 5.62 min. IR (neat) 2934, 2863, 1454, 1375, 1229, 1060, 947, 856 cm-1. IH NMR (CDCI3, 300 MHz) 5 1.31 (s, 3H), 1.30-1.50 (m, 8H), 1.85 (m, J = 9 Hz, 2H), 3.41 (t, J = 9 Hz, 2H), 3.95 (m, 4H). LRMS (m/z) 251 ( 8 1Br, M+ - C H 3 , 8), 249 ( 7 9Br, M + - CH3, 8), 87 (100). HRMS calcd for CioHi 8 0 2 8 1 Br (M+ - CH 3): 251.0470; found: 251.0470. calcd for C i 0 H i 8 O 2 7 9 B r (M + - CH 3): 249.0490; found: 249.0494. 3.7 5-(lS3'-Dithian-2'-yl)-l-(2-tetrahydropyranyloxy)pentane 87 1,3-Propanedithiol (13 mL, 0.18 mol) and 3,4-dihydro-2//-pyran (25 mL, 0.28 mol) were dissolved in CH2CI2 (300 mL) and the solution cooled to 0 °C. Boron trifluoride etherate (27 mL, 0.22 mol) was added dropwise over 30 minutes and the mixture was stirred overnight. After quenching with water, the organic phase was washed successively with 3N NaOH solution, water and saturated NaCl solution, then dried (MgS04) and filtered. To the dry organic layer was then added 3,4-dihydro-2H-pyran (20 mL, 0.22 mol) and a catalytic amount of /?-toluenesulfonic acid monohydrate. The resulting solution was stirred for 2 hours at room temperature after which time it was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgS04) and filtered. Evaporation of the solvent followed by column chromatography (6:1 petroleum ether-ethyl acetate eluant) yielded 26 g (72%) of 41 as a slightly yellow oil. Rt (SE-30, 80 °C/2 min/20 °C mirr^OO °C, 10 psi) 9.62 min. IR (CHCI3) 2943, 2869, 1454, 1352, 1277, 1135, 1030 cm-1. IH NMR (CDCI3, 400 MHz) 8 1.45-1.95 (m, 14H), 2.86 (m, 4H), 3.40 (dt, J = 6.5, 10 Hz, IH), 3.51 (m, IH), 3.75 (dt, J = 6.5, 10 Hz, IH), 3.86 (m, IH), 4.06 (t, J = 7 Hz, IH), 4.58 (t, J = 3.5 Hz, IH). LRMS (m/z) 276 (M+ 2), 191 (100), 119 (45), 85 (59). HRMS calcd for C13H24O2S2: 276.1217; found: 276.1210. c 88 3.8 10-(l',3'-Dithian-2'-yl)-14-(2-tetrahydropyranyloxy)-2-tetradecanone ethylene acetal A solution of 5-(r,3'-dithian-2'-yl)-l-(2-tetrahydropyranyloxy)pentane (3.7 g, 14 mmol) in THF (70 mL) was cooled to -20 °C and n-butyllithium (1.6 M in hexane,8.8 mL, 14 mmol) added dropwise over 15 minutes. After stirring for 2 hours, a solution of 9-bromo-2-nonanone ethylene acetal (3.3 g, 12 mmol) in THF (30 mL) was added dropwise and the resulting mixture was stirred for 1 hour at -20 °C. The cooling bath was removed and, after stirring at room temperature for 1 hour, the reaction was quenched with saturated NH4CI solution. The mixture was washed successively with water and saturated NaCl solution, then dried (MgSCU) and filtered. Evaporation of solvent gave a yellow sirupy oil which was subjected to column chromatography (12:1 petroleum ether-ethyl acetate eluant) to provide 5.0 g (88%) of 42 as a slightly yellow oil. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.37. IR (neat) 2934, 2865, 1451, 1374, 1201, 1137,1034 cm-1. IH NMR (CDCI3, 400 MHz) 5 1.31 (s, 3H), 1.30-2.00 (m, 8H), 2.81 (m, 4H), 3.42 (dt, J = 7, 10 Hz, IH) 3.51 (m, IH), 3.77 (dt, J = 7, 10 Hz, IH), 3.83-3.97 (m, 5H), 4.58 (t, J = 3.5 Hz, IH). LRMS (m/z) 460 (M+, 3), 376 (3), 361 (2), 303 (6), 191 (12), 87 (100), 85 (33), 55 (29), 43 (56). HRMS calcd for C24H44O4S2: 460.2681; found: 460.2683. 89 3.9 10-(l,,3'-Dithian-2'-yl)-14-hydroxy-2-tetradecanone 10-(1^3'-Dithian-2'-yl)-14-(2-terrahydropyranyloxy)-2-tetradecanone ethylene acetal (5.4 g, 12 mmol) was dissolved in a mixture of acetone (60 mL), water (60 mL), and methanol (120 mL), followed by the addition of a catalytic amount of pyridinium p-toluenesulfonate. The solution was heated to 55 °C and stirred for 2 hours. The solvents were evaporated and the residue dissolved in diethyl ether. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 3.8 g (97%) of 43 as a colourless oil. Rf (1:1 petroleum ether-ethyl acetate eluant) 0.25. IR (CHC13) 3687, 3620, 2937, 2858, 1710, 1458, 1362, 1246, 1052 cm-1. *H NMR (CDCI3, 300 MHz) 6 1.20-1.70 (m, 14H), 1.82-2.00 (m, 6H), 2.14 (s, 3H), 2.43 (t, J = 7.5 Hz, 2H), 2.81 (t, J = 5.5 Hz, 2H), 2.82 (t, J = 5.5 Hz, 2H), 3.68 (t, J = 6.5 Hz, 2H). LRMS (m/z) 332 (M+, 9), 259 (48), 225 (27), 191 (62), 98 (66), 55 (72), 43 (100). HRMS calcd for C17H32O2S2: 332.1845; found: 332.1845. 3.10 5-(lS3'-Dithian-2'-yI)a3-oxotetradecanaI 44 To a solution of 10-(r,3'-dithian-2'-yl)-14-hydroxy-2-tetradecanone (3.9 g, 12 mmol) in benzene (100 mL) was added successively: DMSO (9.0 mL, 120 mmol), dichloroacetic acid (0.50 mL, 6.0 mmol), and 1,3-dicyclohexylcarbodiimide (DCC) (7.4 g, 36 mmol). The reaction mixture was stirred overnight, then diluted with ethyl acetate and the excess DCC destroyed by slow addition of a solution of oxalic acid dihydrate (3.6 g, 29 mmol) in methanol (25 mL). The resulting suspension was poured into saturated NaCl solution and the mixture filtered to remove the urea precipitate. The layers were separated and the organic phase was washed successively with water and saturated NaCl (more urea precipitated, removed again by fitration), then dried (MgS04) and filtered. Removal of solvents and column chromatography (6:1 petroleum ether-ethyl acetate eluant) afforded 2.7 g (70%) of 44 as a yellowish oil. Rf (5:3 petroleum ether-ethyl acetate eluant) 0.38. IR (CHCI3) 2935, 2857, 2725, 1714, 1419, 1358, 1275 cm"1. *H NMR (CDCI3, 400 MHz) 5 1.32 (m, 6H), 1.44 (m, 2H), 1.58 (m, 2H), 1.73-2.00 (m, 8H), 2.14 (s, 3H), 2.43 (t, J = 8 Hz, 2H), 2.49 (m, J = 7, 2 Hz, 2H), 2.82 (m, 4H), 9.1 (t, J = 2 Hz, IH). LRMS (m/z) 330 (M+, 7), 259 (22), 189 (28), 145 (25), 55 (55), 43 (100). HRMS calcd for C17H30O2S2: 330.1687; found: 330.1687. 3.11 5-(r,3'-Dithian-2'-yl)-13-oxotetradecanoic acid 5-(r,3'-dithian-2'-yl)-13-oxotetradecanal (2.7 g, 8.1 mmol) was dissolved in THF (75 mL) and water (47 mL). Silver nitrate (6.9 g, 41 mmol) was added to this solution, followed by 3M NaOH (28 mL, 81 mmol). After stirring for 2 hours, the black precipitate was removed by filtration through a pad of Celite® and the Celite® washed with diethyl ether and 3M NaOH. The aqueous phase was separated and acidified with 3M HCl, then extracted with chloroform. The chloroform phase was washed with saturated NaCl solution, then dried (MgS04) and filtered. Evaporation of the solvent provided 2.7 g (96%) of pure 45 as a viscous oil. Rf (9:1 chloroform-methanol eluant with one drop of acetic acid) 0.38. IR (CHCI3) 3200, 2989, 2939, 2860, 1711, 1455, 1360 cm"1. *H NMR (CDCI3, 300 MHz) 5 1.32 (m, 6H), 1.45 (m, 2H), 1.85 (m, 4H), 1.95 (m, 4H), 2.18 (s, 3H), 2.45 (m, 4H), 2.85 (m, 4H). LRMS (m/z) 346 (M+, 8), 259 (44), 205 (74), 145 (43), 107 (32), 97 (35), 79 (50), 71 (53), 67 (51), 55 (69), 45 (51), 43 (100), 41 (78). HRMS calcd for C17H30O3S2: 346.1636; found: 346.1639. 