Synthesis of Thiacrown and Azacrown Ethers Based on the Spiroacetal Framework

Home , Crown ether, Nonactin, Retrosynthetic analysis, Total synthesis

Marica Nikac

A Thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

University of Western Sydney February 2005.

To my family and friends

Acknowledgements

I would like to express my enormous gratitude to my supervisors Dr. Robyn Crumbie, Prof. Margaret Brimble and Dr. Trevor Bailey for all their advice, support and encouragement throughout this project. I would also like to thank Assoc. Prof. Paul Woodgate and Dr. Vittorio “Cappy” Caprio for their suggestions. I would also like to thank the members of the Brimble group particularly Rosliana Halim, Michelle Lai, Christina Funnell, Jae Hyun Park and Kit Sophie Tsang for their encouragement and especially their friendship. I would not have made it without them. I would also like to thank Daniel Furkert and Adrian Blaser for their assistance and friendship. I would like to express my gratitude to Dr. Narsimha Reddy and Mike Walker for their help with everything to do with NMR. I also wish to thank the technical staff at the Universities of Auckland and Western Sydney particularly, Noel Renner, Chris Myclecharane and Gloria Tree. I would like to thank my family (especially my sister Milka) and friends for their support, encouragement and love throughout all these years. Finally I wish to thank all the students and staff at the Universities of Auckland and Western Sydney.

Statement of Authentication

The work presented in this thesis is, to the best of my knowledge and belief, original except as acknowledged in the text. I hereby declare that I have not submitted this material, either in whole or in part, for a degree at this or any other institution.

………………………………………….

Table of Contents

Page Table of Contents i List of Tables v List of Figures vi Abbreviations vii Abstract ix

Chapter 1: Crown Ethers

1.0 Discovery of Crown Ethers 1 1.1 Crown Ethers Containing Carbohydrate Scaffolds 4 1.1.1 Crown Ethers Containing Pyranose Units 6 1.1.2 Lactose Trehalose and Other Di- and Trisaccharides 10 1.1.3 Furanoside Derivatives 12 1.2 Crown Ethers Containing the Spiroacetal Framework 13 1.2.1 Starands 13 1.2.2 Incorporation of the 1,4,7,10-Tetraoxaspiro[5.5]undecane Ring System 14 1.2.3 Polyspiroacetal Ligands 15 1.2.4 Incorporation of 1,7-Dioxaspiro[5.5]undecane Ring System 15

Chapter 2: Synthetic Strategies for the Preparation of Crown Ethers

2.0 General Synthetic Principles for the Preparation of Crown Ethers 20 2.0.1 The Effect of the Chain Length 20 2.0.2 Nature of the Atoms 21 2.0.3 Type Of Cyclisation 21 2.0.4 Ring Closure Methods 22 2.1 Selectivity of Crown Ethers for Metal Ions 23 2.2 Nitrogen Macrocycles (Azacrown Ethers) 26

2.2.1 Templated Syntheses 27 2.2.2 The Sulfonamide Method 29 2.2.3 Azacrown Ethers via Amide Formation 31 2.2.4 Peptide Chemistry 34 2.2.5 Crab-like Cyclisation 34 2.3 Sulfur Macrocycles (Thiacrown Ethers)

2.3.1 Synthesis of Thiacrown Compounds Using Cs2CO3 38 2.3.2 The Cesium Effect 38 2.3.3 Templated Syntheses 39

Chapter 3: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.0 Synthetic Targets 42 3.1 Synthesis of Spiroacetal Diol (50) 43

3.1.1 Synthesis of (±)-1,7-Dioxaspiro[5.5]undec-4-ene 43 3.1.2 Epoxidation of Olefin (114) 45 3.1.3 Base-Induced Ring opening of Epoxide (115) and Epoxidation of Allylic Alcohol (117) 46 3.1.4 Reduction of syn-Epoxy Alcohol (119) 48 3.2 Synthesis of the Spiroacetal Thiacrown Ethers via Path A 49

3.2.1 Synthesis of the β-Chloroethyl Sulfides (104), (127) and (128) 49 3.2.2 Reaction Between Spiroacetal Diol (50) and β-Chloroethyl Sulfide (104) 51 3.2.3 Attempted Synthesis of Thiacrown Ether (134) via Olefin 52 Metathesis of Diene (132) with Spiroacetal Diene (133) 52 3.2.3.1 Synthesis of Spiroacetal Bisallyl Ether (133) 53 3.2.3.2 Olefin Metathesis 54 3.4 Synthesis of the Spiroacetal thiacrown Ethers via Path B 61 3.4.1 Synthesis of Spiroacetal Diol (145) 62 3.4.2 Ditosylation of Spiroacetal Diol (145) 64 3.4.3 Synthesis of Spiroacetal Thiacrown Ethers (55), (56) and (57) 65 3.5 Synthesis of Spiroacetal Azacrown Ethers 68 3.5.1 Attempted Synthesis of Spiroacetal Azacrown Ethers (58), (59) And (60) via Imine Formation from Aldehyde (150) 69

3.5.2 The Sulfonamide Method 70 3.6 Summary 74

Chapter 4: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

4.0 Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands 75 4.0.1 Crown Ethers As Primary Ligands 75 4.0.2 Second-Sphere Coordination 76 4.1 Binding Studies 78 4.1.1 Determination of Association Constants Using Picrate Salts 78 4.1.2 Interaction of Thiacrown Ethers (55), (56) and (57) with Neutral and Ionic Complexes (Second-Sphere Coordination) 82

4.1.2.1 Interaction With [Co(NH3)NO2](BPh4)2 (165) and

[Co(en)3](BPh4)3 (166) 83

4.1.2.2 Interaction With [Al(acac)3] (164) 84 4.2 Summary 86

Chapter 5: Kinetic Resolution of the Spiroacetal Moiety

5.0 Kinetic Resolutions 87 5.1 Base-Induced Rearrangement Of Epoxides 88

5.1.1 Ring Opening of α-Epoxide (115) Using Chiral Non-Racemic Lithium Amide Bases 89 5.2 Hydrolytic Kinetic Resolution 91 5.2.1 Jacobsen Hydrolytic Resolution Reaction 91

5.2.2 Hydrolytic Kinetic Resolution of α-Epoxide (115) 92 5.3 Sharpless Epoxidation 94 5.3.1 Sharpless Epoxidation of Allylic Alcohol (117) 96 5.4 Summary 98

iii

Chapter 6: Conclusion

Conclusion 100

Chapter 7: Experimental

7.0 General Details 105 7.1 Preparation of [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diol (50) 106 7.2 Synthesis of [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bis(ethyl p-toluenesulfonate) (146) 115

7.3 Preparation of β-Chloroethyl Sulfides (104) and (127) and Dithiols (122) and (123) 120 7.4 Synthesis of Spiroacetal Thiacrown Ethers (55), (56) and (57) 125 7.5 Synthesis of Spiroacetal Azacrown Ether (58) 128 7.6 Olefin Metathesis of Spiroacetal Bisallyl Ether (142) 132 7.7 Kinetic Resolution Reactions 134 7.8 Second-Sphere Complexation 142

Chapter 8: References

References 143

List of Tables

Page Table 1: Selected 1H Chemical Shifts and Coupling For Dimers (142a), (142b) and (142c) 60 Table 2: Reaction Conditions Used in the Synthesis of Diester (149) 62 Table 3: Reaction Conditions Used in the Synthesis of Protected Azacrown (159) 72 Table 4: Association Constants For Spiroacetal Thiacrown Ethers (55), (56) and (57) 79

Table 5: Base-Induced Rearrangement of α-Epoxide (115) Using Chiral Lithium Amide Bases 90

List of Figures

Page Figure 1: Association Constants of the Axial Crown Ethers 16 Figure 2: Association Constants of the Equatorial Crown Ethers 17 Figure 3: Resolution of Spiroacetal Diol (50) 19 Figure 4: Common Approaches to Synthesising Macrocycles 20 Figure 5: Sandwich Effect 24 Figure 6: Common Structure of Complexed Azacrown Ethers 25 Figure 7: Effect of the Longer Carbon-Sulfur Bond on the Cavity of Thiacrown Compounds 26 Figure 8: Cesium Effect 39 Figure 9: Hydrogen Bonding in the Epoxidation of Allylic Alcohol (117) using meta -chloroperoxybenzoic acid 48 Figure 10: Proton Driven Transport of Metal Ions Through Membranes 76 Figure 11: Second-Sphere Coordination 77 Figure 12: Second-Sphere Coordination of Carboplatin 77 Figure 13: Association Constants of the Thiacrown Ethers (55), (56) and (57) 80 Figure 14: Expanded Region 80 Figure 15: Holodirected and Hemidirected Geometry of Pb(II) 82

Figure 16: Electrostatic Potential Map of [Al(acac)3] 85 Figure 17: Electrostatic Potential Map of Thiacrown (55) 85

Figure 18: Differentiation Between Two syn β-Protons in Cyclohexene Oxide 88 Figure 19: Mechanism of the Sharpless Epoxidation 95 Figure 20: Kinetic Resolution of Stereochemically Rigid Compounds Using Sharpless Epoxidation 96 Figure 21: Determination of the Appropriate Tartrate in the Epoxidation of Allylic Alocohol (117) 97

Abbreviations

2D. two dimensional 18-Crown-6 1,4,7,10,13,16-hexaoxacyclooctadecane 18-S-6 1,4.7,10,13,16-hexathiacyclooctadecane 9-S-3 1,4,7-trithiacyclononane Ar. Aryl aq. aqueous ax. axial n-Bu, Bun. n-butyl t-Bu, But. tert-butyl CI. chemical ionisation (mass spectroscopy) cm3. cubic centimetres conc. concentration COSY. correlation spectroscopy D. deuterium deg. degrees DEPT. distortionless enhancement by polarisation transfer dil. dilute DMDO. dimethyldioxirane DMAP. 4-dimethylaminopyridine DMF. N,N-dimethylformamide DMSO. dimethylsulfoxide ee. enantiomeric excess EI. electron impact ionisation (mass spectroscopy) eq. equatorial equiv. equivalent Et. ethyl h. hour HMQC. Heteronuclear Multiple Quantum Correlation HMBC. Heteronuclear Multiple Bond Correlation HSQC. Heteronuclear Single Quantum Correlation

vii

IR. infrared spectroscopy J. NMR coupling constant

Ka. association constant L. ligand m-CPBA. meta-chloroperoxybenzoicacid Me. methyl min. minute mmol. millimoles mol. moles mp. melting point NMR. nuclear magnetic resonance Ns. nosyl, o-nitrobenzenesulfonyl Oxone® potassium peroxymonosulfate IPr. isopropyl PPTS. pyridinium p-toluenesulfonate THF. tetrahydrofuran TMS. trimethylsilyl Ts. tosyl, p-toluenesulfonyl

viii

Abstract

This thesis describes the synthesis of novel thiacrown and azacrown ethers based on the 1,7-dioxaspiro[5.5]undecane ring system. Chapter one presents an overview of the discovery and development of crown ethers. The incorporation of naturally occurring compounds into the crown ether framework such as carbohydrates and more recently the spiroacetal functionality is also described. Chapter two describes the synthetic strategies available for the synthesis of crown ethers, azacrown ethers and thiacrown ethers. The ability of crown ethers to bind metal ions is also discussed especially in terms of cavity size (and shape) and donor heteroatom. Chapter three describes the synthesis of the thiacrown ethers (55), (56) and (57) and azacrown ether (58) incorporating the spiroacetal 3,5-diol (50). The synthesis of the spiroacetal moiety was carried out starting from δ-valerolactone (110) and trimethylsilyl protected alcohol (111) to form spiroacetal alkene (114). Subsequent treatment of alkene

(114) with DMDO yielded the β-epoxide (116) and desired α-epoxide (115). The base- induced ring opening of α-epoxide (115) afforded allylic alcohol (117) and homoallylic alcohol (118). Epoxidation of allylic alcohol (117) using m-CPBA, followed by reduction of the desired syn-epoxy alcohol (119) afforded the 3,5-diol (50) and the 4,5-diol (121). Three methods to synthesise spiroacetal thiacrown ethers (55), (56) and (57) are then discussed, (a) the reaction between spiroacetal diol (50) and β-chloroethyl sulfide (104), (b) the cross olefin metathesis reaction between diene (132) and spiroacetal bisallyl ether (133) and (c) the reaction of the spiroacetal ditosylate (146) and the appropriate dithiol (99), (122) or (123). Chapter three also describes two synthetic routes for the analogous spiroacetal azacrown ethers (58), (59) and (60). The first of these involved the reaction of spiroacetal dialdehyde (150) and triamine (151) and the second involved the reaction of the spiroacetal ditosylate (146) with Ts or Ns-protected triamines (157) and (158). Chapter four describes the complexing ability of thiacrown ether compounds (55), (56), (57) and (100) as primary receptors and as ligands in second-sphere coordination. The binding affinity of thiacrown ethers (55), (56), (57) and (100) for alkali metal ions, transition metal ions and heavy metals was determined using the ultraviolet spectroscopic method. The spiroacetal thiacrown ethers (55), (56) and (57) showed a large preference for the heavy metal ions, particularly silver. The large difference in the complexing

ix behaviour of the thiacrown ethers (55), (56) and (57) indicates their selective extraction ability. The interaction between crown compounds (55), (56), (57), (64) and (100) with neutral [Al(acac)3] (164) and ionic [Co(NH3)5NO2](BPh4)2 (165) and [Co(en)3](BPh4)3 (166) was investigated via 1H NMR spectroscopy. It was found that spiroacetal thiacrown ethers (56) and (57) showed interaction with the aluminium complex (164). Chapter five describes the attempted kinetic resolution of the spiroacetal moiety, to provide enantiopure starting material for the synthesis of non-racemic spiroacetal crown ethers. Three different approaches were investigated (a) base-induced rearrangement of the spiroacetal α-epoxide (115) using chiral non-racemic bases, (b) the hydrolytic kinetic resolution of the α-epoxide (115) using a cobalt-acetate complex (177) and (c) Sharpless epoxidation of the allylic alcohol (117). The enantiomeric excesses were determined by 1H NMR spectroscopic analysis of the Mosher’s ester of the appropriate alcohol. Chapter six summarises the results achieved and discusses possible strategies emanating from these results.

INTRODUCTION : Crown Ethers

1.0 The Discovery of Crown Ethers

In 1987, Charles Pedersen, Donald Cram and Jean-Marie Lehn were awarded the Nobel Prize in chemistry for their work with crown ethers and cryptands (bimacrocyclic nitrogen derivatives of crown ethers). Their pioneering work led to the development of a new area of chemistry known as macrocyclic or supramolecular chemistry.1,2 Macrocyclic chemistry deals with the interactions between the atoms of the macrocycle and various metal ions.2 Since the discovery of crown ethers by Charles J. Pedersen in 1967,3 numerous macrocycles have been synthesised and their ability to complex ions has been investigated. A few examples of macrocycles were already known in literature before Pedersen’s synthesis, however he is credited with their discovery because he was the first to recognise the complexation ability of these compounds and describe the specific characteristics of the complexation. Some of the early work carried out in macrocyclic chemistry was by Lüttringhaus4 who was interested in preparing large ring molecules. His idea was to prepare structures that would possess unusual properties. Using resorcinol (1,3-dihydroxybenzene) as a nucleophile, he carried out reactions with a variety of substituted diol derivatives, resulting in the isolation of several macrocyclic polyethers of general structure (1). Unfortunately for Lüttringhaus, the compounds did not possess sufficient donor groups to exhibit cation binding abilities.

O O

The condensation of furan (2) and acetone (3) in the presence of a protic or Lewis acid catalyst lead to the furan-acetone tetramer (4) (Scheme 1). Brown et al5 named these initial compounds anhydrotetramers because the analytical data corresponded to the furan and ketone starting materials with the loss of four water molecules.

INTRODUCTION : Crown Ethers

O O O + O Lewis Acid O O 23

4 Scheme 1

In 1957, Stewart, Wadden and Borrows6 patented a process for the cyclo- oligomerisation of ethylene oxide. They treated an oxirane with alkyl aluminium, zinc and magnesium to produce dioxane and a variety of other cyclic compounds, one of which was the cyclic tetramer of ethylene oxide (5). In the same year Wilkinson et al7 reported the synthesis of a cyclic tetramer from propylene oxide. The workers recognised the interesting properties of the cyclic compounds but did not appreciate their potential.

O O

Then in 1967 Pedersen synthesised dibenzo-18-crown-6 as a by-product in the preparation of biphenol (8).3 Mono-protected catechol (6) was reacted with 2,2’- dichlorodiethyl ether (7) using sodium hydroxide in n-butanol to give the phenolic derivative (8) after deprotection. A small amount of unprotected catechol was also present in the initial reaction mixture and it was this unprotected compound that gave rise to the crown ether (9) (Scheme 2). Pedersen was the first to recognise the complexing ability of this type of compound. He found that dibenzo-18-crown-6 (9) had an increased solubility in methanol in the presence of sodium hydroxide. This was attributed to the complexation between the crown ether and the sodium ion. Pedersen also showed that several other species could be coordinated to the electron rich compounds and that sulfur and nitrogen could be substituted in place of oxygen. Cram et al8 elaborated further on these discoveries and introduced the term host-guest complexation to describe the relationship between the crown ether (host) and the metal ion (guest). 2

INTRODUCTION : Crown Ethers

OR O O O OH

R = THP, H OH HO

6 8 i + O

O O O Cl Cl O O 7 O

Reagents and Conditions: (i) NaOH, n-BuOH, reflux

Scheme 2

In the ensuing years, since the synthesis of dibenzo-18-crown-6, a large number of macrocycles have been synthesised and their impact on the understanding of host- guest interactions and separation science has been significant.9 This has also allowed for a greater understanding of the properties and behaviour of biologically important crown ethers and ionophores. Nonactin (10), a macrotetralide antibiotic, is a naturally occurring ionophore whose cation binding and transport abilities are now better understood in terms of the principles discovered through crown ether research.

O O O O O O O O O O

O 10

The introduction of biological systems into the crown ether structure has provided scientists with an even greater insight into biological ionophores, such as cyclic peptides10 and macrolide antibiotics.11 Alternatively, macrocycles have been synthesised on the basis of their resemblance to biological systems. This has provided molecules 3

INTRODUCTION : Crown Ethers potentially capable of mimicking various aspects of macromolecular biological systems.12 Macrocycles that incorporate rings identical or closely related to various ring systems found in nature are known as classical macrocycles.13 The carbohydrate structure is commonly incorporated into the crown ether framework because they introduce chirality to a macrocycle and have multiple interaction sites making them attractive as chiral receptors.14 The spiroacetal ring system has also been investigated more recently. The description of the structure of monensin (11) in 196715 and the discovery of its cation binding properties instigated extensive interest in spiroacetal ionophores. Monensin is an acyclic ionophore, which has been used in the synthesis of its cyclised lactone derivative.16

O H O O H O H O H H O O

HO HO

HO 11

1.1 Crown Ethers Containing Carbohydrate Scaffolds

Interest in crown ethers containing the carbohydrate moiety was initiated with the isolation of cyclic oligosaccharides, which are more commonly known as cyclodextrins. Cyclodextrins are produced by the action of Bacillus macerans amylase on starch.17 The most studied representitives of this class of compounds are the α-, β- and γ-cyclodextrins and are composed of six (12), seven and eight α-1,4-linked D-glucopyranose units, 4 resectively. The α-D-glucopyranose units exist as C1 chair conformations and are joined to each other by glycosidic linkages involving axial C-1-O and equatorial C-4-O bonds. Ogawa and Takahashi18 reported the first total synthesis of two cyclic oligosaccharides. Starting from maltose they synthesised both (12) and the 8-membered compound in 21 steps. These syntheses represented pioneering investigations and have led to the synthesis

INTRODUCTION : Crown Ethers of many cyclodextrins including the bicyclic 3,6-anhydro analogues of type (13) that 1 19 exist as the C4 chair conformations.

OH HO O 4 1 O OH O O O O O 1 4 HO O O O HO HO OH HO O O HO OH HO O O OH O O HO O O OH HO OH O O HO OH OH O OH O OH O O O O HO O O O OH O HO 12 13

The carbohydrate structure can provide relatively inexpensive chiral starting materials for the synthesis of optically active macrocycles.20 This is important in terms of chiral recognition, as it is a fundamental property of biological molecules. Chiral recognition deals with the ability of a macrocycle to complex only one of the two enantiomers in a racemic mixture. Chiral barriers are required in the molecule for chiral discrimination to occur. Chiral crown ether compounds have been used as catalysts in enantioselective syntheses,21 chromatographic separations,22,23 1H NMR spectroscopy24 and differentiating between protonated enantiomers of amines, amino acids, amino alcohols and other derivatives.25 Cram and co-workers26 pioneered this area of research by incorporating 1,1’- binaphthyl chiral units into crown ether structure (14). Stoddart et al27 expanded on this work by synthesising a number of compounds derived from D-mannitol (15). Both sets of compounds showed a good degree of enantiomeric differentiation of amine salts based on 1H NMR spectroscopic data. Stoddart and co-workers28 were also interested in macrocycles having two different chiral centres. They synthesised 18-crown-6 possessing both the binaphthyl and D-mannitol derived units. These compounds also showed their capability of acting as chiral hosts.

INTRODUCTION : Crown Ethers

O O O

O O Me O Me H H O O Me Me O O

O O Me O O Me H H O Me Me O O

1.1.1 Crown Ethers Containing Glucose, Mannose, Galactose and Other Pyranose Units

The majority of compounds synthesised, which possess the carbohydrate structure, are composed of D-hexopyranoside units. The most commonly used D- hexopyranosides are D-glucoside, D-mannoside and D-galactoside. Stoddart et al20a,29 synthesised a series of 18-crown-6 α- and β- analogues containing the glycosides. They also investigated α-D-altroside. They found that α-D-mannoside (16) and D-galactoside (17) compounds showed a greater binding ability to the ammonium thiocyanate salt than the α-D-glucoside crown ether (18). They also found that the α-D-altroside compound (19) formed very weak complexes. This was due to the introduction of one anti C–C bond into the 18-crown-6 structure which removed the opportunity for all six oxygens to act co-operatively in the binding. This work has been reviewed by Stoddart.20a

OMe O OMe O H H O O O O O O H H O O O O H H OOH O OOH O

Ph Ph 16 17

INTRODUCTION : Crown Ethers

OMe O OMe O H H O O O O O O H H O O O O H H OOH O OOH O

Ph Ph 18 19

Joly and co-workers23b built upon the work by Stoddart20a and synthesised a number of carbohydrate crown ethers based on the 18-crown-6 structure (20-22). They evaluated the extraction ability of these compounds toward racemic phenylglycine and found that no enantiomeric differentiation was exhibited. The naphthalene analogue of (22a), synthesised by Pietraszkiewicz et al,30 showed a marked increase in enantioselectivity. This observation suggested that the increased aromatic content 31 favoured chiral recognition. The α-D-mannoside derivative (23) also showed good enantioselectivity. It was suggested that the α-D-mannoside moiety contributed to the enantioselectivity by providing extra binding strength through the hydrogen bonding of the hydroxyl groups. Part of this work has been reviewed by Bradshaw et al.20b

OMe O OMe O O O O O O O O O O O O O O OO O OO O OMe

Ph Ph 20 21

OMe O OMe O O O O O O O

O O O O OR OR' O OH OH O

22 23 (a) R = R' = OH (b) R = OH, R' = OMe

INTRODUCTION : Crown Ethers

Tõke and co-workers32 synthesised a number of glucose derived 18-crown-6 compounds having the structure of (24) to act as catalysts in asymmetric Michael addition. The reaction of methyl phenylacetate (25) and methyl acrylate (26) in the presence of (24) as a complex with potassium tert-butoxide lead to the formation of the new C–C bond in compound (27) (Scheme 3). The enantiomeric excesses were determined by 1H NMR spectroscopy and found to be relatively high for some of the reactions (76 – 85%).

OMe O OR' OR O O O O O O O O O OR OR' O OMe O O 24 + O O *

25 26 O

* R or S 27 Scheme 3

24c Wenzel et al were interested in synthesis of crown ether (28) containing a β-D- galactoside moiety. The crown ether was investigated as a chiral shift reagent in 1H NMR spectroscopy. With the addition of an achiral lanthanide species crown ether (28) proved to be effective in determining the ee of an L- enriched DL-alanine salt.

O O O O OMe O

O O

INTRODUCTION : Crown Ethers

Much of the research into carbohydrate-based crown ethers discussed thus far has dealt with the incorporation of the carbohydrate moiety through the 2,3-hydroxyl positions onto an 18-crown-6 framework. In 1994, Mani and co-workers33 synthesised novel chiral macrocycles, which incorporated the glycoside unit bound to the 1,4- hydroxyl groups. Miethchen et al34 synthesised similar compounds via a slightly different synthetic pathway. The macrocycles (29) and (30) were based on the D-glucose unit. Mani et al33a reported the binding ability of the crown hosts for lithium, sodium, potassium, cesium and the ammonium ion. They found that the results were comparable to those published in the literature for monosaccharide crown ethers.

OR OBn

O O O O RO BnO OR OBn O O O O

O O O O O

29 BnO O OBn O O

OBn 30

Incorporation of macromolecular polyether derivatives into lipids has been shown to result in cation transport rates comparable to those ion channels formed by natural and synthetic oligopeptides or antifungal macrolides.35 Martin et al35 were interested in synthesising a polyether based ion channel incorporating a D-glucose unit (31) and investigating whether it could mimic the essential functional features of natural transport processes. They postulated that the inner macroring would embed in the membrane and the two carbohydrate units would be near the bilayer surfaces. They found that compounds with a length greater than 26 Å were active in the lipid bilayer.

OBn OBn BnO OBn BnO O O O O O O O O O O O O O O O O BnO O O O O O O

INTRODUCTION : Crown Ethers

Calixarene chemistry has been well documented and has had applications in analytical and separation chemistry,36 however limited literature exists relating to chiral calixcrowns. Recently, Bitter and co-workers37 described the synthesis and binding ability of novel chromogenic monosaccharide-appended calixcrowns. Compound (32) was synthesised as part of their investigations into the development of optical sensors. The authors established that (32) exhibited noticeable optical recognition towards (R)- and (S)-α-methylbenzylamine enantiomers.

HO O HO OMe

OO OH HO

1 1 R R 32

1.1.2 Lactose, Trehalose and Other Di- and Trisaccharides

Penadés and co-workers21c,d,38 introduced both lactose and trehalose units into the crown ether framework. They synthesised a number of macrocycles with different cavity shapes and sizes from the disaccharides. The idea behind the disaccharides was to increase the rigidity of the macrocycles and thereby the stability of the complexes formed in order to achieve chiral recognition. The lactose derivatives including (33) and (34) were evaluated for their binding ability and used as catalysts in asymmetric Michael addition. The results were comparable to those of the monosaccharides. Investigation of the trehalose derivatives, for example (35), was based on molecular energy calculations.

INTRODUCTION : Crown Ethers

OR OR O O O O OR' O O O O O OR O O O O O OR' O

O O O O OR O OR OR 33 34

R'O OR'

R'O OR' O O

O O O OR OR O O

O O

O O OR OR OOO

O O R'O OR'

R'O OR' 35

Joly et al39 synthesised a series of 18-crown-6 crown ethers possessing di- and trisaccharide units to investigate whether these crown compounds would be recognised by lectins (a carbohydrate-binding protein). The authors tested the compounds including (36) and (37) against the galactose-specific lectin Kluyveromyces bulgaricus.

OR R' OR R' RO OR O O O RO O O OMe RO O O O Me

RO OR O O RO OR O O OR O O O O 36 37

O O O O

INTRODUCTION : Crown Ethers

1.1.3 Furanoside Derivatives

Unlike the pyranoside analogues, there are only limited examples of crown ethers containing the furanoside units. Some of the early work carried out in relation to crown ethers containing a five-membered carbohydrate was by Gross and co-workers.40 They synthesised a number of 18-crown-6 macrocycles incorporating various carbohydrate structures, one of which was glucofuranoside. Compound (38) was then investigated for its complexing behaviour towards primary alkylammonium and α-amino acid ester salts. It was found that (38) was not very successful as a complexing or discriminating agent.

O O

OMe O O

O O

Miethchen et al41 also utilised the glucofuranoside, as well as the galactofuranoside moieties in their syntheses of 12-crown-4 derivatives. They were able to synthesise (39) from glucofuranosyl fluoride as part of their investigations into organofluorine compounds and fluorinating agents.

MeO OMe O OOO MeO OMe O O MeO OMe 39

In recent years, Sharma and co-workers42 have investigated D-xylose as a chiral source in the 18-crown-6 structure. Their interest primarily lay in the synthesis of four isomeric macrocycles. Using a substitution reaction they synthesised all four compounds including (40) and (41) in 40 – 60% yields.

INTRODUCTION : Crown Ethers

O OMeO OMe MeO MeO

O O O O

MeO MeO O OMe O OMe 40 41

1.2 Crown Ethers Containing the Spiroacetal Framework

The increasing pharmacological importance of compounds containing the spiroacetal structure/s has increased interest in both their synthesis and reactivity.43 The spiroacetal ring system enjoys widespread occurrence in insects, plants, microbes, fungi and marine organisms. The vast majority of chemistry in this area is focused on the three common spiroacetal structures: 1,7-dioxaspiro[5.5]undecane (42), 1,6- dioxaspiro[5.4]decane (43) and 1,6-dioxaspiro[4.4]nonane (44). The conformation of the spiroacetal ring system is influenced by three factors: steric effects, anomeric effects44 and intramolecular hydrogen bonding or other chelation effects. The anomeric effect deals with the preference of C-O bond of the rings to be in an axial orientation with respect to each other.

766 8 5 4 7 4 4 O O 3 7 O 3 9 6 3 8 5 5 O 2 8 2 10 11 O 2 9 10 9 O 1 1 1 42 43 44

Much of the early research into ionophores containing the spiroacetal moiety has dealt with naturally occurring compounds. In recent years however, the spiroacetal system has been incorporated into crown ethers not biologically derived.

1.2.1 Starands

In 1993, Lee and co-workers45 synthesised a macrocycle containing a polyspiroacetal framework. At the time they were investigating a new branch of

INTRODUCTION : Crown Ethers orthocyclophane chemistry. The initial idea was to synthesise a polyoxa compound (45) via a series of oxidations. However, the isolated product proved to be the polyacetal compound (46) and not the desired polyoxa compound. Since the oxidation was carried out under acidic conditions it was proposed that an acid catalysed cyclisation occurred after the final oxidation (Scheme 3). They synthesised a series of compounds, which they named starands because of their star-shaped structure.

O O O O O O O H O O O O O HO O HO OO OO O O OO O

45 46 Scheme 4

1.2.2 Incorporation of the 1,4,7,10-Tetraoxaspiro[5.5]undecane Ring System

Garcia and co-workers46 synthesised a number of enantiopure spiroacetal crown ethers of the type (47). They were also able to obtain a small amount of the dimeric products (48) as a result of the 2+2 condensation. They investigated the binding ability of the crown ether compounds (47) towards group IA and IIA cations. In the smaller ring systems (n = 1, 2) they found that the distortions of the spiran functionality adversely affected the spatial distribution of the donor oxygen atoms and thus the binding ability of the macrocycle.

n O OOO O n O X O O O X O O O O O O X

n = 1, 2, 3 OOO X = O, NR n

47 n = 1, 2 X = O, NR 48

INTRODUCTION : Crown Ethers

1.2.3 Polyspiroacetal Ligands

McGarvey et al47 synthesised a number of polyspiroacetal ligands, including (49), that offer a 1,3,5-triaxial orientation of the ring system in their search for nonmacrocyclic alternatives to the crown ethers. That axial orientation of the donor groups creates the coordination cavity. They found that the compounds with the hydroxy or methoxy substituents were poor ligands. However, when the substituent was benzylamine the binding was considerably greater.

OO M OOM X X O O

X = OH, OMe, NHBn 49

1.2.4 Incorporation of the 1,7-Dioxaspiro[5.5]undecane Ring System

Brimble et al48 were the first to deliberately incorporate the 1,7- dioxaspiro[5.5]undecane spiroacetal ring system into the crown ether structure. They synthesised a number of spiroacetal crown ethers starting from spiroacetal (50) and evaluated their potential to act as pH dependent ionophores. The initial spiroacetal structure has the hydroxyl groups at C-3 and C-5 in an axial orientation. Upon treatment with acid the structure can undergo a ring opening followed by reclosure to form the diequatorial isomer (51) with the two hydroxyl substituents adopting the more thermodynamically stable equatorial position (Scheme 5).

O O

O H O 5 3 HO OH 53 OH OH 51 50

Scheme 5

The diaxial (53a-c) and diequatorial (54a-c) crown ethers were synthesised by reacting the dianion of spiroacetal diol (50) or (51) with ethylene glycol ditosylate (52) (Schemes 6 and 7). The crown ethers were formed in moderate yields. The association 15

INTRODUCTION : Crown Ethers constants of the crown ethers were evaluated for cesium, lithium, sodium, potassium and the ammonium ion. The results of the binding studies are represented in Figures 1 and 2.

O O O O + n i, ii O OTs OTs OH OH n = 1, 2, 3 OO 50 52 OO O n 53a, n = 1 (16-Crown-5) 53b, n = 2 (19-Crown-6) 53c, n = 3 (22-Crown-7) Reagents and Conditions: (i) KH, THF, reflux, 30 min; (ii) 52 in THF over 3 h, reflux 24 h; 53a (48%), 53b (42%), 53c (32%)

Scheme 6

Associations Constants (Axial Crown Ethers) 25000

20000

3 15000

10000 Ka x 10

5000

0 Li Na K NH4 Cs Picrate Salts 16-Crown-5 (53a) 19-Crown-6 (53b) 22-Crown-7 (53c)

Figure 1: Association Constants of the Spiroacetal Axial Crown Ethers

INTRODUCTION : Crown Ethers

O O O n O + i, ii O HO OH OTs OTs O O n = 1, 2, 3 51 52 O O O n 54a, n = 1 (16-Crown-5) 54b, n = 2 (19-Crown-6) 54c, n = 3 (22-Crown-7) Reagents and Conditions: (i) KH, THF, reflux, 30 min; (ii) 52 in THF over 3 h, reflux 24 h; 54a (34%), 54b (28%), 54c (42%)

Scheme 7

Association Constants (Equatorial Crown Ethers) 100 90 80 70 3 60 50 40 Ka x 10 30 20 10 0 Li Na K NH4 Cs Picrate Salts 16-Crown-5 (54a) 19-Crown-6 (54b) 22-Crown-7 (54c)

Figure 2: Association Constants of the Spiroacetal Equatorial Crown Ethers

The results showed that the diaxial crown ethers had a much greater binding ability than the diequatorial crown ethers. For example, in the case of potassium the binding of the axial crown ethers was on average 300 times greater than that of the diequatorial analogues. This was explained by the fact that the two oxygen atoms at C-3 and C-5 in the equatorial crown ethers were further away for binding to take place (Scheme 8). Therefore, it was theoretically possible for the diaxial crown ethers to complex metals, which would then be released upon exposure to acid when the diaxial crown ethers underwent ring opening and reclosure to the equatorial crown ethers. These 17

INTRODUCTION : Crown Ethers results were also compared to 18-crown-6. The axial spiroacetal crown ethers proved to have a stronger complexing ability than 18-crown-6 itself, in some cases.

O O H M O O + O O

OO O M O OO O n O n = 1, 2, 3 n n = 1, 2, 3 Scheme 8

This unique ability of spiroacetal crown ethers to selectively bind metal ions is desirable in terms of separation science as they offer the potential to act as pH dependent ionophores. It was proposed that the spiroacetal thiacrown and azacrown ethers would behave in a similar fashion but would have affinities for different metal ions, such as the transition and heavy metals. This would provide compounds with interesting possibilities for environmental and medicinal chemistry. The aim of the present work was to synthesise the sulfur (55, 56, 57) and nitrogen (58, 59, 60) analogues and evaluate their binding ability. A review of the synthetic methods available to construct azacrown and thiacrown compounds is given in the sections 2.0, 2.2 and 2.3. A further course for investigation was directed towards the synthesis of enantiopure spiroacetal crown ethers via the kinetic resolution of the spiroacetal functionality (Figure 3) because of the increased interest in enantiopure crown compounds.

O O O

OO O O OO

SSS S SS S S S 55 S S 56 S 57

INTRODUCTION : Crown Ethers

O O O

OO O O OO

NH NH NH HN HN NH H N NH HN 58 HN NH 59 N H 60

O O

OH OH OH OH

Figure 3

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

2.0 General Synthetic Principles for the Preparation of Crown Ethers

The synthesis of macrocyclic systems can be achieved starting from a chain or a linear group of segments in which the final step is the joining of the two ends of the chain to form the desired macrocycle.49 This chain can be the result of an in situ assembly of bifunctional units or it can be directly from the starting material. There are three common approaches for synthesising macrocycles: (a) simple cyclisation, (b) cyclisation in conjunction with another molecule, known as capping and (c) condensation of two or four identical or non-identical units (Figure 4).

(a) (b)

Figure 4

In the synthesis of crown ethers, the desired cyclisation reaction must compete with the polycondensation reaction.49 Cyclisation involves the terminal chain ends reacting with each other in an intramolecular fashion while polycondensation involves the terminal ends reacting with another molecule, i.e. an intermolecular reaction. Cyclisation is influenced by four main factors: (1) chain length, (2) nature of the atoms, (3) type of cyclisation and (4) ring closure methods.