3.12 5-(l',3'-Dithian-2'-yl)-13-hydroxytetradecanoic acid 92 5-(r,3'-Dithian-2'-yl)-13-oxotetradecanoic acid (2.2 g, 6.4 mmol) was dissolved in ethanol (50 mL) and the solution was cooled to 0 °C. Sodium borohydride (0.49 g, 13 mmol) was added slowly and the reaction mixture was stirred at 0 °C for 1 hour. The excess borohydride was destroyed with IM HC1, the solution was diluted with water and extracted with CH2CI2. The organic phase was washed with saturated NaCl solution, then dried (MgS04) and filtered. Removal of solvent gave 2.2 g (100%) of 46 as a viscous colourless oil. Rf (9:1 chloroform-methanol eluant with one drop of acetic acid) 0.30. IR (CHCI3) 3684, 2933, 1747, 1715, 1425, 1217, 1021, 928 cm"1. *H NMR (CDCI3, 400 MHz) 5 1.20 (d, J = 6 Hz, 3H), 1.25-1.55 (m, 12H), 1.82 (m, 4H), 1.97 (m, 4H), 2.41 (t, J = 7 Hz, 2H), 2.82 (m, 4H), 3.82 (m, J = 6 Hz, 2H). LRMS (m/z) 348 (M+, 17), 261 (35), 205 (100), 145 (32), 107 (24). HRMS calcd for C17H30O3S2: 348.1793; found: 348.1789. 93 3.13 5-(lS3'-Dithian-2'-yl)-13-tetradecanolide 47 l-Phenyl-l//-tetrazole-5-thiol (0.61 g, 3.4 mmol) was suspended in toluene (25 mL). tert-Butyl isocyanide (0.39 mL, 3.4 mmol) was added and the solution was stirred until homogeneous. 5-(r,3'-Dithian-2'-yl)-13-hydroxytetradecanoic acid (990 mg, 2.84 mmol) in toluene (25 mL) was added. This solution was diluted with toluene (825 mL) and refluxed for 1 hour. The toluene was removed to give a dark golden oil which was submitted to column chromatography (9:1 petroleum ether-ethyl acetate eluant) to yield 736 mg (78%) of 47 as a yellowish oil. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.47. IR (CHCI3) 2936, 2860, 1718, 1458, 1371, 1258, 1167, 1100 cm-1. J H NMR (CDCI3, 400 MHz) 5 1.23 (d, J = 6 Hz, 3H), 1.25-1.75 (m, 12H), 1.80-2.20 (m, 8H), 2.53 (m, J = 12, 6.5, 3 Hz, IH), 2.81 (m, 5H), 4.94 (m, J = 6, 10, 2 Hz, IH). LRMS (m/z) 330 (M+ 62), 315 (4), 255 (21), 203 (49), 169 (22), 145 (65), 135 (23), 106 (62), 79 (33), 67 (47), 55 (53), 41 (100). HRMS calcd for C17H30O2S2: 330.1688; found: 330.1686. 94 3.14 5-Oxo-13-tetradecanoIide 2 8 N-bromosuccinimide (3.0 g, 17 mmol) was dissolved in a mixture of acetone (45 mL) and water (5 mL). The solution was cooled to 0 °C and 5-(l\3'-dithian-2'-yl)-13-tetradecanolide (621 mg, 1.9 mmol) in acetone (15 mL) was added slowly with stirring. After 10 minutes, the solution was shaken with a mixture of saturated Na2SC*3 solution and 1:1 CH2Cl2-hexane. The organic phase was washed successively with saturated NaHC03 solution, water and saturated NaCl solution, then dried (MgSC>4) and filtered. Removal of solvents and column chromatography (12:1 petroleum ether-ethyl actetate eluant) gave 390 mg (86%) of 28 as a colourless oil. R t (DB-210, 140 °C, 20 psi) 9.10 min. IR (CHC13) 2936, 2862, 1727, 1710, 1458, 1374, 1262, 1132 cm"1. * H NMR (CDC13, 400 MHz) 8 1.20 (d, J = 6 Hz, 3H), 1.20-1.40 (m, 8H), 1.45-1.70 (m, 4H), 1.90 (m, 2H), 2.20 (m, J = 14, 7, 5 Hz, IH), 2.28-2.55 (m, 4H), 2.79 (m, J = 17, 8, 7 Hz, IH), 5.02 (m, J = 6, 8, 4 Hz, IH). LRMS (m/z) 240 (M+ 6), 225 (6), 130 (42), 125 (29), 115 (100), 112 (95), 97 (38), 87 (42), 84 (50), 69 (61), 67 (25), 55 (62), 43 (36), 41 (64). HRMS calcd for C14H24O3: 240.1725; found, 240.1723. 3.15 6-Bromohexanal Oxalyl chloride (9.1 mL, 0.10 mol) was dissolved in CH2CI2 (200 mL) and the solution was cooled to -60 °C. DMSO (16 mL, 0.23 mol) in CH2CI2 (40 mL) was added and the mixture was stirred for 15 minutes. A solution of 6-bromo-l-hexanol (15.7 g, 87 mmol) in CH2CI2 (80 mL) was added dropwise over 15 minutes and the reaction mixture was stirred at -60 °C for 30 minutes, after which time triethylamine (61 mL, 0.44 mol) was added. After 10 minutes stirring, the cold bath was removed and water added. The organic phase was washed successively with water and saturated NaCl solution, then dried (MgSC»4) and filtered. Removal of solvents gave a yellow oil which was purified by column chromatography (9:1 petroleum ether-ethyl acetate eluant) to give 12.7 g (82%) of 48 as a colourless oil. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.52. IR (neat) 2931, 2857, 2719, 1724, 1460, 1390, 1253 cm"1. *H NMR (CDCI3, 200 MHz) 5 1.45 (m, 2H), 1.65 (m ,J = 7 Hz, 2H), 1.95 (m, J = 7.5 Hz, 2H), 2.45 (m, J = 7, 2 Hz, 2H), 3.40 (t, J = 7.5 Hz, 2H), 9.75 (t, J = 2 Hz, IH). LRMS (m/z) 181 ( 8 1Br, M+ + 1, 1), 179 (79Br, M+ + 1, 1), 81 (46), 57 (20), 55 (42), 45 (25), 44 (100), 41 (35). HRMS calcd for C 5Hn8lBr (M+ - CO): 152.0025; found: 152.0040. calcd for C6Hi 0 O 7 9 Br (M+ - 1): 176.9916; found: 176.9914. 96 3.16 6-Bromohexanal ethylene acetal To a solution of 6-bromohexanal (6.5 g, 36 mmol) in benzene (180 mL) was added ethanediol (10.1 mL, 181 mmol) and a catalytic amount of pyridinium p-toluenesulfonate. The mixture was refluxed with water separation by a Dean-Stark trap for 3.5 hours. The solvent was removed and the oil obtained was dissolved in diethyl ether. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. Evaporation of the solvent yielded 7.8 g (96%) of 51 as a colourless oil. Rf (3:1 petroleum ether-acetone eluant) 0.54. IR (CHC13) 2945, 2879, 1461, 1401, 1132, 1044, 950 cm"1. J H NMR (CDCI3, 400 MHz) 8 1.48 (m, 4H), 1.67 (m, 2H), 1.89 (m, J = 7 Hz, 2H), 3.42 (t, J = 7 Hz, 2H), 3.86 (m, 2H), 3.97 (m, 2H), 4.86 (t, J = 5 Hz, IH). LRMS (m/z) 223 ( 8 1Br, M+ - 1, 1), 221 ( 7 9Br, M+ - 1, 1), 73 (100), 45 (17). HRMS calcd for C 8 H i 4 0 2 8 1 B r (M+ - 1): 223.0158; found: 223.0158. calcd for C8Hi40 2 7 9Br (M+ - 1): 221.0177; found: 221.0186. 97 3.17 7-Bromo-2-heptanol Magnesium turnings (0.52 g, 21 mmol) were suspended in diethyl ether (50 mL), along with a few crystals of iodine. Iodomethane (1.3 mL, 21 mmol) was added and the mixture was gently refluxed for 30 minutes. The reaction mixture was then cooled to 0 °C and a solution of 6-bromohexanal (3.5 g, 19 mmol) in diethyl ether (40 mL) was added dropwise over 15 minutes. The resulting solution was stirred for 30 minutes, after which time the cooling bath was removed and the reaction mixture was quenched with saturated NH4CI solution. The organic phase was separated, washed successively with water and saturated NaCl solution, then dried (MgSCU) and filtered. Removal of the solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 3.0 g (79%) of 49 as a colourless oil. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.29. IR (CHCI3) 3613, 2934, 2861, 1458, 1380, 1069 cm-l. ! H NMR (CDCI3, 200 MHz) 5 1.15 (d, J = 6 Hz, 3H), 1.40 (m, 6H), 1.70 (s, IH), 1.85 (m, J = 6 Hz, 2H), 3.40 (t, J = 6 Hz, 2H), 3.75 (m, J = 6 Hz, IH). LRMS (m/z) 195 (8lBr, M+ - 1, 0.1), 193 ( 7 9Br, M+ - 1, 0.2), 181 (2), 179 (2), 97 (17), 45 (100). HRMS calcd for C 7 Hi 4 08lBr (M+ - 1): 195.0209; found: 195.0180. calcd for C7Hi4079Br (M+ - 1): 193.0228; found: 193.0229. 98 3.18 7-Bromo-2-(2-tetrahydropyranyloxy)heptane To a solution of 7-bromo-2-heptanol (10.1 g, 52 mmol) in CH2CI2 (300 mL) was added 3,4-dihydro-2//-pyran (5.7 mL, 62 mmol) and a catalytic amount of pyridinium p-toluenesulfonate. The resulting mixture was stirred at room temperature for 3 hours, after which time it was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC«4) and filtered. Removal of solvent and column chromatography (12:1 petroleum ether-ethyl acetate eluant) yielded 14.1 g (98%) of 50 (a clear, colourless oil) as a mixture of diastereomers. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.69. IR (CHCI3) 2939, 2860, 1454, 1377, 1125, 1074, 1027, 981 cm-1. *H NMR (CDCI3, 400 Mhz) 5 1.12, 1.24 (2t, J = 6 Hz, 3H), 1.30-2.00 (m, 14H), 3.41, 3.42 (2t, J = 7 Hz, 2H), 3.50 (m, IH), 3.74, 3.80 (2m, J = 6 Hz, IH), 3.91 (m, IH), 4.64, 4.70 (2t, J = 5 Hz, IH). LRMS (m/z) 236 ( 8 1Br, M+ - C2H4O, 4.4), 234 (79Br, M+ - C 2 H 4 0 , 4.5), 179 (38), 177 (38), 103 (25), 102 (36), 101 (68), 97 (53), 57 (50), 56 (64), 55 (60), 43 (54). HRMS calcd for Ci2H 2 20 2 8 1 Br (M+ - 1): 279.0784; found: 279.0812. calcd for Ci2H 220 2 7 9Br (M+ - 1): 277.0804; found: 277.0803. 3.19 7-(l\3'-Dithian-2'-yl)heptanal ethylene acetal 99 53 1,3-Dithiane (4.6 g, 38 mmol) was dissolved in THF (75 mL) and the solution was cooled to -20 °C. n-Butyllithium (1.6 M in hexane, 24 mL, 38 mmol) was added over 15 minutes and the solution was stirred for 2 hours. 6-Bromohexanal ethylene acetal (7.7 g, 34 mmol) in THF (100 mL) was then added and the reaction mixture stirred for 1 hour, after which time the cooling bath was removed. After stirring for an additional hour, the reaction was quenched with saturated NH4CI solution. The mixture was washed successively with water and saturated NaCl solution, then dried (MgSCU) and filtered. Evaporation of solvent gave a yellow oil which was subjected to column chromatography (9:1 petroleum ether-acetone eluant) to provide 7.5 g (84%) of 53 as a colourless oil. Rf (3:1 petroleum ether-acetone eluant) 0.49. IR (CHCI3) 2982, 2940, 2861, 1461, 1363, 1277, 1135, 1037 cm"1. *H NMR (CDCI3, 400 MHz) 8 1.32-1.50 (m, 4H), 1.54 (m, J = 7.5 Hz, 2H), 1.67 (m,2H), 1.77 (m, J = 8 Hz, 2H), 1.88 (m, 2H), 2.85 (m, 4H), 3.85 (m, 2H), 3.97 (m, 2H), 4.04 (t, J = 7 Hz, IH), 4.85 (t, J = 5 Hz, IH). LRMS (m/z) 262 (M+, 20), 119 (49), 73 (100), 45 (20). HRMS calcd for C12H22O2S2: 262.1061; found: 262.1058. 100 3.20 7-(l',3'-Dithian-2'-yl)-13-(2-tetrahydropyranyIoxy)tetradecanal ethylene acetal 54 7-(r,3'-Dithian-2'-yl)heptanal ethylene acetal (7.4 g, 28 mmoles) was dissolved in THF (100 mL), the mixture cooled to -20 °C and H-butyllithium (1.6 M in hexane, 18 mL, 29 mmol) was added. After 2 hours stirring, a solution of 7-bromo-2-(tetrahydropyranyloxy)heptane (8.7 g, 31 mmol) was added and the reaction mixture was stirred for 1 hour, after which time the cooling bath was removed. Stirring was continued for another hour, then the reaction was quenched with saturated NH4CI solution. The organic phase was washed successively with water and saturated NaCl solution, then dried (MgSC«4) and filtered. Removal of solvent and column chromatography (12:1 petroleum ether-acetone eluant) gave 10.6 g (82%) of 54 (a pale yellow oil) as a mixture of diastereomers. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.39. IR (CHCI3) 2940, 2861, 1433, 1276, 1132, 1074, 1028 cm"1. IH NMR (CDCI3, 400 MHz) 8 1.12, 1.23 (2d, J = 6 Hz, 3H), 1.30-1.75 (m, 22H), 1.85 (m, 4H), 1.96 (m, 2H), 2.82 (m, 4H), 3.50 (m, IH), 3.72, 3.78 (2m, J = 6 Hz, 1H),3.84-4.00 (m, 5H), 4.64, 4.71 (2t, J = 3.5 Hz, IH), 4.86 (2 overlapping triplets, J = 5 Hz, IH). LRMS (m/z) 460 (M+, 4), 376 (7), 359 (1), 261 (20), 119 (20), 85 (37), 73 (100), 67 (20), 55 (41), 45 (26), 41 (33). HRMS calcd for C24H44O2S2: 460.2681; found: 460.2684. 101 3.21 7-(lS3,-Dithian-2^yI)-13-hydroxytetradecanaI a. From 7-(l\3'-dithian-2'-yl)-13-(2-tetrahydropyranyloxy')tetradecanal ethylene acetal 7-(r,3'-dithian-2'-yl)-13-(2-tetrahydropyranyloxy)tetradecanal ethylene acetal (10.2 g, 22 mmol) was dissolved in a mixture of acetone (75mL), methanol (150 mL) and water (75 mL), followed by the addition of a catalytic amount of pyridinium p-toluenesulfonate. The solution was heated to 55 °C and stirred for 4 hours. The solvents were evaporated and the residue dissolved in diethyl ether. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC«4) and filtered. Removal of solvent and column chromatography (6:1 petroleum ether-ethyl acetate eluant) gave 2.4 g (33%) of 55 as a colourless oil. The major product was 7-(l',3'-dithian-2'-yl)-13-hydroxytetradecanal dimethyl acetal 56 (2.99 g, 36%). Data for 7-(r)3'-dithian-2'-yl)-13-hydroxytetradecanal (55) Rf (3:1 petroleum ether-ethyl acetate eluant) 0.17. IR (CHC13) 3612, 2937, 2860, 2725, 1723, 1439, 1275, 1083 cnr*. ! H NMR (CDCI3, 400 MHz) 5 1.21 (d, J = 6 Hz, 3H), 1.30-1.55 (m, 12H), 1.68 (m, J = 7 Hz, 2H), 1.88 (m, 4H), 1.94 (m, 2H), 2.44 (m, J = 7, 2 Hz, 2H), 2/81, 2.82 (2t, J = 5.5 Hz, 4H), 3.81 (m, J = 6 Hz, IH), 9.78 (t, J = 2 Hz, IH). LRMS (m/z) 332 (M+, 25), 233 (56), 219 (20), 217 (100), 145 (32), 119 (46), 109 (26), 95 (29), 81 (42), 79 (29), 67 (32), 55 (37), 43 (38), 41 (41). HRMS calcd for C17H32O2S2: 332.1844; found: 332.1845. 102 Data for 7-(r,3l-dithian-2'-yl)-13-hydroxytetradecanal dimethyl acetal (56) Rf (3:1 petroleum ether-ethyl acetate eluant) 0.31. IR (CHC13) 3613, 2938, 2860, 1444, 1275, 1127, 1057 cm"1. J H NMR (CDCI3, 400 MHz) 5 1.20 (d, J = 6 Hz, 3H), 1.30-1.50 (m, 14H), 1.61 (m, 2H), 1.86 (m, 4H), 1.95 (m, 2H), 2.81, 2.82 (2t, J = 5.5 Hz, 4H), 3.32 (s, 6H), 3.80 (m, J = 6 Hz, IH), 4.36 (t, J = 7 Hz, IH). LRMS (m/z) 378 (M+ 13), 346 (11), 249 (45), 233 (49), 231 (38), 145 (29), 137 (29), 125 (26), 106 (40), 97 (79), 81 (58), 79 (44), 75 (87), 71 (64), 67 (63), 55 (58), 45 (44), 43 (46), 41 (100). HRMS calcd for C19H38O3S2: 378.2263; found: 378.2256. b. From 7-(r.3'-dithian-2'-yl)-13-hydroxytetradecanal dimethyl acetal 7-(l',3'-Dithian-2'-yl)-13-hydroxytetradecanal dimethyl acetal (2.9 g, 7.7 mmol) was dissolved in acetone (60 mL) and a catalytic amount of p-toluenesulfonic acid monohydrate was added. The solution was stirred at 55 °C for 3 hours. The solution was concentrated to 20 mL and the residue dissolved in diethyl ether. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSCU) and filtered. Removal of solvent afforded 2.6 g (100%) of 55. 3.22 7-(lS3'-Dithian-2^yl)-13-hydroxytetradecanoic acid 103 7-(r,3'-Dithian-2'-yl)-13-hydroxytetradecanal (2.6 g, 7.8 mmol) was dissolved in THF (75 mL) and water (16 mL). Silver nitrate (15.1 g, 89 mmol) was added to this solution, followed by 3M NaOH (59 mL, 177 mmol). After stirring for 2 hours, the black precipitate was removed by filtration through a pad of Celite® and the Celite® washed with diethyl ether and 3M NaOH. The aqueous phase was separated and acidified with 3M HC1, then extracted with chloroform. The chloroform phase was washed with saturated NaCl solution, then dried (MgS04) and filtered. Evaporation of the solvent provided 2.2 g (81%) of pure (NMR) 57 as a viscous oil. Rf (9:1 chloroform-methanol eluant with one drop of acetic acid) 0.22. IR (CHCI3) 3618, 2975, 2838, 2864, 1746, 1716, 1455, 1388, 1263, 1047 cm"1. *H NMR (CDCI3, 400 MHz) 5 1.21 (d, J = 6 Hz, 3H), 1.30-1.55 (m, 12H), 1.68 (m, J = 7 Hz, 2H), 1.85 (m, 4H), 1.96 (m, 2H), 2.36 (m, J = 7 Hz, 2H), 2.81, 2.82 (2t, J = 5.5 Hz, 4H), 3.81 (m, J = 6 Hz, IH). LRMS (m/z) 348 (M+, 10), 330 (3.1), 233 (100), 145 (19), 107 (27), 106 (26), 81 (21). HRMS calcd for C17H32O3S2: 348.1793; found: 348.1799. 