2.0.1 The Effect of Chain Length

The ease of ring formation is dependent on the chain length.50 In the cyclisation process, the interaction of the atoms and the ensuing entropy change leads to different degrees of ring strain and this particularly affects the yield.49 Ring strain can be expressed in terms of heat of combustion, with a maximum for small rings (n = 3, 4) and as a result of angular strain with a minimum for six membered rings. The ring strain increases for medium sized rings and then decreases for the larger rings (n > 14), becoming almost zero.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

The concept effective molarity (EM)51 is another influence relating to the chain length. Effective molarity is the concentration needed in order for the intramolecular reaction to take place at the same rate as the intermolecular reaction.49 At this concentration, equal amounts of both the cyclic and open chain product are formed; therefore a dilute solution will give predominantly the desired macrocycle while a concentrated solution will give predominantly polymers. The greater the effective molarity, as seen with four, five and six membered rings, the more prevalent the cyclisation reaction.50

2.0.2 Nature of the Atoms

The presence of oxygen, nitrogen or sulfur in a chain favours the formation of rings.52 Replacement of a methylene group with the less bulky heteroatoms leads to a decrease in transannular interactions.49 This is especially true in the synthesis of smaller and medium ring systems. In larger rings, however, the effect of the heteroatoms is not as pronounced since even in their absence the transannular interactions are weak or non- existent.53 The effect of incorporating rigid groups (aromatic, double and triple bonds) into chains increases the probability of cyclisation.54 As the chain length and internal entropy increase, the probability of an intramolecular reaction between the two ends decreases. This is due to the greater flexibility of the chains and the increased number of possible conformations. Because rigidity in a molecule decreases the flexibility and therefore the number of conformations, the probability of an intramolecular reaction are increased.

2.0.3 Type of Cyclisation

As mentioned previously, there are three common approaches to synthesising macrocycles. However, the desired cyclisation type (one component, two component, etc) is influenced by various factors.49 For example, in the synthesis of dibenzo-18- crown-6 (9), catechol (6) and dichlorodiethyl ether (7) can react in a one, two or three component fashion to form the monomer (61), the dimer (62) or the trimer (63) respectively.1 By varying the reaction conditions, such as the type of nucleophile, base, leaving groups, solvent and concentration, the synthesis of different sized macrocycles may be targeted.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

O OO O O O O O O O O O O O 61 O 62 OOO

The type of base (weak or strong) used in a reaction will depend on the particular nucleophile being generated. Oxygen nucleophiles can be generated as alkoxide or phenoxide ions.1 The phenoxide ion is much easier to generate so weaker bases such as sodium and potassium hydroxide can be used. Deprotonation of alkoxides by hydroxide is less efficient so stronger bases such as potassium tert-butoxide or sodium or potassium hydride are used. Amines can act as nucleophiles based on their basicity or they can be converted to sulfonamides and deprotonated using metal carbonates or sodium or potassium hydride (Section 2.2). Sulfur nucleophiles are commonly formed using cesium carbonate.55 The use of cesium carbonate creates what is termed as the ‘cesium effect’ (Section 2.3).56 Specific bases sometimes require certain solvents but it is the nucleophile that more often dictates the solvent. In most cases, the choice of solvent is dependent on its ability to solubilise the anion formed. Polar aprotic solvents such as tetrahydrofuran and dimethylformamide are most commonly used for this purpose.1 The amount of solvent used can be critical to the size of the ring formed. A concentrated solution favours more intermolecular reactions and thus larger rings (or polymers). The question of leaving groups in macrocyclic reactions is based on economy, efficiency and reactivity. The most common choices are the halides, mesylates and tosylates. The tosylate group is the most reactive followed by mesylate, iodine, bromine and finally the chloride ion.

2.0.4 Ring Closure Methods

The reaction between the two ends of a chain to form the cyclic compound depends greatly on chain end proximity, i.e. the closer the two chain ends are to each other the greater the chance for an intramolecular reaction.49 The template effect deals

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers directly with this issue. It involves the use of a cation to act as a temporary or permanent coordination site allowing the negative heteroatoms of the chain to assemble themselves around the positive ion in a circle or semicircle, thus bringing the two reacting ends closer together.1 An example is shown in Scheme 9.

TsO OTs O O O

O O O O O O M M OTs O O O O O O O O O

Scheme 9

Another commonly employed method used to promote cyclisation is the high dilution technique. It follows that the intramolecular reaction is first order and its rate is proportional to concentration, while the intermolecular reaction is second order and its rate is proportional to the square of the concentration. Dilution should therefore favour the intramolecular reaction.49 In principle, the high dilution technique can be carried out under homogeneous or heterogeneous conditions. It requires the use of a large amount of solvent with a small amount of reactants, so that once the linear chain has formed the remaining reactants are so dilute that the intermolecular collisions are decreased and the two functional ends of the chain react with each other.57 In practice, the conditions involve the use of a reduced amount of solvent with the reactants added over a period of time. If the reaction rate is greater than the addition rate, the reactant concentration is kept low and the dilution is high.1

2.1 Selectivity of Crown Ethers for Metal Ions

One of the properties of crown ethers is their ability to complex metal ions. A complexed macrocycle can be defined by the spatial arrangement of its components, its superstructure and the nature of the intermolecular bonds that hold the components together.2 There are many factors that influence the selectivity of the macrocycle for different ions; they include the cavity size and shape, substituent effects, conformational flexibility/rigidity and the type of heteroatom.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

The formation of the most stable complex between a metal and macrocycle, in respect to cavity size, is explained in terms of the size match selectivity hypothesis.58 This is when the ionic radius of the metal and the cavity size of the macrocycle are the closest. This rule applies very well to the complex of 18-crown-6 and the potassium ion. Models show that the potassium ion fits in the cavity best and that potassium forms the most stable complex compared with the other cations. This rule however, does not always apply. Macrocycles of the rigid type (smaller macro ring) tend to discriminate between cations that are either smaller or larger than the one that exactly fits in to the cavity.59 This is known as peak selectivity. In some cases the macrocycle creates a sandwich effect, that is, two macrocycles form a complex with the one cation. This type of interaction is seen between 12-crown-4 (cavity size, 1.5 Å) and the sodium ion (ionic radius, 2 Å) (Figure 5).60 The incorporation of groups such as benzene, cyclohexane, pyridine rings or other constituents into the crown ether framework can lead to more rigid macrocycles which can alter the strength and selectivity of the crown ether. Macrocycles of the flexible type (larger macro ring) prefer the smaller cations. This is known as plateau selectivity.59 A flexible crown ether can accommodate a wider range of metals than the rigid crown ethers as such factors like ligand conformations and the cation solvation enthalpy become important.58

O O O O

Na+

O O O

O Figure 5

Crown ethers fall into three different categories with respect to metal binding affinities. The first group contains the small macrocycles and usually coordinate to metal ions outside the cavity as in the case of 12-crown-4. The second group contains only 18- crown-6. When 18-crown-6 is complexed to certain metal ions it exhibits the idealised 61 62 D3d geometry (64). In the uncomplexed compound (65) the methylene groups occupy the central area, however they can rotate outward to create a cavity. The third group incorporates the larger macrocycles. Due to their flexibility they can wrap around the

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers cation. This phenomenon is seen in nature with valinomycin, which envelops the potassium ion and transports it in vivo.49 In this way it protects the potassium from the lipid environment.

O O O O O H O H HH O O O O O O 64 65

The substitution of nitrogen or sulfur leads to crown compounds capable of targeting different cations. Oxygen donor groups have an affinity for alkali, alkaline earth and lanthanide ions.57 Nitrogen and sulfur donor groups are considered soft bases and therefore prefer the transition metal ions (eg, nickel, copper, iron) and heavy metal ions (eg, silver, lead, mercury), which are classified as soft acids.63 It has been shown through X-ray crystal analysis64 that complexed azacrown ethers tend to adopt the structure shown in Figure 6.63 It shows the metal at the centre coordinated by the surrounding nitrogen groups on almost the same plane and the anions are located on the axis normal to the plane. Many reviews are available on the different complexes formed by azacrown compounds64 and thiacrown compounds.65

N N

Figure 6

The macrocyclic effect is based on the ability of the crown ether to complex a metal ion more strongly than its open-chain analogue. The macrocyclic effect is observed quite strongly in the oxa and aza macrocycles but is much less pronounced in thiacrown

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

66 ethers. This is related to the tendency of the sulfide linkages (-SCH2CH2S-) to assemble themselves outside (exodentate) the cavity.67 This is because of the longer carbon-sulfur bond and the 1,4-interactions in gauche C-C-E-C and E-C-C-E units (E = O, S) (Figure 7).68 In the oxa analogues this orientation gives anti C-O and gauche C-C bonds while in thiacrowns the opposite effect is observed, therefore the sulfur lone pairs point out. It follows that a considerable amount of reordering is necessary for metal binding to take place, which is energetically not favoured. Due to this phenomenon, some thiacrown ethers tend to bridge metal ions rather than chelate to them.69 It has been observed that the binding ability of some thiacrown compounds is only slightly greater than the open chain ligand.66a There are, however, exceptions with (96) and (100) (Section 2.3) having the endodentate orientation of their sulfur atoms.

O OSS CC CC weakly destabilising stabilising gauche C-C bonds

2.4 Å 1.8 Å H H H H C C C 1.4 Å C CO CS1.8 Å gauche is gauche is not destabilised disfavoured

gauche C-X bonds

Figure 7

2.2 Nitrogen Macrocycles (Azacrown Ethers)

Macrocycles containing nitrogen have been known for over 100 years. Examples include a number of biologically important compounds such as the green pigment of chorophyll and the haem of haemoglobin. They were recognised as tetrapyrrole porphyrin macrocycles (68) and were later synthesised by a series of condensations between pyrrole (66) and formaldehyde (67) (Scheme 10). During this early period in azacrown chemistry, Baeyer70 synthesied tetraazaquaterene, a compound resembling the porphyrins, via an acid catalysed condensation between pyrrole (66) and acetone (3) in an analogous fashion to the furan-acetone reaction (Section 1.0). The reaction is believed

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers to take place under a hydrogen bonding network.57 Since these early discoveries and syntheses there have been many synthetic approaches to the assembly of macrocycles containing nitrogen.

H N HN N O + HH NH N 66 67

68 Scheme 10

A second wave of interest in nitrogen macrocycles occurred in the 1930s. The early part of that decade saw significant interest in complexed azacrowns because of their industrial importance as pigments and dyeing agents. It is also interesting to note that in 1937, the same year that Lüttringhaus synthesised the first cyclic polyether (1), Alphen obtained the first saturated macrocyclic polyamine, cyclam (69).71

NH HN

2.2.1 Templated Syntheses

The first observed templated synthesis of a macrocyclic ligand was by Braun and Tcheriac72 in 1907 when they obtained metal derivatives of phthalocyanines from the reaction of pyrrole and o-diaminobenzene or o-cyanobenzamide or their related compounds with metals. In 1928, a dark by-product was isolated during the synthesis of phthalimide in an iron reaction vessel and proved to be the ferrous iron complex of phthalocyine.57 This area of research became the province of the pigments/dyestuffs industry. The term ‘template effect’ was not applied to the synthesis of macrocycles until the 1960s when Busch et al73 and Hurley et al74 recognised the role of the ferrous ion.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

Busch73 noted that the coordination sphere of the metal ion could hold the reactive groups in the correct orientation for cyclisation to occur. In 1960, Curtis et al75 used Ni(II) and Cu(II) to promote the cyclo condensation of ethylenediamine with acetone to afford a bisimine. Acetone (3) was added to the complexed diamine (70) to form the complexed tetraazadienes (71) in both the cis and trans forms (Scheme 11). The complexes were precipitated as the perchlorate salts. In order to explain the mechanism of formation of the macrocycles the compounds and their intermediates were decomposed. The authors found that ethylenediamine, mesityl oxide and/or acetone were generated depending on the conditions used for decomposition. They proposed that the acetone reacted with an N-isopropylidene imine to form either a mesityl oxide imine, which would then undergo a Michael type reaction with the adjacent amine group. Alternatively, a β-amino ketone would react with the adjacent amino group to form an imine. 2+

H H 2 2 O N N NNH NH N 4 MII i + H3CCH3 M + M N N 3 HN N NH N H2 H2 70 71

Reagents and Conditions: (i) room temp. 1 week, 80%

Scheme 11

In many cases the use of a metal template is needed for cyclisation to occur. Examples of non-template cyclised macrocyclic-Schiff bases76 are known, however a requirement of these reactions is that the starting materials contain rigid groups, such as benzene and pyridine rings. Jackels et al77 reported that the reaction of 1,3- diaminopropane monohydrochloride (72) and biacetyl (73) to form (74) was only possible in the presence of cobalt acetate tetrahydrate (Scheme 12). The use of metal ions in reactions has also led to an increase in the yield of the macrocycle being formed.57

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

OO H3C NNCH3 i + CoII H CCH NH2 NH3Cl 3 3 H3C NNCH3 72 73 74

Reagents and Conditions: (i) CH3OH, KOH, 15 min 2 °C,

Co(OAc)2•4H2O, CH3OH, room temp. 12 h, 30%

Scheme 12

The macrocyclic imine ligands can be reduced to form the cyclic amines using hydrogen and a catalyst, sodium borohydride, nickel-aluminium alloy or cathodic reduction.57 Subsequent demetallation can be effected by the addition of acid, a ligand exchange process whereby cyanide, sulfide or EDTA are added, or by the reduction of the metal if it has a suitable redox couple.78 Removal of the metal proved to be difficult in some cases.77

2.2.2 The Sulfonamide Method

The so-called Richman-Atkins reaction does not utilise a metal template but rather relies on the conformational constraints of its reactants to promote cyclisation. In 1974, Richman and Atkins79 reported the generality of the sulfonamide reaction to synthesise polyazacrowns. Conversion of primary amines to sulfonamides increases the acidity, which makes deprotonation of the nitrogen easier and allows it to react in an SN2 fashion. In the initial reaction, the disodium salt of a polysulfonamide (75) was reacted with a ditosylate or dimesylate (76) in dimethylformamide to yield the polytosylated cyclen (77) (Scheme 13). Detosylation with sulfuric acid afforded the polyamine macrocycle in 80% yield. The use of a tosylate or mesylate ester leaving group was found to give greater yields.80 Stetter et al81 and Koyama et al82 previously prepared similar macrocycles by reacting polysulfonamides with dihalides. The yields obtained were poor. This result was in contrast to the dihalide derivatives of the polyoxygen systems.3 The sulfonamide method can be used to synthesise small and large ring macrocycles including azacrowns possessing other heteroatoms.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

Ts Ts N Na X N i Ts N + NTs Ts N N Ts

N N Na X Ts Ts X = OTs, OMs 75 76 77

Reagents and Conditions: (i) DMF, 100 °C, 76 was added over 1-2 hr, 80%

Scheme 13

Richman and Atkins used preformed sulfonamide salts in their reaction, however the sulfonamide nucleophiles can be generated in situ by using a base. Metal carbonates are commonly used to form the nucleophile. Changing the metal carbonate can affect the cyclisation yield of the protected azacrown. Chavez and Sherry83 found using potassium and cesium carbonate in dimethylformamide gave the best results while the use of lithium and sodium carbonate yielded no product. Sodium and potassium hydride are often used when bases stronger than the carbonates are needed to deprotonate the sulfonamide. The diminished template effect in certain polytosylated azacrowns has been attributed to the effects of restricted rotational freedom of the starting material due to the bulky tosyl groups.83,84 It was postulated that the resulting small change in entropy allowed cyclisation to occur without a need for preorganisation of the starting materials or intermediates. These same reasons are used to explain why high dilution was not needed in some cases. It is interesting to note that replacement of the sodium ion for the tetramethylammonium ion in a Richman-Atkins reaction resulted in a decrease in yield of the macrocycle formed. This was explained as a slight template effect by the sodium ion.85 The tosylation of the nitrogen not only increases the acidity but it also protects the nitrogen from further reaction. For example, the reaction between 2-aminophenol (78) and tetraethylene glycol dichloride (79) can yield two monoaza compounds. Lockhart and co-workers86 formed a 12-membered ring (80) and a 15-membered ring (81) (Scheme 14). In this case the authors found that by changing the solvent they could obtain the desired crown ether. However, in many cases it is not that straightforward.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

NH2 O O NH O 3 ++Cl O Cl O O OH NO O 78 79 OH 80 81 Scheme 14

There are three main methods for the removal of the tosyl group: (1) acid hydrolysis with concentrated sulfuric acid, (2) reductive cleavage using a mixture of hydrobromic and acetic acid and (3) reduction using lithium aluminium hydride. The choice of method for removing the protecting group is based on the sensitivity of the macrocycle and its substituents for the particular conditions. More recently, Fukuyama et al87 showed that removal of a p- or o-nitrobenzenesulfonyl group could be achieved with relative ease using thiophenol and potassium carbonate in dimethylformamide. However, it was found that some thiolate addition occurred at the nitro group when it was in the para position, forming the phenylthioether. This effect was not observed using the ortho derivative.88 The sulfonamide method has been prominent in the synthesis of azacrown ethers, however there are other protecting groups that can be used. The diethoxyphosphoryl group has been shown to be an excellent activator of primary amines and can be removed using gaseous hydrochloric acid in tetrahydrofuran.89 The trifluroacetyl group has also been used.90 It can be easily prepared and readily removed after the reaction. The disadvantage of this group is the poor cyclisation yields due to the electronegativity of the trifluoroacetyl moiety, which reduces the nucleophilicity of the nitrogen. When activation is not a consideration, the benzyl protecting group can be used and easily removed by hydrogenolysis.

2.2.3 Azacrown Ethers via Amide Formation

Macrocycles with the amide (lactam) functional group are commonly used intermediates in the synthesis of various azacrowns. The ring-closure reaction can be effected by nitrogen addition to a diester or diacid chloride, or by Michael addition to an

α,β-unsaturated ester. Addition of nitrogen to a diester has led to the synthesis of many cyclic bislactams. The cyclisation process, often referred to as the Tabushi method, does not require high dilution in many cases. An example of this is the reaction between 1,3- 31

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers bis(2’-aminoethylpropane) (82) and diethyl malonate (83) which afforded the bisamide (84). Reduction of the amide groups using borane/tetrahydrofuran yielded cyclam (69) (Scheme 15). 91 Lithium aluminium hydride has also be used to reduce the amides.57

OO NH HN NH HN NH HN + iii EtO OEt N N NH HN NH HN 82 H2 H2 83 OO 84 69

Reagents and Conditions: (i) EtOH, reflux 3 days,

(ii) B2H6/THF, reflux 24 h, HCl, KOH/CH3OH, 80%.

Scheme 15

In alcoholic solvents esters react with primary amines but not with secondary amines, therefore avoiding the need for nitrogen protection.91 This is in contrast to the use of diacid chlorides, which require nitrogen protection. Stetter and Marx92 were the first to utilise the reaction between a diamine and diacid chloride in the synthesis of tetraaza macrocycles. The cyclisation proceeded cleanly and no by-products were observed. Dietrich et al53,93 first used this method to synthesise the diazacrown compounds, which are precursors to the cryptands. Later Dietrich and co-workers94 showed that macrocycles containing large numbers of nitrogen atoms could also be synthesised via this method. The bisamide (87) was formed by the reaction of a protected diamine (85) and a diacid chloride (86). The protected azacrown (88) was obtained after reduction (Scheme 16).

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

O O

NH2 Cl N N Ts Ts Ts H Ts Ts H Ts N N N N N N + i ii N N N N N N Ts Ts TsH Ts TsH Ts NH2 Cl N N

85O 86 87O 88

Reagents and Conditions: (i) NEt3, CH2Cl2/toluene (3:2),

room temp. 7 h (ii) B2H6/THF, reflux 20 h, 85%.

Scheme 16

Reaction of an α,β-unsaturated ester (89) with a diamine (90) has lead to the synthesis of cyclic compounds containing one amide group (91) (Scheme 17). The reaction proceeds via Michael addition of one amine to the β-unsaturated carbon followed by amide formation by the reaction of the other amine and the ester functional group. Kimura and co-workers95 have used this method to synthesise triamines, tetraamines and pentaamines. The lactam functional group was reduced to yield the desired azacrown. The yields obtained, however, were generally low.

H2 H2 O N N NH HN i + OEt NH HN NH HN 89

90 91

Reagents and Conditions: (i) CH3OH, room temp. 12 h→reflux 24 h, 50%.

Scheme 17

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

2.2.4 Peptide Chemistry

Peptide chemistry has also proved to be useful in azacrown synthesis. A bisamide can be formed in the reaction between a dicarboxylic acid and a diamine. In these cases the carboxylic acid functionality has to be activated. The use of dicyclohexylcarbodimide (DCC), diphenylphosphoryl azide (DPPA) and many other activating reagents has led to the synthesis of azacrown ethers in good yields. Qian and co-workers96 reacted (85) and the carboxylic acid derivative of (86) to form (87) (Scheme 16) using DPPA in dimethylformamide. Other work has included the use of DCC with 1- hydroxybenzotriazole (HOBT) to form azacrowns in 50-55% yields.97 An example of the activating properties of DCC is shown in Scheme 18.

O O HN ROH RO I R NH2 N NCN

(DCC)

O O HN R NHRI RO N N RI H

Scheme 18

2.2.5 Crab-Like Cyclisation

The reaction between a bis-α-chloramide (92) and a diamine (93) is another method that has been used to form cyclic bisamides (94) (Scheme 19).98 Subsequent reduction of the amide functionalities using borane/tetrahydrofuran yielded the substituted azacrown (95). The cyclisation is different from other reactions to form bislactams in that the amide functional groups are not formed in the ring closure step. The advantages of using this method are: (1) nitrogen protection can be avoided because the secondary amide functionality on the starting material is unreactive as a nucleophile;

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

(2) the reaction is short and the overall yields are good; and (3) the chloride is activated by the amide group but the molecule does not possess the blistering properties of the β- 57 chloroamines. The ‘crab-like’ bis-α-chloramide starting material (92) can be synthesised from simple primary and secondary diamines and oligooxadiamines but not from polyamines with terminal secondary and internal tertiary amine groups. One problem is that the starting material is susceptible to strong acids and bases as the chloroacetamide group can be cleaved. This method has been used to prepare azacrowns of different sizes with various nitrogen substituents.

O RRO N N O R R O R R Cl Cl N N N N R = CH3 iii I N I I N I 92 R N R R N R + N N II II RI NH NH RI R R N 94 95 RII I II R = CH2CH3, R = (CH2)2O(CH2)2OH 93

Reagents and Conditions: (i) Na2CO3 CH3CN, reflux 24 h, (ii) B2H6/THF, reflux, 60%

Scheme 19

2.3 Sulfur Macrocycles (Thiacrown Ethers)

Much of the early research into cyclic sulfur compounds was in relation to understanding ring formation and theories of ring strain.99 The reaction between a dihalide and a dithiol yielded high polymers and small amounts of the cyclic sulfur compounds. Ray et al100 was the first to report the synthesis of a trithia macrocyclic compound from ethanedithiol, in 1920. In the ensuing years subsequent research proved that the isolated compound was not 1,4,7-trithiacyclononane (9-S-3) (96) but rather p- dithiane (97).101,102 The first reported synthesis of (96) was by Ochrymowycz et al102 in 1977. This will be discussed later.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

S SS

S S 96 97

In 1934, Meadow and Reid103 synthesised a number of cyclic sulfur compounds containing two, four and six sulfur atoms. The reaction between various dithiols and dibromides lead to the formation of polymers and small amounts of the cyclic compounds. One of the macrocycles formed was a hexathia compound (18-S-6) (100) via a tetra-component cyclisation. Ethylene bromide (98) was added to 3-thiapentane-1,5- dithiol (99) in absolute ethanol containing an equivalent amount of base to afford the thiacrown, albeit in 1.7% yield (Scheme 20).

S S S i + Br BrHS S SH S S 98 99 S

100

Reagents and Conditions: (i) NaOEt, EtOH, room temp., 1.7%

Scheme 20

Thirty-five years later, Black and McLean104 managed to improve the yield significantly to 31%. In this case, the reaction was carried out under high dilution conditions. They also managed to synthesise cyclic sulfur compounds containing oxygen and/or nitrogen groups using these same reaction conditions. In the same year, Rosen and Busch105 synthesised a number of cyclic tetrathia compounds (101), (102) and (103) in 7.5%, 4% and 16% respectively.

S S S S S S

101 102 103

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

As interest in crown ethers increased so did the interest in macrocyclic compounds containing sulfur atoms. Following on from the work by Pedersen3, Bradshaw et al106 were interested in the synthesis of crown ethers containing both oxygen and sulfur atoms. Bradshaw and co-workers106 found that the reaction between an ethylene glycol dichloride and a dithiol or sodium sulfide in basic ethanol afforded the desired oxathiacrown ethers. At the same time Ochrymowycz and co-workers107 were investigating the synthesis of several polythiacrown compounds. Their initial attempt at the synthesis of (100) proved to be less successful than previously reported. They were only able to obtain the desired product in 8% yield following the procedure by Black and McLean104 however by adopting a two-component approach they were able to improve the yield to 33%. They reacted a β-chlorothioether (104) with (99) in n-butanol using sodium (Scheme 21). The disadvantage of this procedure is the use of the β- chlorothioethers, which are powerful vesicants as they are analogues of mustard gas (blister-causing agent).

S S S i n + Cl S Cl HS S SH S S n = 3 99 104 S 100

Reagents and Conditions: (i) NaOnBu, room temp. 2 days, 33%

Scheme 21

Using these conditions, Ochrymowycz and co-workers107 synthesised a number of thiacrown compounds and investigated their metal-binding properties as a function of ring size and the number of donor atoms. To further understand the relationship, Ochrymowycz et al102 became interested in the synthesis of the smaller nine-membered ring (96). The synthesis of the thia analogue proved difficult even though the oxa, aza or mixed oxa-aza-thia nine membered ligands had been synthesised in good yields. The product was obtained in only 0.04% yield. The reaction was initially carried out in n- butanol but the desired product was not observed. The nine membered ring only formed when ethanol was used. This was explained by the solvating effect of ethanol at a particular stage in the cyclisation process.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

2.3.1 Synthesis of Thiacrown Compounds Using Cs2CO3

Much of the early research into sulfur macrocycles was hampered by the lack of a good yielding general procedure. In 1980, Kellogg and co-workers108 revolutionised the synthesis of thiacrown compounds using cesium carbonate in dimethylformamide. In general, the procedure required that the dithiol (105) and the dibromide (106) be added over a period of time to a stirred suspension of cesium carbonate in dimethylformamide at 50-60 °C to form (107) (Scheme 22). Using this procedure they were able to form a number of macrocycles containing sulfur and oxygen in yields of 65% and higher. They later applied this method to the synthesis of polythia compounds by reacting a dithiol and a β-chlorothioether.

O SO i n + HS O SH Br O Br O O n = 3 106 105 S

107

Reagents and Conditions: (i) CsCO3, DMF, 50 °C, 105 and 106 were added over 12-15 hr.

Scheme 22

2.3.2 The Cesium Effect

Substitution of cesium carbonate with rubidium, potassium, sodium and lithium carbonates gave reduced yields or no product at all.108 This advantageous function of the cesium ion is known as the cesium effect. Based on the ω-halo carboxylate cyclisations, Kellogg et al109 initially postulated that the tight ion pairs formed by the carboxylate anion and the cesium cation promoted the intramolecular cyclisation. It was believed that the cesium ion assisted in the transition state by attaching to the carboxylate while providing electrophilic assistance to the leaving group (Figure 8). In essence the cyclisation occurred on the ‘surface’ of the cesium ion. This theory was the basis for other reactions involving thiols, phenols, sulfonamides and 1,3-dicarbonyl compounds with Cs2CO3.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

(CH2)n C RI R O O X

Figure 8

Subsequent research by Kellogg110 and Galli111 and their co-workers showed that this was not the case. Using 133Cs NMR studies110 and kinetic studies111 both groups determined that the cesium salts exist as free ions or solvent separated ion pairs in dipolar aprotic solvents such as DMSO and DMF. In relation to thiolate chemistry, the effect of the cesium in promoting the cyclisation is believed to occur after the intermediate ω- 65a halo-α-thiolate has been formed. The large cesium cations form weak ion pairs with the thiolate anions, producing ‘naked anions’ which are very soluble and highly reactive. Under high dilution conditions, the enhanced reactivity would therefore favour the intramolecular over the intermolecular reaction.

2.3.3 Templated Syntheses

Unlike the strong template effect exhibited by the oxo and azacrown ethers, low sulfur-active metal ion coordination renders template effects of little consequence in the synthesis of thiacrown ethers. Attempts to use transition metals as templates have not been very successful, although there have been exceptions. Sellman and Zapf112 reported the synthesis of (96) in 60% yield via a templated synthesis. Dithiol (99) was coordinated to a molybdenum(0) carbonyl complex to form (108), which was then reacted with ethylene bromide (98) to form the complexed thiacrown (109). The complexed macrocycle was liberated from the metal under regeneration of the starting complex to form (96) (Scheme 23). Edema et al113 reported more recently that boric acid and a boron/aluminium isopropoxide cocktail were effective in the synthesis of a number of thiacrown ethers.

INTRODUCTION : Synthetic Strategies for the Preparation of Crown Ethers

S S S i S S S ii SS [N(CH3)4]2 + Mo Br Br Mo OC CO OC CO S CO 98 CO 96 108 109

Reagents and Conditions: (i) CH3CN, room temp. 15 min, (ii)

(NCH3)4(SC2H4SC2H4S), DMSO, room temp. 2 days, 60%.

Scheme 23

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.0 Synthetic Targets

As discussed in Chapter 1 the aim of the present work was to synthesise the sulfur (55, 56, 57) and nitrogen (58, 59, 60) analogues of the axial spiroacetal crown ethers (53a-c) in order to evaluate their binding affinities towards specific metal ions and their interactions with metal complexes (second sphere coordination).

O O O

OO O O OO

SSS S SS S S S 55 S S 56 S 57

O O O

OO O O OO

NH NH NH HN HN NH H N NH HN 58 HN NH 59 N H 60

According to the retrosynthesis outlined (Scheme 24) it was envisaged that the synthesis of the thiacrown and azacrown ethers could be carried out via two possible pathways. The first approach (Path A) involves the reaction between the spiroacetal diol and the respective β-chloroethyl sulfide or amine. It was envisaged that the cyclisation would be effected in a similar fashion to the method used by Brimble et al48 to synthesise spiroacetal crown ethers (53a-c) and (54a-c). The second approach (Path B) involves the reaction between a dithiol or diamine with an electrophilic group attached to the spiroacetal moiety. The electrophilic group could be a leaving group such as a tosylate or mesylate for reaction with a dithiol or a protected diamine. Alternatively, the electrophilic group could be a carbonyl group such as an ester or aldehyde, which could undergo nucleophilic addition with an amine. This

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers latter method is more amenable for the construction of the azacrown ethers than the thiacrown ethers. The spiroacetal containing the electrophilic side chain can be synthesised from the parent spiroacetal diol (50), which had previously been prepared by Brimble et al.114 The synthesis of the spiroacetal diol (50) is described in the next section.

A OOA

B B XX X Path An Path B

O O

O + n O + n Cl X Cl HX X XH X = S, NH X = S, NH, NR OH OH OO

R = CH2OMs, CH2OTs or R = CHO or R = C Y (Y = leaving group) O Scheme 24

3.1 Synthesis of Spiroacetal Diol (50)

3.1.1 Synthesis of (±)-1,7-Dioxaspiro[5.5]undec-4-ene (114)

The synthesis of spiroacetal diol (50) has previously been reported by Brimble et al114 starting from alkene (114) (Scheme 25). Alkene (114) was treated with dimethyldioxirane to form the α-epoxide (115), which in turn was treated with lithium diethyl amide to form the allylic alcohol (117). Epoxidation of the allylic alcohol (117) using meta-chloroperoxybenzoic acid, followed by reduction of the epoxy alcohol (119) formed the spiroacetal diol (50). The incorporation of the hydroxyl groups after cyclisation was necessary so as to avoid the formation of the thermodynamically more stable equatorial isomer.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O O O O

O OH OH O OH OH 114 115 117 119 50 Scheme 25

Alkene (114) was prepared following Scheme 26 starting from δ-valerolactone (110) and trimethylsilyl protected 3-butyn-1-ol (111). The protected alcohol (111) was treated with n-butyllithium in THF at –78 °C. δ-Valerolactone (110) was then added to the acetylide. The coupled product was isolated and treated with acidic methanol or ethanol to yield the methoxyacetal (112) or ethoxyacetal (113) respectively (Scheme 26). Initially, this reaction proved to be problematic with only low yields (20 - 36%) obtained for the methoxyacetal. Changing the acid source did not increase the yield to the reported 70%. It was hoped that changing the alkoxy group would increase the stability and yield of the product. Indeed using ethanol as the solvent afforded ethoxyacetal in an improved 88% yield. The low resolution mass spectrum of the product showed a molecular ion at m/z 153 which corresponded to the molecular formula of C9H13O2 indicating the loss of the ethoxy group. The infrared spectrum exhibited a broad absorbance at 3419 cm-1 supporting the presence of the alcohol group. Another absorbance at 2249 cm-1 showed the presence of the acetylene. The 1H NMR spectrum was assigned on the basis of 2D

COSY and HSQC experiments. The OH group resonated at δH 2.54 as a triplet with

JOH,4’ 6.2 Hz. A triplet (J3’,4’ 6.5 Hz) at δH 2.51 was assigned to 3’-CH2. The ethoxy CH3 13 group resonated as a triplet at δH 1.24 with JCH3,OCH2 7.1 Hz. The C NMR spectrum was assigned with the aid of HSQC and HMBC experiments. The quaternary carbons at

δC 80.2 and δC 81.7 were characteristic of the acetylenic carbons C-1’ and C-2’ and a quaternary carbon at δC 94.4 was assigned to the acetal carbon. With the ethoxyacetal in hand, the next step involved the semi-hydrogenation of the acetylene over Lindlar catalyst in pentane/ether (4:1) followed by acid catalysed cyclisation using pyridinium p-toluenesulfonate to afford the spiroacetal olefin in 80% yield. The 1H and 13C NMR spectra were in agreement with the literature NMR data reported for spiroacetal (114).115 Notably the spiroacetal carbon C-6 was observed at

δC 92.8 and the vinylic protons 5-H and 4-H resonated at δH 5.62 and δH 5.91-5.97 respectively in the 1H NMR spectra.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O MeO i, MeOH, Amberlite resin O + OSiMe3 O 111 110 112

OH i, EtOH, ii, iii PPTS 3' 4' 2' 8 9 EtO 2 1' O 7 ii, iii 10 3 O 1 6 1 2 11 O 4 6 5 3 4 5 113 114

Reagents and Conditions: (i) n-BuLi, -78 °C, THF, 45 min, then 110, 45 min; (ii) H2,

Lindlar catalyst, pentane/ether (4:1); (iii) CH2Cl2, PPTS, room temp., 30 min, 78%. Scheme 26

3.1.2 Epoxidation of Olefin (114)

Having successfully synthesised the spiroacetal olefin (114) the next step involved epoxidation using dimethyldioxirane to form both the desired α-epoxide (115) together with the minor β-epoxide (116) (Scheme 27). The dimethyldioxirane/acetone solution was added to the spiroacetal olefin and the reaction was monitored by TLC until completion. The α-epoxide (115) was obtained in 71% yield and the β-epoxide (116) in 11% yield after purification by flash chromatography. The 1H and 13C NMR spectra for both epoxides were in agreement with the literature.114

O O O i or ii + O O O O

114O 115 116 Reagents and Conditions: (i) Dimethyldioxirane, acetone, room

® temp., 18 h, 115 (71%), 116 (11%) (ii) NaHCO3, acetone, Oxone

(1.2 equiv), H2O, 0 °C→room temp., 18 h, 115 (41%), 116 (36%)

Scheme 27

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

Dimethyldioxirane is a three membered cyclic peroxide that can be prepared under buffered conditions from Oxone® (potassium peroxymonosulfate) and acetone and subsequently isolated by distillation116 or it can be generated in situ.117 When the 0.1 M solution of dimethyldioxirane was used in the epoxidation reaction the desired α-epoxide formed preferentially, however in situ formation of dimethyldioxirane led to the synthesis of the α- and β-epoxides in a 1:1 ratio. A possible explanation for this observation is that Oxone®, which itself is used as an oxidising agent, competes with dimethyldioxirane in the epoxidation of olefin (114). Since Oxone® is a less sterically sensitive oxidant than dimethyldioxirane it does not discriminate between the α- or β-faces. Dimethyldioxirane, on the other hand, delivers the oxygen primarily to the less sterically hindered α-face, which is reflected by the α:β product ratio. There are two proposed transition states for the epoxidation of olefins using dimethyldioxirane. Transition state A shows the direct delivery of the oxygen atom to the double bond proceeding via formation of a three membered ring, whereas, transition state B shows the addition of the oxygen atom via a ring opened diradical (Scheme 28).118 The authors suggest that oxidation by the diradical is the most likely mechanism however further study needs to be carried out to conclusively determine the mechanism of epoxidation.

OOO O

O O R1 O R3 O + O O 2 4 Me Me R1 R3 R1 R3 R R

R2 R4 R2 R4 AB Scheme 28

3.1.3 Base-Induced Ring Opening of Epoxide (115) and Epoxidation of Allylic Alcohol (117)

With the α-epoxide (115) successfully in hand, attention turned to its base- induced ring opening to allylic alcohol (117). The mechanism involves the removal of a

β-proton syn to the epoxide oxygen and is thought to proceed via a cyclic six-membered 119 transition state. This mechanism can be applied to the reaction of the α-epoxide with lithium diethyl amide (Scheme 29). The base coordinates to a lone pair of electrons on

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers the epoxide oxygen and promotes syn β-proton removal rather than anti β-proton removal. In this case the β-epoxide cannot react because it does not possess a syn β-proton. The base-induced ring opening of α-epoxide (115) yielded the allylic alcohol (117) in 76% yield and the homoallylic alcohol (118) in 3% yield (Scheme 30). Base- induced ring opening has also been performed using chiral lithium amide bases and this will be discussed in more detail in Chapter 5.

O O O

O H O H OH N Li Et Li Et N Et Et Scheme 29

O O O i + O O O

O OH OH 115 117 118

Reagents and Conditions: (i) Lithium diethylamide (1.1 equiv),

hexane, -35 °C→room temp., 20 h, 117 (76%), 118 (3%)

Scheme 30

Allylic alcohol (117) was next treated with meta-chloroperoxybenzoic acid buffered with sodium acetate in dichloromethane affording the syn-epoxy alcohol (119) and anti-epoxy alcohol (120) in 88% and 9% yields respectively (Scheme 31). Meta- chloroperoxybenzoic acid is capable of hydrogen bonding with hydroxyl groups and other oxygen substituents in the substrate. It is envisaged that adoption of the stabilised transition state depicted in Figure 9 together with the fact that the lower α-face is less sterically hindered allows for the fact that syn-epoxy alcohol (119) is the major product. The 1H and 13C NMR spectra recorded for the syn-epoxy alcohol (119) was in agreement with those reported in literature.114

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O O i + O O O O

OH OH O OH 117 119 120

Reagents and Conditions: (i) m-CPBA (2.0 equiv),

NaOAc, CH2Cl2, 0 °C→room temp., 119 (88%), 120 (9%)

Scheme 31

O O H O O R Figure 9

3.1.4 Reduction of syn-Epoxy Alcohol (119)

Finally, the reduction of the syn-epoxy alcohol (119) with lithium aluminium hydride in tetrahydrofuran afforded the desired 3,5-diaxial diol (50) in 81% yield together with the 4,5-diol (121) in 10% yield (Scheme 32). In order to establish that the 1,3-diaxial relationship of the hydroxyl groups was obtained, the coupling constants observed for the resonances assigned to the methine protons 3-H and 5-H were examined. 5-H resonated as a triplet at δH 3.45 with J5,4 3.0 Hz. This observation established that 5-H adopted an equatorial position because the coupling constant was within the range for typical equatorial-equatorial or axial-equatorial coupling.120 3-H resonated as a multiplet hence an indirect assignment of the stereochemistry at C-3 was obtained from the coupling patterns of 4ax-H and 4eq-H. The magnitude of the coupling constants J4ax,3 3.0 Hz and J4eq,3 3.0 Hz established that 3-H adopted the equatorial position.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O 7 50 O i Hax 6 2 + O O O Heq Heq 5 Heq 3 OH OH O OH OH OH J 121 119 4ax,3 = 3.0 Hz J4eq,3 = 3.0 Hz J5,4 = 3.0 Hz

Reagents and Conditions: (i) LiAlH4, THF, room temp., 12 H, 50 (81%), 121 (10%)

Scheme 32

3.2 Synthesis of the Spiroacetal Thiacrown Ethers via Path A

With the successful synthesis of spiroacetal diol (50) in hand, attention next focused on the synthesis of the spiroacetal thiacrown ethers (55), (56) and (57). Adopting the retrosynthetic strategy depicted for Path A in Scheme 33 it was envisaged that the reaction between the spiroacetal diol (50) and the β-chloroethyl sulfides (104), (127) and (128) would afford the thiacrown ethers (55), (56) and (57).

O O Path A + n O O Cl S Cl 104, n = 3 OO OH OH 127, n = 4 50 128, n = 5 SS S n 55, n = 3 56, n = 4 57, n = 5 Scheme 33

3.2.1 Synthesis of the β-Chloroethyl Sulfides (104), (127) and (128)

β-Chloroethyl sulfides (104) and (127) were synthesised starting from commercially available 2-mercaptoethyl sulfide (99) and 3,6-dithia-1,8-octanediol by well-established methods reported in the literature (Scheme 34).68,121 The general procedure involved the reaction between the dithiols (99) and (122) with 2-chloroethanol

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers to form diols (124) and (125) which were then treated with thionyl chloride to yield the 1 desired β-chloroethyl sulfides (104) and (127) (Scheme 34). The H NMR spectra were in agreement with the literature.68,121a

iii n n n HSS SH HOS OH ClS Cl 99, n = 1 124, n = 3 104, n = 3 122, n = 2 125, n = 4 127, n = 4 123, n = 3 126, n = 5 128, n = 5

Reagents and Conditions: (i) Absolute EtOH, Na(s), 2-chloroethanol (2.0 equiv), reflux, 18 h, 124 (80%), 125 (77%); (ii) thionyl chloride (3.0 equiv),

CH2Cl2, 18 h, 104 (90%), 127 (85%).

Scheme 34

Dithiols (122) and (123) required for the synthesis of β-chloroethyl sulfides (127) and (128) were prepared by reduction of the corresponding thioacetates (130) and (131) using lithium aluminium hydride (Scheme 35). Thioacetate (130) was prepared from

β-chloroethyl sulfide (129), which in turn was synthesised from commercially available 3,6-dithia-1,8-octanediol. Thioacetate (131) was prepared from β-chloroethyl sulfide (104) by reaction with cesium thioacetate. The direct synthesis of dithiols (122) and

(123) from the respective β-chloroethyl sulfides (129) and (104) using acid and thiourea was found to give a mixture of products that were difficult to separate. It was found that synthesis via the thioacetate species was much more effective. Edema et al122 did not purify the intermediate thioacetates (130) and (131), however it was found that purification led to an increased yield of the respective dithiols.

O O ii2 3 5 i n n n Cl SCS lH3C SSS SCH3 HSS SH 1 4 129, n = 1 130, n = 1 122, n = 2 104, n = 2 131, n = 2 123, n = 3

Reagents and Conditions: (i) cesium thioacetate, DMF, 16 h, 130

(86%), 131 (79%); (ii) LiAlH4, ether, 122 (89%), 123 (68%). Scheme 35

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

The high resolution mass spectrum for thioacetate (130) exhibited a molecular ion at m/z 298.01821 corresponding to the molecular formula C10H18O2S4. The infrared spectrum exhibited an absorbance at 1685 cm-1 supporting the presence of a carbonyl group of the thioacetate moiety. The 1H NMR spectrum was assigned with the aid of 2D

COSY and HMQC experiments. A singlet at δH 2.34 was assigned to the methyl group. 13 5-CH2 and 6-CH2 also resonated as a singlet at δH 2.83. The C NMR spectrum exhibited resonances at δC 30.6 and δC 195.3 assigned to the methyl groups and the carbonyl carbons respectively.

Thioacetate (131) analysed correctly for C12H22O2S5 with a molecular ion at m/z 158.02229 in the high resolution spectrum. A molecular ion m/z at 360.01736 was 34 also observed supporting the molecular formula C12H22O2S4 S. The infrared spectrum exhibited a carbonyl group absorbance at 1691 cm-1. The 1H NMR spectrum exhibited a singlet at δH 2.33 representing the methyl groups. An eight-proton singlet at δH 2.80 was 13 assigned to the CH2S groups. The C NMR spectrum showed the carbonyl group at

δC 195.3. Treatment of thioacetate (130) with lithium aluminium hydride in diethyl ether afforded dithiol (122) after work-up and purification by flash chromatography. The 1H NMR spectrum was in agreement with the literature.68 Using a similar procedure dithiol (123) was prepared in 68% yield. The high resolution mass spectrum for dithiol (123) exhibited a molecular ion at m/z 274.00046 corresponding to the molecular formula 1 C8H18S5. The H NMR spectrum exhibited a multiplet at δH 1.68-1.77 assigned to the SH groups. A sixteen proton multiplet at δH 2.67-2.81 was assigned to the CH2S protons.

3.2.2 Reaction Between Spiroacetal Diol (50) and β-Chloroethyl Sulfide (104)

Having successfully synthesised both the spiroacetal diol (50) and the

β-chloroethyl sulfide (104), the synthesis of spiroacetal thiacrown ether (55) was next attempted. It was envisaged that the high dilution technique could be employed in the synthesis of the thiacrown ethers. Methods used successfully by Kellogg108 and Brimble48 and their co-workers were investigated. The method by Kellogg et al108 involved the addition of the spiroacetal diol (50) and the β-chloroethyl sulfide (104) to a suspension of Cs2CO3 in DMF at 60 °C for 3 h. Subsequent work-up afforded the recovered diol (50) and diene (132) (Scheme 36). Alternatively diol (50) was treated with sodium hydride under reflux for 30 min then the dichloride (104) was added over 3 h following a similar method previously reported by Brimble et al.48 Once again, the

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers reaction yielded diene (132) and the recovered diol (50). Clearly the dianion generated from diol (50) acted as a base rather than a nucleophile.

O HA 1 2 45 i or ii O + n ClS Cl HB SSS 3 6 OH OH 104, n = 3 132 50

Reagents and Conditions: (i) Cs2CO3, DMF, 60°C, 50 and 104 in DMF added over 3 h, 132 (85%); (ii) KH (1.2 equiv), THF, reflux, 30 min, then 104 in THF over 3 h, reflux, 24 h, 132 (90%)

Scheme 36

The high resolution mass spectrum recorded for (132) showed a protonated molecular ion at m/z 207.03352 which corresponded to the molecular formula C8H15S3. The 1H NMR spectrum was assigned on the basis of 2D COSY and HSQC experiments.

The vinylic proton at C-2 resonated as a double doublet at δH 6.33 with JCH,CHB 16.7 and

JCH,CHA 10.1 Hz. The terminal vinylic protons resonated at δH 5.17 and δH 5.26 as doublets with JCHB,CH 16.7 and JCHA,CH 10.1 Hz respectively, consistent with the trans and cis relationship between the terminal vinylic protons and the CHS proton. The 13C NMR spectrum exhibited resonances at δC 112.1 and δC 131.3 that were assigned to the two sets of vinylic carbons (CH2 and CH respectively).

3.2.3 Attempted Synthesis of Thiacrown Ether (134) via Olefin Cross Metathesis of Diene (132) with Spiroacetal Diene (133)

The synthesis of diene (132) prompted an investigation into the possibility of synthesising the spiroacetal thiacrown ethers via an olefin cross metathesis reaction. This would provide crown ethers with additional double bonds that could either be functionalised further or reduced to give the saturated thiacrown ethers. It was envisaged that a reaction between diene (132) and bisallyl ether (133) would result in the formation of the unsaturated crown ether (134) (Scheme 37).

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O

+ O O SSS 132 OO OO

133 S S S 134 Scheme 37

3.2.3.1 Synthesis of the Spiroacetal Bisallyl Ether (133)

In order to investigate the cross metathesis of diene (132) with spiroacetal diene (133) a synthesis of the bisallyl ether (133) was required. Thus, treatment of a solution of 3,5-diaxial diol (50) in THF with NaH followed by the addition of allyl bromide (135) yielded the desired bisallyl ether (133) in 84% yield (Scheme 38).

O O7

6 1 2 O + Br i O 5 3 135 OH OH OO 1' 1' 50 2' HA 2' 3' 3'

HB 133 Reagents and Conditions: (i) NaH, THF, reflux, 30 min, then 135 (2.0 equiv), 18 h, 133 (84%)

Scheme 38

The high resolution mass spectrum for bisallyl ether (133) exhibited a protonated molecular ion at m/z 269.17533 in the high resolution mass spectrum, supporting the molecular formula of C15H25O4. The infrared spectrum exhibited an absorbance at 1646 cm-1 due to the unsaturated double bonds and the absence of an OH absorbance was 1 noted. In the H NMR spectrum 5-H resonated as a triplet at δH 3.16 with J5,4 4.7 Hz suggesting that 5-H is in an equatorial orientation. 3-H resonated as a double double double doublet at δH 3.45 with J3,2ax 4.1, J3,2eq 4.1, J3,4ax 4.1 and J3,4eq 4.1 Hz, confirming 53

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

that 3-H is also in an equatorial position. 1’-HA resonated as a double double triplet at

δH 4.13 with Jgem 12.9, J1’A,2’ 5.4 and J1’A,3’ 1.5 Hz. The double double doublet at δH 5.25 with J3’A,2’ 17.2, J3’A,1’A 1.5 and J3’A,1’B 1.5 Hz was assigned to 3’-HA, suggesting that it is trans to 2’-H while 3’-HB resonated as a doublet at δH 5.13 with J3’B,2’ 10.3 Hz consistent with a cis relationship with 2’-H. The 13C NMR spectrum exhibited methine resonances

δC 70.7 and δC 76.0, assigned to C-3 and C-5 respectively. The characteristic quaternary spirocarbon C-6 resonated at δC 97.4. The vinylic C-2’ carbons were observed at

δC 116.54 and 116.55 while the resonances at δC 135.27 and 135.29 were assigned to the terminal vinylic carbons C-3’.

3.2.3.2 Olefin Metathesis

Olefin metathesis has become a powerful tool for carbon-carbon double bond formation with the development of well-defined new catalysts, such as Shrock’s molybdenum catalyst (136)123 and Grubbs’ ruthenium catalysts (137)124 and (138).125 The use of molybdenum catalyst (136) in olefin metathesis reactions was first reported by Grubbs,126 Wagener,127 Forbes128 and their co-workers. A limitation of (136) is that it is extremely sensitive to oxygen, water and acid functionalities, including, alcohols and carboxylic acids.129 In contrast, Grubbs’ first generation ruthenium catalyst (137) has displayed a much greater tolerance for a wider range of functional groups and solvents. However, the range of substrates susceptible to olefin metathesis has been limited because of the lower reactivity exhibited by (137) compared to that of the molybdenum catalyst (136). The more recent introduction of N-heterocyclic carbene-coordinated catalysts, such as (138), has considerably improved the reactivity of the ruthenium based catalysts while retaining the functional group tolerance of (137).130

R NNR N PCy3 Cl Ph Cl Ph Mo Ru Ru O F C Cl H Cl 3 O Ph PCy H CF3 3 PCy3 F3C 137 138 CF3 136

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

Olefin metathesis forms part of three closely related reactions: ring-opening metathesis polymerisation (ROMP), acyclic cross metathesis and ring-closing metathesis (RCM).131 The general mechanism involves a series of alternating [2+2] cycloadditions and cycloreversions between the metal alkylidene and the metallacyclobutane species.129 This mechanism was first postulated by Chauvin132 for polymerisation reactions and is now generally accepted for olefin metatheses. Using RCM as an example, the first step involves the cycloaddition between the metal catalyst and an alkene to form the metallacyclobutane species. This then undergoes a cycloreversion to form the metal alkylidene with the elimination of ethene. Another cycloaddition followed by a cycloreversion regenerates the activated catalyst and forms the new carbon-carbon double bond as either the E isomer or Z isomer or a mixture of the two. The stereochemical outcome of the reaction is governed by the conformation of the substrate and the stability of the possible products. The mechanism is illustrated in Scheme 39 which also shows the overall reactions.129,131 The mechanism is in reverse in the case of ROMP.

Overall Reaction Ring-Closing Metathesis n Polymerisation

RII I II Overall Reaction I R + R Acyclic Cross Metathesis R

Overall Reaction Ring-Closing Metathesis

LnMCHR LnMCHR

LnMCHR LnM

LnM

H2CCHR Scheme 39

Cyclic Cross Metathesis Reaction

In cross metatheses three unique products can be formed: one desired heterodimeric product and two homodimeric products, each as a mixture of olefin isomers.133 Our aim was to determine whether the cyclic cross metathesis reaction of

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers diene (132) that contains three sulfur atoms, with spiroacetal diene (133) would proceed to give the cross metathesis product (134) as depicted in Scheme 40.

O O

+ O O SSS 132 OO OO

133 S S S 134 Scheme 40

Much of the literature suggests that compounds containing sulfur atoms are not very susceptible to olefin metathesis reactions because the sulfur atoms and the metal centres of the catalysts interact favourably.134 The sulfur atoms can coordinate to the metal and deplete the active catalyst and thereby terminate the catalysis. Despite this observation several authors have shown that olefin metathesis has been successful in compounds containing one or two sulfur atoms. 135 In 1995 Basset et al135a were the first to report the formation of a cyclic sulfur compound via RCM using a tungsten alkylidene catalyst (139) (Scheme 41). Armstrong,135b Lee135c and their co-workers have since reported the application of olefin metathesis to a range of sulfide and disulfide compounds.

139

W O OEt2 ArO Cl S S (Ar = 2,6-diphenyl-C6H3)

Scheme 41

Initially the cross metathesis between diene (132) and the bisallyl ether (133) was attempted using Grubbs’ catalyst (137) (5 mol%) in dichloromethane. It was observed that the colour of the reaction mixture changed almost immediately from a deep purple/red to a yellow/black as soon as the reactants were added to the catalyst. This was a probable indication that the catalyst had been poisoned.129 It has been reported that Shrock’s molybdenum catalyst (136) is much more stable towards sulfide groups. This is because the sulfur atoms do not coordinate as well to the molybdenum as they do to the 56

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers ruthenium metal. Using Shrock’s catalyst (136) in benzene also proved to be disappointing with no change in either starting materials being detected after 5 h at room temperature or reflux. The catalyst eventually deteriorated and only the spiroacetal bisallyl ether (133) starting material was recovered. In order to conclusively determine the effect of each of the substrates on the catalysts, a ring-closing metathesis reaction was attempted using each of the individual reactants (132) and (133). The nine-membered cyclic sulfide (140) and spiroacetal olefin (141) were the envisaged products (Scheme 42) from the ring-closing metathesis of (132) and (133) respectively.

O O S O O SSS S S 132 140 OO OO

133 141

Scheme 42

Attempted ring-closing metathesis of diene (132) using Grubbs’ catalyst (137) was not successful and was again indicated by a colour change of the reaction mixture. The use of Shrock’s catalyst (136) also proved to be ineffective. On the other hand, the RCM reaction of the bisallyl ether (133) led to the formation of an interesting set of spiroacetal compounds with a general structure represented by compound (142) (Scheme 43). The initial reaction was performed using Grubbs’ catalyst (137) (5 mol%) in dichloromethane for 6 h at reflux and yielded three relatively polar compounds after purification by flash chromatography. These same products were obtained using Shrock’s catalyst (136) in similar yields.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

8 9

O 7 10 6 1 2 11 O 5 4 3 OO O 4' 1'

3' 2' O i or ii 2' 3' OO 1' 4' 142a OO142b 3 5 142c O 1 6 7 O

Reagents and Conditions: (i) Grubbs’ catalyst (137), CH2Cl2, reflux, 6 h, 142a (15%), 142b (40%), 142c (10%); (ii) Shrock’s catalyst (136), benzene, room temp., 5 h, 142a (18%), 142b (45%), 142c (9%)

Scheme 43

Initial examination of the 1H and 13C NMR spectra of the three compounds suggested they were structurally very similar. The spectra also resembled that of the bisallyl ether with the exception that the terminal carbons of the alkene were absent. This initial analysis suggested the formation of three RCM products as a mixture of isomers. However, further analysis by high resolution mass spectroscopy indicated that the compounds obtained were not the desired RCM product (141) since the mass spectra exhibited molecular ions with twice the mass expected for the desired RCM compound (141). This was consistent with the formation of spiroacetal dimers (142a-c) arising from a cross metathesis reaction. Only one set of signals was observed in the 1H and 13C NMR spectra, which suggested that all three compounds were symmetrical. This also implied each dimeric compound had the same stereochemistry on each of its two double bonds. Detailed assignment of the exact stereochemistry for the three isomers of the dimers (142a-c) obtained proved to be elusive. However, the three isomeric products, dimer (142a), (142b) and (142c) were readily separated and purified by flash chromatography and exhibited unique 1H and 13C NMR spectra. Dimer (142a) was obtained as fine white needes in 15% yield. A molecular ion at m/z 480.27238 in the high resolution mass spectrum established the molecular formula 1 C26H40O8. The H NMR spectrum was assigned with the aid of 2D COSY and HSQC experiments. 5-H resonated as a triplet at δH 3.05 with J5,4 3.4 Hz. Two broad singlets

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

were observed for 3-H at δH 3.32 and the vinylic protons resonated at δH 5.89. 4-Hax resonated as a double double doublet at δH1.85 with J4ax,4eq 15.0, J4ax,3 3.4 and

J4ax,5 3.4 Hz confirming that both 5-H and 3-H were in an equatorial orientation. The 13 C NMR spectrum exhibited resonances at δC 71.3 and δC 76.2, which are characteristic of C-3 and C-5 respectively. The quaternary spirocarbon C-6 resonated at δC 96.3. The vinylic carbons were observed at δC 129.0 and δC 129.2. Attempted crystallisation of dimer (142a) using several solvent systems did not afford crystals suitable for X-ray crystallographic analysis. Dimer (142b) was obtained as a viscous oil in 40% yield with a molecular ion at m/z 480.27275 which supported the molecular formula C26H40O8. 5-H resonated as a triplet at δH 3.07 with J5,4 3.4 Hz suggesting that 5-H was in an equatorial orientation.

4-Hax resonated as a double double doublet at δH 1.86 with J4ax,4eq 14.9, J4ax,3 3.4 and

J4ax,5 3.4 Hz confirming that 5-H adopts an equatorial position. The vinylic protons 2’-H 13 and 3’-H resonated as a triplet at δH 5.89 with JCH,OCH2 2.8 Hz. The C NMR spectrum exhibited resonances at δC 71.0 and δC 76.4 assigned to C-3 and C-5 respectively and the vinylic carbons resonated at δC 129.1 and δC 129.2. Dimer (142c) was synthesised in 10% yield as a viscous oil and the high resolution mass spectrum exhibited a molecular ion at m/z 480.27248 establishing the 1 molecular formula C26H40O8. In the H NMR spectrum 4-Hax resonated slightly further downfield at δH 1.95 with J4ax,4eq 14.6, J4ax,3 3.7 and J4ax,5 3.7 Hz compared to the analogous protons in dimers (142a) and (142b). 4-Heq was observed as multiplet at

δH 2.05 and 3-H resonated as a multiplet at δH 2.05. 5-H was observed as a triplet at

δH 3.09 with J5,4 3.7 Hz. The vinylic protons resonated as triplets at δH 5.89 and δH 5.91 13 with JCH,OCH2 2.6 Hz. The C NMR spectrum exhibited resonances at δC 70.7 and δC 76.1 representing C-3 and C-5 respectively and the quaternary spirocarbon C-6 resonated at

δC 96.9. The vinylic carbons resonated at δC 128.9 and δC 128.4. A summary of the 1H NMR data is presented in Table 1.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O 5 O 6 O 142a 4 7 1 142b O 3' 2' O O O 3 142c 4' 1'

Dimer 3-H 4-Hax 5-H OCH2CH=CH CH=CH

142a 3.32, br s 1.85, ddd, J4ax,4eq 15.0, 3.05, t, J5,4 3.4 3.86-4.06, m 5.89, br s

J4ax,5 3.4, J4ax,3 3.4

142b 3.31, br s 1.86, ddd, J4ax,4eq 14.9, 3.07, t, J5,4 3.4 3.84-4.07, m 5.89, t, J 2.8

J4ax,5 3.4, J4ax,3 3.4

142c 3.35, m 1.95, ddd, J4ax,4eq 14.6, 3.09, t, J5,4 3.7 3.84-4.10, m 5.89, t, J 2.5

J4ax,5 3.7, J4ax,3 3.7 5.91, t, J 2.5

Table 1: Selected 1H NMR Chemical Shifts and Coupling Constants for Dimers (142a), (142b) and (142c).

As mentioned previously, the NMR data indicates that all three dimers are symmetrical. It is proposed there are two possible forms of symmetry inherent in dimers (142a),

(142b) and (142c). Structure (143) shows a Ci centre of symmetry while structure (144) shows a C2 axis of symmetry. Each symmetrical conformation is also capable of having both unsaturated bonds in either E or Z orientations.

O O

OO OO

Ci Centre of Symmetry C2 Axis of Symmetry

OO OO

O O

143 144

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.4 Synthesis of the Spiroacetal Thiacrown Ethers via Path B

Disappointed by the inability to effect reaction of the spiroacetal diol (50) with

β-chloroethyl sulfide (104) via Path A, our attention next focused on the alternative disconnection summarised by Path B in Scheme 44.

O O Path B O O + n HS S SH OO OO 99, n = 3 122, n = 4 SS OR RO 123, n = 5 S 146, R = Ts n 147, R = Ms 55, n = 3 56, n = 4 57, n = 5 Scheme 44

Adoption of Path B in the retrosynthetic analysis involved the use of dithiols (99), (122) and (123) as nucleophiles in the displacement of two leaving groups attached to the present on the C-3 and C-5 on the spiroacetal ring system. The thiols (99), (122) and (123) had previously been prepared using the method by Edema et al122 (Scheme 35). It was envisaged that spiroacetal diol (145) could be synthesised from spiroacetal diol (50) and in turn the hydroxyl groups of (145) could be tosylated or mesylated to yield (146) or (147) respectively (Scheme 45).

O O O O O O O O

OO OH OH OO OO 50 SS OH HO OR RO 145 146, R = Ts S 147, R = Ms n 55, n = 3 56, n = 4 57, n = 5 Scheme 45

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.4.1 Synthesis of Spiroacetal Diol (145)

In order to pursue the synthetic strategy outlined in Scheme 45, initial synthesis of the spiroacetal diol (145) was required starting from the previous diol (50). In the synthesis of diol (145) two different approaches were investigated. It was envisaged that reduction of a diester (149) (Scheme 46) or ozonolysis of the bisallyl ether (133) (Scheme 47) would provide effective methods for the synthesis of diol (145). Several procedures were evaluated for the synthesis of the diester (Table 2). Treatment of the spiroacetal diol (50) with sodium hydride for 30 min in THF at 0 °C followed by the addition of tert-butyl bromoacetate (148) gave the desired diester in 52% yield after work-up and purification by flash chromatography.

O O O 7 O i 6 1 O + O O Br O 5 3 OH OH OO OO 50 148 O O OO OH HO 145 149

Reagents and Conditions: (i) NaH, THF, 0 °C, 30 min, then 18-Crown-6 and 148 (2.0 equiv), 0 °C→room temp, 18 h, 149 (52%) Scheme 46

Entry Base Temperature Reagent Product

(°C)

1 NaH Reflux BrCH2CO2Et Complex mixture

2 n-BuLi -78 BrCH2CO2Et Recovered starting material

3 NaH 0 BrCH2CO2Et Recovered starting material t 4 NaH 0 BrCH2CO2 Bu Diester (149) 52% yield

Table 2: Reactions Conditions Used in the Synthesis of Diester (149)

The spiroacetal diester (146) analysed correctly for C21H36O8 with a molecular ion at m/z 416.24092 being observed in the high resolution mass spectrum. The 1H NMR spectrum exhibited an eighteen-proton singlet δH 1.46 assigned to the methyl groups of the tert-butyl group. 5-H resonated as a triplet at δH 3.29 with J5,4 4.6 Hz. 3-H resonated

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

as a double double double doublet at δH 3.53 with J3,2ax 3.5, J3,2eq 3.5, J3,4ax 3.5 and 13 J3,4eq 3.5 Hz. The C NMR spectrum exhibited resonances at δC 71.9 and δC 76.8 assigned to C-3 and C-5. The quaternary spirocentre C-6 resonated at δC 97.4. The two carbonyl carbons resonated very close to each other at δC 170.001 and δC 170.011. Due to the moderate yield of the diester (149) obtained, the alternative ozonolysis method was investigated to provide diol (145). Ozone was bubbled through a solution of the bisallyl ether (133) in methanol until a pale blue colour persisted (ca 10-15 min) indicating that an excess of ozone was present (Scheme 47). The intermediate ozonide was reduced using sodium borohydride. Subsequent work-up and purification by flash chromatography afforded diol (145) in 75% yield.

O O 7

6 1 O i O 5 3 OO OO 1' 1'

2' 2' OH HO 133 145

Reagents and Conditions: (i) ozone, MeOH, -78 °C, 15 min

then NaBH4, room temp, 18 h, 145 (75%).

Scheme 47

A molecular ion m/z at 276.15725 in the high resolution mass spectrum supported the molecular formula C13H24O6. The infrared spectrum exhibited a broad absorbance at 3628-3290 cm-1 supporting the formation of a diol group. A double double doublet at

δH 1.98 with J4ax,4eq 15.1, J4ax,3 3.6 and J4ax,5 3.6 Hz was assigned to the axial proton at

C-4. The equatorial proton at C-4 also resonated as a multiplet at δH 2.11-2.29 together with 11-Heq. 5-H resonated as a triplet at δH 3.11 with J5,4 3.6 Hz. The equatorial 3-H 13 resonated as a multiplet at δH 3.37-3.41. The C NMR spectrum exhibited resonances at

δC 26.0 assigned to C-4, δC 72.2 assigned to C-3 and δC 77.2 assigned to C-5. The methine carbons bearing the OH groups resonated at δC 70.9 and 71.3 and the quaternary spirocarbon C-6 resonated at δC 96.1.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.4.2 Ditosylation of Spiroacetal Diol (145)

The synthesis of the ditosylate (146) was next attempted. The tosylate group was chosen over the mesylate because of its greater stability. The synthesis was achieved by the addition of n-butyllithium to a solution of diol (145) in THF at –78 °C followed by the addition of p-toluenesulfonyl chloride after 30 min. The reaction mixture was then warmed to room temperature and the solution left to stir for 16 h (Scheme 48). Subsequent work-up and purification by flash chromatography yielded the ditosylate (146) in 90% yield.

O O7

6 1 O i O 5 3 OO OO 1' 1'

2' 2' OH HO OTs TsO 145 146

Reagents and Conditions: (i) n-BuLi, THF, -78 °C, 30 min, then TsCl, 30 min, room temp, 16 h, 143 (90%)

Scheme 48

Spiroacetal ditosylate (146) analysed correctly for C27H36O10S2 with a molecular ion m/z at 585.18336 being observed in the high resolution mass spectrum. The 1H NMR spectrum exhibited a six-proton singlet δH 2.44 assigned to the methyl groups of the tosyl moiety. The proton at C-5 resonated as a triplet at δH 2.44 with J5,4 3.9 Hz and the equatorial proton at C-3 resonated as a double double double doublet at δH 3.34 with

J3,2ax 3.2, J3,2eq 3.2, J3,4ax 3.2 and J3,4eq 3.2 Hz confirming, that 3-H and 5-H were both 13 equatorial. The aromatic protons were observed as a multiplet at δH 7.31-7.79. The C

NMR spectrum exhibited resonances at δC 69.5 and δC 69.6 assigned to C-1’. C-3 and

C-5 resonated at δC 72.2 and δC 77.2 respectively. A quaternary resonance at δC 96.5 confirmed the presence of the spiroacetal carbon.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.4.3 Synthesis of Spiroacetal Thiacrown Ethers (55), (56) and (57)

Thiacrown ether (55)

Having successfully prepared ditosylate (146) and the thiols (99), (122), and (123), the synthesis of the thiacrown ethers was next undertaken. It was decided to use the method reported by Kellogg et al108,136 for related systems. Thus, ditosylate (146) and 2-mercaptoethyl sulfide (99) were added simultaneously from separate addition funnels to a suspension of Cs2CO3 in DMF over 2-3 h at 60 °C (Scheme 49). The reaction was allowed to proceed for 18 h at 60 °C after the final addition. Subsequent work-up and purification by flash chromatography afforded the first thiacrown ether (55) in 86% yield.

6' 5' O O 1' 4' i 182' 17 16 O + n 3' O HS S SH 1 19 15 n = 1 OO 2 OO14 99 3 13 4 12 OTs TsO 5 SS11

146 6 S 10 8 7 9 55

Reagents and Conditions: (i) Cs2CO3, DMF, 60 °C, 99 and 146 in DMF added over 2.5 h, 55 (86%)

Scheme 49

A molecular ion at m/z 394.13104 in the high resolution mass spectrum supported 1 the molecular formula C17H30O4S3. The H NMR spectrum was assigned with the aid of 2D COSY and HMQC experiments. Characteristically, 3’-Heq resonated as a double double doublet at δH 1.26 with J3’ax,3’eq 13.6, J3’ax,4’ax 13.6 and J3’ax,4’eq 4.4 Hz. 1-H resonated as a triplet at δH 3.09 with J1,19 3.4 Hz suggesting that 1-H is equatorial. 15-H was observed as a multiplet at δH 3.37 however the double double doublet at δH 1.91 with

J19ax,19eq 15.2, J19ax,1 3.4 and J19ax,15 3.4 Hz assigned to 19-Hax established that 15-H is also equatorial. 3-HA resonated as a double double doublet at δH 3.41 with Jgem 6.8,

J3A,4A 9.0 and J3A,4B 9.0 Hz. It is unusual for geminal coupling to be smaller than vicinal coupling however it was reasoned that the coupling constants of the same magnitude (9.0 Hz) are representative of protons on the same carbon. The 13C NMR spectrum

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

exhibited resonances at δC 31.4, 31.8, 32.3, 32.4, 33.2 and 33.5 assigned to the six CH2S carbons. C-3 showed a resonance at δC 70.3, whilst C-15 and C-1 resonated at δC 72.0 and δC 77.2 respectively. The characteristic resonance at δC 96.6 was assigned to the spirocentre.

Thiacrown Ether (56)

The synthesis of the larger 19-thiacrown-6 (56) analogue was next undertaken using similar methodology to that described for the synthesis of thiacrown (55). Spiroacetal thiacrown ether (56) was prepared in 68% yield by the reaction of ditosylate

(146) with dithiol (122) in the presence of cesium carbonate in DMF at 60 °C for 18 h (Scheme 50). Thiacrown ether (56) was isolated as a colourless oil after purification by flash chromatography.

6' 5' O 1' 4' O

2' 21' 20 19 i 3' O + n O HS S SH 1 22 18

n = 2 2 17 OO 3 O O 16 122 4 15

OTs TsO 5 S S 14 146 6 13 S S 12 7 8 11

910 56

Reagents and Conditions: (i) Cs2CO3, DMF, 60 °C, 122 and 146 in DMF added over 2.5 h, 56 (68%)

Scheme 50

A molecular ion at m/z 454.13397 in the high resolution mass spectrum supported 1 the molecular formula C19H34O4S4. In the H NMR spectrum 1-H resonated as a triplet at

δH 3.09 with J1,22 3.7 Hz. 18-H resonated as a double double double doublet at δH 3.37 with J18eq,22ax 3.7, J18eq,22eq 3.7, J18eq,19ax 3.7 and J18eq,19eq 3.7 Hz, suggesting that 1-H and

18-H were in equatorial orientations. A double double doublet at δH 1.97 with

J22ax,22eq 14.8, J22ax,18 3.7 and J22ax,1 3.7 Hz was assigned to the axial proton 22-Hax.

22-Heq resonated as a double double double doublet at δH 2.07 with J22eq,22ax 14.8, 13 J22eq,1 3.7, J22eq,18 3.7 and J22eq,19eq 1.9 Hz. The C NMR spectrum exhibited resonances

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

for C-1 and C-18 at δC 77.2 and δC 71.9 respectively whilst the characteristic resonance at

δC 96.5 indicated the presence of the spirocentre.

Thiacrown Ether (57)

Finally, the synthesis of the 22-thiacrown-7 (57) was carried out using a similar procedure to that used for the preparation of thiacrown ethers (55) and (56) (Scheme 51). Thiacrown ether (57) was obtained by treatment of ditosylate (146) with dithiol (123) in the presence of cesium carbonate in DMF at 60 °C. 22-Thiacrown-7 (57) was isolated as a pale yellow oil in 64% yield after purification by flash chromatography.

6' 5' O O 1' 4' 2' 24 23 22 O n i 3' + HS S SH O 1 25 21 n = 3 OO 2 OO20 123 3 19

4 18 OTs TsO 5 SS17

6 146 16 7 15

8 S S 14 10 S 12 9 13 11 57

Reagents and Conditions: (i) Cs2CO3, DMF, 60 °C, 123 and 146 in DMF added over 2.5 h, 57 (64%)

Scheme 51

Spiroacetal thiacrown ether (57) analysed correctly for C21H38O4S5 with a molecular ion m/z 514.13764 in the high resolution mass spectrum supporting this 1 molecular formula. In the H NMR spectrum 1-H resonated as a triplet at δH 3.08 with

J1,25 3.8 Hz . 25-Hax resonated as a double double doublet at δH 1.95 with J25ax,25eq 14.8,

J25ax,1 3.8 and J25ax,21 3.8 Hz, confirming that 1-H and 21-H adopt equatorial positions. A 13 twenty proton multiplet at δH 2.69-2.84 was assigned to the CH2S protons. The C NMR spectrum exhibited resonances at δC 71.9 and δC 77.1 assigned to C-21 and C-1 respectively and the quaternary spirocentre resonated at δC 96.5.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.5 Synthesis of the Spiroacetal Azacrown Ethers

The synthesis of azacrown ethers (58), (59) and (60) proved to be much more difficult than anticipated. The choice of method was dependent on the sensitivity of the spiroacetal functionality to the corresponding reaction conditions. Having established previously that the synthesis of the thiacrown ethers via Path A was not successful it was envisaged that the synthesis of azacrown ethers could be carried out following the retrosynthetic outline for Path B (Scheme 24). The reaction between the appropriate spiroacetal compound and diamine was predicted to yield the desired azacrown ethers (58), (59) and (60).

O O O

OO O O OO

NH NH NH HN HN NH H N NH HN 58 HN NH 59 N H 60

It has been reported that the addition of unprotected triamine (151) to a dialdehyde without the use of a template formed the desired azacrown ether after reduction of the intermediate diimine.137 It was believed that this method would yield the unsaturated azacrown ethers (152), (153) and (154) which would then be reduced to give (58), (59) and (60) respectively (Scheme 52). The spiroacetal dialdehyde (150) required for this reaction was prepared from bisallyl ether (133) via an ozonolysis reaction. The sulfonamide method is one of the most commonly used synthetic methodologies in the synthesis of azacrown ethers. The reaction between the spiroacetal ditosylate (146) and the protected polyamines, such as (155) and (156), was envisaged to afford the spiroacetal azacrown ethers (58), (59) and (60) after deprotection (Scheme 53).