1 0 4 3.23 7-(lS3^Dithian-2'-yl)-13-tetradecanoIide 58 1 -Phenyl-l//-tetrazole-5-thiol (0.90 g, 5.1 mmol) was suspended in toluene (25 mL). tert-Butyl isocyanide (0.54 mL, 5.1 mmol) was added and the solution was stirred until homogeneous. 7-(r,3'-Dithian-2'-yl)-13-hydroxytetradecanoic acid (1.18 g, 2.33 mmol) in toluene (25 mL) was added and the solution was diluted with toluene (900 mL) and refluxed for 1 hour. The toluene was removed to give a dark golden oil which was submitted to column chromatography (12:1 petroleum ether-ethyl acetate eluant) to yield 614 mg (55%) of 58 as colourless crystals. R t (DB-210, 200 °C, 20 psi) 7.38 min. mp 74.5-75.5 °C (hexane). IR (CHCI3) 2941, 2861, 1719, 1441, 1337, 1228, 1159 cm"1. *H NMR (CDCI3, 400 MHz) 5 1.22 (d, J = 6 Hz, 3H), 1.30-1.64 (m, 12H), 1.71 (m, 2H), 1.95 (m, 4H), 2.07 (m, 2H), 2.36 (m, 2H), 2.80 (m, 4H), 5.14 (m, IH). LRMS (m/z) 331 (M+ + 1, 20), 330 (M+, 100), 255 (49), 231 (87), 223 (33), 214 (32), 145 (66), 106 (61), 105 (37), 95 (39), 81 (66), 79 (45), 67 (60), 55 (98), 41 (99). HRMS calcd for C17H30O2S2: 330.1687; found: 330.1690. 105 3.24 7-Oxo-13-tetradecanolide 29 N-bromosuccinimide (1.5 g, 8.3 mmol) was dissolved in a mixture of acetone (20 mL) and water (5 mL). The solution was cooled to 0 °C and 7-(r,3'-dithian-2'-yl)-13-tetradecanolide (306 mg, 0.93 mmol) in acetone (10 mL) was added slowly with stirring. After 10 minutes, the solution was shaken with a mixture of saturated Na2SC>3 solution and 1:1 CH2Cl2-hexane. The organic phase was washed successively with saturated NaHCC>3 solution, water and saturated NaCl solution, then dried (MgS04) and filtered. Removal of solvents and column chromatography (12:1 petroleum ether-ethyl actetate eluant) gave 147 mg (66%) of 29 as a colourless oil. R t (DB-210, 160 °C, 20 psi) 5.47 min. IR (CHCI3) 2939, 2863, 1719, 1712, 1439, 1362, 1257, 1127 crrr1. *H NMR (CDCI3, 400 MHz) 5 1.22 (d, J = 5 Hz, 3H), 1.10-1.45 (m, 4H), 1.50-1.75 (m, 6H), 1.79 (m, 2H), 1.91 (m, 2H), 2.20-2.42 (m, 4H), 2.50-2.62 (m, 2H), 4.88 (m, J = 6, 9.5, *3 Hz, IH). LRMS (m/z) 240 (M+, 6), 140 (42), 125 (35), 98 (40), 97 (42), 82 (27), 69 (31), 55 (100), 43 (29), 42 (29), 41 (48). HRMS calcd for C14H24O3: 240.1725; found: 240.1731. 106 3.25 5-Hydroxy-13-tetradecanolide HOA 61 62 a. Reduction of 5-oxo-13-tetradecanolide with excess Li-Selectride 5-Oxo-13-tetradecanolide (54 mg, 0.23 mmol) was dissolved in THF (5 mL) at -78 °C. Li-Selectride (IM in THF, 0.68 mL, 0.68 mmol) was added in one portion. The solution was stirred for 3 hours and then quenched by addition of 30% H2O2 (2 mL), followed by 3M NaOH (2 mL). The solution was diluted with water, then extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 28 mg (52%) of 61 as colourless crystals. R t (DB-210, 140 °C, 20 psi) 7.11 (62), 8.03 (61) min. mp 47-49 °C (hexane). IR (CHCI3) 3611, 2935, 2882, 1718, 1455, 1374, 1259, 1132, 1056, 974 cm"1. *H NMR (CDCI3, 400 MHz) 8 1.22 (d, J = 6 Hz, 3H), 1.25-2.00 (m, 18H), 2.30 (m, J = 15, 8.5, 4 Hz, IH), 2.51 (m, J = 15, 9.5, 3.5 Hz, IH), 3.77 (m, IH), 5.02 (m, J = 6.5 Hz, IH). LRMS (m/z) 242 (M+ 1), 224 (6), 117 (55), 99 (100), 95 (28), 81 (34), 71 (21), 69 (29), 67 (21), 55 (39), 41 (35). HRMS calcd for C14H26O3: 242.1882; found: 242.1876. ANAL calcd: C, 69.38; H, 10.85 % found: C, 69.48; H, 10.85 %. 107 b. Reduction of 5-oxo-13-tetradecanoIide with Li-Selectride 5-Oxo-13-tetradecanolide (60 mg, 0.25 mmol) was dissolved in THF (5 mL) at -78 °C. Li-Selectride (1M in THF, 0.26 mL, 0.26 mmol) was added over 15 minutes. The solution was stirred for 30 minutes and then quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). The solution was diluted with water, then extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 59 mg (97%) of a 73:27 mixture (GC) of 6 1 and 6 2 as colourless crystals. c. Reduction of 5-oxo-13-tetradecanolide with K-Selectride 5-Oxo-13-tetradecanolide (45 mg, 0.19 mmol) was dissolved in THF (5 mL) at -78 °C. K-Selectride (1M in THF, 0.21 mL, 0.21 mmol) was added over 15 minutes. The solution was stirred for 30 minutes, and then quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). The solution was diluted with water, then extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 13 mg of 5-oxo-13-tetradecanolide and 26 mg (58%) of a 84:16 mixture (GC) of 6 1 and 6 2 as colourless crystals. d. Reduction of 5-oxo-13-tetradecanolide with sodium borohydride 5-Oxo-13-tetradecanolide (37 mg, 0.15 mmol) was dissolved in methanol at -78 °C and NaBH4 (2mg, 0.05 mmol) added. The solution was stirred for 6 hours, with addition of NaBFLj (4 mg) each hour. The mixture was then quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). The solution was diluted with water, then extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 36 mg (100%) of a 43:57 mixture (GC) of 6 1 and 6 2 as colourless crystals. 108 e. Reduction of 5-oxo-13-tetradecanolide with tetramethylammonium borohydride 5-Oxo-13-tetradecanolide (57 mg, 0.24 mmol) was dissolved in methanol (5 mL) and tetramethylammonium borohydride (62 mg, 0.24 mmol) added at -78 °C with stirring. After 2 hours more reducing agent (64 mg, 0.24 mmol) was added. After 3 hours, the mixture was then quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). The solution was diluted with water, then extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) gave 56 mg (98%) of a 41:59 mixture (GC) of 61 and 62 as colourless crystals. f. Reduction of 5-oxo-13-tetradecanolide with methylaluminium bis(2.6-di-rm-butyl-4- methylphenoxide) (MAD) and terf-butylmagnesium chloride 5-Oxo-13-tetradecanolide (43 mg, 0.18 mmol) was dissolved in a solution of MAD (0.26M in toluene, 4.2 mL, 1.1 mmol). The solution was cooled to -78 °C and rm-butylmagnesium chloride (2M in THF, 0.54 mL, 1.1 mmol) added in one portion. After 45 minutes stirring, the reaction mixture was quenched by the addition of 3M HC1 and diluted with diethyl ether. It was washed successively with water and saturated NaCl solution, then dried (MgS04) and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl acetate eluant) yielded 40 mg (93%) of a 20:80 mixture (GC) of 61 and 62 as colourless crystals. 