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

O O O O 150 146 O O O O

OO OO OO OO

OTs TsO NR NR OOH H NN R + H + N N n n n n RHN N NHR H2N N NH2 H 154, n = 1 R 159, n = 1 151, n = 1 155, n = 2 160, n = 2 152, n = 2 157, R = Ts 156, n = 3 158, R = Ns 161, n = 3 153, n = 3 Scheme 52 Scheme 53

3.5.1 Attempted Synthesis of Spiroacetal Azacrown Ethers (58), (59) and (60) via formation of Imines from Aldehyde (150)

Starting with the bisallyl ether (133) it was envisaged that a reductive ozonolysis using dimethyl sulfide would furnish spiroacetal dialdehyde (150) required for the subsequent reaction with the diamines. In a protic solvent, such as methanol, ozonolysis yields the aldehyde and a hydroperoxide by-product. Dimethyl sulfide readily reduces the hydroperoxide at low temperatures but not the aldehyde to form dimethyl sulfoxide, which is easily removed by purification.138 Thus, treatment of the bisallyl ether (133) in methanol with ozone at –78 °C followed by the addition of dimethyl sulfide afforded the dialdehyde (150) in 86% yield as an oil (Scheme 54).

8 9 O O 7 10 6 1 O i 11 O 5 4 3 OO OO

O HHO 133 150

Reagents and Conditions: (i) ozone, MeOH, -78 °C,

15 min then (CH3)2S, room temp, 18 h, 150 (86%)

Scheme 54

Spiroacetal dialdehyde (150) analysed correctly for C13H20O6 with a molecular ion m/z at 272.12576 in the high resolution mass spectrum. The infrared spectrum

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers exhibited a strong absorbance at 1733 cm-1 supporting the formation of the dialdehyde.

The aldehyde protons resonated as a multiplet at δH 9.68-9.74. The characteristic axial proton 11-Hax resonated as a double double doublet at δH 1.43 with J11ax,11eq 13.5,

J11ax,10ax 13.5 and J11ax,10eq 4.8. 5-H resonated as a triplet at δH 3.21 with J5,4 3.8 Hz, confirming that 5-H is in the equatorial orientation. 3-H resonated as a multiplet at

δH 3.44-3.46 and 4-CH2 also resonated as a multiplet with 11-Heq at δH 2.04-2.22. The 13 C NMR spectrum exhibited characteristic resonances at δC 72.7 and δC 76.8 assigned to

C-3 and C-5 respectively. The quaternary spirocarbon resonated at δC 96.6 and the two aldehyde carbonyl groups resonated at δC 201.1 and 201.3. Having successfully synthesised dialdehyde (150), its subsequent condensation reaction with diamine (151) was investigated (Scheme 55). Dialdehyde (150) and commercially available diethylenetriamine (151) were added to a solution of refluxing benzene. The reaction was heated for 18 h with removal of the water by azeotropic distillation. Disappointingly, this procedure did not afford the desired unsaturated crown ether (154) and only polymeric material was obtained.

O O i O + O H2NNN H2 H OO OO 151

O HHO NN H 150 N 154 Reagents and Conditions: (i) benzene, reflux

Scheme 55

3.5.2 The Sulfonamide Method

The sulfonamide method was considered to be a more effective route for the synthesis of the spiroacetal azacrown ethers (58), (59) and (60). The first consideration was the protection of the nitrogen atoms in the triamine compound. The choice of protecting group was primarily based on the conditions required for the subsequent removal of the protecting groups because of the sensitivity of the spiroacetal functionality. Initially the tosyl protected triamine (157) was used in the reaction with

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers spiroacetal ditosylate (Table 3, entries 1 and 2). However, due to the unsuccessful attempt to effect cyclisation and the usually harsh conditions required to remove the tosyl group (namely, use of strong acids) other protecting groups were investigated. It was thought that the o-nosylate group utilised by Fukuyama et al87 would be more successful. Thus, triamine (151) was treated with 2-nitrobenzenesulfonyl chloride in dichloromethane at 0 °C. The reaction was allowed to proceed for 18 h at room temperature. Subsequent work-up and purification afforded the desired o-nosyl protected triamine (158) in 80% yield.

O2N O 1 2 O 3 S NH N HN S O O S O O NO2 NO2

158

A protonated molecular ion at m/z 659.05478 in the high resolution mass 1 spectrum supported the molecular formula C22H23N6O12S3. In the H NMR spectrum the

NH resonated as a triplet at δH 5.72 with JNH,CH2 6.1 Hz. The CH2NH group resonated as a multiplet at δH 3.30-3.37. The CH2N group resonated as triplet at δH 3.54 with 13 JCH2N,CH2NH 6.1 Hz. The C NMR spectrum exhibited resonances at δC 42.3 and δC 49.0, which were assigned to CH2NH and CH2N respectively. The reaction between the Ns-protected triamine (158) and the spiroacetal ditosylate (146) was next attempted. The ideal conditions for sulfonamide reactions include using cesium or potassium carbonate in dimethylformamide at 20-50 °C. The reaction of ditosylate (146) with trinosylate (158) was attempted using reaction conditions successfully utilised by several other groups (Table 3). The slow (ca 2 h) addition of the spiroacetal ditosylate (146) to a suspension of (158) and NaH in THF eventually afforded the Ns-protected azacrown ether (159), albeit in 27% yield (Scheme 56).

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

6' 5' O O 1' 4' i 182' 17 16 O n 3' O HN N NH Ns Ns Ns 1 19 15 OO n = 1 2 OO14 158 3 13 Ns = O 4 12

OTs TsO S 5 NNNs 11 Ns 8 O 6 146 7 N 9 10 NO2 Ns 159 Reagents and Conditions: (i) NaH, reflux, 30 min then 146 over 3 h, 18 h, 159 (27%)

Scheme 56

Entry Base Temperature Solvent Product

(°C) 1 NaHa Reflux THF Complex Mixture a 2 K2CO3 60 DMF Complex Mixture

3 Cs2CO3 RT DMF Recovered Starting Material

4 Cs2CO3 60 DMF Complex Mixture

5 Cs2CO3 40 DMF Starting Material and Mixture 6 NaH 45 DMF Complex Mixture

7 Cs2CO3 60 CH3CN Starting Material 8 NaH Reflux THF Protected Azacrown (159) 27% yield

Table 3: Reaction Conditions Used in the Synthesis of Protected Azacrown (159)

The high resolution mass spectrum for the protected spiroacetal azacrown ether (159) exhibited a protonated molecular ion at m/z 899.18910 consistent with the 1 molecular formula C35H43N6O16S3. The H NMR spectrum was assigned with the aid of

2D COSY and HMQC experiments. 1-H resonated as triplet at δH 3.03 with J1,19 3.0 Hz, affirming that 1-H adopted an equatorial position. 15-H resonated as a broad singlet at

δH 3.24. 19-CH2 resonated as a multiplet at δH 2.03. The equatorial 3’-H resonated as a double double doublet at δH 1.96 with J3’eq,3’ax 13.5, J3’eq,4’ax 2.1 and J3’eq,4’eq 2.1 Hz. The 13 C NMR spectrum exhibited resonances at δC 48.4-51.0 assigned to the CH2N carbons.

a Reaction carried out with Ts-protected triamine (157) 72

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

The spirocarbon resonated characteristically at δC 96.1 whilst C-15 and C-1 resonated at

δC 72.0 and δC 77.5 respectively. The deprotection of (159) was carried out using thiophenol and potassium carbonate in DMF at room temperature for 18 h to afford the spiroacetal azacrown ether (58) in 84% yield (Scheme 57). It is believed that the removal of the 2-nitrosulfonyl groups using potassium carbonate and thiophenol is achieved via the formation of the Meisenheimer complex (162).87

6' 5' O O O 1' 4' i 2' 18 17 16 O O 3' O 1 19 15 OO OO 2 OO14 3 13

4 12

NNNs Ns NNNs Ns 5 NH11 NH H 6 N N 7 N 9 10 Ns 8 O SO 159 58 PhS NO2

162

Reagents and Conditions: (i) K2CO3, PhSH, DMF, 58 (84%)

Scheme 57

A protonated molecular ion at m/z 344.25423 in the high resolution mass 1 spectrum supported the molecular formula C17H34N3O4. The H NMR spectrum was assigned with the aid of 2D COSY experiment. 1-H resonated as triplet at δH 3.04 with

J1,19 3.2 Hz whilst 15-H resonated as a broad singlet at δH 3.29. The CH2N protons 13 resonated as a multiplet at δH 2.63-2.89. A C NMR spectrum for this compound could not be obtained as there was insufficient material to give a satisfactory signal to noise ratio. Disappointingly, subsequent attempts to synthesise the protected spiroacetal azacrown ether (159) proved difficult to repeat. Attempts to vary the conditions, the concentration of the reactants, addition time and nature of the bases used, did not lead to any improvement in the reaction. In light of the difficulties to synthesise azacrown ether (58) efficiently, subsequent attempts to prepare the homologues azacrown ethers (59) and (60) were not undertaken.

DISCUSSION: Synthesis of Spiroacetal Thiacrown and Azacrown Ethers

3.6 Summary

In summary, the synthesis of the target spiroacetal thiacrown ethers (55), (56), and (57) was successfully achieved following the retrosynthetic outline in Scheme 24 via Path B. Reaction between the requisite dithiols (99), (122) and (123) and the spiroacetal ditosylate (146) afforded the desired thiacrown ethers in good yields. The diol (145) precursor to the ditosylate (146) was obtained via a reductive ozonolysis of the bisallyl ether (133), which in turn was synthesised from the (±)-1,7-dioxaspiro[5.5]undec-4-ene (114). During attempts to synthesise thiacrown ethers (55), (56) and (57) a novel set of spiroacetal dimers (142a), (142b) and (142c) were synthesised by a cross metathesis reaction of the bisallyl ether (133). Disappointingly, the synthesis of the desired spiroacetal azacrown ethers (58), (59) and (60) was achieved with limited success. The synthesis of (58) was carried out by the reaction of the protected diamine (158) and the spiroacetal ditosylate (146) to afford the Ns-protected azacrown ether (159). Subsequent removal of the protecting groups afforded the azacrown ether (58).

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

4.0 Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

Having successfully synthesised thiacrown ethers (55), (56) and (57) attention then turned to their binding ability. Crown compounds can serve not only as primary receptors for simple organic and inorganic cations, anions and neutral molecules but also as second-sphere ligands for metal complexes.139 These different types of interactions can be investigated using various spectroscopic methods such as UV-Visible spectrometry and NMR spectroscopy.

O O O

OO O O OO

SSS S SS S S S 55 S S 56 S 57

4.0.1 Crown Ethers As Primary Ligands

The complex between a crown compound and an inorganic salt is formed by ion- dipole interactions between the cation and the donor atoms in the polyether ring.63 The complexation ability of crown compounds and the stability of the resulting complex is dependent on several factors, including the type of heteroatom in the ring and the cavity size and shape. This is particularly true for the oxa- and azacrown compounds, which have been shown to coordinate the cation within the crown ether cavity. In these cases, the lone pairs of electrons are directed toward the inside (endodentate) of the cavity and the donor atoms are located at an equal distance from the cation. Conversely, thiacrown compounds tend to bridge metal ions rather than chelate to them because the sulfur atoms have been shown to exist outside (exodentate) the cavity in many cases (Section 2.1).68,69 The design of biomimetic macromolecules has led to synthesis of many crown compounds capable of transporting ions through cell membranes.140 Crown ethers play a pivotal role in modifying the properties and behaviour of the metal compounds with which they complex. For example, ions do not easily diffuse through the lipid bilayers that surround cells, yet the transport of cations across cellular membranes is crucial in

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

such physiological processes as nerve impulse transmission, the control of muscular function and protein biosynthesis.141 One way to achieve permeability is through the use of ion carriers, such as crown ethers. Membrane cation transport has also been applied in the selective separation of metal ions.142 Figure 10 illustrates the proton-driven transport of metal ions through a membrane impregnated with water-immiscible solvent containing the crown ether ligand.142b

M OOC12H25 NO2 M

O HH OO M

OOC12H25 NO2 Aqueous Source Phase Membrane Phase Aqueous Receiving Phase Figure 10

4.0.2 Second-Sphere Coordination

The idea of second-sphere coordination was first postulated by Alfred Werner143 in 1912 to explain a number of phenomena he observed through research with metal complexes. In the ensuing years, advances in optical, spectroscopic and crystallographic techniques have revealed a much wider range of phenomena that can only be explained in terms of second-sphere coordination.144 Second-sphere coordination can generally be described as the “non-covalent bonding of chemical entities to the first coordination sphere of a transition metal complex.”145 The non-covalent interactions responsible include electrostatic interactions, hydrogen bonding, charge transfer and hydrophobic interactions.144,146 Second-sphere coordination can significantly affect the electronic state of the metal centre and result in a modification of the properties of the metal complexes and macrocycles in terms of photochemical, magnetic and electrochemical characteristics.144,146 For example, the photochemical properties of the second-sphere adduct shown in Figure 11 are altered from those of the free compounds.147 Dibenzo-30- crown-10 and [Pt(bpy)(NH3)2](PF6)2 interact through π-π stacking of the pyridine rings and the aromatic units in the crown ether and hydrogen bonding of the amine groups to the crown ether oxygen atoms. The absorption spectrum of the adduct showed a strong

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

charge transfer band as a result of the π-π stacking interaction. In addition, the luminescence properties were altered, showing a new, broad, low-intensity emission band.

N N + O O Pt2 O O O

O NH3 O N H H H O O O

Figure 11

The adducts formed through second-sphere interactions have found applications in a variety of areas including the possible treatment of cancer. For example, the increased solubility of the neutral anti-cancer platinum complex, carboplatin (163), in aqueous solutions of α-cyclodextrin (12) has lead to its possible application in cancer chemotherapy.145 It has been shown that the cyclobutane ring of carboplatin (163) is directed inside the ring while the amine ligands are positioned over one of the glucopyranosyl residues. This allows for the formation of two hydrogen bonds with the secondary hydroxyl groups of the glucopyranosyl residues (Figure 12).

O OR O NH3 O Pt HO NH OR O 3 n = 6, R = H O n O 163 12

H3N NH3 HO OH Pt

Figure 12

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

4.1 Binding Studies

4.1.1 Determination of Association Constants Using Picrate Salts

Conditional stability constants for thiacrown ethers (55), (56), (57) and 18-S-6 (100) were determined using the ultraviolet spectroscopic method devised by Cram et al.148 This method relies on the fact that most crown ethers are insoluble in water and most metal salts are insoluble in organic solvents. The procedure involves the partitioning of the crown ether (host) and the metal picrate (guest) between chloroform and water. The molar ratio (R) of host to guest in the chloroform layer is measured by ultraviolet spectroscopy. The association constant (Ka) for the equilibrium (Equation 1) and the free energy of complexation (∆G) (Equation 2) can then be determined from this ratio.

K Host Guesta Host.Guest Complex Equation 1 org + aq organic CDCl3

∆G = -RT ln(Ka) Equation 2

The affinity of thiacrowns (55), (56) and (57) for Li+, Na+, K+, Cs+, Co2+, Cd2+, Ag+ and Pb2+ cations was determined using this method. The alkali metals are considered ‘hard’ ions, silver and cadmium are classified as ‘soft’ ions while lead and cobalt are classed as ‘intermediate’ ions based on Pearson’s classification of hard-soft acid-base theory (HSAB).149 The results are summarised in Table 4. Graphical representations are depicted in Figures 14 and 15.

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

-1 3 ° -1 Host Cation Ka (M x 10 ) ∆G (kJ mol ) Li+ 0.05 -9.37 Na+ 0.21 -13.08 K+ 0.40 -14.71 1,4,7,10,13,16- Cs+ 8.95 -22.32 Hexathiacyclooctadecane Co2+ 0.05 -9.82 (18-Thiacrown-6) Cd2+ 8.25 -22.12 Ag+ 3801.27a -37.16 Pb2+ 1.33 -17.64

Li+ 0.11 -11.52 [1S*,15R*,18S*]- Na+ 0.41 -14.78 Spiro[2,14,17-trioxa- K+ 0.32 -14.16 5,8,11-trithiabicyclo Cs+ 0.51 -15.27 [13.3.1]nonadecane- Co2+ 0.84 -16.51 18,2’-tetrahydropyran] Cd2+ 1.19 -17.39 Ag+ 3683.06 -37.08 Pb2+ 35.71 -25.71

Li+ 0.24 -13.49 [1S*,18R*,21S*]- Na+ 1.83 -18.43 Spiro[2,17,20-trioxa- K+ 0.45 -15.00 5,8,11,14-trithiabicyclo Cs+ 0.66 -15.93 [16.3.1]docosane-21,2’- Co2+ 1.10 -17.18 tetrahydropyran] Cd2+ 1.15 -17.28 Ag+ 1576.61 -35.00 Pb2+ 8.93 -22.31

Li+ 0.09 -11.08 [1S*,21R*,24S*]- Na+ 0.62 -15.78 Spiro[2,20,23-trioxa- K+ 0.70 -16.05 5,8,11,14,17- Cs+ 0.83 -16.49 pentathiabicyclo Co2+ 0.38 -14.60 [19.3.1]pentacosane- Cd2+ 1.30 -17.59 24,2’-tetrahydropyran] Ag+ 922.85 -33.69

Pb2+ 67.71 -27.28

Table 4: Association Constants for Spiroacetal Thiacrown Ethers (55), (56) and (57)

a 3 x 10-3 M solution of silver picrate and 0.075 M solution of 18-S-6 (100) were used in the experiment because at higher concentrations of silver picrate a precipitate formed. 79

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

Association Constants for Spiroacetal Thiacrown Ethers (55), (56) and (57) 4000 3500 3000

3 2500 2000 1500 Ka x 10 1000 500 0 Li Na K CsCoCdAgPb Picrate Salts 16-Crown-5 (55) 19-Crown-6 (56) 22-Crown-7 (57) 18-S-6

Figure 14: Association Constants

Association Constants for Spiroacetal Thiacrown Ethers (55), (56) and (57) 80 70 60

3 50 40 30 Ka x 10 20 10 0 Li Na K Cs Co Cd Pb Picrate Salts 16-Crown-5 (55) 19-Crown-6 (56) 22-Crown-7 (57) 18-S-6

Figure 15: Association Constants (Expanded Region)

An initial comparison between the association constants of the spiroacetal crown ethers (53a-c) and the spiroacetal thiacrown ethers (55), (56), (57) revealed that the replacement of oxygen for sulfur donor atoms significantly decreased the ability of the crown ethers to complex alkali metals. This is in accordance with the HSAB theory. The spiroacetal crown ethers (55), (56), (57) do show a slightly greater ability to bind the alkali metals compared to the 18-S-6 (100). A possible explanation for this is the fact that 80

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

the spiroacetal thiacrown ethers possess two oxygen atoms, which have an affinity for alkali metals. The thiacrown ethers exhibited an expected affinity for the heavy metals, particularly silver. It is believed that electrostatic interactions do not predominate in the complexation of silver by thiacrown macrocycles.150 The interaction is thought to be largely covalent based on the Ag-S bond length.49,151 Spiroacetal thiacrown ethers (55), (56) and (57) showed a lower binding affinity for silver than 18-S-6 (100). The difference in the binding of silver by thiacrown (55) and 18-S-6 (100) is only small. Interestingly, the difference in binding between thiacrown (56) and 18-S-6 (100) is much larger even though (56) can be regarded as having a similar cavity size to 18-S-6 (100). The most likely explanations for the reduced binding of silver are the conformation of thiacrown (56) and the presence of the two oxygen atoms. The spiroacetal thiacrown ethers (55), (56), (57) also showed an affinity toward lead, however the stability of complexation was much lower than that for the silver. The complexing ability was reduced by a factor of approximately 150. The association constants for cadmium and cobalt were low indicating a weak interaction. The large difference in the complexing behaviour of thiacrown ethers (55), (56), (57) with silver compared to that of the alkali metals, cobalt and cadmium can be explained in terms of the HSAB theory. Sulfur donor atoms are considered soft bases and therefore prefer soft acids. Cadmium is classified as a soft acid however it has a low value of softness.152 The large difference in the complexation of silver over lead cannot be completely explained using the HSAB theory. Lead is considered a borderline acid and this may partly explain the lower association constant. However, lead and silver are both thiophilic metals, that is, they have an affinity for sulfur donor groups.153 One possible explanation is the difference in the oxidation states. Generally, metal ions with lower oxidation states exert less electrostatic attraction, however thiacrown ethers are known to stabilise lower 154 oxidation states of metals due to their π-acidity. This results in stronger binding of metals with lower oxidation states (eg. Ag+) than those with higher oxidation states (eg. Pb2+) and in this case could indicate that there may be some covalent (polar covalent) bonding taking place. Another possible reason for the difference in complexing stability of silver over lead is that the conformation and arrangement of the donor atoms in the macrocycles may favour the coordination geometry of the silver ion. Silver(I) has been found to favour a linear or tetrahedral coordination geometry155 while lead(II) has been found to adopt many different coordination geometries because of its ability to coordinate in a

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

holodirected or hemidirected geometry (Figure 13).156 In a holodirected geometry, the bonds are directed throughout the surface of the metal, while in a hemidirected geometry the bonds are only directed throughout a part of the metal.156b The void created is assumed to be occupied by the Pb2+ lone pair of electrons. In hemidirected complexes, the resulting bonds are believed to be weaker because there is less transfer of electrons from the ligands to the bonding orbitals of the metal which may explain the lower stability constant for lead(II). Lead(IV) was found to only exhibit the holodirected coodination geometry for all coodination numbers.

Pb Pb

Holodirected Hemidirected Figure 13

The size-match selectivity theory was also considered as a basis for the discrimination of the silver ion over lead. However in this case it was not judged fundamentally important because silver(I) and lead(II) have ionic radii of similar size.

4.1.2 Interaction Of Thiacrown Ethers (55), (56) and (57) with Neutral and Ionic Complexes (Second-Sphere Coordination)

The aim of these experiments was to determine if any interactions could be observed between metal complexes [Al(acac)3] (164), [Co(NH3)5NO2](BPh4)2 (165) and

[Co(en)3](BPh4)3 (166) and the crown compounds (55), (56), (57), 18-crown-6 (64) and 18-S-6 (100). This would provide a greater understanding of the nature of these 1 thiacrown compounds. The interaction was monitored by H NMR spectroscopy at 25 °C in either chloroform or dimethylsulfoxide. The choice of solvent was dependent primarily on its ability to solubilise the complexes. Proton NMR spectroscopy can be used to measure the kinetic aspects of complexation and dissociation because of the small chemical shift or linewidth differences between the complexed and uncomplexed species. It can be used to detect differences in either the ligand or the cation. The direct

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

observation of the metal ions, such as 7Li, 23Na, 39K and 59Co, has enhanced the versatility of this method.

(BPh ) O NO2 4 2 (BPh4)3 NH2 O NH3 O H2N NH Al H N Co NH 2 O 3 3 Co O H2N H3N N O H2 NH3 NH2

164 165 166

The interaction of (55), (56), (57), (64) and (100) with readily available

[Al(acac)3] (164), [Co(NH3)5NO2](BPh4)2 (165) and [Co(en)3](BPh4)3 (166) complexes were thus examined. It was envisaged that the thiacrown ethers would behave differently towards the neutral [Al(acac)3] (164) than with [Co(NH3)5NO2](BPh4)2 (165) and

[Co(en)3](BPh4)3 (166) species.

4.1.3 Interaction with [Co(NH3)5NO2](BPh4)2 (165) and [Co(en)3](BPh4)3 (166)

The interaction of the crown compounds with metal complexes was first investigated using the cobalt complexes (165) and (166). The advantage of using the cobalt complexes was that they had the potential to provide addition information through the direct observation of the metal ion using 59Co NMR spectroscopy. Thus, a mixture of the crown compounds was shaken for 5 minutes with one equivalent of the metal complex in dimethylsulfoxide and monitored by 1H NMR for 1 h. Disappointingly, the 1H NMR spectra were identical to those of the parent thiacrown ethers (55), (56), (57) and 18-S-6 (100). It was particularly surprising that 18-crown-6 (64) also did not show any interaction with [Co(NH3)5NO2](BPh4)2 (165), since it was envisaged that an interaction based on hydrogen bonding would be observed between the amine groups and the oxygen donor atoms. There are several reasons for the lack of interaction between the cobalt complexes and the thiacrown ethers. Firstly, the electrostatic interactions between these crown compounds and the cobalt complexes may have been weak and difficult to detect. Direct coordination of the crown compounds with the metal centre is not expected. It is possible for ligand substitution to occur, however in these cases it was unlikely. Secondly, the anion may have an effect on the possible interactions, however further investigations are needed to conclusively determine any anionic effect. Finally, solvation may affect the

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

ability of the two species to interact as ion pairs. The concept of ion pairing deals with two ions that are in close proximity to each other for a short time before their thermal motions tear them apart.157 There are several proposed models for representing the possible ion associations that exist in solution.157,158 The three types of ion pairs that exist are solvated ions, solvent separated ion pairs and contact ion pairs. The extent of ionic association is qualitatively related to the dielectric constant of the solvent,157 and it has been found that the extent of ionic association increases as the dielectric constant of a solvent decreases.159 DMSO has a high dielectric constant so it is not unreasonable to assume that the ionic species in these experiments may have existed as solvated ions or solvent separated ion pairs in DMSO.

4.1.4 Interaction with [Al(acac)3] (164)

The experiments carried out with the neutral [Al(acac)3] (164) complex did show second sphere interaction with thiacrown crown compounds (56) and (57). The crown compounds (55), (56), (57), (64) and (100) were again treated with one equivalent of the metal complex and shaken in chloroform for 5 minutes and monitored by 1H NMR for 1 h. Thiacrown ethers (55) and (100) and 18-crown-6 (64) did not show any chemical shift or line width differences in the 1H NMR spectrum. However, thiacrowns (56) and (57) did show a change in their 1H NMR spectra.

In the uncomplexed form, thiacrown (56) exhibits a sixteen-proton multiplet at δH

2.72-2.82 assigned to the protons next to the sulfur atoms (CH2S). However, when 1 [Al(acac)3] (164) is added to the solution, the H NMR spectrum shows a broadening of the signal to give a broad singlet at δH 2.77. Since no other signals were affected, this suggested an interaction between the components of the polythioether segment and the aluminium complex. Thiacrown (57) exhibited a similar type of interaction. A twenty- proton multiplet was observed at δH 2.69-2.84, which was assigned to the CH2S groups in the free crown compound. Once again, the addition of [Al(acac)3] (164) lead to a broadening of the signal to produce a broad singlet at δH 2.78. Changes in line broadening (and chemical shift) provide information on the nuclear environment of the nucleus of interest.160 The line broadening observed in an NMR spectrum is related to the rate of exchange and/or relaxation time of the particular nuclei. In this case, the increase in line broadening was believed to be due to a restriction (decrease) in the rate of exchange of the CH2S groups. The charge distributions of both sets of compounds were determined using electrostatic potential maps to better understand the type of interaction between the

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

thiacrown ethers and the aluminium complex. Electrostatic potential maps are used to calculate the electron rich (red) and electron poor (blue) areas in a molecule. The electrostatic potential map for [Al(acac)3] (164) showed that the area surrounding the Al3+ metal is electron rich and is depicted by a red-yellow-green colour while the blue hydrocarbon area was electron poor (Figure 16). Using thiacrown (55) as an example, the electrostatic potential map (focusing on the thioether groups) indicated that the sulfur atoms are partially negative (green) while the ethylene chain is partially positive (blue) (Figure 17). Based on the electrostatic potential maps of charge distribution it was considered that the most likely interaction observed in the 1H NMR spectrum was an electrostatic interaction between the partially negative (δ-) sulfur atoms of the macrocycle and the partially positive (δ+) hydrocarbon area of the aluminium complex.

[Space filling diagram of [Al(acac)3] (164)] [Diagram with mesh surface (to view atoms)] Figure 16

[Space filling diagram of thiacrown (55)] [Diagram with mesh surface (to view atoms)] Figure 17

DISCUSSION: Spiroacetal Thiacrown Ethers As Primary and Secondary Ligands

This interaction also provided information about the flexibility of the spiroacetal thiacrown ethers (55), (56) and (57). It suggested that thiacrown ethers (56) and (57) are much more flexible than the smaller thiacrown (55). This greater flexibility enables the thiacrown ethers to adopt a conformation which maximises favourable interactions and minimises unfavourable interactions. It is likely that this also includes the conformation of the spiroacetal ring system. The greater flexibility of thiacrown ethers (56) and (57) allows the spiroacetal moiety to arrange itself in the most thermodynamically stable conformation while positioning itself away from the metal complex. The size of the cavity is thought to permit extra flexibility and is consistent with the results.

4.2 Summary

The present work constitutes a detailed analysis of the binding abilities of the three thiacrown ethers (55), (56) and (57). The association constants were determined for lithium, potassium, sodium, cesium, cobalt, silver, cadmium and lead. The superior complexing ability observed for (55), (56) and (57) with silver is encouraging in terms of selective extraction.

The interaction between thiacrown ethers (55), (56) and (57) and [Al(acac)3]

(164), [Co(NH3)5NO2](BPh4)2 (165) and [Co(en)3](BPh4)3 (166) yielded some interesting results. Because of the promising results achieved with the aluminium complex, future work in this area will include investigating other aluminium complexes. Further work will also involve the use of other deuterated solvents to monitor any interactions with the cobalt complexes and the thiacrown ethers.

DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

5.0 Kinetic Resolutions of the Spiroacetal Moiety

The synthesis of enantiopure crown ethers has led to compounds capable of selectively binding enantiomeric species; this is known as chiral recognition. The first reported syntheses of chiral macrocycles was carried out by Wudl et al161 in 1972 starting from enantiopure materials. It was envisaged that the synthesis of optically active spiroacetal crown ethers could be achieved via the kinetic resolution of the initial racemic spiroacetal α-epoxide (115) or allylic alcohol (117) (Scheme 58). Kinetic resolution requires the use of a chiral reagent to promote the selective reaction of one enantiomer over the other to give both the starting material and product in enantiomerically enriched form.162 Kinetic resolution reactions are carried out at approximately 50% conversion of the starting material to the product. In this way, the reaction of the faster reacting enantiomer is promoted while the reaction of the slower reacting enantiomer is retarded. Three different approaches were investigated in the present study: (a) the base induced 163 rearrangement of α-epoxide (115) using chiral lithium amide bases, (b) the hydrolytic 164 kinetic resolution (HKR) of the α-epoxide (115) using a cobalt catalyst and (c) the Sharpless asymmetric epoxidation165 of allylic alcohol (117).

O R O

O O O R S O O

O O O S O O 115 S O OH OO

O SS S S O n

O O OH O R S O O

OH OH O R 117 O

Scheme 58

87 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

5.1 Base-Induced Rearrangement of Epoxides

The rearrangement of an epoxide to an allylic alcohol using lithium amide bases is thought to proceed via a cyclic six-membered transition state involving the coordination of an oxygen lone pair with the electron deficient lithium centre (Scheme 59).119a The proposed mechanism involves the removal of a proton syn to the epoxide oxygen (β-elimination).

R RR N R N Li H HO + OLi Scheme 59

When the epoxide reacted is prochiral, the use of chiral lithium amide bases results in enantioselective rearrangement to afford optically active products.163a Whitesell and Felman166 were the first to recognise the possibility of using chiral lithium amide bases to differentiate between two syn β-protons in cyclohexene oxide (167). Later, Asami167 used an S-proline-derived (169) base that resulted in a considerable improvement in the level of asymmetric induction reported by Whitesell and Felman.166 Deprotonation of the pseudoaxial proton was suggested to occur preferentially through the complex where the steric interactions between the cyclohexane ring and the amide were minimised (Figure 18).

H O O H H H R S H H H H OH HO 167 167

H H N N H H N Li H N Li H H O H O

Favoured Disfavoured Figure 18

In addition to providing enantiopure products from prochiral epoxides, chiral lithium amide bases have been used to generate enantiopure materials from racemic

88 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety epoxides via a kinetic resolution process. The aim is to convert the faster reacting enantiomer present in the mixture to the allylic alcohol while the slower reacting enantiomer remains as unreacted epoxide. Asami et al168 were able to achieve high selectivity in the reaction of cis-disubstituted epoxides such as (168) by using less than the stoichiometric quantity of the S-2-(pyrrolidine-1-yl)methylpyrrolidide base (169) (Scheme 60).

169 O N N O OH H H Li HH + Ph Ph Me Ph Me H 168 Epoxide : Base Yield (%) ee (%) Yield (%) ee (%) 4 : 3 31 95 60 33 3 : 1 67 30 21 72 Scheme 60

Chiral lithium amide bases possessing a second nitrogen or a lithium alkoxide have proved to be the most successful chiral bases in epoxide rearrangement reactions.163d Examples of the bases that have been used, both in stoichiometric and catalytic amounts, in the rearrangement of epoxides are shown below. Some of the bases such as (169), (170), (171) and (172) are commercially available while others including (173)169, (174)170 and (175)171 require lengthy syntheses. H Ph Ph H HO NH N 2

Me N Me Ph Me N H 170H 171 172 H Me NH NH N N N NH Ph 173 174 175

5.1.1 Ring Opening of α-Epoxide (115) Using Chiral Non-Racemic Lithium Amide Bases

It was thought that using readily available chiral non-racemic bases, such as, [R- * * (R ,R )]-(+)-bis(α-methylbenzyl)amine (170), (-)-sparteine (171) and (1S,2R)-(+)- norephedrine (172) would provide access to allylic alcohol (117) in enantioenriched form. The allylic alcohol could then be converted to enantiopure spiroacetal crown ethers

(55), (56) and (57). Racemic α-epoxide (115) was therefore treated with several chiral

89 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety lithium amide bases (Table 5) in an effort to afford allylic alcohol (117) enantioselectively (Scheme 61). The reaction was monitored by TLC and quenched at approximately 50% conversion. The determination of the enantiomeric excess of the allylic alcohol thus produced was carried out by NMR analysis of the α-methoxy-α- (trifluoromethyl)phenylacetate ester (Mosher ester) of the allylic alcohol (117). The synthesis of ester (176) was monitored by TLC to ensure complete consumption of the starting material so as to avoid any possibility that kinetic resolution could occur in this step. Unfortunately, the use of the chiral lithium amide bases derived from amines (170), (171) and (172) only afforded racemic allylic alcohol (117).

O O O 7 7 O i or ii iii 6 1 1 6 O O O + O 5 3 3 5 4 4 O 115OH 117 O O O O

OMe F3C F CPh Ph OMe 3 176

Reagents and Conditions: (i) n-BuLi, 0 °C, THF, (172), 30 min, then (115), 3 h, room temp., 18 h; (ii) n-BuLi, 50-55 °C, hexane, (170) or (171),30 min, then (115),

3 h, room temp., 18 h; (iii) DMAP, triethylamine, (S)-MTPA-Cl, CH2Cl2, 18 h, 88% Scheme 61

Entry Base/n-BuLi Temperature Solvent Yield Ratio a (°C) (%) R:S Ph Ph 59 Me N Me 1 -50 Hexane b 1:1 H 170 (43% conversion)

H H N 48 2 -55 Hexane 1:1 N H (55% conversion) H 171 21 3 HO NH2 0 THF 1:1 Ph Me 172 (35% conversion)

Table 5: Base-Induced Rearrangement of α-Epoxide (115)

a Yield based on conversion after flash chromatography b Conversion calculated from the recovery of α-epoxide (115) by flash chromatography

90 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

The title compound (176) analysed correctly for C19H21F3O5 with a protonated molecular ion m/z at 387.14255 in the high resolution spectrum. The 1H NMR spectrum exhibited two sets of signals representing the two diastereomers. The quartets at δH 3.49 with JOMe,CF3 1.1 Hz and δH 3.55 with JOMe,CF3 1.1 Hz were assigned to the two methoxy groups of the individual diastereomers. The methoxy groups showed long range coupling to fluorine. Integration of these two peaks established the ratio of the diastereomers to be

1:1. The resonances at δH 6.05 with J3,4 10.2 and J3,2 2.6 Hz and 6.12 with J3,4 10.2 and 13 J3,2 2.6 Hz were assigned to vinylic protons 3-H. The C NMR spectrum showed quartets at δC 55.4 and δC 55.5 representing the methoxy carbons. Another set of quartets 19 at δC 127.3 and δC 127.6 were assigned to the CF3 carbons. The F NMR spectrum showed two resonances at δF -72.9 and δF -72.7 which were assigned to the CF3 groups and resonated as quartets. Integration of the two peaks also confirmed the diastereomers were present as a 1:1 mixture in each case.

5.2 Hydrolytic Kinetic Resolution

5.2.1 Jacobsen Hydrolytic Kinetic Resolution Reaction

The Jacobsen hydrolytic kinetic resolution of racemic epoxides catalysed by chiral (salen)Co(III) complexes was first reported in 1997164 and has emerged as a general and effective method for the preparation of highly enantioenriched epoxides and 1,2-diols. The acetate complex [R,R-(177) and S,S-(178)], prepared by aerobic oxidation of (salen)Co(II) in the presence of acetic acid, has been the most commonly used catalyst. Using water as the nucleophile, the (R,R)-catalyst (177) selectively reacts with the S-epoxide to afford the S-1,2-diol while the (S,S)-catalyst (178) yields the R-1,2-diol and S-epoxide (Scheme 62).

H H H H NN NN Co Co t-Bu OO t-Bu t-Bu OO t-Bu OAc OAc

t-Bu t-Bu t-Bu t-Bu 177 178

91 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

OH OH O (R,R)-Catalyst O (S,S)-Catalyst O + OH + OH H2O H2O R-epoxide S-1,2-diol Racemic Mixture S-epoxide R-1,2-diol

Scheme 62

Jacobsen et al172 have postulated that the mechanism follows a second-order dependence on the concentration of the catalyst. Based on this, the authors suggest a cooperative, bimetallic mechanism whereby two separate catalyst molecules cooperate to activate both the electrophile (epoxide) and the nucleophile (water). The proposed mechanism is shown in Scheme 63.

X O X X X Co R Co Co Co O O H2O OH H2O R R OH R OH OH OH Co Co OH Co L L Co L

Scheme 63

5.2.2 Hydrolytic Kinetic Resolution of α-Epoxide (115)

The hydrolytic kinetic resolution of α-epoxide (115) relies on the different rates of hydrolysis of the two enantiomers in the presence of the chiral cobalt catalyst. It was envisaged that the reaction would afford the 1,2-diol and the starting α-epoxide in enantiopure form. The synthesis of enantiopure crown compounds would then be possible starting from the enantiopure α-epoxide (Scheme 58). The first step in the hydrolytic kinetic resolution was the aerobic oxidation of the commercially available (R,R)-(salen)Co(II) complex using acetic acid to form (177) as a brown solid. The α-epoxide was stirred in THF in the presence of the catalyst (177) (0.5 mol%) and water (0.55 equiv) to afford the recovered starting material (115) and the

92 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety diol (179b) in 55% yielda based on 53% conversionb (Scheme 64). Diol (179b) had identical spectroscopic properties to those previously published.173 From previous studies, it has been shown through X-ray crystallographic data that the diol adopts the diequatorial conformer depicted by structure (179b).173 Trans-diaxial opening of the epoxide initially forms diaxial diol (179a), however due to the unfavourable 1,3-diaxial interactions between the 4-OH and the C-O bond of the neighbouring tetrahydropyran, the substituted cyclohexane ring (179b) flips to form the less sterically hindered chair conformation with both OH groups adopting equatorial positions. In diol (179a), stabilisation from two anomeric effects is possible, while in diol (179b) only one anomeric stabilisation is possible. In this case, the decrease in the number of favourable anomeric effects is compensated for by the fact that the hydroxyl groups occupy the more sterically favoured equatorial positions.

O O O O OH ring O i O O flip O + HO 4 O 115O 115OH 179a 179b OH

Reagents and Conditions: (i) (177), H2O, THF, 55% (based on 53% conversion)

Scheme 64

To determine the enantiomeric excess, the recovered starting material was first converted to the allylic alcohol (117) via base-induced rearrangement of the epoxide

(115) using lithium diethylamide. The α-methoxy-α-(trifluoromethyl)phenylacetate ester (Mosher ester) (176) was then synthesised from the allylic alcohol (117) as described earlier (Scheme 61). The spectroscopic data was identical to that previously 1 19 discussed (vide supra). The H and F NMR spectrum of the α-methoxy-α- (trifluoromethyl)phenylacetate derivative once again showed that the diastereomers were present as a 1:1 mixture establishing that the hydrolytic kinetic resolution of the racemic

α-epoxide (115) had not taken place.

a Yield based on conversion after flash chromatography b Conversion calculated from the recovery of α-epoxide (115) by flash chromatography

93 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

5.3 Sharpless Epoxidation

In 2001, K. Barry Sharpless was awarded the Nobel Prize in chemistry for his contribution to asymmetric syntheses. The Sharpless asymmetric epoxidation relies on an in situ formation of a complex of titanium tetraisopropoxide, dialkyl tartrate, tert-butyl hydroperoxide and the allylic alcohol.165a The formation of epoxy alcohols proceeds with high enantiomeric purity and the enantioface selection is determined by the chirality of the dialkyl tartrate used (Scheme 65). When the allylic alcohol is arranged with the hydroxymethyl group at the bottom right, oxygen is delivered from the bottom side using (R,R)-(+)-dialkyl tartrate and from the top side using (S,S)-(-)-dialkyl tartrate. This empirical rule applies to all reactions of achiral alcohols174, however epoxidation of some chiral allylic alcohols has shown unusual face selectivity.175

(S,S)-(-)-dialkyl tartrate

1 R 3 O R 1 R 3 R R i 2 OH Ti(O Pr)4 R2 t OH BuOOH, CH2Cl2 1 R 3 O R

R2 OH

O (R,R)-(+)-dialkyl tartrate Scheme 65

The mechanism has been proposed to occur via the dimeric transition state represented by complex (180).176 In the conformation of substrate (180) shown the olefinic moiety, having a small dihedral angle (O-C-C=C, ca 30°), is arranged in an appropriate space for epoxidation to occur. The coordination of the distal oxygen (O2) to the titanium activates the peroxide bond and promotes the nucleophilic attack by the 177 1 double bond (Figure 19). The olefin approaches the proximal (O ) oxygen in an SN2- type reaction. The electrophilic centre (O1) and the leaving group (O2) are chiral in the transition state. The conformation and therefore the enantioselectivity can be affected by the size of the substituents on the alkene R2 and R3 (181).165b The R2 substituent is in the vicinity of the tartrate ligand while the R3 group is directed toward the ligand. When the allylic R4

94 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety is a substituent other than hydrogen, the epoxidation is strongly retarded due to steric hindrance, which is attributed to the poor reactivity of tertiary allylic alcohols.

R CO2R 5 R5 R4 O OR O R4 E R3 2 RO O O R O O R O Ti E Ti E 2 O O 3 O1R 1 R O O2 R 1 O CO R t-Bu t-Bu 2 RO E = CO2R 180 (The View Down O1-Ti Bond Axis) 181

O MLn O2 O1 R

Figure 19

The kinetic resolution of racemic secondary allylic alcohols using the titanium- tartrate complex deals with the rates of epoxidation of the two enantiomers.178 When the (R,R)-(+)-tartrate is used as the chiral source, the (S)-enantiomer of the allylic alcohol reacts faster to form the anti-epoxy alcohol, while the (R)-enantiomer reacts faster when the (S,S)-(-)-tartrate is used as the chiral source (Scheme 66). For example, the reaction of (E)-1-trimethylsilyl-1-octen-3-ol with (R,R)-(+)-diisopropyl tartrate at 50% conversion affords both the anti-epoxy alcohol and the unreacted alcohol in enantiomerically 179 enriched forms. The relative rates (krel) of epoxidation can be improved as the bulkiness of the ester alkyl group increases, for example, diisopropyl tartrate is better than diethyl tartrate which is better than dimethyl tartrate.

O Me3Si C5H11 (R,R)-(+)-diisopropyl tartrate Me3Si C5H11 Me3Si C5H11 + OH OH OH Racemic Mixture Recovered Substrate Epoxide >99% ee, 42% >99% ee, 42%

Scheme 66

95 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

Kinetic resolution is not restricted to allylic alcohols in which chirality exists at the carbinol carbon. Sharpless et al175 have shown the kinetic resolution of a variety of compounds with chiral centres present at carbons other than the carbinol carbon. The authors were able to selectively react one enantiomer to obtain highly enantioenriched epoxide and starting material. In stereochemically rigid cases, the faster reacting enantiomer is predicted to be the one in which the reacting olefin face is also the less- hindered olefin face. Hamon and Tuck180 have been able to kinetically resolve the allylic alcohol (182) depicted in Figure 20 using diisopropyl L-tartrate. The tartrate enantiofacial selection rule dictates that diisopropyl L-tartrate attacks one enantiomer from the top and the other from the bottom. In this case, the faster reacting enantiomer is found to be (182a) because attack at the less sterically hindered convex side is favoured.

H H L-(+)-DIPT (attack fast to unhindered convex side) 182b

HO OH 182a L-(+)-DIPT (attack slow to hindered concave side) Figure 20

5.3.1 Sharpless Epoxidation of Allylic Alcohol (117)

The Sharpless epoxidation of allylic alcohol (117) was expected to provide enantiopure syn-epoxy alcohol (119) by relying on the different rates of epoxidation of the two enantiomers with the respective tartrate ligands. This would lead to the synthesis of enantiopure thiacrown ethers (55), (56) and (57). The choice of tartrate ligand was dependent on the model developed by Sharpless (Scheme 65). When the spiroacetal allylic alcohol (117) is positioned correctly on the template, a complex containing (-)-diisopropyl D-tartrate delivers the oxygen to the top face while the (+)-diisopropyl L-tartrate delivers the oxygen to the bottom face (Figure 21). When (-)-diisopropyl D-tartrate is used the (S)-enantiomer is expected to react faster to form the syn-epoxy alcohol (119) while the (R)-enantiomer should yield the anti- epoxy alcohol (120). Alternatively, using (+)-diisopropyl L-tartrate the (R)-enantiomer is predicted to react faster to form the syn-epoxy alcohol (119) while the slower reacting (S)-enantiomer should produce the anti-epoxy alcohol (120) (Scheme 67). At 50%

96 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety conversion of the allylic alcohol (117) the slower reacting enantiomer is recovered as the starting allylic alcohol (117). The presence of steric barriers in a molecule contribute to the stereochemical outcome. In this case, the presence of the axial tetrahydropyran ring is expected to hinder formation of the anti-epoxy alcohol using either tartrate ligand.

(S,S)-(-)-dialkyl tartrate

O O S R O O

OH OH

(R,R)-(+)-dialkyl tartrate Figure 21

Thus, treatment of allylic alcohol (117) with titanium tetraisopropoxide, diisopropyl D-tartrate and anhydrous tert-butyl hydroperoxide in dichloromethane at a b -20 °C afforded the syn-epoxy alcohol in 77% yield based on 40% conversion. The formation of the anti-epoxy alcohol (120) was not observed. The determination of the 1 19 enantiomeric excess was carried out by the H and F NMR analysis of the α-methoxy- α-(trifluoromethyl)phenylacetate (Mosher ester) derivative (183) of the syn-epoxy alcohol (119) and found to be a 1:1 mixture of the diasteromers.

O O O 7 7 O i ii 6 1 1 6 O O O + O 5 3 3 5 4 4 OH OH O O O O O 117 119 O O

OMe F3C F3CPh Ph OMe 183

i Reagents and Conditions: (i) Ti(O Pr)4, diisopropyl D-tartrate, 4Å molecular t a b sieves, BuCOOH, -20 °C, CH2Cl2, 18 h; 77% (based on 40% conversion)

(ii) DMAP, triethylamine, (S)-MTPA-Cl, CH2Cl2, 18 h, 93%

Scheme 67

a Yield based on conversion after flash chromatography b Conversion calculated from the recovery of allylic alcohol (114) by flash chromatography

97 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

The Mosher ester derivative (183) analysed correctly for C19H22F3O6 with a protonated molecular ion m/z at 403.13675 in the high resolution spectrum. The 1H NMR spectrum exhibited two sets of signals representing the two diastereomers. The quartets at

δH 3.51 with JOMe,CF3 1.0 Hz and δH 3.64 with JOMe,CF3 1.0 Hz were assigned to the methoxy groups. Integration of these two peaks showed a 1:1 mixture of the two diasteromers. 3-H resonated as a double double doublet at δH 3.24 with J3,4 4.0, J3,2ax 1.2 and J3,2eq 1.2 Hz, and δH 3.29 with J3,4 4.0, J3,2ax 1.2 and J3,2eq 1.2 Hz. 5-H resonated as a doublet at δH 4.71 with J5,4 4.8 Hz and at δH 4.73 with J5,4 4.8 Hz. This suggested that both 3-H and 5-H adopted equatorial positions. The 13C NMR spectrum exhibited quartets at δC 55.5 and δC 55.7 representing the methoxy carbons. Another set of quartets 19 at δC 127.5 and δC 127.9 were assigned to the CF3 carbons. The F NMR spectrum showed two quartets at δF –73.4 and δF -72.9, which were assigned to the CF3 groups. Integration of these two signals confirmed the two diastereomers of the Mosher ester were present in a 1:1 ratio. Disappointingly, attempts to effect the kinetic resolution of the allylic alcohol via Sharpless epoxidation were not successful.

5.4 Summary

The three different attempts used to promote the kinetic resolution of the

α-epoxide (115) and the allylic alcohol (117) proved to be unsuccessful (Scheme 68). The based-induced rearrangement of the α-epoxide (115) using lithium amide bases * * derived from [R-(R ,R )]-(+)-bis(α-methylbenzyl)amine (170), (-)-sparteine (171) and (1S,2R)-(+)-norephedrine (172) yielded the desired allylic alcohol (117) and recovered starting material (115). Analysis of the Mosher ester derivative (176) of the allylic alcohol by 1H and 19F NMR established that the allylic alcohol was a 1:1 mixture of the two diasteromers.

The Jacobsen hydrolytic kinetic resolution of the α-epoxide (115) using a chiral cobalt catalyst (177) afforded the 1,2-diol (179a) and recovered starting material (115).

The recovered α-epoxide (115) was converted to the allylic alcohol (117) which, in turn, was converted to the Mosher ester derivative (176). Analysis of the Mosher ester derivative (176) by 1H and 19F NMR established that the allylic alcohol was a 1:1 mixture of the two diasteromers. This supported the unsuccessful kinetic resolution of the

α-epoxide (115).

98 DISCUSSION: Kinetic Resolution of the Spiroacetal Moiety

The Sharpless epoxidation of allylic alcohol (117) was carried out using (-)-diisopropyl D-tartrate as the chiral source. The epoxidation was expected to occur faster with the (S)-enantiomer to form the syn-epoxy alcohol (119) while the (R)- enantiomer remained as unreacted α-epoxide (115). Conversion of the syn-epoxy alcohol (119) to the Mosher ester (183) and its subsequent analysis by NMR spectroscopy established that the diastereomers were present in a 1:1 ratio.

O O Base-Induced No Kinetic Epoxide Rearrangement O O Resolution O 115 OH 114

O O Jacobsen Hydrolytic Kinetic Resolution O No Kinetic O HO Resolution 179b O 115 OH

Sharpless Epoxidation O O No Kinetic O Resolution O 119 OH 117 OH O Scheme 68

99 DISCUSSION: Conclusion

6.0 Conclusion

In conclusion, the synthesis of the target spiroacetal thiacrown ethers (55), (56) and (57) was successfully achieved. The reaction between the spiroacetal ditosylate (146) and the appropriate dithiol (99), (122) or (123) in dimethylformamide containing cesium carbonate yielded thiacrown ethers (55), (56) and (57) in 86%, 68% and 64% respectively (Scheme 69). The spiroacetal ditosylate (146) was prepared from the diol (145) precursor, which in turn was obtained by the reductive ozonolysis of the bisallyl ether (133). The bisallyl ether was synthesised starting from (±)-1,7- dioxaspiro[5.5]undec-4-ene (114).

O O

O + n O HS S SH OO 99, n = 1 OO 122, n = 2 OTs TsO 123, n = 3 SS 146 S n 55, n = 1 56, n = 2 57, n = 3 Scheme 69

The binding ability of thiacrown ethers (55), (56) and (57) with lithium, sodium, potassium, cesium, cobalt, cadmium, silver and lead showed an affinity for the heavy metals. This can largely be explained in terms of the HSAB theory, the large difference in the complexing behaviour of the thiacrown ethers with silver compared to that of lead being the exception. Both silver and lead are considered to be thiophilic. One possible explanation for their different complexing abilities is the difference in oxidation states. Thiacrown ethers can thermodynamically stabilise lower oxidation by binding more strongly to them than those with higher oxidation states. Another explanation is that the conformation and arrangement of the donor atoms favours the coordination geometry of the silver ion. The size-match selectivity was also considered, however silver and lead have radii of similar size. The ability of thiacrown compounds (55), (56) and (57) to act as second-sphere ligands with [Al(acac)3] (164), [Co(NH3)5NO2](BPh4)2 (165) and [Co(en)3](BPh4)3 (166) revealed some interesting results. Thiacrown ether (55) did not show any interaction with

100

DISCUSSION: Conclusion the three complexes while thiacrown ethers (56) and (57) did show interaction with the aluminium complex. This interaction was observed using 1H NMR spectroscopy and was represented by the appearance of a broad singlet of the CH2S multiplet at δH 2.77 and

δH 2.78 for (56) and (57) respectively. The electrostatic potential map for [Al(acac)3] (164) and spiroacetal thiacrown (55) suggested an electrostatic interaction between the hydrocarbon area (δ+) of the metal complex and the sulfur atoms (δ-) of the thiacrown ether.

(BPh ) O NO2 4 2 (BPh4)3 NH2 O NH3 O H2N NH Al H N Co NH 2 O 3 3 Co O H2N H3N N O H2 NH3 NH2

164 165 166

The synthesis of the target spiroacetal azacrown ethers (58), (59) and (60) was only achieved with limited success. Initially, the reaction between diethylenetriamine (151) and spiroacetal dialdehyde (150) was carried out in an attempt to avoid nitrogen protection, with azeotropic removal of water, however, the unsaturated crown ether (154) was not obtained. The spiroacetal dialdehyde (150) was prepared via ozonolysis of the bisallyl ether (133). Reduction of the intermediate ozonide with dimethyl sulfoxide afforded the desired spiroacetal dialdehyde (150). The successful synthesis of azacrown (58) was achieved after removal of the protecting groups of azacrown ether (159). The slow addition of spiroacetal ditosylate (146) to a solution of Ns-protected triamine (158) in tetrahydrofuran containing sodium hydride afforded the protected azacrown ether (159) in 27% yield. Subsequent deprotection yielded azacrown (58) in 84% yield (Scheme 70). Attempts to apply a similar methodology in the synthesis of (59) and (60) were not successful.

O O O

+ n O HN N NH O O Ns Ns Ns OO n = 1 OO OO 158 O OTs TsO Ns = S NNNs Ns NH NH O H 146 NO2 N N Ns 58 159 Scheme 70

101

DISCUSSION: Conclusion

Kinetic resolution reactions of the spiroacetal moiety were also carried out to provide enantiopure starting materials for the synthesis of the spiroacetal crown ethers (Scheme 71). Theoretically enantiopure compounds can be obtained by relying on the different rates of reaction of the two enantiomers. The base-induced rearrangement of the

α-epoxide (115) afforded the allylic alcohol (117) and recovered starting epoxide (115). The 1H and 13F NMR analysis of the Mosher ester derivative (176) of the allylic alcohol (117) revealed a 1:1 ratio of the diastereomers. Jacobsen’s hydrolytic kinetic resolution of α-epoxide (115) using a cobalt catalyst (177) and water as the nucleophile afforded diol (179b) and the starting α-epoxide (115). Analysis of the Mosher ester derivative (176) showed a racemic mixture. The Sharpless epoxidation of allylic alcohol (117) was performed using titanium tetraisopropoxide, diisopropyl D-tartrate and anhydrous tert-butyl hydroperoxide. The 1H and 19F NMR analysis of the syn-epoxy alcohol (183) obtained showed a 1:1 mixture of the two diastereomers.

O O Base-Induced No Kinetic Epoxide Rearrangement O O Resolution O 115 OH 114

O O Jacobsen Hydrolytic Kinetic Resolution O No Kinetic O HO Resolution 179b O 115 OH

Sharpless Epoxidation O O No Kinetic O Resolution O 119

OH 117 OH O

Scheme 71

A novel set of spiroacetal dimers (142a), (142b) and (142c) were synthesised during attempts to synthesise crown compounds via a cross metathesis reaction. Treatment of bisallyl ether (133) with Grubbs’ ruthenium catalyst (137) or Shrock’s molybdenum catalyst (136) afforded the three isomeric compounds represented by the general structure (142). Detailed assignment of the stereochemistry for the three dimers proved to be elusive, however each compound exhibited unique 1H and 13C NMR spectra.

102

DISCUSSION: Conclusion

142a OO142b 142c O

6.1 Future Work

Future work in this area will involve a further investigation into the synthesis of the spiroacetal azacrown ethers (58), (59) and (60) using other methodologies. One option is using the method developed by Tabushi et al,91 which involves the addition of a diamine to a diester. Synthesis of the spiroacetal diamine could be achieved via the reaction of the spiroacetal ditosylate (146) and sodium azide followed by reduction of the diazide. The spiroacetal diamine could then be reacted with the appropriate diester to yield the macrocyclic diamide. Reduction of the amide group would afford the spiroacetal crown ethers. The advantage of this method is that the need for nitrogen protection is avoided. The use of peptide chemistry is another option in the synthesis of azacrown ethers (58), (59) and (60). The reaction of the spiroacetal diamine with the appropriate dicarboxylic acid using an activating agent could be used to yield the protected diamide. Reduction of the amide group would afford the spiroacetal crown ethers. Crown ethers can be designed to target specific metal ions by changing the number, type and location of the macrocycle donor groups. The synthesis of spiroacetal crown ethers having a mixture of oxygen, nitrogen and sulfur atoms may also be undertaken based on this idea. Similar synthetic methods to those already used could be applied in these cases. The successful synthesis of the spiroacetal azacrown ethers and mixed donor crown ethers would then allow an analysis of their binding abilities as primary and second-sphere ligands. A greater investigation will also be undertaken in the ability of

103

DISCUSSION: Conclusion the spiroacetal crown compounds to bind a variety of other metals. Thiacrown ligands are noted for their significant affinity for copper, platinum, gold and mercury and have a wide application in medicine and the environment. The ability of the spiroacetal crown compounds to act as pH dependent ionophores may allow them to be used to carry metal ions across a membrane, which upon interaction with acid would release the ion. The addition of fatty acid substituents to the spiroacetal moiety may provide ion channels. The promising results obtained using thiacrown ethers (56) and (57) as second- sphere ligands with [Al(acac)3] (164) suggests that this work should be extended to include a larger range of complexes and include a larger number of metals. The interaction of spiroacetal crown ethers with platinum complexes is of particular interest because of their anti-cancer abilities. Another area of interest is the synthesis of optically active spiroacetal crown ethers. The asymmetric epoxidation of spiroacetal alkene (114) to afford α-epoxide (115) may be achieved by using a fructose-derived ketone in the presence of Oxone®.181 This method may be applied to the kinetic resolution of spiroacetal alkene (114) whereby one enantiomer reacts faster than the other yielding the recovered starting material and desired epoxide.

104

EXPERIMENTAL

7.0 General Details

Melting points were determined using a Gallenkamp Melting Point Apparatus and are uncorrected. Ultraviolet measurements were determined on a Varian CARY 1E ultraviolet spectrophotometer at 380 nm in a 1 cm cell at the concentrations specified. All measurements were performed at 22 °C. Infrared spectra were recorded with a Shimadzu FTIR-8300 spectrometer or Perkin Elmer Spectrum One Fourier Transform IR spectrometer as thin films between sodium chloride plates. Absorption maxima are expressed in wavenumbers (cm-1). Thin layer chromatography (TLC) was performed using 0.2 mm thick precoated silica gel plates (Merck Kieselgel 60 F254 or Riedel-de Haen Kieselgel S F254). Compounds were visualised by ultraviolet fluorescence or by staining with vanillin in methanolic sulfuric acid. Flash chromatography was performed using Merck Kieselgel 60 (230-400 mesh) with the indicated solvents. 1H nuclear magnetic resonance spectra were recorded on a Bruker Advance 300 (300 MHz), Bruker DRX 400 (400 MHz), Varian Unity Plus 300 (300 MHz) or Varian

Mercury 400 (400 MHz) spectrometer at 25 °C. Data is expressed in parts per million using tetramethylsilane or the solvent peak as the internal reference and reported as position (δH), relative integral, multiplicity (s = singlet, br s = broad singlet, d = doublet, dd = double doublet, ddd = double double doublet, dddd = double double double doublet, t = triplet, dt = doublet of triplets, ddt = double double triplet, q = quartet or m = multiplet), coupling constant and assignment (aided by COSY, HMBC and HSQC (or HMQC)). 13C nuclear magnetic resonance spectra were recorded on a Bruker Advance 300 (75 MHz), Bruker DRX 400 (100 MHz), Varian Unity Plus 300 (75 MHz) or Varian

Mercury 400 (100 MHz) spectrometer at 25 °C. Data is expressed in parts per million using tetramethylsilane or the solvent peak as the internal reference and reported as position (δC), multiplicity and assignment. Mass spectra were recorded using a VG70-SE spectrometer operating at a nominal voltage of 70eV. Chemical ionisation (CI) mass spectra were obtained with the indicated reagent gas and fast atom bombardment (FAB) mass spectra were obtained using 3-nitrobenzyl alcohol as the matrix. All reactions were performed in dry glassware under an inert atmosphere unless * * otherwise stated. 3-Butyn-1-ol, δ-valerolactone, diethylenetriamine, [R-(R ,R )]-(+)- bis(α-methylbenzyl)amine, (-)-sparteine (1S,2R)-(+)-norephedrine, (R,R)-N,N’-bis(3,5- di-tert-butylsalicylidene)-1,2-cyclohexanediaminocobalt(II), benzylidene 105

EXPERIMENTAL bis(tricyclohexylphosphine)-dichlororuthenium, diisopropyl D-tartrate, 2-mercaptoethyl sulfide and 3,6-dithia-1,8-octanediol were purchased from Aldrich Chemical Company and purified as necessary. 2,6-Diisopropylphenylimidoneophylidenemolybdenum(VI) bis(hexafluoro-t-butoxide) was purchased from Strem Chemicals. [Al(acac)3]

[Co(NH3)5NO2](BPh4)2 and [Co(en)3](BPh4)3 were provided by Dr. Trevor Bailey. Solvents were purified and dried according to the procedures outlined by Leonard, Lygo and Procter.182 Molecular modelling was executed using Spartan ’02 (Wavefunction Inc, Irvine, CA.) operating on a Macintosh G5 (operating system 10.3). The electrostatic potential maps for compounds [1S∗, 15R∗, 18S∗]-Spiro[2,14,17-trioxa-5,8,11- trithiabicyclo[13.3.1]-nonadecane-18,2’-tetrahydropyran] (55) and [Al(acac)3] (164) were calculated from semi-empirical PM3 calculations.

7.1 Preparation of [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan- 3,5-diol (50)

1-Trimethylsilyloxy-3-butyne (111)

Chlorotrimethylsilane (7.75 g, 71.3 mmol) was added to a mixture of 3-butyn-1- ol (5.0g, 71.3 mmol) and triethylamine (14.44 g, 142.7 mmol) in tetrahydrofuran (150 cm3) using the procedure reported by Brimble et al.115 Removal of the solvent yielded a pale yellow oil that was purified by flash chromatography using pentane- diethyl ether (4:1) as eluent to give the trimethylsilyl ether as a colourless oil (9.01 g, 89%). The 1H and 13C NMR spectra were in agreement with that reported in the literature.115

HC OSi(CH3)3

111

106

EXPERIMENTAL

2-Methoxy-1-(Tetrahydropyran-2-yl)-1-butyn-4-ol (112) and 2- Ethoxy-1-(Tetrahydropyran-2-yl)-1-butyn-4-ol (113)

A solution of n-butyllithium (13.2 cm3 of a 1.6 mol dm-3 solution in hexane, 21.1 mmol) was added to a solution of 1-trimethylsiloxy-3-butyne (2.50 g, 17.8 mmol) in 3 tetrahydrofuran (250 cm ) at –78 °C. The mixture was stirred for 30 min at which time 3 δ-valerolactone (110) (2.28 g, 22.8 mmol) in tetrahydrofuran (7 cm ) was added. The reaction mixture was left to stir for a further 30 min. It was quenched with a saturated aqueous ammonium chloride (15 cm3) and allowed to warm to room temperature. After extraction with ethyl acetate (5 x 50 cm3) the combined extracts were washed with water and dried over magnesium sulfate. Removal of the solvent in vacuo yielded an orange oil which was dissolved in methanol or ethanol (20 cm3) and stirred for 16 h with catalytic acid. Triethylamine (5 drops) was added and the solvent removed under reduced pressure to give an oil which was purified by flash chromatography using hexane-ethyl acetate (4:1) as eluent to yield the desired alkoxy acetals.

(i) 2-Methoxy-1-(tetrahydropyran-2-yl)-1-butyn-4-ol (112). Prepared following the procedure reported by Brimble et al.115 Using methanol and Amberlite resin yielded a pale yellow oil (1.17 g, 36%). The 1H and 13C NMR spectra were in agreement with that reported in literature.115

H3C O

112

(ii) 2-Ethoxy-1-(tetrahydropyran-2-yl)-1-butyn-4-ol (113). Prepared using ethanol and -1 pyridinium p-toluenesulfonate to yield a pale yellow oil (3.06 g, 88%); νmax(film)/cm

3419 (br, OH), 2249 (w, C≡C); δH (300 MHz; CDCl3; Me4Si) 1.24 (3 H, t,

JCH3,OCH2 7.1 Hz, CH3), 1.45-1.70 (3 H, m, 5-CH2 and 4-HA), 1.77-1.89 (3 H, m, 3-CH2 and 4-HB), 2.51 (2 H, t, J3’,4’ 6.5 Hz, C-9), 2.54 (1 H, t, JOH,4’ 6.2 Hz, OH), 3.60-3.82

(6 H, m, 3 x OCH2); δC (75 MHz; CDCl3; Me4Si) 15.7 (CH3, CH2CH3), 19.1 (CH2, C-4),

22.8 (CH2, C-3’), 24.6 (CH2, C-5), 36.8 (CH2, C-3), 58.6 (CH2, OCH2CH3), 60.8 (CH2,

107

EXPERIMENTAL

C-4’), 62.1 (CH2, C-6), 80.2, 81.7, 94.4 (quat, C-1’, C-2’, C-2); m/z (EI) 153 + (M -OCH2CH3, 28%), 129 (89), 101 (100), 97 (54), 83 (36), 67 (29), 55 (68).

OH 3' 4' 2' H3C O 2 1' 3 O 1

4 6 5 113

(±)-1,7-Dioxaspiro[5.5]undec-4-ene (114)

Potassium carbonate (0.20 g) and Lindlar catalyst [5 wt. % Palladium on calcium carbonate, poisoned with lead (0.02 g)] were added to a solution of the 2-ethoxy-1- (tetrahydropyran-2-yl)-1-butyn-4-ol (113) (3.25 g, 16.4 mmol) in 4:1 pentane-diethyl ether (50 cm3) and left to stir for 16 h under a balloon of hydrogen following the procedure reported by Brimble et al.115 The reaction mixture was then filtered through a short pad of Celite and the solvent removed under reduced pressure to afford a colourless oil. The oil was dissolved in dichloromethane (20 cm3), pyridinium p-toluenesulfonate (0.01 g) was added and the mixture was left to stand at room temperature for 1 h. The solvent was removed under reduced pressure and the residue purified by flash chromatography using pentane-diethyl ether (9:1) as eluent, to afford the desired spiroacetal as a colourless oil (1.97 g, 78%); δH (300 MHz; CDCl3; Me4Si) 1.42-1.94

(7H, m, 3 x CH2 and 3-HA), 2.12-2.38 (1H, m, 3-HB), 3.51-3.99 (4H, m, 2 x OCH2), 5.62

(1H, ddd, J5,4 10.3, J5,3ax 2.9 and J5,3eq 1.5 Hz, 5-H), 5.91-5.96 (1H, m, 4-H); δC (75 MHz;

CDCl3; Me4Si) 18.6, 24.7, 25.0 (C-9, C-10 and C-11), 34.8 (C-3), 57.7 (C-2), 60.8 (C-8), 92.8 (C-6), 127.7 (C-5), 130.6 (C-4). The 1H and 13C NMR spectra were in agreement with that reported in literature.115

8 9 O7 10 1 6 2 11 O

5 3 4 114

108

EXPERIMENTAL

[4S∗, 5S∗, 6S∗] and [4R∗, 5R∗, 6S∗]-4,5-Epoxy-1,7- dioxaspiro[5.5]undecane (115) and (116)

(a) Preparation of Dimethyldioxirane

Oxone® (120 g, 195.2 mmol) was added to a cooled mixture of water (254 cm3), acetone (192 cm3) and sodium hydrogen carbonate (58 g) in five portions following the procedure described by Adam et al.116 The dimethyldioxirane/acetone solution was distilled and collected in a cooled (-78 °C) receiving flask as a pale yellow liquid (150 cm3, ca 0.1 mol dm-3).

(b) Epoxidation Using Dimethyldioxirane

Dimethyldioxirane (135 cm3, 0.1 mol dm-3 solution in acetone) was added to a solution 3 of (±)-1,7-dioxaspiro[5.5]undec-4-ene (114) (1.90 g, 12.3 mmol) in acetone (10 cm ) and the mixture was left to stir for 18 h. Removal of the solvent under reduced pressure afforded a colourless oil which was dissolved in dichloromethane and dried over sodium sulfate. The solvent was removed and the oil purified by flash chromatography using hexane-ethyl acetate (4:1) to afford:

(i) [4S∗, 5S∗, 6S∗]-4,5-Epoxy-1,7-dioxaspiro[5.5]undecane (115) (1.49 g, 71%) as 114 colourless needles mp 47-49 °C (lit., 48-49 °C); δH (400 MHz; CDCl3; Me4Si) 1.47-

1.78 (7H, m, 9-CH2, 10-CH2, 11-CH2 and 3-Heq), 1.97-2.06 (1H, m, 3-Hax), 2.76 (1H, dd, J5,4 4.0 and J5,3eq 0.9 Hz, 5-H), 3.29 (1H, dd, J4,5 4.0 and J4,3ax 4.5 Hz, 4-H), 3.39 (1H, ddd, J2eq,2ax 11.0, J2eq,3ax 7.0 andJ2eq,3eq 0.9 Hz, 2-Heq), 3.59-373 (3H, m, 8-CH2 and

2-Hax); δC (100 MHz; CDCl3; Me4Si) 17.6, 22.7, 25.0, 31.9 (CH2, C-3, C-9, C-10 and

C-11), 50.5 (CH, C-4), 52.9 (CH, C-5), 54.9 (CH2, C-2), 60.9 (CH2, C-8), 93.7 (quat, C-6). The 1H and 13C NMR spectra were in agreement with that reported in literature.114

8 9 O7 10 1 6 2 11 O 5 3 4 O 115

109

EXPERIMENTAL

(ii) [4R∗, 5R∗, 6S∗]-4,5-Epoxy-1,7-dioxaspiro[5.5]undecane (116) (0.23 g, 11%) as a colourless oil. δH (400 MHz; CDCl3; Me4Si) 1.45-1.78 (6H, m, 9-CH2, 10-CH2 and

11-CH2), 1.84-1.92 (2H, m, 3-CH2), 3.01 (1H, d, J5,4 4.0 Hz, 5-H), 3.26-3.29 (1H, m, 4-H), 3.36-3.41 (1H, m, 2-Heq), 3.57-369 (2H, m, 2-Hax and 8-Heq), 3.74 (1H, ddd, J8ax,8eq 11.3, J8ax,9ax 11.3 and J8ax,9eq 2.7 Hz, 8-Hax); δC (100 MHz; CDCl3; Me4Si)

18.1, 24.8, 25.0, 34.5 (CH2, C-3, C-9, C-10 and C-11), 50.2 (CH, C-4), 54.9 (CH2, C-2), 1 13 55.3 (CH, C-5), 60.9 (CH2, C-8), 92.8 (quat, C-6). The H and C NMR spectra were in agreement with that reported in literature.114

8 9 O7 10 1 6 2 11 O O 5 3 4 116

A solution of Oxone® (0.49 g, 0.78 mmol) in water (cm3) was added to a stirred solution of (±)-1,7-dioxaspiro[5.5]undec-4-ene (114) (0.10 g, 0.65 mmol), NaHCO3 (0.11 3 183 g, 1.36 mmol) and acetone (5 cm ) at 0 °C using the procedure by Ferraz et al. The resulting slurry was then stirred at room temperature for 18 h. The acetone was removed under reduced pressure and the residue extracted with ethyl acetate (3 x 10 cm3). The combined extracts were washed with brine and died over magnesium sulfate. Purification by flash chromatography afforded the α-epoxide (115) (45 mg, 41%) and the β-epoxide (116) (40 mg, 36%). The 1H and 13C NMR spectra were in agreement with that reported in literature.114

[5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-3-en-5-ol (117) and [5S∗, 6S∗]- 1,7-Dioxaspiro[5.5]undec-2-en-5-ol (118)

n-Butyllithium (2.0 cm3 of a 1.6 mol dm-3 solution in hexane, 3.23 mmol) was added dropwise to a solution of diethylamine (0.36 cm3, 3.55 mmol) in hexane (25 cm3) at –35 °C and the suspension was left to stir for 30 min. [4S∗, 5S∗, 6S∗]-4,5-Epoxy-1,7-

110

EXPERIMENTAL dioxaspiro[5.5]undecane (115) (0.5 g, 2.93 mmol) as a solution in hexane (5 cm3) was added to the mixture. The solution was allowed to warm slowly (ca 3 h) to room temperature and then left to stir for a further 18 h. The reaction mixture was then quenched with sodium dihydrogen phosphate solution (5 cm3, 10% w/v) and extracted with ethyl acetate (5 x 25 cm3). The combined extracts were washed with water (10 cm3) and dried over sodium sulfate. Removal of the solvent under reduced pressure yielded an orange oil that was purified by flash chromatography using hexane-ethyl acetate (3:2) as eluent to afford:

(i) [5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-3-en-5-ol (117) (0.38 g, 76%) as colourless 114 needles mp 54-56 °C (from pentane) (lit., 54-56 °C ); δH (300 MHz; CDCl3; Me4Si)

1.49-1.60 (3H, m, 9-CH2 and 11-Hax), 1.65-1.82 (2H, m, 10-CH2), 2.03-2.07 (3H, m,

11-Heq and OH), 3.59 (1H, s, 5-H), 3.71 (1H, ddd, J8ax,8eq 11.3, J8ax,9ax 11.3 and

J8ax,9eq 3.3 Hz, 8-Hax), 3.75-3.80 (1H, m, 8-Heq), 4.13 (1H, d, J2eq,2ax 16.4 Hz, 2-Heq),

4.17 (1H, dd, J2ax,2eq 16.4 and J2ax,3 1.8 Hz, 2-Hax), 5.96-5.97 (2H, m, 3-H and 4-H);