1 0 9 3.26 (5S*,13S*)-5-(p-Bromobenzenesulfonyloxy)-13-tetradecanoIide To a solution of (5S*,13S*)-5-hydroxy-13-tetradecanolide (28 mg, 0.12 mmol) in pyridine (3 mL) was added p-bromobenzenesulfonyl chloride (0.18 g, 0.70 mmol). The solution was stirred at room temperature overnight, then poured into 1M HCl. The aqueous solution was then extracted with CH2CI2. The organic phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. After removal of solvent, the residue was submitted to column chromatography (19:1 petroleum ether-ethyl acetate eluant) to give 49 mg (92%) of 63 as colourless crystals. Rf (3:1 petroleum ether-ethyl acetate) 0.58. mp 78.5-80.5 °C (hexane). IR (CHCI3) 2935, 2861, 1720, 1620, 1578, 1446, 1356, 1257, 1186, 1176, 1070, 824 cm-'1. * H N M R (CDCI3, 400 MHz) 6 1.21 (d, J = 6 Hz, 3H), 1.20-1.48 (m, 10H), 1.56 (m, 6H), 1.78 (m, 2H), 2.24 (m, IH), 2.42 (m, IH), 4.72 (m, IH), 5.02 (m, J = 6, 8.5, 3.5 Hz, IH), 7.70 (d, J = 9 Hz, 2H), 7.78 (d, J = 9 Hz, 2H). L R M S (m/z) 447 ( 8 1Br, M+ - C H 3 , 0.4), 445 ( 7 9Br, M + - C H 3 ) 0.5), 241 (22), 238 (38), 236 (36), 224 (75), 174 (22), 164 (30), 155 (32), 110 (38), 109 (36), 99 (97), 95 (64), 82 (81), 81 (100), 68 (70), 67 (84), 55 (79), 41 (69). H R M S calcd for Ci9H260sS 8 1Br (M+ - CH3): 447.0665; found: 447.0675 calcd for Ci 9H260 5S 7 9Br (M+ - CH3): 445.0684; found: 445.0685. A N A L calcd: C, 52.06; H, 6.33; S, 6.95 % found: C, 52.22; H, 6.31; SV, 7.00 %. 63 110 3.27 7-Hydroxy-13-tetradecanolide 66 67 a. Reduction of 7-oxo-13-tetradecanolide with Li-Selectride 7-Oxo-13-tetradecanolide (63 mg, 0.26 mmol) was dissolved in T H F (5 mL) and the solution was cooled to -78 °C. Li-Selectride (IM in THF, 0.29 mL, 0.29 mmol) was added in one portion and the mixture was stirred for 30 minutes. The cooling bath was removed and the reaction was quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). Water (15 mL) was added and the resulting mixture was extracted with diethyl ether. The ether phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent gave 64 mg (100%) of a 9:1 mixture (NMR) of 67 and 66 as colourless crystals. Data for 67 R t (DB-210, 160 °C, 20 psi) 3.97 min. mp 77-78 °C (hexane). IR (CHCI3) 3611, 2936, 2863, 1715, 1447, 1331, 1254, 1149 cm"1. *H NMR (CDCI3, 400 MHz) 8 1.22 (d, J = 6 Hz, 3H), 1.25-1.75 (m, 18H), 2.20-2.40 (m, 2H), 3.78 (m, J = 8.5, 4, 4.5 Hz, IH), 5.02 (m, J = 8.5, 6, 2.5 Hz, IH). LRMS (m/z) 242 (M+, 0.1), 224 (10), 145 (20), 127 (100), 116 (49), 109 (72), 98 (40), 81 (61), 67 (35), 55 (35). HRMS calcd for C14H26O3: 242.1882; found: 242.1880. 111 b. Reduction of 7-oxo-13-tetradecanolide with K-Selectride 7-Oxo-13-tetradecanolide (48 mg, 0.20 mmol) was dissolved in THF (5 mL) and the solution was cooled to -78 °C. K-Selectride (IM in THF, 0.22 mL, 0.22 mmol) was added in one portion and the mixture was stirred for 45 minutes. Then, more K-Selectride (0.22 mL, 0.22 mmol) was added and the mixture was stirred for an additional 45 minutes. The cooling bath was removed and the reaction was quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). Water (15 mL) was added and the resulting mixture was extracted with diethyl ether. The ether phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography gave 40 mg (83%) of an 80:20 mixture (NMR) of 67 and 66 as colourless crystals. c. Reduction of 7-oxo-13-tetradecanolide with sodium borohydride 7-Oxo-13-tetradecanolide (30 mg, 0.13 mmol) was dissolved in methanol (5 mL) at 0 °C. NaBH4 (5 mg, 0.13 mmol) was added in one portion and the mixture stirred for 2 hours. More NaBH4 (5 mg, 0.13 mmol) was added and stirring was continued for 1 hour. The cooling bath was removed and the mixture was then quenched by addition of 30% H2O2 (1 mL), followed by 3M NaOH (1 mL). The solution was diluted with water and extracted with diethyl ether. The organic phase was washed with saturated NaCl solution, dried (MgS04), and filtered. Removal of solvent and column chromatography (6:1 petroleum ether-ethyl acetate eluant) gave 27 mg (91%) of a 50:50 mixture (NMR) of 67 and 66 as colourless crystals. d. Reduction of 7-oxo-13-tetradecanolide with methylaluminium bis(2.6-di-rm-butyl-4- methylphenoxide) (MAD) and rm-butylmagnesium chloride 7-Oxo-13-tetradecanolide (39 mg, 0.16 mmol) was dissolved in a solution of MAD in toluene (0.13M, 3.7 mL, 0.49 mmol). The solution was cooled to -78 °C and tert-butylmagnesium chloride (2M in THF, 0.24 mL, 0.49 mmol) was added in one portion. After 1.5 hours stirring, the reaction mixture was quenched by the addition of 3M HC1 and diluted with diethyl ether. It was washed successively with water and saturated NaCl solution, then dried (MgS04) and filtered. Removal of solvent and column chromatography (3:1 petroleum ether-ethyl 112 acetate eluant) yielded 29 mg (74%) of a 50:50 mixture (NMR) of 67 and 66 as colourless crystals. 3.28 5-(l\3'-Dithian-2^yl)-2-methyl-13-tetradecanoIide Diisopropylamine (0.46 mL, 3.3 mmol) was dissolved in THF (5 mL) at -78 °C. n-Butyllithium (1.68M in THF, 2.0 mL, 3.3 mmol) was added and the solution was stirred for 10 minutes. A solution of 5-(l',3'-dithian-2'-yl)-13-tetradecanolide (725 mg, 2.2 mmol) in THF (10 mL) was added and the mixture was stirred at -78 °C for 1 hour. Methyl iodide (0.68 mL, 11 mmol) was added and stirring was continued for 1 hour, after which time the cooling bath was removed. The reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The organic phase was washed successively with saturated NaHC03 solution, water and saturated NaCl solution, then dried (MgSC>4) and filtered. Removal of solvents gave 721 mg (95%) of 68 (an oily yellowish solid) as an 85:15 mixture (GC) of diastereomers. R t (DB-210, 200 °C, 20 psi) 4.89 (minor product, 72), 5.60 (major product, 71) min. IR (CHCI3) 2937, 2863, 1714, 1455, 1275, 1251, 1190, 1077 cnr1. *H NMR (CDCI3, 400 MHz) 5 1.20 (m, 6H), 1.36-2.20 (m, 20H), 2.37, 2.65 (2m, IH each), 2.82 (m, 4H), 4.93 (m, IH). LRMS (m/z) 344 (M+ 53), 269 (24), 217 (26), 145 (100), 106 (51), 95 (29), 81 (41), 79 (32), 55 (60), 41 (94). HRMS calcd for C18H32O2S2: 344.1844; found: 344.1850. 68 113 3.29 2-Methyl-13-tetradecanolide 69 70 5-(r,3'-Dithian-2'-yl)-2-methyl-13-tetradecanolide (from experiment 3.28, 78 mg, 0.23 mmol) was dissolved in THF (10 mL) and Raney nickel (W-2 in aqueous solution of pH=10, 1.5 g) was added. The suspension was refluxed for 2 hours, then cooled to room temperature and diluted with diethyl ether. It was washed successively with water and saturated NaCl solution, then dried and filtered. Evaporation of solvents gave 38 mg (70%) of an 85:15 mixture (GC) of 69 and 70 by comparison with authentic materials.65 R t (DB-210, 130 °C/2 min/2 °C min-Vl40 °C, 20 psi) 5.56 (70), 6.25 (69) min. 114 3.30 2,2-Dimethyl.5-(l',3'-dithian-2'-yl)-13-tetradecanoIide 74 a. From alkylation of 2-methyl-5-(l\3'-dithian-2'-yl)-13-tetradecanolide with methyl iodide Diisopropylamine (0.41 mL, 2.9 mmol) was dissolved in THF (5 mL) at -78 °C. n-Butyllithium (1.68M in THF, 1.7 mL, 2.9 mmol) was added and the solution was stirred for 10 minutes. A solution of 2-methyl-5-(r,3'-dithian-2'-yl)-13-tetradecanolide (from experiment 3.28, 666 mg, 1.9 mmol) in THF (5 mL) was added and the mixture was stirred at -78 °C for 1.5 hour. Methyl iodide (0.60 mL, 9.7 mmol) was added and stirring was continued for 1 hour, after which time the cooling bath was removed. The reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The organic phase was washed successively with saturated NaHCC*3 solution, water and saturated NaCl solution, then dried (MgSC<4) and filtered. Removal of solvents and column chromatography (19:1 petroleum ether-ethyl acetate as eluant) gave 601 mg (87%) of an 85:15 mixture (GC) of 74 and 72. The mixture was used without further purification or characterization. b. From reaction of 2.2-dimethyl-5-oxo-13-tetradecanolide with 1.3-propanedithiol 1,3-Propanedithiol (10 uL, 0.10 mmol) and 2,2-dimethyl-5-oxo-13-tetradecanolide (21 mg, 0.08 mmol) were dissolved in CH2CI2 (5 mL) and the solution was cooled to 0 °C. Boron trifluoride etherate (20 uL, 0.16 mmol) was added and the solution was stirred for 4 hours. After quenching with water, the organic phase was washed successively with 3N NaOH solution, water 11 and saturated NaCl solution, then dried (MgSC<4) and filtered. Removal of solvent and column chromatography (19:1 petroleum ether-ethyl acetate as eluant) gave 25 mg (89%) of 74 as colourless crystals. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.60. mp 87-89 °C (hexane). IR (CHCI3) 2937, 2863, 1714, 1451, 1277, 1202, 1153, 1121, 1046 cm"1. *H NMR (CDCI3, 400 MHz) 8 1.18 (s, 3H), 1.20 (d, J = 6 Hz, 3H), 1.22 (s, 3H), 1.40-1.70 (m, 16H), 1.97 (m, 3H), 2.11 (m, J = 14, 5 Hz, IH), 2.80 (m, 4H), 4.89 (m, J = 6, 10, 2 Hz, IH). LRMS (m/z) 358 (M+, 33), 145 (89), 107 (47), 106 (31), 95 (26), 81 (31), 79 (25), 69 (36), 67 (37), 55 (64), 43 (47), 41 (100). HRMS calcd for C19H34O2S2: 358.2000; found: 358.1998. 116 3.31 2,2-Dimethyl-5-oxo-13-tetradecanolide 75 N-bromosuccinimide (2.7 g, 15 mmol) was dissolved in a mixture of acetone (36 mL) and water (4 mL). The solution was cooled to 0 °C and 2,2-dimethyl-5-(l',3'-dithian-2'-yl)-13-tetradecanolide (from experiment 3.30, 593 mg, 1.7 mmol) in acetone (15 mL) was added slowly with stirring. After 10 minutes, the solution was shaken with a mixture of saturated Na2SC«3 solution and 1:1 CH2Cl2-hexane. The organic phase was washed successively with saturated NaHCC>3 solution, water and saturated NaCl solution, then dried (MgSCU) and filtered. Removal of solvents and careful column chromatography (24:1 petroleum ether-ethyl acetate as eluant) gave 24 mg (5%) of 75 as colourless crystals and 311 mg of a mixture of 75 and 76. R t (DB-210, 140 °C, 20 psi) 13.80 min. mp 57-59 °C (hexane). IR (CHCI3) 2936, 2864, 1709, 1447, 1264, 1187, 1124 cm"1. *H NMR (CDCI3, 400 MHz) 8 1.13 (s, 3H), 1.22 (d, J = 6 Hz, 3H), 1.27 (s, 3H), 1.30-1.45 (m, 7H), 1.50-1.75 (m, 6H), 1.93 (m, J = 13.5, 3.5 Hz, IH), 2.22 (m, J = 14, 10.5, 5 Hz, IH), 2.34 (m, J = 18, 12.5, 5.5 Hz, IH), 2.50 (m, 2H), 4.96 (m, IH). LRMS (m/z) 268 (M + , 6), 181 (19), 143 (30), 140 (31), 125 (34), 115(82), 97 (75), 88 (35), 70 (52), 69 (71), 56 (31), 55 (75), 43 (50), 41 (100). HRMS calcd for C16H28O3: 268.2039; found: 268.2034. 117 3.32 7-(l,,3'-Dithian-2l-yl)-2-methyl-13-tetradecanolide 77 Diisopropylamine (71 uL, 0.51 mmol) was dissolved in THF (5 mL) at -78 °C. n-Butyllithium (1.56M in THF, 0.33 mL, 0.51 mmol) was added and the solution was stirred for 10 minutes. A solution of 7-(r,3'-dithian-2'-yl)-13-tetradecanolide (111 mg, 0.34 mmol) in THF (5 mL) was added and the mixture was stirred at -78 °C for 1 hour. Methyl iodide (0.11 mL, 1.7 mmol) was added and stirring was continued for 1 hour, after which time the cooling bath was removed. The reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The organic phase was washed successively with saturated NaHCC<3 solution, water and saturated NaCl solution, then dried (MgSC»4) and filtered. Removal of solvents gave 116 mg (100%) of 77 (an oily yellowish solid) as a 75:25 mixture (GC) of diastereomers. Rt (DB-210, 200 °C, 20 psi) 7.02 (minor product, 77b), 8.18 (major product, 77a) min. IR (CHCI3) 2939, 2860, 1721, 1711, 1446, 1243, 1194 cm-1. *H NMR (CDCI3, 400 MHz) 8 1.12 (m, 3H), 1.22 (m, 3H), 1.20-1.75 (m, 14H), 1.80-2.10 (m, 6H), 2.40-2.60 (m, IH), 2.78 (m, 4H), 4.95 (m, IH). LRMS (m/z) 344 (M+, 54), 269 (24), 231 (31), 145 (40), 107 (34), 106 (32), 95 (73), 81 (38), 79 (25), 67 (44), 55 (65), 43 (30), 41 (100). HRMS calcd for C18H32O2S2: 344.1844; found: 344.1889. 118 3.33 2-Methyl-13-tetradecanoIide (69,70) from 2-methyl-7-(ll,3,-dithian-2'-yl)-13-tetradecanoIide (77) 2-Methyl-7-(r,3'-dithian-2'-yl)-13-tetradecanolide (from experiment 3.32, 53 mg, 0.15 mmol) was dissolved in acetone (10 mL) and Raney nickel (W-2 in aqueous solution of pH=10, 1.5 g) was added. The suspension was refluxed for 2 hours, then cooled to room temperature and diluted with diethyl ether. It was washed successively with water and saturated NaCl solution, then dried and filtered. Evaporation of solvents gave 33 mg (89%) of a 75:25 mixture (GC) of 69 and 70 by comparison with authentic materials.65 3.34 2,2-Dimethyl-7-(l,,3'-dithian-2,-yl)-13-tetradecanolide Diisopropylamine (93 uL, 0.66 mmol) was dissolved in THF (5 mL) at 0 °C. n-Butyllithium (1.57M in THF, 0.42 mL, 0.66 mmol) was added and the solution was stirred for 10 minutes. A solution of 2-methyl-7-(r,3'-dithian-2'-yl)-13-tetradecanolide (from experiment 3.32, 77 mg, 0.22 mmol) in THF (5 mL) was added and the mixture was stirred at 0 °C for 1 hour. Methyl iodide (0.14 mL, 2.2 mmol) was added and stirring was continued for 1 hour, after which time the cooling bath was removed. The reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The organic phase was washed successively with saturated NaHC03 solution, water and saturated NaCl solution, then dried (MgS04) and filtered. Removal of solvents and column chromatography (24:1 petroleum ether-ethyl acetate as eluant) gave 26 mg 79 119 (33%) of 79 as a colourless oil. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.67. IR (CHC13) 2941, 2862, 1712, 1441, 1244, 1182, 1126 cm-1. *H NMR (CDCI3, 400 MHz) 5 1.13 (s, 3H), 1.17 (s, 3H), 1.18 (d, J = 6.5 Hz, 3H), 1.20-1.80 (m, 14H), 1.85-2.18 (m, 6H), 2.76 (m, 4H), 5.07 (m, IH). LRMS (m/z) 358 (M+, 72), 283 (27), 231 (48), 214 (30), 145 (37), 132 (30), 107 (36), 106 (35), 81 (43), 69 (31), 67 (46), 55 (80), 43 (41), 41 (100). HRMS calcd for C19H34O2S2: 358.2000; found: 358.1995. 3.35 2,2-Dimethyl-7-oxo-13-tetradecanoIide N-bromosuccinimide (77 mg, 0.44 mmol) was dissolved in a mixture of acetone (9 mL) and water (1 mL). The solution was cooled to 0 °C and 2,2-dimethyl-7-(l',3'-dithian-2'-yl)-13-tetradecanolide (26 mg, 0.07 mmol) in acetone (5 mL) was added slowly with stirring. After 10 minutes, the solution was shaken with a mixture of saturated Na2SC*3 solution and 1:1 CH2CI2-hexane. The organic phase was washed successively with saturated NaHCC>3 solution, water and saturated NaCl solution, then dried (MgSCU) and filtered. Removal of solvents and careful column chromatography (24:1 petroleum ether-ethyl acetate as eluant) gave 7 mg (37%) of 80 120 80 as colourless crystals. Rf (3:1 petroleum ether-ethyl acetate eluant) 0.56. mp 47-49 °C (hexane). IR (CHC13) 2940, 2867, 1715, 1459, 1376, 1255, 1121, 908 cnr*. IH NMR (CDCI3, 400 MHz) 5 1.13 (s, 3H), 1.16 (d, J = 6 Hz, 3H), 1.17 (s, 3H), 1.15-1.28 (m, 5H), 1.3-1.6 (m, 8H), 1.71 (m, J = 13, 4 Hz, IH), 1.86 (m, IH), 1.97 (m, IH), 2.28 (m, IH), 2.59 (m, IH), 4.80 (m, J = 6, 10, 3 Hz, IH). LRMS (m/z) 268 (M + , 5), 181 (20), 97 (24), 83 (30), 82 (65), 69 (26), 67 (27), 55 (100), 42 (20), 41 (70), 39 (22). HRMS calcd for C16H28O3: 268.2039; found: 268.2039. 