δC (100 MHz; CDCl3; Me4Si) 18.3, 24.9, 30.3, (CH2, C-9, C-10 and C-11), 60.2 (CH2,

C-2), 62.9 (CH2, C-8), 66.9 (CH, C-5), 96.9 (quat, C-6), 124.6, 128.7 (CH, C-3 and C-4). The spectroscopic data were in agreement with those reported in literature.114

8 9 O7 10 6 1 2 11 O 5 3 4 OH 117

(ii) [5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-2-en-5-ol (118) (17 mg, 3%) as a pale yellow oil; δH (400 MHz; CDCl3; Me4Si) 1.44 (1H, ddd, J11ax,11eq 13.5, J11ax,10ax 13.5 and

J11ax,10eq 4.7 Hz, 11-Hax), 1.50-1.81 (4H, m, 9-CH2 and 10-CH2), 1.91 (1H, dddd,

J4eq,4ax 17.4, J4eq,5 4.0, J4eq,3 4.0 and J4eq,2 1.0 Hz, 4-Heq), 2.01-2.05 (2H, m, 11eq and

OH), 2.42 (1H, dddd, J4ax,4eq 17.4, J4ax,5 4.5, J4ax,3 2.5 and J4ax,2 2.5 Hz, 4-Hax), 3.58 (1H, ddd, J5,OH 8.5, J5,4ax 4.5 and J5,4eq 4.0 Hz, 5-H), 3.66 (1H, dt, J8eq,8ax 11.6 and J8eq,9 1.9 Hz,

8-Heq), 3.79 (1H, ddd, J8ax,8eq 11.6, J8ax,9ax 11.6 and J8ax,9eq 3.2 Hz, 8-Hax), 4.65-4.69

(1H, m, 3-H), 6.26 (1H, d, J2,3 6.2 Hz, 2-H); δC (100 MHz; CDCl3; Me4Si) 17.9, 24.9

(CH2, C-9 and C-10), 25.8 (CH2, C-4), 29.4 (CH2, C-11), 61.8 (CH2, C-8), 68.5 (CH,

111

EXPERIMENTAL

C-5), 96.7 (quat, C-6), 98.6 (CH, C-3), 139.9 (CH, C-2). The spectroscopic data was in agreement with that reported in literature.114

8 9 O7 10 6 1 2 11 O 5 3 4 OH 118

[3S∗, 4S∗, 5S∗, 6S∗]-3,4-Epoxy-1,7-dioxaspiro[5.5]undecan-5-ol (119) and [3R∗, 4R∗, 5S∗, 6S∗]-3,4-Epoxy-1,7-dioxaspiro[5.5]undecan-5-ol (120)

A solution of [5S∗, 6S∗]-1,7-dioxaspiro[5.5]undec-3-en-5-ol (117) (0.20 g, 3 1.18 mmol) in dichloromethane (20 cm ) was cooled to 0 °C in an ice/water bath. Sodium acetate (0.35 g, 4.23 mmol) was added, followed by meta-chloroperoxybenzoic acid (0.58 g, 70% w/w, 2.35 mmol) in five portions over 1 min. The reaction was allowed to warm to room temperature and stirred for 72 h. The suspension was then filtered through a short pad of Celite. The solution was washed with sodium sulfite (10 cm3, 10% w/v), saturated aqueous sodium hydrogen carbonate (2 x 5 cm3), water (5 cm3) and dried over sodium sulfate. The solvent was removed under reduced pressure to yield a pale yellow oil that was purified by flash chromatography using hexane-ethyl acetate (3:2) to afford:

(i) [3S∗, 4S∗, 5S∗, 6S∗]-3,4-Epoxy-1,7-dioxaspiro[5.5]undecan-5-ol (119) (0.19 g, 88%) 114 as colourless prisms mp 130-133 °C (lit., 130-132 °C); (200 MHz; CDCl3; Me4Si) 1.40

(1H, ddd, J11ax,11eq 13.7, J11ax,10ax 13.7 and J11ax,10eq 5.2 Hz, 11-Hax), 1.48-1.67 (4H, m,

9-CH2 and 10-CH2), 1.90 (1H, dt, J11eq,11ax 13.7 and J11eq,10 2.8 Hz, 11-Heq), 2.32 (1H, d,

JOH,5 11.6 Hz, OH), 3.36 (1H, d, J3,4 3.0 Hz, 3-H), 3.46-3.67 (4H, m, 4-H, 5-H and

8-CH2), 3.84 (1H, d, J2eq,2ax 13.4 Hz, 2-Heq), 3.96 (1H, d, J2ax,2eq 13.4 Hz, 2-Hax);

δC (100 MHz; CDCl3; Me4Si) 18.1, 24.8, (CH2, C-9 and C-10), 29.9 (CH2, C-11), 51.2

(CH, C-4), 51.9 (CH, C-3), 57.3 (CH2, C-2), 61.8 (CH2, C-8), 65.7 (CH, C-5), 95.2 (quat, C-6). The spectroscopic data were in agreement with those reported in literature.114

112

EXPERIMENTAL

8 9 O7 10 6 1 2 11 O 5 3 4 OH O 119

(ii) [3R∗, 4R∗, 5S∗, 6S∗]-3,4-Epoxy-1,7-dioxaspiro[5.5]undecan-5-ol (120) (0.02 g, 9%) as a colourless oil; (400 MHz; CDCl3; Me4Si) 1.48-1.83 (6H, m, 9-CH2, 10-CH2 and

11-CH2), 2.35 (1H, d, JOH,5 6.8 Hz, OH), 3.22 (1H, dd, J3,2eq 3.0 and J3,4 4.0 Hz, 3-H),

3.25 (1H, d, J4,3 4.0 Hz, 4-H), 3.66 (1H, d, J5,OH 6.8 Hz, 5-H), 3.70-3.77 (2H, m, 8-CH2),

3.92 (1H, dd, J2eq,2ax 13.7 and J2eq,3 3.0 Hz, 2-Heq), 4.07 (1H, d, J2ax,2eq 13.7 Hz, 2-Hax);

δC (100 MHz; CDCl3; Me4Si) 17.9, 24.7, 26.3 (CH2, C-9, C-10 and C-11), 49.3 (CH,

C-3), 53.2 (CH, C-4), 60.2 (CH2, C-2), 61.9 (CH2, C-8), 68.9 (CH, C-5), 97.1 (quat, C-6). The spectroscopic data were in agreement with those reported in literature.114

8 9 O7 10 6 1 2 11 O 5 O 3 4 OH 120

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diol (50) and [4S∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-4,5-diol (121)

Lithium aluminium hydride (0.13 g, 3.38 mmol) was added in 3 portions over 1 min to a solution of [3S∗, 4S∗, 5S∗, 6S∗]-3,4-epoxy-1,7-dioxaspiro[5.5]undecan-5-ol 3 (119) (0.14 g, 0.77 mmol) in tetrahydrofuran (15 cm ) at 0 °C. The reaction was stirred at 0 °C for 1 h then at room temperature for 16h. The reaction was quenched with sodium dihydrogen phosphate (5 cm3, 10% w/v) and half the tetrahydofuran removed under reduced pressure. The residue was then extracted with ethyl acetate (5 x 20 cm3) and the combined extracts washed with water (15 cm3) and dried over sodium sulfate. The

113

EXPERIMENTAL solvent was removed in vacuo to give a pale yellow oil that was purified by flash chromatography using hexane-ethyl acetate (2:1) as eluent to yield:

(i) [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diol (50) (120 mg, 81%) as a colourless oil; (300 MHz; CDCl3; Me4Si) 1.39 (1H, ddd, J11ax,11eq 13.2, J11ax,10ax 13.2 and

J11ax,10eq 4.7 Hz, 11-Hax), 1.50-1.79 (4H, m, 9-CH2 and 10-CH2), 1.96 (1H, dddd,

J4eq,4ax 14.5, J4eq,5 3.0, J4eq,3 3.0 and J4eq,2eq 3.0 Hz, 4-Heq), 2.09 (1H, dt, J11eq,11ax 13.2 and

J11eq,10 3.2 Hz, 11-Heq), 2.15 (1H, ddd, J4ax,4eq 14.5, J4ax,5 3.0 and J4ax,3 3.0 Hz, 4-Hax),

3.46 (1H, t, J5,4 2.9 Hz, 5-H), 3.65-3.72 (4H, m, 8-CH2, 3-H and 2-Heq), 3.82 (1H, dd,

J2ax,2eq 12.1 and J2ax,3 1.4 Hz, 2-Hax), 3.88-3.90 (2H, m, 3-OH and 5-OH); δC (75 MHz;

CDCl3; Me4Si) 18.1, 24.9 (CH2, C-9 and C-10), 30.8 (CH2, C-4), 31.1 (CH2, C-11), 60.9 1 (CH2, C-8), 64.8 (CH2, C-2), 65.3 (CH, C-3), 70.3 (CH, C-5), 96.6 (quat, C-6). The H and 13C NMR spectra were in agreement with that reported in literature.114

8 9 O 7 10 6 1 2 11 O 5 4 3

OH OH 50

(ii) [4S∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-4,5-diol (121) (15 mg, 10%) as a colourless oil; (300 MHz; CDCl3; Me4Si) 1.35-2.01 (6 H, m, 4 x CH2), 2.51 (1 H, d,

JOH,5 6.2 Hz, OH), 2.58 (1 H, d, JOH,4 5.6 Hz, OH), 3.49-3.54 (1 H, m, 5-Heq), 3.58-3.70

(4 H, m, 2 x OCH2), 4.01-4.11 (1 H, m, 4-Hax); δC (75 MHz; CDCl3; Me4Si) 18.1 (CH2,

C-10), 24.9 (CH2, C-9), 28.4 (CH2, C-3), 31.5 (CH2, C-11), 58.2, 60.4 (CH2, C-2 and C-8), 65.7 (CH, C-4), 72.8 (CH, C-5), 97.9 (quat, C-6). The 1H and 13C NMR spectra were in agreement with that reported in literature.114

8 9 O7 10 6 1 2 11 O 5 4 3 OH OH 121

114

EXPERIMENTAL

7.2 Preparation of [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan- 3,5-diyl bis(ethyl p-toluenesulfonate) (146)

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bis-tert- butylacetate (149)

Sodium hydride (4 mg, 0.16 mmol) was added to a solution of [3R∗, 5S∗, 6S∗]-1,7- 3 dioxaspiro[5.5]undecan-3,5-diol (50) (14 mg, 0.07 mmol) in THF (15 cm ) at 0 °C. The reaction mixture was stirred for 30 min at 0 °C then 18-crown-6 was added followed by tert-butyl bromoacetate (148) (29 mg, 0.15 mmol). The reaction mixture was stirred at

0 °C for 1 h then at room temperature for 16 h. After filtration through a short Celite pad the solvent was removed at reduced pressure to yield a pale yellow oil which was purified by flash chromatography using hexane-ethyl acetate (2:1) to afford the title + compound (149) (16 mg, 52%) as a colourless oil [Found: M (EI) 416.24092. C21H36O8 + -1 requires: M , 416.24102]; νmax(film)/cm 1740 (s, C=O) δH (300 MHz; CDCl3; Me4Si)

1.46 (18 H, s, 6 x CH3), 1.53-1.64 (4 H, m, 9-CH2, 10-HA or 10-HB and 11-Hax),

1.67-1.80 (1 H, m, 10-HA or 10-HB) 2.09-2.19 (3 H, m, 4-CH2 and 11-Heq), 3.29 (1 H, t,

J5,4 4.6 Hz, 5-H), 3.53 (1 H, dddd, J3,2ax 3.5, J3,2eq 3.5, J3,4ax 3.5 and J3,4eq 3.5 Hz, 3-H),

3.62-3.82 (4 H, m, 2-CH2 and 8-CH2), 3.99-4.19 (4 H, m, 2 x OCH2CO); δC (75 MHz;

CDCl3; Me4Si) 17.9 (CH2, C-10), 25.2 (CH2, C-9), 27.0 (CH2, C-4), 28.12, 28.14 (CH3,

Me), 28.4 (CH2, C-11), 61.3 (CH2, C-8), 62.4 (CH2, C-2), 66.5, 67.3 (CH2,

2 x OCH2CO), 71.9 (CH, C-3), 76.8 (CH, C-5), 81.3, 81.4 (quat, 2 x OCMe3), 97.4 (quat, C-6), 170.00, 170.01 (quat, 2 x C=O); m/z (EI) 416 (M+, 0.2%), 101 (86), 69 (28), 57 (100), 41 (34).

8 9 O 7 10 6 1 2 11 O 5 4 3 OO

OO OO

149

115

EXPERIMENTAL

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bisallyl ether (133)

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diol (50) (0.22 g, 1.17 mmol) as a solution in tetrahydrofuran (5 cm3) was added to a suspension of sodium hydride (0.08 g, 3.51 mmol) in tetrahydrofuran (10 cm3). The reaction mixture was heated under reflux for 0.5 h, allyl bromide (135) (0.22 cm3, 2.51 mmol) was then added dropwise and the reaction heated under reflux for a further 18 h. After cooling, the reaction was quenched with sodium dihydrogen phosphate (7 cm3, 10% w/v) and extracted with ethyl acetate (5 x 20 cm3). The combined extracts were washed with water (15 cm3) and dried over sodium sulfate. Removal of the solvent under reduced pressure gave a yellow oil that was purified by flash chromatography using ethyl-acetate (4:1) as eluent to afford the + title compound (133) as a pale yellow oil (0.26 g, 84%); [Found: M H (CI, NH3) + -1 269.17533. C15H25O4 requires: M H, 269.17528]; νmax(film)/cm 1646 (w, C=C);

δH (300 MHz; CDCl3; Me4Si) 1.41 (1 H, ddd, J11ax,11eq 13.6, J11ax,10ax 13.6 and

J11ax,10eq 4.3 Hz, 11-Hax), 1.48-1.82 (4 H, m, 9-CH2 and 10-CH2), 1.91-2.02 (2 H, m,

4-CH2), 2.14 (1 H, dt, J11eq,11ax 13.6 and J11eq,10 2.5 Hz, 11-Heq), 3.16 (1H, t, J5,4 4.7 Hz,

5-H), 3.45 (1 H, dddd, J3,2ax 4.1, J3,2eq 4.1, J3,4ax 4.1 and J3,4eq 4.1 Hz, 3-H), 3.60-3.81

(3 H, m, OCH2), 4.13 (1 H, ddt, Jgem 12.9, J1’A,2’ 5.4 and J1’A,3’ 1.5 Hz, 1’-HA), 5.13 (2 H, d, J3’B,2’ 10.3 Hz, 2 x 3’-HB), 5.25 (2 H, ddd, J3’A,2’ 17.2, J3’A,1’A 1.5 and J3’A,1’B 1.5 Hz,

2 x 3’-HA), 5.85-5.97 (2 H, m, 2 x 2’-H); δC (75 MHz; CDCl3; Me4Si) 17.9 (CH2, C-10),

25.2 (CH2, C-9), 27.1 (CH2, C-4), 28.5 (CH2, C-11), 61.1 (CH2, C-8), 62.8 (CH2, C-2),

69.7, 70.6 (CH2, 2 x OCH2), 70.7 (CH, C-3), 76.0 (CH, C-5), 97.4 (quat C-6), 116.54, + 116.55 (CH2, 2 x C-3’), 135.27, 135.29 (CH, 2 x C-2’); m/z (CI, NH3) 269 (M H, 17%), 211 (100), 127 (38), 101 (78), 84 (42), 71 (47).

8 9

7 10 O

6 1 2 11 O 5 4 3 OO 1' 1'

2' 2' 3' 3' 133

116

EXPERIMENTAL

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bis(2- hydroxyethyl) ether (145)

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bisallyl ether (133) (75 mg, 3 0.28 mmol) was dissolved in methanol (8 cm ) and the solution cooled to -78°C. Ozone was bubbled through the solution until a pale blue colour persisted (10 – 15 min). Excess ozone was removed by passing oxygen through the solution. The reaction was taken out of the cooling bath and sodium borohydride (42 mg, 1.12 mmol) was added. The reaction was allowed to warm to room temperature and left to stir for 16 h. The solution was diluted with brine (5 cm3) and the methanol removed. The aqueous solution was extracted with dichloromethane (5 x 15 cm3). The combined extracts were dried over sodium sulfate and the dichloromethane removed under reduced pressure. The pale yellow oil was purified by flash chromatography using 8% MeOH in CH2Cl2 to afford + the title compound (145) (120 mg, 75%) as colourless needles mp 62-65 °C. [Found: M + -1 (EI) 276.15725. C13H24O6 requires: M , 276.15729]; νmax(film)/cm 3628-3290 (br s,

OH); δH (400 MHz; CDCl3) 1.31 (1 H, ddd, J11ax,11eq 13.6, J11ax,10ax 13.6 and

J11ax,10eq 4.5 Hz, 11-Hax), 1.49-1.62 (3 H, m, 9-CH2, 10-HA or 10-HB), 1.73-1.82 (1 H, m,

10-HA or 10-HB), 1.98 (1 H, ddd, J4ax,4eq 15.1, J4ax,5 3.6 and J4ax,3 3.6 Hz, 4-Hax), 2.11-

2.29 (2 H, m, 4-Heq and 11-Heq), 3.11 (1 H, t, J5,4 3.6 Hz, 5-H), 3.37-3.41 (1 H, m, 3-H),

3.47-3.84 (12 H, m, 2-CH2, 8-CH2, 2 x 1’-CH2 and 2 x CH2OH); δC (75 MHz; CDCl3)

18.1 (CH2, C-10), 25.1 (CH2, C-9), 26.0 (CH2, C-4), 31.1 (CH2, C-11), 60.9 (CH2, C-8),

61.0 (CH2, C-2), 61.5, 61.6 (CH2, 2 x C-1’), 70.9, 71.3 (CH2, 2 x CH2OH), 72.2 (CH, C-3), 77.2 (CH, C-5), 96.1 (quat, C-6); m/z (EI) 276 (M+, 0.4%), 158 (10), 101 (41), 88 (95) 73 (42), 69 (23), 45 (100).

8 9 O 7 10 6 1 2 11 O 5 4 3 OO 1' 1'

2' 2' OH HO 145

117

EXPERIMENTAL

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl diacetaldehyde (150)

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bisallyl ether (133) (0.19 g, 3 0.71 mmol) was dissolved in methanol (10 cm ) and the solution cooled to -78°C. Ozone was bubbled through the solution until a pale blue colour persisted (10 – 15 min). Excess ozone was discharged by passing oxygen and nitrogen through the flask. The reaction was taken out of the cooling bath and dimethyl sulfide (0.16 cm3, 2.1 mmol) was added. The reaction was allowed to warm to room temperature and left to stir for 16 h. The volatiles were removed in vacuo to yield an oil that was purified by flash chromatography using 8% methanol in dichloromethane to yield the title compound (150) + as a colourless viscous oil (0.165 g, 86%). [Found: M (EI) 272.12576. C13H20O6 + -1 requires: M , 272.12599]; νmax(film)/cm 1733 (s, C=O); δH (400 MHz; CDCl3) 1.43

(1 H, ddd, J11ax,11eq 13.5, J11ax,10ax 13.5 and J11ax,10eq 4.8 Hz, 11-Hax), 1.51-1.64 (3 H, m,

9-CH2, 10-HA or 10-HB), 1.69-1.83 (1 H, m, 10-HA or 10-HB), 2.04-2.22 (3 H, m, 4-CH2 and 11-Heq), 3.21 (1 H, t, J5,4 3.8 Hz, 5-H), 3.44-3.46 (1 H, m, 3-H), 3.64-3.79 (4 H, m,

2-CH2 and 8-CH2), 3.96-4.28 (4 H, m, 2 x OCH2CHO), 9.68-9.74 (2 H, m, 2 x HC=O);

δC (75 MHz; CDCl3) 17.9 (CH2, C-10), 25.0 (CH2, C-9), 26.5 (CH2, C-4), 29.8 (CH2,

C-11), 61.2 (CH2, C-8), 61.8 (CH2, C-2), 74.5, 75.4 (CH2, 2 x OCH2CHO), 72.7 (CH, C-3), 76.8 (CH, C-5), 96.6 (quat, C-6), 201.1, 201.3 (quat, 2 x C=O); m/z (EI) 272 (M+, 1%), 172 (10), 101 (100), 69 (81), 43 (27).

8 9 O 7 10 6 1 2 11 O 5 4 3 OO

O HHO 150

118

EXPERIMENTAL

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bis(2-p- toluenesulfonyl) ethyl ether (146)

A solution of n-butyllithium (0.41 cm3 of a 1.6 mol dm-1 solution in hexane, 0.66 mmol) was added to a solution of [3R∗, 5S∗, 6S∗]-1,7-dioxaspiro[5.5]undecan-3,5-diyl bis(2- 3 hydroxyethyl) ether (145) in tetrahydrofuran (8 cm ) at -78 °C. The mixture was stirred for 0.5 h. A solution of tosyl chloride (0.15 g, 0.79 mmol) in tetrahydrofuran (5 cm3) was added and the mixture stirred for a further 0.5 h at –78 °C. The reaction was then taken out of the cooling bath, allowed to warm to room temperature and left to stir for 16 h. The reaction was quenched with sodium dihydrogen phosphate solution (5 cm3, 10% w/v) and extracted with ethyl acetate (5 x 20 cm3). The combined organic layers were dried over magnesium sulfate and the solvent removed in vacuo to yield a pale yellow oil. Purification by flash chromatography using hexane-ethyl acetate (3:2) as eluent yielded the title compound (146) as a colourless oil (0.17 g, 90%) [Found: M+H (FAB) + 585.18336. C27H37O10S2 requires: M H, 585.18282]; δH (300 MHz; CDCl3; Me4Si) 1.25

(1 H, ddd, J11ax,11eq 14.3, J11ax,10ax 14.3 and J11ax,10eq 4.3 Hz, 11-Hax), 1.45-1.56 (3 H, m,

9-CH2 and 10-HA or 10-HB), 1.65-1.76 (1 H, m, 10-HA or 10-HB), 1.83-2.02 (3 H, m,

4-CH2 and 11-Heq), 2.44 (6 H, s, 2 x ArCH3), 3.07 (1 H, t, J5,4 3.9 Hz, 5-H), 3.34 (1 H, dddd, J3,2ax 3.2, J3,2eq 3.2, J3,4ax 3.2 and J3,4eq 3.2 Hz, 3-H), 3.56-3.79 (8 H, m, 2-CH2,

8-CH2 and 2 x C-1’), 4.08-4.14 (4 H, m, 2 x CH2OTs), 7.31-7.79 (8 H, m, Ar-H);

δC (75 MHz; CDCl3; Me4Si) 17.9 (CH2, C-10), 21.6 (CH3, Ar-CH3), 25.1 (CH2, C-9),

26.0 (CH2, C-4), 29.6 (CH2, C-11), 61.1 (CH2, C-8), 62.3 (CH2, C-2), 66.5, 67.5 (CH2,

2 x C-1’), 69.5, 69.6 (CH2, 2 x CH2OTs) 72.2 (CH, C-3), 77.6 (CH, C-5), 96.5 (quat, C-6), 127.9, 127.9, 129.83, 129.84 (CH, 4 x Ar) 133.0, 133.0, 144.75, 144.81 (quat, 2 x + Ar-CH3 and 2 x ArSO2); m/z (FAB) 585 (M H, 6%), 369 (45), 199 (27), 154 (100), 136 (72), 91 (33).

119

EXPERIMENTAL

8 9 O7 10

6 1 2 11 O 5 4 3 OO 1' 1'

2' 2' O O O SSO O O

CH3 CH3 146

7.3 Preparation of β-Chloroethyl sulfides (104) and (127) and Dithiols (122) and (123)

3,6,9-Trithiaundecane-1,11-diol (124) and 3,6,9,12- Tetrathiadodecane-1,14-diol (125)

General Procedure. The appropriate dithiol (6.5 mmol) was added to a solution of sodium ethoxide (13.0 mmol) in refluxing ethanol (20 cm3). 2-Chloroethanol (13.0 mmol) was added dropwise and the mixture heated under reflux for 18 h following the procedure reported by Wolf et al.121a The solvent was removed in vacuo and the product recrystallised from acetone to yield the appropriate diol:

(i) 3,6,9-Trithiaundecane-1,11-diol (124). Prepared from 2-mercaptoethyl sulfide (Aldrich) (1.0 g, 6.5 mmol) (99) and 2-chloroethanol (1.1 g, 13.0 mmol) as a white solid + (1.25 g, 80%) mp 96-98 °C (from acetone) [Found: M (EI) 242.04645 C10H22O2S4, + requires: M , 242.04690]; δH [200 MHz; CDCl3] 2.16 (2 H, br s, OH), 2.74-2.80 (12 H, + m, 6 x CH2S), 3.76 (4 H, t, JCH2,OH 5.8 Hz, 2 x CH2OH); m/z (EI) 242 (M , 0.5%), 164 (21), 138 (23), 105 (100), 61 (76), 45 (43). The 1H NMR was in agreement with the published data.121a

120

EXPERIMENTAL

1 2 4 5 7 8 10 11

HOS S S OH 3 6 9 124

(ii) 3,6,9,12-Tetrathiadodecane-1,14-diol (125). Prepared from 3,6-dithiaoctane-1,8- dithiol (1.4 g, 6.5 mmol) (122) and 2-chloroethanol (1.1 g, 13.0 mmol) as a white solid + (1.51 g, 77%) mp 115-117 °C (from acetone) [Found: M (EI) 302.04914. C10H22O2S4, + requires: M , 302.05027]; δH [300 MHz; (CD3)2SO] 2.30 (2 H, b s, OH) 2.74-2.80 (16 H, + m, 8 x CH2S), 3.75 (4 H, t, JCH2,OH 6.0 Hz, 2 x CH2OH); m/z (EI) 302 (M , 0.2%), 224 (10), 198 (22), 164 (31), 105 (100), 61 (52). The 1H NMR was in agreement with the published data.121a

1 2 4 5 7 8 10 11 13 14

HOS S S S OH 3 6 9 12

125

1,8-Dichloro-3,6-dithiaoctane (129), 1,11-Dichloro-3,6,9- trithiaundecane (104) and 1,14-Dichloro-3,6,9,12- tetrathiadodecane (127)

General Procedure. Thionyl chloride (13.6 mmol) was added to the appropriate diol (4.1 mmol) in dichloromethane (10 cm3) at room temperature using well established literature procedures.68,121 The reaction mixture was stirred for 18 h then treated with methanol (2 cm3) to quench the excess thionyl chloride. The resulting solution was evaporated to dryness. The residue was taken up in dichloromethane (10 cm3) and washed with sodium hydrogen carbonate (10 cm3). The organic layer was then dried with sodium sulfate and the solvent removed in vacuo to yield the appropriate dichloride:

(i) 1,8-Dichloro-3,6-dithiaoctane (129). Prepared from 3,6-dithia-1,8-octanediol + (Aldrich) (0.75 g, 4.1 mmol) as a white solid (0.77 g, 85%) mp 55-57 °C [Found: M (EI) + 219.97258. C6H12Cl2S2, requires: M , 219.97280]; δH (200 MHz; CDCl3) 2.80 (4 H, s, 2 x CH2S), 2.90 (4 H, t, JSCH2,CH2Cl 7.7 Hz, 2 x CH2S), 3.61-3.69 (4 H, m, 2 x CH2Cl). The 1H NMR was in agreement with that reported in literature.68 121

EXPERIMENTAL

1 2 4 5 7 8

Cl S S Cl 3 6 129

(ii) 1,11-Dichloro-3,6,9-trithiaundecane (104). Prepared from 3,6,9-trithiaundecane- 1,11-diol (1.0 g, 4.1 mmol) (124) as a white solid (1.04 g, 90%) [Found: M+ (EI) + 279.97616. C8H16Cl2S3, requires: M , 279.97617]; δH (200 MHz; CDCl3) 2.79 (8 H, s,

4 x CH2S), 2.91 (4 H, t, JSCH2,CH2Cl 7.7 Hz, 2 x CH2S), 3.61-3.69 (4 H, m, 2 x CH2Cl). The 1H NMR spectrum was in agreement with that reported in literature.68

1 2 4 5 7 8 10 11

ClS S S Cl 3 6 9 104

(iii) 1,14-Dichloro-3,6,9,12-tetrathiatetradecane (127). Prepared from 3,6,9,12- tetrathiadodecane-1,14-diol (1.24 g, 4.1 mmol) (125) as a white solid (1.18 g, 85%) + + [Found: M (EI) 339.97882. C10H20Cl2S4, requires: M , 339.97954]; δH (200 MHz;

CDCl3) 2.78 (12 H, s, 6 x CH2S), 2.91 (4 H, t, JSCH2,CH2Cl 7.7 Hz, 2 x CH2S), 3.61-3.69 1 68 (4 H, m, 2 x CH2Cl). The H NMR was in agreement with that reported in literature.

1 2 4 5 7 8 10 11 13 14

ClS S S S Cl 3 6 9 12

127

4,7-Dithiaoctane-1,10-dithiyldiacetate (130) and 4,7,10- Trithiaundecane-1,13-dithiyldiacetate (131)

General Procedure. Cesium carbonate (0.33 g, 1.0 mmol) was added in small portions to a solution of thiolacetic acid (0.15 g, 2.0 mmol) in methanol (5 cm3). The mixture was stirred for 30 min and then evaporated to dryness. The white solid obtained was then

122

EXPERIMENTAL dissolved in dimethylformamide (5 cm3) and the requisite dichloride (1.0 mmol) in dimethylformamide (1 cm3) was added. The mixture was stirred for 16 h at room temperature. The solvent was removed under reduced pressure and the residue was extracted with diethyl ether (3 x 20 cm3). The combined extracts were washed with brine (15 cm3) and dried over sodium sulfate. The solvent was removed in vacuo to yield the appropriate thioester:

(i) 4,7-Dithiaoctane-1,10-dithiyldiacetate (130). Prepared from 1,8-dichloro-3,6- dithiaoctane (0.22 g, 1.0 mmol) (129) as a pale orange solid (0.26 g, 86%) mp 58-60 °C + + -1 [Found: M (EI) 298.01821. C10H18O2S4 requires: M , 298.01897]; νmax(film)/cm 1685

(s, C=O); δH (400 MHz; CDCl3) 2.33 (6 H, s, 2 x CH3C=O), 2.68-2.72 (4 H, m, 2 x

CH2S), 2.81 (4 H, s, 2 x CH2S), 3.04-3.08 (4 H, m, 2 x CH2SC=O); δC (100 MHz;

CDCl3) 29.3 (CH2, 2 x CH2SC=O), 30.6 (2 x CH3C=O), 31.8, 32.0 (CH2, 4 x CH2S), 195.3 (2 x C=O); m/z (EI) 298 (M+, 2%), 222 (20), 103 (25), 43 (100).

OO 2 35 68 9

1 10 H3CCSSSS H3 4 7 130

(ii) 4,7,10-Trithiaundecane-1,13-dithiyldiacetate (131) was prepared from 1,11- dichloro-3,6,9-trithiaundecane (0.28 g, 1.0 mmol) (104) as a pale orange solid (0.28 g, + + 79%) mp 80-83 °C [Found: M (EI) 358.02229. C12H22O2S5 requires: M , 358.02234]; -1 νmax(film)/cm 1691 (s, C=O); δH (400 MHz; CDCl3) 2.32 (6 H, s, 2 x CH3C=O), 2.68-

2.71 (4 H, m, 2 x CH2S), 2.80 (8 H, s, 4 x CH2S), 3.03-3.07 (4 H, m, 2 x CH2SC=O);

δC (100 MHz; CDCl3) 29.3 (CH2, 2 x CH2SC=O), 30.7 (2 x CH3C=O), 31.8, 32.1, 32.2 + (CH2, 6 x CH2S), 195.3 (2 x C=O); m/z (EI) 358 (M , 1.5%), 222 (6.5), 162 (22), 103 (38), 43 (100).

OO 2 35 6 8 9 11 12

1 13 H3CCSSS SSH3 4 7 10 131

123

EXPERIMENTAL

3,6-Dithiaoctane-1,8-dithiol (122) and 3,6,9-Trithiaundecane-1,11- dithiol (123)

General Procedure. Lithium aluminium hydride (0.17 g, 4.5 mmol) was added to a solution of the appropriate thiolate (1.8 mmol) in diethyl ether using the procedure described by Edema et al.122 Excess lithium aluminium hydride was quenched with saturated ammonium chloride (10 cm3). The reaction mixture was extracted with ethyl acetate (3 x 20 cm3), washed with water (15 cm3) and dried over sodium sulfate. The solvent was removed under reduced pressure and the product purified by flash chromatography using dichloromethane as eluent to yield the appropriate dithiol:

(i) 3,6-Dithiaoctane-1,8-dithiol (122). Prepared from 4,7-dithiaoctane-1,10- dithiyldiacetate (0.5 g, 1.8 mmol) (130) as a white solid (0.34 g, 89%) [Found: M+ (EI) + 21399764. C12H22O2S5 requires: M , 213.99784]; δH (200 MHz; CDCl3) 1.69-1.77 (2 H, 1 m, SH), 2.71-2.80 (12 H, m, CH2S). The H NMR spectrum was in agreement with that reported in literature.68

1 2 4 5 7 8

HS S S SH 3 6 122

(ii) 3,6,9-Trithiaundecane-1,11-dithiol (123). Prepared from 4,7,10-Trithiaundecane- 1,13-dithiyldiacetate (0.64 g, 1.8 mmol) (131) as a white solid (0.34 g, 68%) [Found: M+ + (EI) 274.00046. C8H18S5 requires: M , 274.00121]; δH (200 MHz; CDCl3) 1.68-1.77 (2 + H, m, SH), 2.67-2.81 (16 H, m, CH2S); m/z (EI) 274 (M , 0.7%), 120 (39), 61 (100).

1 2 4 5 7 8 10 11

HSSSSSH 3 6 9 123

124

EXPERIMENTAL

3,6,9-Trithiaundeca-1,10-diene (132)

Potassium hydride (0.18 g, 30% dispersion in mineral oil, 1.34 mmol) was added to a solution of [3R*,5S*,6S*]-1,7-dioxaspiro[5.5]undecane-3,5-diol (50) (0.12 g, 0.64 mmol) in tetrahydrofuran (40 cm3) and the resulting solution was heated under reflux for 30 min. A solution of 1,11-dichloro-3,6,9-trithiaundecane (104) (0.21 g, 0.77 mmol) in tetrahydrofuran (10 cm3) was added dropwise over 3 h and the reaction mixture heated under reflux for a further 20 h. The solution was then cooled and filtered through a short pad of Celite. The solvent was removed under reduced pressure and the residue purified by flash chromatography using hexane-ethyl acetate (4:1) as eluent to afford the title + compound (132) (0.12 g, 90%) as a pale yellow oil [Found: M H (CI, NH3) 207.03352. + C8H15S3 requires: M H, 207.03359]; δH (300 MHz; CDCl3; Me4Si) 2.79-2.94 (8 H, m, 4 x CH2S), 5.17 (2 H, d, JCHB,CH 16.7 Hz, HCHB=CH), 5.26 (2 H, d, JCHA,CH 10.1 Hz,

HCHA=CH), 6.33 (2 H, dd, JCHB,CH 16.7, JCHA,CH 10.1 Hz, CH2=CH); δC (75 MHz;

CDCl3; Me4Si) 31.60, 31.64 (CH2, CH2S), 112.1 (CH2, 2 x CH2=CH), 131.3 (CH, 2 x

CH2=CH).

HA 1 2 4 5 7 8 10 11

HB SSS 3 6 9 132

7.4 The Synthesis of Spiroacetal Thiacrown Ethers

[1S∗, 15R∗, 18S∗]-Spiro[2,14,17-trioxa-5,8,11- trithiabicyclo[13.3.1]nonadecane-18,2’-tetrahydropyran] (55)

A solution of the spiroacetal ditosylate (146) (100 mg, 0.17 mmol) in dimethylformamide (5 cm3) and a solution of the 2-mercaptoethyl sulfide (99) (220 mg, 0.17 mmol) in dimethylformamide (5 cm3) were added from separate addition funnels over 2.5 h to a vigorously stirred suspension of cesium carbonate (170 mg, 0.51 mmol) in 3 dimethylformamide (20 cm ) at 60 °C. The mixture was left to stir for 16 h. The reaction mixture was then filtered through a short pad of Celite and the filter cake washed with

125

EXPERIMENTAL dichloromethane (3 x 15 cm3). The solvent was removed under reduced pressure to yield a yellow oil that was purified by flash chromatography using hexane-ethyl acetate (4:1) as eluent to afford the title compound (55) (58 mg, 86%) as a pale yellow oil [Found: M+ + (EI) 394.13104. C17H30O4S3 requires: M , 394.13063]; δH (400 MHz; CDCl3) 1.26 (1 H, ddd, J3’ax,3’eq 13.6, J3’ax,4’ax 13.6 and J3’ax,4’eq 4.4 Hz, 3’-Hax), 1.51-1.57 (3 H, m, 5’-CH2 and 4’-Ha or 4’-HB), 1.73-1.81 (1 H, m, 4’-HA or 4’-HB), 1.91 (1 H, ddd, J19ax,19eq 15.2,

J19ax,15 3.4 and J19ax,1 3.4, Hz, 19-Hax), 2.15-2.25 (2 H, m, 19-Heq and 3’-Heq), 2.72-

2.82 (12 H, m, 6 x CH2S), 3.09 (1 H, t, J1,19 3.4 Hz, 1-H), 3.37 (1 H, m, 15-H), 3.41

(1 H, ddd, JA,B 6.8, J3A,4A 9.0 and J3A,4B 9.0 Hz, 3-HA), 3.55-3.68 (4 H, m, 16-CH2 and

13-CH2), 3.72-3.82 (3 H, m, 3-HB and 6’-CH2); δC (100 MHz; CDCl3) 18.0 (CH2, C-4’),

24.3 (CH2, C-19), 25.2 (CH2, C-5’), 31.2 (CH2, C-3’), 31.4, 31.8, 32.3, 32.4, 33.2, 33.5

(CH2, 6 x CH2S), 60.9 (CH2, C-6’), 62.2 (CH2, C-16), 70.1 (CH2, C-13), 70.3 (CH2, C-3), 72.0 (CH, C-15), 77.2 (CH, C-1), 96.6 (quat, C-18); m/z (EI) 394 (M+, 23%), 120 (54), 103 (57), 87 (100), 61 (76), 41 (46).