3.36 5- (Tr imethylsi Iy Ioxy)-4- tetradecen-13-oI ide and 5-( tr imethylsi ly Ioxy)-5-tetradecen-13-oI ide 81 82 5-Oxo-13-tetradecanolide (40 mg, 0.17 mmol) and triethylamine (46 uL, 0.33 mmol) were dissolved in CH2CI2 (5 mL) and the solution was cooled to 0 °C. Trimethylsilyl trifluoromethanesulfonate (48 uL, 0.25 mmol) was added and the reaction mixture was stirred at 0°C for 1 hour, after which time the cooling bath was removed. After an additional hour, the reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The ether phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC»4) and filtered. Removal of solvent and column chromatography (19:1 petroleum 121 ether-ethyl acetate as eluant) gave 45 mg (87%) of a 14:72:14 mixture (GC) of isomeric enol ethers as a clear oil. R t (DB-210, 160 °C, 10 psi) 3.91, 4.51, 4.69 min. IR (CHC13) 2937, 2858, 1720, 1712, 1665, 1602, 1442, 1375, 1254, 1144, 844 cm"1. !H NMR (CDCI3, 400 MHz) 8 0.16-0.20 (3s, 9H total), 1.21 (m, 3H), 1.25-2.10 (m, 16H), 2.20- 2.50 (m, 2H), 4.48-4.65 (3m, IH total), 4.95-5.10 (2m, IH total). LRMS (m/z) 312 (M+, 7), 297 (2), 222 (1), 143 (39), 130 (35), 113 (22), 75 (51), 73 (100), 55 (24), 45 (26), 41 (26). HRMS calcd for Ci7H32Si03: 312.2121; found: 312.2129. 122 3.37 7-(TrimethyIsiIyloxy)-6-tetradecen-13-oIide and 7-(trimethyIsilyIoxy)-7-tetradecen-13-oIide 85 86 7-Oxo-13-tetradecanolide (44 mg, 0.18 mmol) and triethylamine (51 uL, 0.37 mmol) were dissolved in C H 2 C I 2 (5 mL) and the solution was cooled to 0 °C. Trimethylsilyl trifluoromethanesulfonate (53 uL, 0.28 mmol) was added and the reaction mixture stirred at 0°C for 1 hour, after which time the cooling bath was removed. After an additional hour, the reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The ether phase was washed successively with saturated NaHCC«3 solution and saturated NaCl solution, then dried (MgSC«4) and filtered. Removal of solvent and column chromatography (19:1 petroleum ether-ethyl acetate as eluant) gave 50 mg (88%) of a 74:12:6:8 mixture (GC) of isomeric enol ethers as a clear oil. R t (DB-210, 160 °C, 10 psi) 4.49, 4.85, 5.10, 5.55 min. IR (CHCI3) 2939, 2858, 1720, 1711, 1671, 1602, 1442, 1359, 1254, 844 cnr 1. *H NMR (CDCI3, 400 MHz) 8 0.15-0.20 (4s, 9H total), 1.20-1.70 (m, 16H), 1.85-2.10 (m, 3H), 2.30 (m, 2H), 4.30-4.70 (4m, IH total), 5.00 (m, IH). LRMS (m/z) 312 (M+, 12), 143 (26), 130 (62), 75 (50), 73 (100), 55 (20), 43 (20). HRMS calcd for Ci 7 H 3 2 Si03: 312.2121; found: 312.2114. 123 3.38 5-( /£r/-Butyldimethyls i lyloxy)-4-tetradecen-13-ol ide and 5-(tert-butyldimethyIsilyIoxy)-5-tetradecen-13-o!ide ((CH3)3C)(CH3)2Si 83 84 5-Oxo-13-tetradecanolide (52 mg, 0.22 mmol) and triethylamine (60 uL, 0.43 mmol) were dissolved in CH2CI2 (5 mL) and the solution was cooled to 0 °C. rm-Butyldimethylsilyl trifluoromethanesulfonate (75 uL, 0.33 mmol) was added and the reaction mixture was stirred at 0°C for 1 hour, after which time the cooling bath was removed. After an additional hour, the reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The ether phase was washed successively with saturated NaHCC«3 solution and saturated NaCl solution, then dried (MgSCU) and filtered. Removal of solvent and column chromatography (19:1 petroleum ether-ethyl acetate as eluant) gave 70 mg (90%) of a 19:70:11 mixture (GC) of isomeric enol ethers as a clear oil. R t (DB-210, 160 °C, 10 psi) 8.63, 9.79, 10.11 min. IR (CHCI3) 2936, 2858, 1720, 1712, 1669, 1608, 1364, 1256, 1148, 1039, 839 cm"1. * H NMR (CDCI3, 400 MHz) 8 0.10-0.16 (3s, 6H total), 0.90-0.98 (3s, 9H total), 1.20 (m, 3H), 1.25-2.10 (m, 14H), 2.37 (m, 4H), 4.45-4.65 (4m, IH total), 4.95-5.13 (2m, IH total). LRMS (m/z) 354 (M+ 0.3), 297 (46), 205 (22), 199 (28), 163 (28), 145 (30), 75 (100), 73 (67), 41 (21). HRMS calcd for C2oH 3 8Si0 3: 354.2590; found: 354.2590. 124 3.39 7-(ferf-ButyldimethyIsiIyloxy)-6-tetradecen-13-olide and l-(tert-butyldimethylsilyloxy)-7-tetradecen-13-olide 87 88 7-Oxo-13-tetradecanolide (54 mg, 0.23 mmol) and triethylamine (78 uL, 0.56 mmol) were dissolved in CH2CI2 (5 mL) and the solution was cooled to 0 °C. rm-Butyldimethylsilyl trifluoromethanesulfonate (130 uL, 0.56 mmol) was added and the reaction mixture was stirred at 0°C for 1 hour, after which time the cooling bath was removed. After an additional hour, the reaction was quenched with saturated NH4CI solution and diluted with diethyl ether. The ether phase was washed successively with saturated NaHC03 solution and saturated NaCl solution, then dried (MgSC>4) and filtered. Removal of solvent and column chromatography (19:1 petroleum ether-ethyl acetate as eluant) gave 59 mg (74%) of a 69:6:16:9 mixture (GC) of isomeric enol ethers as a clear oil. Rt (DB-210, 160 °C, 20 psi) 5.05, 5.45, 5.58, 6.10 min. IR (CHCI3) 2939, 2858, 1720, 1712, 1670, 1602, 1446, 1360, 1255, 835 cm-1. *H NMR (CDCI3, 400 MHz) 8 0.10-0.15 (4s, 6H total), 0.90-0.97 (3s, 9H total), 1.18-2.20 (m, 15H), 2.03 (m, 4H), 2.30 (m, 2H), 4.28-4.70 (4m, IH total), 4.83-5.05 (2m, IH total). LRMS (m/z) 354 (M+ 0.3), 239 (2), 199 (78), 157 (32), 129 (29), 75 (100), 55 (22), 41 (18). HRMS calcd for C20H38SiO3: 354.2590; found: 354.2584. References 125 (1) Hassel, O . Tidsskrift for Kjemi, Bergvesen og Mettalurgi 1942,3, 32. (2) Barton, D.H.R. Experientia 1950,6,316. (3) (a) Eliel, E.L.; Allinger, N.; Angyal, S.J.; Morrison, G.A. Conformational Analysis; Wiley: New York, 1965. (b) Eliel, E.L. Acc. Chem. Res. 1 9 7 0 , 5 , 1. (c) Zevirov, N.S. Tetrahedron 1977,33, 3193. (d) Ridell, F.G. The Conformational Analysis of Heterocyclic Compounds; Academic Press: London, 1980. (4) (a) Ruzicka, L. Helv. Chim. Acta 1926, 9, 715. (b) Ruzicka, L. Helv. Chim. Acta 1930,13, 1152. (5) Brockmann, H.; Henkel, W. Naturwissenschaften 1950,37,138. (6) Masamune, S.; Bates, G.S.; Corcoran, J.W. Angew. Chem. Int. Ed. Engl. 1977, 16, 585. (7) (a) Ruzicka, L.; Stoll, M.; Huyser, H.W.; Boekenoogen, H.A. Helv. Chim. Acta 1930,13, 1152. (b) Ruzicka, L; Plattner, P.A.; Wild, H. Helv. Chim. Acta 1946,29, 1611. (c) Kobelt, M.; Barman, P.; Prelog, V.; Ruzicka, L. Helv. Chim. Acta 1949, 32, 256. (8) Prelog, V. / . Chem. Soc. 1950, 420. (9) Huber-Buser, E.; Dunitz, J.D.; Venkatesan, K. Proc. Chem. Soc. 1961, 463. (10) (a) Dale, J. / . Chem. Soc. 1963, 93. (b) Dale, J. Angew. Chem. Int. Ed. Engl. 1966, J , 1000. (c) Borgen, G ; Dale, J.; Teien, G. Acta Chemica Scand. 1979, B33, 15. (11) Groth, P. Acta Chemica Scand. 1976, A30, 155. (12) Dale, J. Acta Chemica Scand. 