6' 5' O 1' 4' 182' 17 16 3' O 1 19 15 2 OO14 3 13

4 12 5 SS11

6 S 10 8 7 9 55

[1S∗, 18R∗, 21S∗]-Spiro[2,17,20-trioxa-5,8,11,14- tetrathiabicyclo[16.3.1]docosane-21,2’-tetrahydropyran] (56)

The title compound (56) was prepared from the spiroacetal ditosylate (146) (140 mg, 0.24 mmol), 3,6-dithiaoctane-1,8-dithiol (122) (50 mg, 0.24 mmol) and cesium carbonate (230 mg, 0.72 mmol) using a similar procedure to that described above for crown ether (55). The crude product was purified by flash chromatography using hexane- ethyl acetate (4:1) as eluent to afford the title compound (56) (75 mg, 68%) as a + + colourless oil [Found: M (EI) 454.13397. C19H34O4S4 requires: M , 454.13400]; δH (400

MHz; CDCl3) 1.32 (1 H, ddd, J3’ax,3’eq 13.6, J3’ax,4’ax 13.6 and J3’ax,4’eq 4.4 Hz, 3’-Hax),

126

EXPERIMENTAL

1.49-1.60 (3 H, m, 5’-CH2 and 4’-HA or 4’-HB), 1.71-1.79 (1 H, m, 4’-HA or 4’-HB), 1.97

(1 H, ddd, J22ax,22eq 14.8, J22ax,18 3.7 and J22ax,1 3.7 Hz, 22-Hax), 2.07 (1 H, dddd,

J22eq,22ax 14.8, J22eq,1 3.7, J22eq,18 3.7 and J22eq,19eq 1.9 Hz, 22-Heq), 2.12 (1 H, dt,

J3’eq,3’ax 13.6 and J3’eq,4’ 2.8 Hz, 3’-Heq), 2.72-2.82 (16 H, m, 8 x CH2S), 3.09 (1 H, t,

J1,22 3.7 Hz, 1-H), 3.37 (1 H, dddd, J18eq,22ax 3.7, J18eq,22eq 3.7, J18eq,19ax 3.7 and

J18eq,19eq 3.7 Hz, 18-H), 3.55-3.78 (8 H, m, 6’-CH2, 19-CH2, 16-CH2 and 3-CH2);

δC (100 MHz; CDCl3) 17.9 (CH2, C-4’), 25.2 (CH2, C-5’), 26.4 (CH2, C-22), 30.2 (CH2,

C-3’), 32.01, 32.04, 32.41, 32.46, 32.49, 32.52, 33.15, 33.18 (CH2, 8 x CH2S), 61.0 (CH2,

C-6’), 62.1 (CH2, C-19), 69.2 (CH2, C-16), 70.5 (CH2, C-3), 71.9 (CH, C-18), 77.2 (CH, C-1), 96.5 (quat, C-21); m/z (EI) 454 (M+, 19%), 131 (42), 120 (71), 87 (100), 61 (74), 41 (35).

6' 5'

1' 4' O 2' 21 20 19 3' O 1 22 18

2 17 3 O O 16

4 15

5 S S 14

6 13

7 S S 12 8 11 910 56

[1S∗, 21R∗, 24S∗]-Spiro[2,20,23-trioxa-5,8,11,14,17- pentathiabicyclo[19.3.1]pentacosane-24,2’-tetrahydropyran] (57)

The title compound (57) was prepared from the spiroacetal ditosylate (146) (0.27 g, 0.46 mmol), 3,6,9-Trithiaundecane-1,11-dithiol (123) (0.13 g, 0.46 mmol) and cesium carbonate (0.45 g, 1.4 mmol) using a similar procedure to that described above for crown ether (55). The crude product was purified by flash chromatography using hexane-ethyl acetate (4:1) as eluent to afford the title compound (57) (0.15 g, 64 %) as a pale yellow + + oil [Found: M (EI) 514.13764. C21H38O4S5 requires: M , 514.13737]; δH (400 MHz;

CDCl3) 1.32 (1 H, ddd, J3’ax,3’eq 13.3, J3’ax,4’ax 13.3 and J3’ax,4’eq 4.3 Hz, 3’-Hax), 1.46-

1.60 (3 H, m, 5’-CH2 and 4’-HA or 4’-HB), 1.66-1.82 (1 H, m, 4’-HA or 4’-HB), 1.95

(1 H, ddd, J25ax,25eq 14.8, J25ax,21 3.8 and J25ax,1 3.8 Hz, 25-Hax), 2.05-2.14 (2 H, m, 127

EXPERIMENTAL

3’-Heq and 25-Heq), 2.69-2.84 (20 H, m, 10 x CH2S), 3.08 (1 H, t, J1,25 3.8 Hz, 1-H),

3.53 (1 H, br s, 21-H), 3.55-3.79 (8 H, m, 6’-CH2, 22-CH2, 3-CH2 and 19-CH2);

δC (100 MHz; CDCl3) 17.9 (CH2, C-4’), 25.2 (CH2, C-5’), 26.1 (CH2, C-25), 30.3 (CH2,

C-3’), 31.9, 32.1, 32.30, 32.32, 32.4, 32.5, 32.6, 32.7, 32.9, 33.0 (CH2, 10 x CH2S), 61.0

(CH2, C-6’), 62.0 (CH2, C-22), 69.2 (CH2, C-19), 70.1 (CH2, C-3), 71.9 (CH, C-21), 77.1 (CH, C-1), 96.5 (quat, C-24); m/z (EI) 514 (M+, 13%), 120 (72), 105 (54), 87 (100), 61 (72), 41 (32).

6' 5'

O 1' 4' 2' 24 23 22 3' O 1 25 21 2 OO20 3 19

4 18 5 SS17

6 16 7 15

8 S S 10 S 12 14 9 13 11 57

7.5 The Synthesis of Spiroacetal Azacrown Ethers

1 3 5 N ,N ,N -Tris(p-tosylsulfonyl)diethylenetriamine (157)

A solution of p-toluenesulfonyl chloride (2.77 g, 14.5 mmol) in ether (12 cm3) was added dropwise to a vigorously stirred solution of diethylenetriamine (151) (0.50 g, 4.8 mmol) and NaOH (0.58 g, 14.5 mmol) in water (5 cm3). The mixture was then stirred for 1 h at room temperature then heated at 65 °C for 3 h. The precipitate, collected by filtration, was then suspended in EtOH and refluxed for 4 h to afford a white crystalline + + solid (2.21 g, 80%) [Found: M H (EI) 566.14504. C25H31N3O6S3 requires: M H,

566.14533]; δH (300 MHz; CDCl3; Me4Si) 2.50 (9 H, s, 3 x CH3), 2.91 (4 H, m, 2 x

CH2NH), 3.12 (4 H, t, JCH2N,CH2NH 6.7 Hz, 2 x CH2N), 7.46-7.80 (12 H, m, Ar); m/z (FAB) 566 (M+H, 21%), 154 (100), 136 (72). The 1H NMR was in agreement with that reported in the literature.184

128

EXPERIMENTAL

O 124 5 O 3 H3C S NH N HN S CH3 O O S O O

CH3 157

N1,N3,N5-Tris(2-nitrobenzenesulfonyl)-1,5-diamino-3-azapentane (158)

A solution of diethylenetriamine (151) (0.25 g, 2.4 mmol) and triethylamine (0.5 cm3, 3.9 mmol) was added to a stirred solution of 2-nitrobenzenesulfonyl chloride 3 (1.61 g, 7.7 mmol) in dichloromethane (20 cm ) at 0 °C. The reaction mixture was stirred for 20 h. The solvent was removed in vacuo and the residue dissolved in H2O-CHCl3.

The organic layer was separated and the aqueous layer extracted with CHCl3 (3 x 10 3 3 cm ), washed with saturated NaHCO3 (15 cm ) and dried over Na2SO4. Removal of the solvent and purification by flash chromatography using hexane-ethyl acetate (3:2) afforded the title compound (158) (1.27 g, 80%) as a yellow solid mp 70-74 °C [Found: + + -1 M H (FAB) 659.05478. C22H23N6O12S3 requires: M H, 659.05361]; νmax(film)/cm 3323

(w, NH); δH (400 MHz; CDCl3) 3.33 (4 H, m, 2 x CH2NH), 3.54 (4 H, t, JCH2N,CH2NH 6.1

Hz, 2 x CH2N), 5.72 (2 H, t, JNH,CH2 6.1 Hz, NH); δC (100 MHz; CDCl3) 42.3 (C-1), 49.0 (C-2), 124.5, 124.7, 125.6, 131.0, 131.4, 132.1, 132.6, 133.0, 133.9, 134.5, 135.9 (18 x Ar); m/z (FAB) 659 (M+H, 10%), 154 (100), 136 (73).

O2N O 124 5 O 3 S NH N HN S O O S O O NO2 NO2

158

129

EXPERIMENTAL

[1S∗, 15R∗, 18S∗]-Spiro[2,14,17-trioxa-5,8,11-tris(2-nitrobenzenesulfonyl)-5,8,11-triazabicyclo[13.3.1]nonadecane-18,2’- tetrahydropyran] (159)

To a solution of the Ns-protected triamine (158) (46 mg, 0.07 mmol) in tetrahydrofuran (20 cm3) was added sodium hydride (5 mg, 0.17 mmol) and the resulting mixture heated under reflux for 30 min. A solution of [3R∗, 5S∗, 6S∗]-1,7- dioxaspiro[5.5]undecan-3,5-diyl bis(2-p-toluenesulfonyl) ethyl ether (146) (41 mg, 0.07 mmol) in tetrahydrofuran (5 cm3) was added over 3 h and the reaction mixture was heated under reflux for a further 20 h. The solvent was removed under reduced pressure to yield a tan oil which was purified by flash chromatography using ethyl acetate-hexane (4:1) as eluent to afford the title compound (159) (17 mg, 27%) as a colourless oil + + [Found: M H (FAB) 899.18910. C35H43N6O16S3 requires: M H, 899.18977];

δH (300 MHz; CDCl3; Me4Si) 1.30-1.34 (1 H, m, 3’-Hax), 1.51-1.59 (4 H, m, 4’-CH2 and

5’-CH2), 1.96 (1 H, dt, J11eq,11ax 13.5 and J11eq,10 2.1 Hz, 3’-Heq), 2.03-2.07 (2 H, m,

19-CH2), 3.03 (1 H, t, J1,19 3.0 Hz, 1-H), 3.24 (1 H, br s, 15-H), 3.38-3.88 (20 H, m,

6’-CH2, 16-CH2, 3-CH2, 13-CH2 and 6 x CH2N), 7.61-8.11 (12 H, m, Ar); δC (75 MHz;

CDCl3; Me4Si) 18.2 (CH2, C-4’), 25.1 (CH2, C-5’), 27.5 (CH2, C-19), 31.6 (CH2, C-3’),

48.4, 48.6, 49.8, 49.9, 50.9, 51.0 (CH2, 6 x CH2N), 60.5 (CH2, C-6’), 60.9 (CH2, C-16),

71.1 (CH2, C-13), 71.9 (CH2, C-3), 72.0 (CH, C-15), 77.2 (CH, C-1), 96.1 (quat, C-18), 124.2, 124.4, 131.2, 131.5, 132.0, 132.0, 132.1, 132.2, 132.4, 133.6, 133.7, 133.8 (18 x Ar); m/z (EI) 899 (M+H, 0.6%), 368 (11), 194 (57), 181 (60), 167 (91), 101 (85), 71 (100).

6' 5' O 1' 4' 182' 17 16 3' O 1 19 15 2 OO14 O2N 3 13 O O 4 12 5 S S NN11 8 6 O O 7 N 9 10 O SO NO2

NO2

159

130

EXPERIMENTAL

[1S∗, 15R∗, 18S∗]-Spiro[2,14,17-trioxa-5,8,11- triazabicyclo[13.3.1]nonadecane-18,2’-tetrahydropyran] (58)

Thiophenol (5 mg, 0.05 mmol) was added to a stirred mixture of the [1S∗, 15R∗, 18S∗]-spiro[2,14,17-trioxa-5,8,11-tris(2-nitro-benzenesulfonyl)-5,8,11- triazabicyclo[13.3.1]nonadecane-18,2’-tetrahydropyran] (159) (10 mg, 0.01 mmol) and 3 K2CO3 (21 mg, 0.15 mmol) in DMF (5 cm ). The resulting solution was stirred at room temperature for 18 h. The solvent was removed under reduced pressure to yield a tan residue that was redissolved in H2O-CHCl3 (3:5). The organic phase was separated and 3 the aqueous layer extracted with CHCl3 (5 x 5 cm ). The combined organic layers were 3 washed with brine (5 cm ), dried over Na2SO4 and the solvent concentrated. The residue was purified by flash chromatography using CH2Cl2:MeOH (20:1), then CH2Cl2:MeOH-

30% NH4OH (20:1) as eluent to afford the title compound (58) (3.2 mg, 84%) as a + + viscous tan oil [Found: M H (FAB) 344.25423. C17H34N3O4 requires: M H, 344.25493];

δH (300 MHz; CDCl3; Me4Si) 1.32-1.36 (1 H, m, 3’-Hax), 1.55-1.70 (4 H, m, 4’-CH2 and

5’-CH2), 1.96 (1 H, m, 3’-Heq), 2.17 (2 H, m, 19-CH2), 3.04 (1 H, m, 1-H), 3.29 (1 H, + br s, 15-H), 3.55-3.84 (8 H, m, 3-CH2, 6’-CH2, 13-CH2 and 16-CH2); m/z (EI) 344 (M H, 2%), 299 (52), 287 (73), 226 (61), 153 (58), 99 (68), 56 (100), 44 (75). 13C NMR data was not acquired for this compound as there was insufficient material to give a satisfactory signal to noise ratio.

6' 5' O 1' 4' 2' 18 17 16 3' O 1 19 15 2 OO14 3 13

4 12

5NH11 NH 8 6 N 10 H 7 9 58

131

EXPERIMENTAL

7.6 Olefin Metathesis of the Spiroacetal Bisallyl Ether (133)

Bis[1,7-dioxaspiro[5.5]undec-3,5-diyl 2-buten-1,4-diyl] ether (142)

(a) Using Grubbs’ First Generation Ruthenium Catalyst

[3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl bisallyl ether (133) (89 mg, 0.33mmol) in dichloromethane (10 cm3) was slowly added via syringe or cannula to a solution of benzylidene-bis(tricyclohexylphosphine)- dichlororuthenium (137) (Aldrich) (14 mg, 5 mol%) in dichloromethane (20 cm3) over 2 h at reflux. The reaction mixture was heated under reflux for 6 h. The solvent was then removed in vacuo and the brown residue was purified by flash chromatography using hexane-ethyl acetate (3:2) as eluent to afford three diastereomers of bis[1,7-dioxaspiro[5.5]undec-3,5-diyl 2-buten-1,4-diyl] ether (142) for which the stereochemistry was not defined:

8 9 O7 10 6 1 2 11 O 5 4 3 OO

4' 1'

3' 2' 142a 142b 2' 3' 142c 1' 4' OO

3 4 5 O 11 2 1 6 10 7O 9 8

+ (i) Dimer (142a) (12 mg, 15%) as fine needles mp 164-168 °C [Found: M (EI) + 480.27238. C26H40O8 requires: M , 480.27232]; δH (300 MHz; CDCl3; Me4Si) 1.28 (2 H, ddd, J11ax,11eq 13.4, J11ax,10ax 13.4 and J11ax,10eq 4.3 Hz, 11-Hax), 1.51-1.64 (8 H, m, 9-CH2 and 10-CH2), 1.85 (2 H, ddd, J4ax,4eq 15.0, J4ax,5 3.4 and J4ax,3 3.4 Hz, 4-Hax), 2.18-2.24 (4

H, m, 11-Heq and 4-Heq), 3.05 (2 H, t, J5,4 3.4 Hz, 5-H), 3.32 (2 H, br s, 3-H), 3.60-3.81

(8 H, m, 2-CH2 and 8-CH2), 3.86-4.06 (8 H, m, 4 x CH2CH=CH), 5.89 (4 H, br s, 4 x

132

EXPERIMENTAL

CHCH2O); δC (75 MHz; CDCl3; Me4Si) 18.1 (CH2, C-10), 24.7 (CH2, C-4), 25.3 (CH2,

C-9), 31.0 (CH2, C-11), 60.8 (CH2, C-8), 62.3 (CH2, C-2), 69.1, 70.1 (CH2, 4 x

CH2CH=CH), 71.3 (CH, C-3), 76.2 (CH, C-5), 96.3 (quat, C-6), 129.0, 129.2 (CH, 4 x + CHCH2O); m/z (EI) 480 (M , 2.6%), 171 (33), 153 (41), 111 (43), 101 (100), 69 (64), 54 (58), 41 (63).

+ (ii) Dimer (142b) (32 mg, 40%) as a viscous oil [Found: M (EI) 480.27275. C26H40O8 + requires: M , 480.27232]; δH (300 MHz; CDCl3; Me4Si) 1.25-1.33 (2H, m, 11-Hax),

1.49-1.59 (8 H, m, 9-CH2 and 10-CH2), 1.86 (2 H, dt, J4ax,4eq 14.9, J4ax,3 3.4 and J4ax,5 3.4 Hz, 4-Hax), 2.17-2.24 (4 H, m, 4-Heq and 11-Heq), 3.07 (2 H, t, 3.4 Hz, 5-H), 3.31 (2 H, br s, 3-H), 3.60-3.83 (8 H, m, 2-CH2 and 8-CH2), 3.84-4.07 (8 H, m, 4 x CH2CH=CH),

5.89 (4 H, t, JCH,OCH2 2.8 Hz, 4 x CHCH2O); δC (75 MHz; CDCl3; Me4Si) 18.1 (CH2,

C-10), 24.7 (CH2, C-4), 25.3 (CH2, C-9), 31.0 (CH2, C-11), 60.8 (CH2, C-8), 62.4 (CH2,

C-2), 68.9, 70.3 (CH2, 4 x CH2CH=CH), 71.0 (CH, C-3), 76.4 (CH, C-5), 96.4 (quat, + C-6), 129.1, 129.2 (CH, 4 x CHCH2O); m/z (EI) 480 (M , 2%), 171 (34), 153 (52), 111 (48), 101 (100), 85 (40), 69 (83), 54 (75), 41 (99).

+ (iii) Dimer (142c) (8 mg, 10%) as a viscous oil [Found: M (EI) 480.27248. C26H40O8 + requires: M , 480.27232]; δH (300 MHz; CDCl3; Me4Si) 1.20-1.39 (2 H, m, 11-Hax),

1.47-1.63 (8 H, m, 9-CH2 and 10-CH2), 1.95 (2 H, ddd, J4ax,4eq 14.6, J4ax,5 3.7 and

J4ax,3 3.7 Hz, 4-Hax), 2.05-2.20 (2 H, m, 4-Heq and 11-Heq), 3.09 (2 H, t, J5,4 3.7 Hz,

5-H), 3.35 (2 H, m, 3-H), 3.61-3.78 (8 H, m, 2-CH2 and 8-CH2), 3.84-4.10 (8 H, m, 4 x

CH2CH=CH), 5.89 (2 H, t, JCH,OCH2 2.5 Hz, CHCH2O) 5.91 (2 H, t, JCH,OCH2 2.5 Hz,

CHCH2O); δC (75 MHz; CDCl3; Me4Si) 18.1 (CH2, C-10), 25.3 (CH2, C-4), 26.6 (CH2,

C-9), 30.0 (CH2, C-11), 60.9 (CH2, C-8), 61.8 (CH2, C-2), 68.4, 69.9 (CH2, 4 x

CH2CH=CH), 70.7 (CH, C-3), 76.1 (CH, C-5), 96.9 (quat, C-6), 128.4, 128.9 (CH, 4 x + CHCH2O); m/z (EI) 480 (M , 2%), 171 (31), 153 (51), 101 (96), 69 (83), 55 (73), 41 (100).

(b) Using Shrock’s Molybdenum Catalyst

A solution of 2,6-diisopropylphenylimidoneophylidenemolybdenum(VI) bis(hexafluoro-t-butoxide) (136) (Strem Chemicals) (17 mg, 25 mol%) in benzene (5 cm3) was added to a solution of [3R∗, 5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undecan-3,5-diyl

133

EXPERIMENTAL bisallyl ether (133) (25 mg, 0.09 mmol) in benzene (5 cm3) at room temperature. The reaction mixture was left to stir for 6 h. The solvent was then removed in vacuo and the brown residue was purified by flash chromatography using hexane-ethyl acetate (3:2) as eluent to afford dimer (142a) (4 mg, 18%), dimer (142b) (10 mg, 45%) and dimer (142c) (2 mg, 9%) for which the spectroscopic data was in agreement with that reported above.

7.7 Kinetic Resolution of Spiroacetal α-Epoxide (155) and Spiroacetal Allylic Alcohol (117)

7.7.1 Base-Induced Rearrangement of α-Epoxide (155)

[5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-3-en-5-ol (117)

General Procedure. n-Butyllithium (0.21 cm3 of a 1.6 mol dm-3 solution in hexane, 0.36 mmol) was added dropwise to a solution of the appropriate chiral non-racemic base 3 3 (0.36 mmol) in hexane (8 cm ) at –55 °C or in tetrahydrofuran at 0 °C (8 cm ) and the suspension was left to stir for 30 min. [4S∗, 5S∗, 6S∗]-4,5-Epoxy-1,7- dioxaspiro[5.5]undecane (115) (56 mg, 0.33 mmol) was added to the mixture as a solution in hexane (5 cm3). The solution was allowed to warm slowly (ca 3 h) to room temperature and then left to stir for a further 18 h. The reaction mixture was then quenched with sodium dihydrogen phosphate solution (5 cm3, 10% w/v) and extracted with ethyl acetate (3 x 10 cm3). The combined extracts were washed with water (10 cm3) and dried over sodium sulfate. Removal of the solvent under reduced pressure yielded an orange oil that was purified by flash chromatography using hexane-ethyl acetate (3:2) as eluent to afford [5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-3-en-5-ol (117). The 1H NMR spectrum was in agreement with that previously reported.114

* * (a) [R-(R ,R )]-(+)-bis(α-methylbenzyl)amine (170) (81 mg, 0.36 mmol) in hexane (10 -3 cm ) at –50 °C to yield (117) as colourless needles (14 mg, 59%, based on recovered starting material, 32 mg).

-3 (b) (-)-sparteine (171) (84 mg, 0.36 mmol) in hexane (8 cm ) at –55 °C to yield (117) as colourless needles (15 mg, 48%, based on recovered starting material, 25 mg).

134

EXPERIMENTAL

(c) (1S,2R)-(+)-norephedrine (172) (54 mg, 0.36 mmol), n-butyllithium (0.43 cm3 of a -3 3 1.6 mol dm solution in hexane, 0.69 mmol) in tetrahydrofuran (8 cm ) at 0 °C as colourless needles (4 mg, 21%, based on recovered starting material, 36 mg).

[5S∗, 6S∗]-1,7-Dioxaspiro[5.5]undec-3-en-5-yl (R)-α-methoxy-α- (trifluoromethyl)phenylacetate (176)

3 Triethylamine (0.015 cm , 0.11 mmol) and (S)-α-methoxy-α- (trifluoromethyl)phenylacetyl chloride (0.015 cm3, 0.084 mmol) were added to a solution of [5S∗, 6S∗]-1,7-dioxaspiro[5.5]undec-3-en-5-ol (117) (12 mg, 0.07 mmol) and 4-(dimethylamino)pyridine (34 mg, 0.28 mmol) in dichloromethane (1 cm3). The reaction mixture was left to stir for 16 h at room temperature. Removal of the solvent under reduced pressure gave a brown oil that was purified by flash chromatography using ethyl- acetate (4:1) as eluent to afford a tan oil (24 mg, 88%) as a 1:1 mixture of diastereomers + + [Found: M H (CI, NH3) 387.14255. C19H21F3O5 requires: M , 387.14193]; -1 νmax(film)/cm 1790 (C=O); δH (400 MHz; CDCl3) 1.18 (0.5 H, ddd, J11ax,11eq 13.1 and

J11ax,10ax 13.1, J11ax,10eq 4.1 Hz, 11-Hax), 1.40 (0.5 H, ddd, J11ax,11eq 13.1, J11ax,10ax 13.1 and

J11ax,10eq 4.7 Hz, 11-Hax), 1.43-1.91 (6 H, m, 9-CH2, 10-CH2, and 11-CH2), 3.49

(1.5 H, q, JOMe,CF3 1.1 Hz, OMe), 3.55 (1.5 H, q, JOMe,CF3 1.1 Hz, OMe), 3.60-3.80 (2 H, m, 8-CH2), 4.10-4.16 (2 H, m, 2-CH2), 4.94-4.99 (1 H, m, 5-H), 5.87-5.99 (1 H, m, 4-H),

6.05 (0.5 H, dt, J3,4 10.2 and J3,2 2.6 Hz, 3-H), 6.12 (0.5 H, dt, J3,4 10.2 and J3,2 2.6 Hz,

3-H), 7.35-7.39 (3 H, m, Ar), 7.48-7.54 (2 H, m, Ar); δC (75 MHz; CDCl3) 18.1, 18.2

(CH2, C-10), 24.7, 24.8 (CH2, C-9), 30.3, 30.5 (CH2, C-11), 55.4, 55.5 (CH3, OMe), 60.1,

60.2 (CH2, C-2), 62.8, 62.8 (CH2, C-8), 70.3, 70.6 (CH, C-5), 95.1, 95.2 (quat, C-6),

119.2, 119.3 (CH, Ar), 121.3,121.4 (quat, C-CF3) 127.3, 127.6 (CF3), 128.3, 128.4 (CH, Ar), 129.5, 129.6 (CH, Ar), 132.0, 132.3 (quat, Ar), 165.9, 166.1 (quat, C=O); + δF (300 MHz; CDCl3) -72.9 (CF3), -72.7 (CF3); m/z (CI, NH3) 387 (M H, 4%), 189 (31), 153 (100).

135

EXPERIMENTAL

8 9 O 7 10 6 1 2 11 O 5 3 4 O O OMe F3CPh 176

7.7.2 Jacobsen Hydrolytic Kinetic Resolution of α-Epoxide (115)

(a) Preparation of the Activated Catalyst

A mixture of (R,R)-N,N’-bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediaminocobalt(II) (Aldrich) (50 mg, 0.08 mmol) and acetic acid (10 mg, 0.16 mmol) in toluene (1 cm3) was stirred open to the atmosphere at room temperature for 1 h using the procedure by Jacobsen et al.185 Removal of the solvent under reduced pressure yielded (R,R)-N,N’-bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediaminocobalt(III) acetate (177) as a brown solid (50 mg, 91%) that was used without further purification.

(b) Reaction using the Activated Catalyst (177)

(R,R)-N,N’-Bis(3,5-di-tert-butylsalicylidene)-1,2-cyclohexanediaminocobalt(III) acetate

(177) (2 mg, 3.0 µmol, 0.5 mol%) was added to a mixture of α-epoxide (115) (100 mg, 0.60 mmol) and water (5.0 mg, 0.30 mmol) in tetrahydrofuran (0.5 cm3) at room temperature. The mixture was allowed to stir for 18 h. The solvent was removed at reduced pressure and the residue purified by flash chromatography using hexane-ethyl acetate (4:1) as eluent to afford [4R∗, 5R∗, 6S∗]-1,7-dioxaspiro[5.5]undecan-4,5-diol (179b) (32 mg, 55%, based on recovered starting material, 47 mg) as colourless needles 173 mp 134-136 °C (lit., mp 135-137 °C); δH (300 MHz; CDCl3; Me4Si) 1.47 (1 H, ddd,

J11ax,11eq 13.5, J11ax,10ax 13.5 and J11ax,10eq 4.4 Hz, 11-Hax), 1.53-1.85 (5 H, m, 9-CH2,

10-CH2 and 3-Hax), 1.91 (1 H, dt, J11eq,11ax 13.5 and J11eq, 10 2.7 Hz, 11-Heq), 2.05-2.14

(1 H, m, 3-Heq), 3.41 (1 H, dd, J5,4 7.4 and J5,OH 3.7 Hz, 5-H), 3.62 (1 H, ddd,

J2eq,2ax 12.0, J2eq,3ax 5.5 and J2eq,3eq 2.3 Hz, 2-Heq), 3.71-4.00 (6 H, m, 8-CH2, 4-H, 2-Hax,

136

EXPERIMENTAL

4-OH, 5-OH); δC (75 MHz; CDCl3; Me4Si) 17.6, 24.8, 27.6, 30.6 (CH2, C-3, C-9, C-10 and C-11), 55.6, 61.3 (CH2, C-2 and C-8), 68.6 (CH, C-4), 71.1 (CH, C-5), 98.4 (quat, C-6). The 1H and 13C NMR were in agreement with published data.173 The enantiomeric excess was determined to be 1:1 mixture of the two diastereomers by conversion of the recovered starting material (115) to the allylic alcohol (117) and its subsequent conversion to the Mosher ester (176). The 1H, 13C and 19F NMR spectra of the Mosher derivative (176) were in agreement with that previously discussed.

8 9 O 7 10 6 11 O 1 HO 5 2

4 3 OH 179b

7.7.3 Sharpless Epoxidation of Allylic Alcohol (117)

Titanium tetraisopropoxide (28 mg, 0.10 mmol) was added to a solution of diisopropyl D-tartrate (34 mg, 0.15 mmol), allylic alcohol (117) (166mg, 0.98 mmol) and 3 powdered 4Å molecular sieves (25 mg) in dichloromethane (15 cm ) at –20 °C and the mixture stirred for 30 min following the procedure by Sharpless et al.165a Anhydrous tert- butyl hydroperoxide solution in dichloromethane (2.4 M, 0.28 cm3, 0.68 mmol) was then added and the reaction stirred at –20 °C for 5 h. The reaction mixture was allowed to stand for a further 18 h in the freezer at –20 °C. The cold reaction mixture was quenched with 10% NaOH solution in brine (1 cm3). Ether (5 cm3) was added to the mixture and the solution was allowed to warm to room temperature, whereupon MgSO4 (1.5 g) and Celite (0.14 g) were added. The mixture was stirred for 15 min, diluted with dichloromethane and filtered through a short pad of Celite. Removal of the solvent under reduced pressure and purification by flash chromatography using hexane-ethyl acetate (3:2) as eluent afforded the syn-epoxy alcohol (119) (55 mg, 77%, based on recovered starting material, 101 mg) as colourless prisms. The 1H NMR was in agreement with published data.114

137

EXPERIMENTAL

[3S∗, 4S∗, 5S∗, 6S∗]-3,4-Epoxy-1,7-dioxaspiro[5.5]undecan-5-yl (R)-

α-methoxy-α-(trifluoromethyl)phenylacetate (183)

3 Triethylamine (0.03 cm , 0.2mmol) and (S)-(+)-α-methoxy-α- (trifluromethyl)phenylacetyl chloride (0.07 cm3, 0.29 mmol) were added to a solution of [3S∗, 4S∗, 5S∗, 6S∗]-3,4-epoxy-1,7-dioxaspiro[5.5]undecan-5-ol (119) (25 mg, 0.13 mmol) and 4-(dimethylamino)pyridine (66 mg, 0.54 mmol) in dichloromethane (1 cm3). The reaction mixture was left to stir for 16 h at room temperature. Removal of the solvent under reduced pressure gave a brown oil that was purified by flash chromatography using ethyl-acetate (4:1) as eluent to afford a tan oil (50 mg, 93%) as a 1:1 mixture of + + diastereomers [Found: M (EI) 403.13675. C19H22F3O6 requires: M , 403.13685]; -1 νmax(film)/cm 1790 (C=O); δH (400 MHz; CDCl3) 1.03 (0.5 H, ddd, J11ax,11eq 14.0,

J11ax,10ax 14.0 and J11ax,10eq 4.4 Hz, 11-Hax), 1.31 (0.5 H, ddd, J11ax,11eq 13.2, J11ax,10ax 13.2 and J11ax,10eq 4.8 Hz, 11-Hax), 1.46-1.74 (4.5 H, m, 2 x 9-CH2, 2 x 10-CH2, and 11-Heq),

1.84 (0.5 H, dt, J11eq,11ax 14.0 and J11eq,10 2.8 Hz, 11-Heq), 3.24 (0.5 H, dt, J3,4 4.0 and

J3,2 1.2 Hz, 3-H), 3.29 (0.5 H, dt, J3,4 4.0 and J3,2 1.2 Hz, 3-H), 3.51 (1.5 H, q, JOMe,CF3 1.0

Hz, OMe), 3.53-3.60 (1 H, m, 8-CH2), 3.64 (1.5 H, q, JOMe,CF3 1.0 Hz, OMe), 3.68-3.74

(2 H, m, 2 x 4-H and 8-CH2), 3.83 (1 H, ddd, J2eq,2ax 13.2, J2eq,3 1.2 and J2eq,4 1.2 Hz, 2 x

2-Heq), 3.97 (0.5 H, d, J2ax,2eq 13.2 Hz, 2-Hax), 4.00 (0.5 H, d, J2ax,2eq 13.2 Hz, 2-Hax),

4.71 (0.5 H, d, J5,4 4.8 Hz, 5-H), 4.73 (0.5 H, d, J5,4 4.8 Hz, 5-H), 7.36-7.41 (2 H, m, Ar),

7.57-7.65 (4 H, m, Ar); δC (75 MHz; CDCl3) 17.9, 18.0 (CH2, C-10), 24.5, 24.6 (CH2,

C-9), 29.6, 30.0 (CH2, C-11), 47.9, 48.0 (CH, C-4), 50.0, 50.1 (CH, C-3), 55.5, 55.7

(CH3, OMe), 57.0, 57.1 (CH2, C-2), 61.7, 61.8 (CH2, C-8), 71.2, 71.3 (CH, C-5), 93.7,

93.6 (quat, C-6), 127.5, 127.9 (CF3), 128.3, 128.4 (CH, Ar), 129.6, 129.7 (CH, Ar),

131.6, 132.1 (quat, Ar), 166.1, 166.3 (quat, C=O); δF (300 MHz; CDCl3) -73.4 (CF3), + -72.9 (CF3); m/z (CI, NH3) 403 (M , 20%), 189 (100), 169 (60), 114 (47).

8 9 O 7 10 6 1 2 11 O 5 3 4 O O O OMe F3CPh 183

138

EXPERIMENTAL

6.8 Determination of the Association Constants

Lithium, potassium, sodium, cesium, lead, cobalt and cadmium picrates were prepared from picric acid and the appropriate carbonate in distilled water following the procedure by Silberrad et al.186 Silver picrate was prepared from picric acid and silver oxide in distilled water using the procedure described by Silberrad et al.186 The picrate salts were recrystallised twice from distilled water and dried in vacuo at room temperature. Association constants were determined using the ultraviolet spectroscopic method developed by Cram et al.148 Solutions of the picrate salts were prepared in distilled water at a concentration of 0.015 mol dm-3. Solutions of the hosts were prepared in 2.0 cm3 volumetric flasks as 0.075 mol dm-3 solutions in chloroform. General procedure. A solution of the host (0.2 cm3) was added to a solution of the guest (0.5 cm3) in a 2 cm3 centrifuge tube. The tubes were stoppered, shaken for 2 minutes and centrifuged for 2 minutes. An aliquot of the chloroform layer (0.15 cm3) was transferred to a 50 cm3 volumetric flask and diluted to the mark with acetonitrile. An appropriate blank was also prepared. Ultraviolet measurements were carried out at 380 nm at 22 °C. Calculations were based on the Beer’s law relationship, a = εbc, where a is the absorbance, ε is the extinction coefficient, b is the path length of the cell and c is the concentration of the measured species. The total number of millimoles of picrate salt extracted into the chloroform layer could be determined from Beer’s law. The millimoles of the host was calculated from the initial host concentration and the aliquot volume. The guest to host molar ration (R) was given by the millimoles of picrate salt divided by the millimoles of host. The extraction constant (Ke) could be calculated from Equation 3 using the experimentally determined values for R.

R Ke = (1-R){[Guest] - R[Host] (V /V )}2 Equation 3 i i organic aq

R = molar ratio of guest to host in the chloroform layer determined by UV.

[Guest]i = initial concentration of the guest in the aqueous layer.

[Host]i = initial concentration of the host in the chloroform layer.

Vorganic and Vaq are the volumes of the organic and aqueous phases respectively.

139

EXPERIMENTAL

The association constant (Ka) is defined by Equation 1 and the distribution constant (Kd) by Equation 4.

K Host Guesta Host.Guest Complex Equation 1 org + aq organic CDCl3

Kd Guest Guest Equation 4 aq organic

Ka is then related to Ke and Kd by Equation 5

Ka = Ke/Kd Equation 5

148 The Kd values were determined using the procedure described by Cram. Picrate solutions (20 cm3, 0.015 M) were shaken in a separatory funnel with chloroform (30 cm3). The layers were allowed to separate and clarify. The lower chloroform layer was carefully transferred to a flask and the solvent was removed in vacuo. The residue was quantitatively transferred with acetonitrile to a volumetric flask (10 cm3) and diluted to the mark with acetonitrile. The amount of picrate salt extracted was calculated using the UV techniques discussed above.

The Gibbs free energy could then be calculated using Equation 2.