1973,27,1130. (13) Anet, F.A.L.; Cheng, K. / . Am. Chem. Soc. 1975, 97, 2420. (14) Celmer, W.D. Antimicrob. Agents Chemother. 1965, 144. (15) Perun, T.J.; Egan, R.S. Tetrahedron Lett. 1969,10, 387. 126 (16) (a) Egan, R.S.; Perun, TJ.; Martin, J.R.; Mitscher, L.A. Tetrahedron 1973,29, 2525. (b) Egan, R.S.; Martin, J.R.; Perun, T.J.; Mitscher, L.A. J. Am. Chem. Soc. 1975, 97, 4578. (17) Nourse, J.G.; Roberts, J.D. / . Am. Chem. Soc. 1975, 97, 4584. (18) Omura, S.; Neszmelyi, A.; Sangare, M.; Lukacs, G. Tetrahedron Lett. 1975,16, 2939. (19) Shannon, V.L.; Strauss, H.L.; Snyder, R.J.; Elliger, C.A.; Mattice, W.L. / . Am. Chem. Soc. 1989, 111, 1947. (20) Dale, J. Acta Chemica Scand. 1973,27, 1115. (21) (a) Still, W.C. / . Am. Chem. Soc. 1979,101, 2493. (b) Corey, E.J.; Hopkins, P.B.; Kim, S.; Yoo, S.; Nambiar, K.P.; Falck, J.R. / . Am. Chem. Soc. 1979,101, 7131. (c) Kuroda, C ; Hirota, H.; Takahashi, T. Chem. Lett. 1982, 249. (d) Still, W.C; Galynker, I. J. Am. Chem. Soc. 1982,104, \11A. (22) Still, W.C; Gennari, C ; Noguez, J.A.; Pearson, D.A. / . Am. Chem. Soc. 1984, 106, 260. (23) Schreiber, S.L.; Hawley, R.C. Tetrahedron Lett. 1985,26, 5971. (24) Graden, D.W.; Lynn, D.G. / . Am. Chem. Soc. 1984,106, 1119. (25) Paterson, I.; Rawson, DJ. Tetrahedron Lett. 1989, 30, 7463. (26) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull. Chem. Soc. Jpn. 1979,52, 1989. (27) Chamberlin, A.R.; Reich, S.H. J. Am. Chem. Soc. 1985,107, 1440. (28) (a) Still, W.C; MacPherson, L.J.; Harada, T.; Callahan, J.F.; Rheingold, A.L. Tetrahedron 1984, 40, 2275. (b) Still, W.C. Curr. Trends Org. Synth., Proc. Int. Conf., 4 t h 1983, 233. (29) (a) Neeland, E.G. Ph.D. Thesis, University of British Columbia, November 1987. (b) Keller, T.H. Ph.D. Thesis, University of British Columbia, June 1988. (30) Dale, J. Top. in Stereochem. 1976, 9, 199. (31) Saunders, M. Tetrahedron 1967,23, 2105. (32) Burkert, U.; Allinger, N.L. Molecular Mechanics, ACS Monograph 177: Washington, 1982. (33) (a) Ogura, H.; Furuhata, K.; Kuwano, H.; Harada, N. / . Am. Chem. Soc. 1975, 97, 1930. (b) Ogura, H.; Furuhata, K.; Kuwano, H.; Suzuki, M. Tetrahedron 1981, 37, 165. (c) Furusaki, R.; Matsumoto, T.; Furuhata, K.; Ogura, H. Bull. Chem. Soc. Jpn. 1982,55, 59. (d) Ogura, H.; Furuhata, K.; Harada, Y.; Iitaka, Y. Chem. Pharm. Bull. 1977,25, 2385. (34) Ounsworth, J.P.; Weiler, L. / . Chem. Ed. 1987, 64, 568. (35) Ogura, H.; Furuhata, K.; Harada, Y.; Yoichi, I. / . Am. Chem. Soc. 1978,100, 6733. (36) Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry 1983, Pergamon Press, Willowdale. (37) Schweizer, W.B.; Dunitz, J.D. Helv. Chim. Acta 1982, 65, 1547. (38) Jones, T.K.; Reamer, R.A.; Desmond, R.; Mills, S.G. J. Am. Chem. Soc. 1990, 112, 2998. (39) Still, W.C; Galynker, I. Tetrahedron 1981,57,3981. (40) (a) Corey, E.J.; Nicolaou, K . C J. Am. Chem. Soc. 1974, 96, 5614. (b) Corey, E.J.; Nicolaou, K . C ; Melvin, L.S. Jr. / . Am. Chem. Soc. 1975, 97, 653, 654. (c) Corey, EJ . ; Nicolaou, K . C ; Torn, T. / . Am. Chem. Soc. 1975, 97, 2287. (d) Corey, E.J.; Ulrich, P.; Fitzpatrick, J.M. / . Am. Chem. Soc. 1976, 98, 222. (41) (a) Masamune, S.; Yamamoto, H.; Kamata, S.; Fukuzawa, A. / . Am. Chem. Soc. 1975, 97, 3513. (b)Masamune, S.; Kamata, S.; Schilling, W. J. Am. Chem. Soc. 1975, 97, 3515. (42) Schmidt, U.; Dietsche, M. Angew. Chem. Int. Ed. Engl. 1981, 20, 111. (43) Kang, K.; Kim, W.; Moon, B. Synthesis 1985,1161. (44) Michelot, D. Synthesis 1983, 130. (45) Marvel, C.S.; Hager, F.D. Org. Synth. Coll. Vol. 1 1941, 248. (46) Brillon, D.; Deslongchamps, P. Can. J. Chem. 1987, 65, 43. (47) Rama Rao, A.V.; Deshmukh, M.N.; Sharma, G.V.M. Tetrahedron 1987,43, 779. 128 (48) Safaryn, J.E.; Chiarello, J.; Chen, K.M.; Joullie, M.M. Tetrahedron 1986,42, 2635. (49) Wood, G.F. Jr. Org. Synth. Coll. Vol. 3 1955, 1470. (50) Broden, K.; Heilbron, I.M.; Jones, E.R.H.; Weedon, B.C.L. / . Chem. Soc. 1946, 39. (51) (a) Corey, E.J.; Scmidt, G. Tet. Letters 1979, 20, 399. (b) Corey, E.J.; Suggs, J.W. Tet. Letters 1975,16, 2647. (52) Swern, D.; Omura, K. Tetrahedron 1978,54, 1651. (53) Dess, D.M.; Martin, J.C. J. Org. Chem. 1983,48, 4155. (54) Pfitzner, K.E.; Moffat, J.G. / . Am. Chem. Soc. 1965, 87, 5661. (55) Meyers, A.L; Reider, P.J.; Campbell, A.L. / . Am. Chem. Soc. 1980,102, 6597. (56) Corey, E.J.; Erickson, B.W. / . Org. Chem. 1971, 36, 3553. (57) Sterzycki, R. Synthesis 1979, 724. (58) Schreiber, S.L.; Liew, W. J. Am. Chem. Soc. 1985,107, 2980. (59) Keller, T.H.; Neeland, E.G.; Rettig, S.; Trotter, J.; Weiler, L. / . Am. Chem. Soc. 1988,110, 7858. (60) Maruoka, K.; Itoh, T.; Sakurai, M.; Nonoshita, K; Yamamoto, H. / . Am. Chem. Soc. 1988,110, 3588. (61) Wu, Y.D.; Houk, K.N. / . Am. Chem. Soc. 1987,109, 908. (62) Evans, D.A. Aldrichimica Acta 1982,15, 23 and references therein. (63) Masamune, S.; Choy, W. Aldrichimica Acta 1982,15, 47 and references therein. (64) (a) Meyers, A.L; Williams, D.R.; White, S.; Erickson, G.W. / . Am. Chem. Soc. 1981,103, 3088. (b) Meyers, A.L; Smith, R.K.; Whitten, C E . / . Org. Chem. 1979,44, 2250. (65) -Ferreira, J.T.B.; Neeland, E.G.; Ounsworth, J.P.; Weiler, L. Can. J. Chem. 1987, 65, 2314. (66) (a) Corey, E.J. / . Am. Chem. Soc. 1954, 76, 175. (b) Ireland, R.E.; Mueller, R.H.; Willard, A.K. / . Am. Chem. Soc. 1976, 98, 2868. (67) (a) Stork, G.; Hudrlick, J. / . Am. Chem. Soc. 1968, 90, 4462. (b) Stork, G ; Hudrlick, J. / . Am. Chem. Soc. 1968, 90, 4464. (c) Emde, H.; Gotz, A.; Hofmann, K.; Simchen, G. Liebigs Ann. Chem. 1981, 1643. (d) Emde, H.; Domsch, D.;Feger, H.; Frick, U.; Gotz, A.; Hergott, H.; Hofmann, K.; Kober, W.; Kragelohn, K.; Oesterle, T.; Steppan, W.; West, W.; Simchen, G. Synthesis 1982, 1. (68) Perrin, D.D. and Armarego, W.L.F. Purification of Laboratory Chemicals, third ed. Permagon Press, Toronto, 1988. SPECTRAL APPENDIX 131 32 133 134 137 8-i.OO—t t t i i i i i i i • i r r ' i i - " i i r i i i 1 3360. 2720. 2080. 1440. 800. CM-i 138 1.0300-1— i i 1 1" i — i t r 1 - t - - I r 1 1 r r T 1 1 1 r 1 .9780-.9260— .8740-.8220-7700_j_ 4000. 4C.9000 3400. i r 2800. , r 2200. T — r • 1 1600. 1000. CM-1 139 1 4 0 142 143 144 51 400 MHz 50 53 153 0.0676—1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r i ~T i i i I 1 1 1 1—^-i 1 1 1 1 1 1 1 1 1 1 1 1 1 4000. 3400. 2800. 2200. 1600. 1000. CM-1 157 4000. 3400. 2800. 2200. 1600. 1000. CM~1 , 1 1 j 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 p I I 4000. 3400. 2800. 2200. 1600. 1000. CM~1 162 163 164 T < 1 1 I ' • i . | • i I I | i i i | i i t | t - l 4000. 3400. 2800. ' 2200. 1600. 1000. CM-1 166 —i—•—i—•—i—•—i—'—i— '— i— •—i—•—i—•—i—*—l—•—r—•—i—• ~i—•—i—-—i—•—i—>—i—>—i—•—i—•—i—>—i—>—i—•—i—*—i—t—i—•—i—•—r~ 5.2 S.C 1.6 4.6 «.« 4 . J tl S .t 1.6 J . « 5.1 it J .B 2.6 2.4 2.2 2.1 I B 1.6 1.4 : ; l . c .S .6 .4 .1 t c 1.0394—1 I I I I I I I I I | I I I t - l I ' f t I — t 1 1 4000. 3400. 2800. 2200. 1600. 1000. CM-1 167 

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.831.1-0059773/manifest

Comment

Related Items