∆G = -RT ln(Ka) Equation 2

140

EXPERIMENTAL

A sample calculation for spiroacetal thiacrown (55) and silver picrate is illustrated below.

-1 [Guest]i = 0.015 mol L -1 [Host]i = 0.075 mol L -1 -1 ε (silver picrate) = 16111 cm mol L Sample absorbance at 380 nm = 1.64 Sample volume = 50 mL Aliquot volume = 0.15 mL

Vorganic = 0.5 mL

Vaqueous = 0.2 mL + -1 Kd (Ag .picrate) = 0.108 M mmols of Ag+.picrate in host.guest complex =

Sample absorbance x sample volume (mL) x Vorganic (mL) -1 -1 ε (silver picrate) (cm mol L) x aliquot volume (mL) = 1.64 x 50 x 0.2 16111 x 0.15 = 6.786 x 10-3

mmol of host (55) in the organic phase = 0.075 mmol mL-1 x 0.2 mL = 0.015 mmol

R = 6.786 x 10-3 0.015 = 0.4524

From Equation 3:

Ke = 0.4524 (1 – 0.4524){0.015 – (0.452 x 0.075 x 0.2/0.5)}2 = 397.77 x 103

141

EXPERIMENTAL

From Equation 5:

Ka = Ke/Kd

= 397770 0.108 = 3683.06 x 103

From Equation 2:

∆G = -RT ln(Ka) = -8.3145 x 295 x ln 3683.06 x 103 = -37.08 kJ mol-1

6.9 Second-Sphere Complexation

General procedure. [Al(acac)3] (164), [Co(NH3)5NO2](BPh4)2 (165) or

[Co(en)3](BPh4)3 (166) (0.011 mmol ) were added to a solution of deuterated chloroform or dimethylsulfoxide (0.8 cm3) containing crown compounds (55), (56), (57), (64) or (100) (0.011 mmol). The solution was shaken for 5 minutes and the 1H NMR spectrum 1 was measured at 25 °C. The H NMR spectrum was monitored for 1 h.

142

REFERENCES

1 G. W. Gokel, Crown Ethers and Cryptands, in Monographs in Supramolecular Chemistry, ed. J. F. Stoddart, The Royal Society of Chemistry, Cambridge, England, 1991.

2 J-M. Lehn, in Supramolecular Chemistry, Concepts and Perspectives, VCH, Weinheim, Federal Republic of Germany, 1995.

3 C. J. Pedersen, J. Am. Chem. Soc., 1967, 89, 2495, 7017.

4 A. Lüttringhaus, K. Ziegler, Ann., 1937, 528, 155.

5 (a) R. G. Ackman, W. H. Brown, G. F. Wright, J. Org. Chem., 1955, 20, 1147. (b) R. E. Beals, W. H. Brown, J. Org. Chem., 1956, 21, 447. (c) W. H. Brown, W. N. French, Can. J. Chem., 1958, 36, 537

6 D. G. Stewart, D. Y. Wadden, E. T. Borrows, Brit. Patent, 1957, 229, 785.

7 J. Down, J. Lewis, B. Moore, G. Wilkinson, Proc. Chem. Soc., 1957, 209.

8 (a) E. P. Kyba, R. C. Helgeson, K. Madan, G. W. Gokel, T. L. Tarnowski, S. S. Moore, D. J. Cram, J. Am. Chem. Soc., 1977, 99, 2564. (b) J. M. Timko, S. S. Moore, D. M. Walba, P. C. Hiberty, D. J. Cram, J. Am. Chem. Soc., 1977, 99, 4207. (c) M. Newcomb, J. M. Timko, D. M. Walba, D. J. Cram, J. Am. Chem. Soc., 1977, 99, 6392.

9 (a) Host-Guest Complex I, ed. F. Vogtle, Springer-Verlag, Berlin, 1981. (b) Host-Guest Complex II, ed. F. Vogtle, Springer-Verlag, Berlin, 1982. (c) Host-Guest Complex III, ed. F. Vogtle, Springer-Verlag, Berlin, 1984.

10 Chemistry of Peptides and Proteins, ed. W. Voelter, E. Bayer, Y. A. Ovchinnikov, V. T. Ivanov, Walter de Gruyter, Berlin, vol. 3, 1986.

11 Macrolide Antibiotics: Chemistry, Biology and Practice, ed. S. Omura, Academic Press, New York, 1984.

143

REFERENCES

12 H. Kantekin, Ö. Hasançebi, R. Abbasoglu, Y. Gök, New. J. Chem., 2001, 25, 879.

13 L. F. Lindoy, D. H. Busch, Complexes of Macrocyclic Ligands, in Preparative Inorganic Reactions, ed. W. L. Jolly, Wiley-Interscience, New York, 1971, vol. 6, p. 4.

14 X. X. Zhang, J. S. Bradshaw, R. M. Izatt, Chem Rev., 1997, 97, 3313.

15 A. Agtrap, J. W. Chamberlin, M. Pinkerton, L. Steinrauf, J. Am. Chem. Soc., 1967, 89, 5737.

16 E. J. Corey, K. C. Nicolaou, S. L. Melvin, J. Am. Chem. Soc., 1975, 97, 653.

17 (a) D. French, Adv. Carbohydr. Chem., 1957, 12, 189. (b) R. J. Clarke, J. H. Coates, S. F. Lincoln, Adv. Carbohydr. Chem. Biochem., 1988, 46, 205.

18 (a) T. Ogawa, Y. Takahashi, Carbohydr. Res., 1985, 138, C5. (b) T. Ogawa, Y. Takahashi, Carbohydr. Res., 1987, 164, 277. (c) T. Ogawa, Y. Takahashi, Carbohydr. Res., 1987, 169, 127.

19 (a) P. R. Ashton, P. Ellwood, I. Stanton, J. F. Stoddart, Angew. Chem. Int. Ed. Engl., 1991, 30, 80. (b) P. R. Ashton, P. Ellwood, I. Stanton, J. F. Stoddart, J. Org. Chem., 1991, 56, 7274. (c) A. Gadelle, J. Defaye, Angew. Chem. Int. Ed. Engl., 1991, 30, 78.

20 (a) J. F. Stoddart, Chem. Soc. Rev., 1979, 8, 85. (b) X. X. Zhang, J. S. Bradshaw, R. M. Izatt, Chem. Rev., 1997, 97, 3313.

144

REFERENCES

21 (a) D. J. Cram, G. D. Y. Sogah, J. Chem. Soc. Chem. Commun., 1981, 625. (b) D. J. Cram, G. D. Y. Sogah, J. Chem. Soc. Chem. Commun., 1985, 107, 8301. (c) M. Alonso-Lopéz, M. Martin-Lomas, S. Penadés, Tetrahedron Lett., 1986, 27, 3551. (d) M. Alonso-Lopéz, J. Jiminez-Barbera, M. Martin- Lopez, S. Penadés, Tetrahedron, 1988, 44, 1535. (e) S. Aoki, S. Sasaki, K. Koga, Tetrahedron Lett., 1989, 30, 7229. (f) D. A. H. van Maarschalkerwaart, N. P. Willard, U. K. Pandit, Tetrahedron, 1992, 48, 8825.

22 L. R. Sousa, G. D. I. Sogath, D. H. Hoffman, D. J. Cram, J. Am. Chem. Soc., 1978, 100, 4569.

23 (a) J-P. Joly, N. Moll, J. Chromatogr., 1990, 521, 134. (b) J-P. Joly, M. Nazhaoui, B. Dumont, Bull. Soc. Chim. Fr., 1994, 131, 369. (c) J-P. Joly, B. Gross, Tetrahedron Lett., 1989, 30, 4231.

24 (a) R. C. Helgeson, J. M. Timko, P. Moreau, S. C. Peacock, J. M. Mayer, D. J. Cram, J. Am. Chem. Soc., 1974, 96, 6762. (b) W. D. Curtis, D. A. Laidler, J. F. Stoddart, G. H. Jones, J. Chem. Soc. Chem. Commun., 1975, 835. (c) T. J. Wenzel, J. E. Thurston, D. C. Sek, J-P. Joly, Tetrahedron: Asymmetry, 2001, 12, 1125.

25 R. M Izatt, C. Y. Zhu, J. S. Bradshaw, Enantiomeric Recognition in Macrocycle-Primary Ammonium Cation Systems, in Crown Compounds: Towards Future Applications, ed. S. R. Cooper, VCH, New York, 1992, p 207.

26 E. B. Kyba, K. Koga, L. R. Sousa, M. G. Siegel, D. J. Cram, J. Am. Chem. Soc, 1973, 95, 2692.

27 W. D. Curtis, D. A. Laidler, J. F. Stoddart, G. H. Jones, J. Chem. Soc. Chem. Commun., 1975, 833.

28 W. D. Curtis, R. M. King, J. F. Stoddart, G. H. Jones, J. Chem. Soc. Chem. Commun., 1976, 284.

145

REFERENCES

29 D. A. Laidler, J. F. Stoddart, J. B. Wolstenholme, Tetrahedron Lett., 1979, 465.

30 M. Pietraszkiewicz, M. Kozbial, J. Inclusion Phenom. Mol. Recognit. Chem., 1992, 14, 339.

31 M. Pietraszkiewicz, M. Kozbial, O. Pietraszkiewicz, J. Membr. Sci., 1998, 138, 109.

32 L. Tõke, P. Bakó, G. M. Keserű, M. Albert, L. Fenichel, Tetrahedron, 1998, 54, 213.

33 (a) N. S. Mani, P. P. Kanakamma, Tetrahedron Lett., 1994, 35, 3629. (b) P. P. Kanakamma, N. S. Mani, V. Nair, Synth. Commun.,1995, 25, 3777.

34 R. Miethchen, V. Fehring, Synthesis, 1997, 94.

35 C. G. Espínola, R. Pérez, J. Martín, Org. Lett., 2000, 2, 3161.

36 V. Böhmer, Angew. Chem. Int. Ed. Engl., 1995, 34, 713.

37 I. Bitter, E. Kőszegi, A. Grün, P. Bakó, K. Pál, A. Grofcsik, M. Kubinyi, B. Balázs, G. Tóth, Tetrahedron: Asymmetry, 2003, 14, 1025.

38 C. Vicent, C. Bosso, F. H. Cano, J. L. G. de Paz, C. Foces-Foces, J. Jiménez- Barbero, M. Martin-Lomas, S. Penadés, J. Org. Chem., 1991, 56, 3614.

39 (a) B. Dumont-Hornebeck, J-P Joly, J. Coulon, Y. Chapleur, Carbohydr. Res., 1999, 320, 147. (b) B. Dumont-Hornebeck, J-P Joly, J. Coulon, Y. Chapleur, Carbohydr. Res., 1999, 321, 214.

40 D. Gehin, P. Di Cesare, B. Gross, J. Org. Chem., 1986, 51, 1906.

41 R. Miethchen, G. Torsten, Chem. Ber., 1993, 126, 2309.

146

REFERENCES

42 G. V. M. Sharma, V. G. Reddy, P. R. Krishna, Tetrahedron: Asymmetry, 1999, 10, 3777.

43 F. Perron, K. F. Albizati, Chem. Rev., 1989, 89, 1617.

44 (a) A. J. Kirby, in The Anomeric and Related Stereoelectronic Effects at Oxygen, Springer-Verlag, New York, 1983. (b) E. Juaristi, G. Cuevas, Tetrahedron, 1992, 48, 5019. (c) P. P. Graczyk, M. Mikolajczyk, Top. Stereochem., 1994, 21, 159.

45 (a) W. Y. Lee, C. H. Park, S. S. Kim, J. Am. Chem. Soc., 1993, 115, 1184. (b) W. Y. Lee, C. H. Park, J. Org. Chem., 1993, 58, 7149.

46 C. Garcia, Y. Pointud, G. Jeminet, D. Dugat, J. L. Beltran, Tetrahedron: Asymmetry, 1998, 9, 4253.

47 G. J. McGarvey, M. W. Stepanian, A. R. Bressette, M. Sabat, Org. Lett., 2000, 2, 3453.

48 M. A. Brimble, A. D. Johnston, J. Chem. Soc., Perkin Trans. 1, 1998, 265.

49 B. Dietrich, P. Viout, J-M. Lehn, in Macrocyclic Chemistry, Aspects of Organic and Inorganic Supramolecular Chemistry, ed. K. van-der Saal, VCH, Weinheim, Federal Republic of Germany, 1993.

50 C. Galli, G. Illuminati, L. Mandolini, P. Tamborra, J. Am. Chem. Soc., 1977, 99, 2591.

51 A. J. Kirby, Adv. Phys. Org. Chem., 1980, 17, 183.

52 J. Dale, Tetrahedron, 1974, 30, 1683

53 B. Dietrich, J. M. Lehn, J. P. Sauvage, J. Blanzat, Tetrahedron, 1973, 29, 1629.

147

REFERENCES

54 (a) W. Baker, J. F. W. McOmie, W. D. Ollis, J. Chem. Soc., 1951, 200. (b) G. Illuminati, L. Mandolini, B. Masci, J. Am. Chem. Soc., 1975, 97, 4960. (c) G. Illuminati, L. Mandolini, B. Masci, J. Am. Chem. Soc., 1974, 96, 1422. (d) L. Mandolini, B. Masci, S. Rolens, J. Org. Chem., 1977, 42, 3733.

55 (a) J. Buter, R. M. Kellogg, J. Chem. Soc. Chem. Commun., 1980, 466. (b) J. Buter, R. M. Kellogg, J. Org. Chem., 1981, 46, 4481.

56 (a) A. Ostrowicki, F. Vögtle, E. Koepp, Top. Curr. Chem., 1992, 161, 37. (b) C. Galli, Org. Prep. Proced. Int., 1992, 24, 287.

57 J. S. Bradshaw, K. E. Krakowiak, R. M. Izatt, in Aza-Crown Macrocycles, Wiley-Interscience, New York, 1993.

58 (a) R. D. Hancock, A. E. Martell, Chem. Rev., 1989, 89, 1875. (b) R. D. Hancock, J. Chem. Educ., 1992, 69, 615. (c) B. P. Hay, J. R. Rustad, C. J. Hostetler, J. Am. Chem. Soc., 1993, 115, 11158.

59 F. Vögtle, E. Weber, J. L. Toner, I. Goldberg, D. A. Laidler, J. F. Stoddart, R. A. Bartsch, C. L. Liotta, Crown Ethers and Analogs, in Updates from the Chemistry of the Functional Groups, ed. S. Patai, Z. Rappoport, Wiley- Interscience, New York, 1989, p 207.

60 (a) F. P. van Remoortere, F. P. Boer, Inorg. Chem., 1974, 13, 2071. (b) F. P. Boer, M. A. Neuman, F. P. van Remoortere, E. C. Steiner, Inorg. Chem., 1974, 13, 2826.

61 (a) P.K. Weiner, S. Profeta, G. Wipff, T. Havel, I. D. Kuntz, R. Langridge, P. A. Kollman, Tetrahedron, 1983, 39, 1113. (b) P. Groth, Acta Chem. Scand., 1971, 25, 3189. (c) P. Seiler, M. Dobler, J. D. Dunitz, Acta Crystallogr., 1974, B30, 2744. (d) M. Dobler, R. P. Phizacherly, J. D. Dunitz, Acta Crystallogr., 1974, B30, 2746.

62 (a) J. D. Dunitz, M. Dobler, P. Seiler, R. P. Phizacherly, Acta Crystallogr., 1974, B30, 2733. (b) J. D. Dunitz, P. Seiler, Acta Crystallogr., 1974, B30, 2739.

148

REFERENCES

63 M. Hiraoka, Crown Compounds, Their Characteristics and Applications, in Studies in Organic Chemistry, Kodasha, Tokyo, vol. 12, 1982.

64 J. J. Christensen, D. J. Eatough, R. M. Izatt, Chem Rev., 1974, 74, 351.

65 (a) S. R. Cooper, Acc. Chem. Res., 1988, 21, 141. (b) A. J. Blake, M. Schroder, Adv. Inorg. Chem., 1990, 35, 1.

66 (a) G. F. Smith, D. W. Margerum, J. Chem. Soc., Chem Commun., 1975, 807. (b) L. S. Sokol, L. A. Ochrymowycz, D. B. Rorabacher, Inorg. Chem., 1981, 20, 3189. (c) R. D. Bach, H. B. Vardhan, J. Org. Chem., 1986, 51, 1609.

67 (a) R. E. DeSimone, M. D. Glick, J. Am Chem. Soc., 1976, 98, 762. (b) N. K. Dalley, J. S. Smith, S. B. Matheson, K. L. Larson, J. J. Christensen, R. M. Izatt, J. Chem. Soc., Chem. Commun., 1975, 85. (c) N. K. Dalley, J. S. Smith, S. B. Matheson, K. L. Larson, J. J. Christensen, R. M. Izatt, J. Heterocycl. Chem., 1981, 18, 463.

68 R. E. Wolf, J. R. Hartman, B. M. Foxman, S. R. Cooper, J. Am. Chem. Soc., 1987, 109, 4328.

69 (a) R. E. DeSimone, M. D. Glick, J. Am. Chem. Soc., 1975, 97, 942. (b) R. E. DeSimone, M. D. Glick, J. Coord. Chem., 1976, 5, 181.

70 A. Baeyer, Chem. Ber., 1886, 19, 2184.

71 V. J. Alphen, Recl. Trav. Chim., 1937, 56, 343.

72 A. Braun, J. Tcherniac, Chem. Ber., 1907, 40, 2709.

73 (a) M. C. Thompson, D. H Busch, J. Am. Chem. Soc., 1964, 84, 1762. (b) M. C. Thompson, D. H Busch, J. Am. Chem. Soc., 1964, 86, 213. (c) M. C. Thompson, D. H Busch, J. Am. Chem. Soc., 1964, 86, 3651.

74 T. J. Hurley, M. A. Robinson, S. J. Trotz, Inorg. Chem., 1967, 6, 389.

149

REFERENCES

75 (a) N. F. Curtis, J. Chem. Soc., 1960, 4409. (b) N. F. Curtis, D. A. House, Chem. Ind., 1961, 1708.

76 (a) L. F. Lindoy, in The Chemistry of Macrocyclic Ligand Complexes, Cambridge University Press, 1989, p. 48. (b) P. G. Owsten, R. Peters, E. Ramsammy, P. A. Tasker, J. Trotter, J. Chem. Soc., Chem. Commun., 1980, 1218.

77 S. C. Jackels, E. K. Barefield, N. J. Rose, D. H. Busch, Inorg. Chem., 1972, 11, 2893.

78 D. Parker, Macrocycle Synthesis, in The Practical Approach in Chemistry Series, ed. L. M. Harwood, C. J. Moody, Oxford University Press, New York, 1996.

79 J. E. Richman, T. J. Atkins, J. Am. Chem. Soc., 1972, 96, 2268.

80 (a) J. Dale, P. O. Kristiansen, J. Chem Soc. Commun., 1971, 670. (b) J. Dale, P. O. Kristiansen, Acta Chem. Scand., 1972, 26, 1471.

81 H. Stetter, E. E. Roos, Chem. Ber., 1954, 87, 566.

82 H. Koyama, T. Yoshino, Bull. Chem. Soc. Jpn., 1972, 45, 481.

83 F Chavez, A. Sherry, J. Org. Chem., 1989, 54, 2990.

84 (a) B. L. Shaw, J. Am. Chem. Soc., 1975, 97, 3856. (b) W. Rasshofer, F. Voegtle, Justus Liebigs Ann. Chem., 1978, 552.

85 G. W. Gokel, D. M. Dishong, R. A. Schultz, V. J. Gatto, Synthesis, 1982, 997.

150

REFERENCES

86 (a) C. J. Lockhart, A. C. Robson, M. E. Thompson, S. D. Furtado, C. K. Kaura, A. R. Allan, J. Chem. Soc., Perkin Trans. 1, 1973, 577. (b) C. J. Lockhart, M. E. Thompson, J. Chem. Soc., Perkin Trans. 1, 1977, 202

87 (a) T. Fukuyama, C-K. Jow, M. Cheung, Tetrahedron Lett., 1995, 36, 6373. (b) T. Fukuyama, M. Cheung, C-K. Jow, Y. Hidai, T. Kan, Tetrahedron Lett., 1997, 38, 5831.

88 P. G. Wuts, J. M. Northuis, Tetrahedron Lett., 1998, 39, 3889.

89 (a) L. Qian, Z. Sun, M. Mertes, K. Mertes, J. Org. Chem., 1991, 56, 4904. (b) L. Qian, Z. Sun, K. Mertes, J. Gao, B. Movassagh, L. Morales, J. Org. Chem., 23, 155.

90 (a) A. P. King, C. G. Krespan, J. Org. Chem., 1974, 39, 1315. (b) J. A. Pratt, I. O. Sutherland, R. F. Newton, J. Chem. Soc., Perkin Trans. 1, 1988, 13.

91 (a) I. Tabushi, H. Okino, Y. Kuroda, Tetrahedron Lett., 1976, 4339. (b) I. Tabushi, Y. Taniguchi, H. Kato, Tetrahedron Lett., 1977, 547.

92 H. Stetter, J. Marx, Liebigs Ann. Chem., 1957, 59.

93 B. Dietrich, J-M. Lehn, J. P. Sauvage, Tetrahedron Lett., 1969, 2885.

94 B. Dietrich, J-M. Lehn, J. Guilhem, C. Pascard, Tetrahedron Lett., 1989, 30, 4125.

95 E. Kimura, R. Machida, M. Kodama, Inorg. Chem., 1983, 22, 2055.

96 L. Qian, Z. Sun, T. Deffo, K. B. Mertes, Tetrahedron Lett., 1990, 31, 6469.

97 (a) F. Vellacio, R. V. Punzar, D. S. Kemp, Tetrahedron Lett., 1977, 547. (b) K. E. Krakowiak, J. S. Bradshaw, N. K. Dalley, W. Jiang, R. M. Izatt, Tetrahedron Lett., 1989, 30, 2897.

151

REFERENCES

98 (a) J. S. Bradshaw, K. E. Krakowiak, G. Wu, R. M. Izatt, Tetrahedron Lett., 1988, 29, 5589. (b) J. S. Bradshaw, K. E. Krakowiak, R. M. Izatt, Tetrahedron Lett., 1989, 30, 803. (c) K. E. Krakowiak, J. S. Bradshaw, R. M. Izatt, J. Heterocyclic Chem., 1990, 27, 1585. (d) K. E. Krakowiak, J. S. Bradshaw, N. K. Dalley, W. Jiang, G. Wu, R. M. Izatt, J. Org. Chem., 1991, 56, 2675.

99 N. B. Tucker, E. E. Reid, J. Am. Chem. Soc., 1933, 55, 775.

100 P. C. Ray, J. Chem. Soc., 1920, 117, 1090.

101 G. M. Bennett, W. A. Berry, J. Chem. Soc., 1925, 127, 910.

102 L. A. Ochrymowycz, D. Gerber, R. Chongsawangvirod, A. K. Leung, J. Org. Chem., 1977, 42, 2644.

103 J. R. Meadow, E. E. Reid, J. Am. Chem. Soc., 1934, 56, 2177.

104 D. St. C. Black, I. A. McLean, Tetrahedron Lett., 1969, 45, 3961.

105 W. Rosen, D. H. Busch, J. Am. Chem. Soc., 1969, 91, 4694.

106 J. S. Bradshaw, J. Y. Hui, Y. Chan, B. L. Haymore, R. M. Izatt, J. J. Christensen, J. Heterocycl. Chem., 1974, 11, 45.

107 L. A. Ochrymowycz, C. P. Mak, J. D. Michna, J. Org. Chem., 1974, 39, 2079.

108 (a) J. Buter, R. M. Kellogg, J. Chem. Soc., Chem. Commun., 1980, 466. (b) J. Buter, R. M. Kellogg, J. Org. Chem., 1981, 46, 4481.

109 W. H. Kruizinga, R. M. Kellogg, J. Am. Chem. Soc., 1981, 103, 5183.

110 G. Dijkstra, W. H. Kruizinga, R. M. Kellogg, J. Org. Chem., 1987, 52, 4230.

111 C. Galli, L. Mandolini, J. Org. Chem., 1991, 56, 3045.

152

REFERENCES

112 D. Sellman, L. J. Zapf, J. Organomet. Chem., 1985, 289, 57.

113 (a) J. J. Edema, H. T. Stock, J. Buter, R. M. Kellogg, W. J. J. Smeets, A. L. Spek, Tetrahedron, 1993, 49, 4355. (b) J. J. Edema, H. T. Stock, J. Buter, R. M. Kellogg, W. J. J. Smeets, A. L. Spek, F. Van Bolhuis, Angew. Chem. Int. Ed. Engl., 1993, 32, 436.

114 M. A. Brimble, A. D. Johnston, R. H. Furneaux, J. Org. Chem., 1998, 63, 471.

115 M. A. Brimble, M. K. Edmonds, G. M. Williams, Tetrahedron, 1992, 48, 6455.

116 W. Adams, J. Bialas, L. Hadjiorapogloui, Chem. Ber., 1991, 124, 2377.

117 R. Curci, M. Fiorentino, L. Troisi, J. O. Edwards, R. H. Pater, J. Org. Chem., 1980, 45, 4758.

118 A. Waldemar, L. Hadjiarapoglou, Dioxiranes: Oxidation Made Easy, in Topics in Current Chemistry, Springer-Verlag, Berlin, 1993, vol 164, p 55.

119 (a) R. P. Thummel, B. Rickborn, J. Am. Chem. Soc., 1970, 92, 2064. (b) J. K. Crandall, M. Apparu, Org. React., 1983, 29, 345.

120 J. B Lambert, H. F. Shurvell, D. Lightner, R. G. Cooks, in Introduction to Organic Spectroscopy, Macmillian Publishing Company, New York, 1987.

121 (a) R. E. Wolf Jr, J. A. R. Hartman, L. A. Ochrymowycz, S. R. Cooper, M. Sabahi, R. S. Glass, Inorg. Synth., 1988, 25, 122. (b) B. de Groot, S. J. Loeb, Inorg. Chem., 1989, 28, 3573.

122 J. J. H. Edema, J. Buter, R. M. Kellogg, Tetrahedron, 1994, 50, 2095.

123 R. R. Schrock, J. S. Murdzek, G. C. Bazan, J. Robbins, M. DiMare, M. O’Regan, J. Am. Chem. Soc., 1990, 112, 3875.

153

REFERENCES

124 S. T. Nguyen, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc., 1993, 115, 9858.

125 (a) M. Scholl, T. M. Trnka, J. P. Morgan, R. H. Grubbs, Tetrahedron Lett., 1999, 40,2247. (b) M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett., 1999, 1, 953.

126 (a) G. C. Fu, R. H. Grubbs, J. Am. Chem. Soc., 1992, 114, 5426. (b) G. C. Fu, R. H. Grubbs, J. Am. Chem. Soc., 1992, 114, 7324.

127 K. B. Wagener, K. Brzezinska, C. G. Bauch, Makromol. Chem. Rapid Commun., 1992, 13, 75.

128 M. D. Forbes, J. T. Patton, T. L. Myers, H. D. Maynard, D. W. Smith, G. R. Schulz, K. B. Wagener, J. Am. Chem. Soc., 1992, 114, 10978.

129 S. K. Armstrong, J. Chem. Soc, Perkin Trans. 1, 1998, 371.

130 J. P. Morgan, R. H. Grubbs, Org. Lett., 2000, 2, 3153.

131 R. H. Grubbs, S. Chang, Tetrahedron, 1998, 54, 4413.

132 J-P. Hérrison, Y. Chauvin, Makromol. Chem., 1970, 141, 161.

133 H. E. Blackwell, D. J. O’Leary, A. K. Chatterjee, R. A. Washenfelder, D. A. Bussman, R. H. Grubbs, J. Am. Chem. Soc., 2000, 122, 58.

134 J. A. Smulik, A. J. Giessert, S. T. Diver, Tetrahedron Lett., 2002, 43, 209.

135 (a) M. P. S. Leconte, A. Mutch, F. Lefebvre, J-M. Basset, Bull. Soc. Chim. Fr., 1995, 132, 1069. (b) S. K Armstrong, B. A. Christie, Tetrahedron Lett., 1996, 37, 9373. (c) Y. S. Shon, T. R. Lee, Tetrahedron Lett., 1997, 38, 1283. (d) J. F. Miller, A. Termin, K. Koch, A. D. Piscopio, J. Org. Chem., 1998, 63, 3158. (e) A. J. Ashe III, X. Fang, J. W. Kampf, Organometallics, 2000, 19, 4935.

154

REFERENCES

136 J. Buter, R. M. Kellogg, Org. Synth., 1987, 65, 1550.

137 J. K. Ray, M. K. Haldar, S. Gupta, G. K. Kar, Tetrahedron, 2000, 56, 909.

138 J. J. Pappas, W. P. Keaveney, E. Gancher, M. Berger, Tetrahedron Lett., 1966, 36, 4273.

139 M. D. Todd, Y. Dong, J. Horney, D. I. Yoon, J. T. Hupp, Inorg. Chem., 1993, 32, 2001

140 (a) M. F. M. Roks, R. J. M. Nolte, Macromolecules, 1992, 25, 5398. (b) V. E. Carmichael, P. J. Dutton, T. M. Fyles, T. D. James, J. A. Swan, M. Zojaji, J. Am. Chem. Soc., 1989, 111, 767. (c) F. J. Urban, L. R. Chappel, A. E. Girard, B. L. Mylari, I. J. Pimblett, J. Med. Chem., 1990, 33, 765. (d) G. R. Brown, A. J. Foubister, J. Med. Chem., 1979, 22, 997. (e) J. D. Lamb, R. M. Izatt, P. A. Robertson, J. J. Christensen, J. Am. Chem. Soc., 1980, 102, 2452.

141 (a) A. J. Wright, S. E. Matthews, W. B. Fischer, P. D. Beer, Chem. Eur. J., 2001, 7, 3474. (b) A. D. Pechulis, R. J. Thompson, J. P. Fotjik, H. M. Schwartz, C. A. Lisek, L. L. Frye, Bioorg. Med. Chem., 1997, 5, 1893. (c) Transport Through Membranes: Carriers, Channels and Pumps, ed. A. Pulman, J. Jortner, B. Pullman, Kluwer Academic, Netherlands, vol. 21. 1988.

142 (a) M. Shamsipur, S. Y. Kazemi, G. Azimi, S. S. Madeni, V. Lippolis, A. Garau, F. Isaia, J. Membr. Sci., 2003, 215, 87. (b) H. Sakamoto, K. Kimura, T. Shono, Anal. Chem., 1997, 59, 1513.

143 A. Werner, Ann. Chem., 1912, 386, 1.

144 H. M. Colquhoun, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl., 1983, 25, 487.

145 D. R. Alston, P. R. Ashton, T. H. Lilley, J. F. Stoddart, R. Zarzycki, A. M. Z. Slawin, D. J. Williams, Carbohydr. Res., 1989, 192, 259.

155

REFERENCES

146 M. F. Raymo, J. F. Stoddart, Chem. Ber., 1996, 129, 981.

147 R. Ballardini, M. T. Gandolfi, L. Prodi, M. Ciano, V. Balzani, F. H. Kohnke, H. Shahriari-Zavareh, N. Spencer, J. F. Stoddart, J. Am. Chem. Soc., 1989, 111, 7072.

148 S. S. Moore, T. L. Tarnowski, M. Newcomb, D. J. Cram, J. Am. Chem. Soc., 1977, 99, 6398.

149 R. G. Pearson, J. Am. Chem. Soc., 1963, 85, 3533.

150 H. K. Frensdorf, J. Am. Chem. Soc., 1971, 93, 600.

151 (a) M. L. Campbell, S. B. Larson, N. K. Dalley, Acta Crystallogr., 1981, B37, 1741, 1744, 1750. (b) M. L. Campbell, N. K. Dalley, S. H. Simonsen, Acta Crystallogr., 1981, B37, 1747.

152 S. Yamada, M. Tanaka, J. Inorg. Nucl. Chem., 1975, 37, 587.

153 G. Wu, W. Jiang, J. D. Lamb, J. S. Bradshaw, R. M. Izatt, J. Am. Chem. Soc., 1991, 113, 6538.

154 (a) J. R. Hartman, S. R. Cooper, J. Am. Chem. Soc., 1986, 108, 1202. (b) J. R. Hartman, E. J. Hintsa, S. R. Cooper, J. Am. Chem. Soc., 1986, 108, 1208.

155 F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochman, Advanced Inorganic Chemistry, (6th Ed), John Wiley and Sons, New York, 1999.

156 (a) J. Parr, Polyhedron, 1997, 16, 551. (b) L. Shimoni-Livny, J. P. Glusker, C. W. Bock, Inorg. Chem., 1998, 37, 1853.

157 Y. Marcus, in Ion Solvation, John Wiley and Sons, New York, 1985, p. 245.

156

REFERENCES

158 (a) H. A. Berman, T. R. Stengle, J. Phys. Chem., 1975, 79, 1001. (b) E. H. White, K. W. Field, W. H. Hendrickson, P. Dzadzic, D. F. Roswell, S. Paik, P. W. Mullen, J. Am. Chem. Soc., 1992, 114, 8023. (c) H. J. Reich, J. P. Borst, R. R. Dykstra, D. P. Green, J. Am. Chem. Soc., 1993, 115, 8728. (d) T. Yabe, J. K. Kochi, J. Am. Chem. Soc., 1992, 114, 4491.

159 D. W. James, R. E. Mayes, J. Phys. Chem., 1984, 88, 637.

160 (a) L. Wittenkeller, W. Lin, C. Diven, A. Ciaccia, F. Wang, D. M. de Freitas, Inorg. Chem., 2001, 40, 1654. (b) J. H. Nelson, in Nuclear Magnetic Resonance Spectroscopy, ed. N. Folchetti, Prentice Hall, New Jersey, 2003.

161 F. Wudl, F. Gaeta, J. Chem. Soc., Chem. Comm., 1972, 107.

162 (a) H. B. Kagan, Tetrahedron, 2001, 57, 2449. (b) H. B. Kagan, J. C. Fiaud, Kinetic Resolution, in Topics in Stereochemistry, ed. E. L Eliel, A. L. Allinger, Wiley, New York, 1988, vol 18, p 249.

163 (a) P. J. Cox, N. S. Simpkins, Tetrahedron: Asymmetry, 1991, 2, 1. (b) M. Asami, J. Synth. Org. Chem. Jpn., 1996, 54, 188. (c) D. M. Hodgson, A. R. Gibbs, G. P. Lee, Tetrahedron, 1996, 52, 14361. (d) P. O’Brien, J. Chem. Soc., Perkin Trans. 1, 1998, 1439.

164 M. Tokunaga, J. F. Larrow, F. Kakiuchi, E. N. Jacobsen, Science, 1997, 277, 936.

165 (a) K. Katsuki, K. B. Sharpless, J. Am. Chem. Soc., 1980, 102, 5974. (b) T. Katsuki, in Comprehensive Asymmetric Catalysis, chapter 18, ed. E. N. Jacobsen, A. Pfaltz, H. Yamamoto, Springer, 1999, vol. 2, 621. (c) A. F. Pfenniger, Synthesis, 1986, 2, 89. (d) M. G. Finn, K. B. Sharpless, Asymmetric Synthesis, 1985, 5, 247.

166 J. K. Whitesell, S. W. Felman, J. Org. Chem., 1980, 45, 755.

167 (a) M. Asami, Chem Lett., 1984, 829. (b) M. Asami, Bull. Chem. Soc. Jpn., 1990, 63, 721.

157

REFERENCES

168 M. Asami, N. Kanemaki, Tetrahedron Lett., 1989, 30, 2125.

169 D. Bhuniya, A. DattaGupta, V. K. Singh, J. Org. Chem., 1996, 61, 6108.

170 (a) M. J. Södergren, P. G. Andersson, J. Am. Chem. Soc., 1998, 120, 10760. (b) M. J. Södergren, S. K. Bertilsson, P. G. Andersson, J. Am. Chem. Soc., 2000, 122, 6610.

171 M. Asami, T. Suga, K. Honda, S. Inoue, Tetrahedron Lett., 1997, 38, 6425.

172 L. P. C. Neilson, C. P. Stevenson, D. G. Blackmond, E. N. Jacobsen, J. Am. Chem. Soc., 2004, 126, 1360.

173 A. Johnston, PhD. Thesis, University of Sydney, NSW, Australia, 1997.

174 (a) R. Johnson, K. B. Sharpless, in Comprehensive Organic Synthesis, ed. B. M. Trost, I. Fleming, Pergamon Press, New York, 1991, vol. 7, p. 389. (b) R. Johnson, K. B. Sharpless, in Catalytic Asymmetric Synthesis, ed. I. Ojima, VCH, New York, 1993, p. 103. (c) T. Katsuki, V. S. Martin, Org. React., 1996, 48, p. 1.

175 K. B. Sharpless, C. H. Behrens, T. Katsuki, A. W. M. Lee, V. S. Martin, M. Takatani, S. M. Viti, F. J. Walker, S. Woodward, Pure Appl. Chem., 1983, 55, 589.

176 (a) I. D. Williams, S. F. Petersen, K. B. Sharpless, S. J. Lippard, J. Am. Chem. Soc., 1984, 106, 6430. (b) K. B. Sharpless, Chem. Scripta, 1987, 27, 521.

177 A. O. Chong, K. B. Sharpless, J. Org. Chem., 1977, 42, 1587.

178 V. S. Martin, S. S. Woodard, T. Katsuki, Y. Yamada, M. Ikeda, K. B. Sharpless, J. Am. Chem. Soc., 1981, 103, 6237.

179 Y. Kitano, T. Matsumoto, F. Sato, Tetrahedron, 1988, 44, 4073.

158

REFERENCES

180 D. P. G. Hamon, K. L. Tuck, Tetrahedron Lett., 1999, 40, 7569.

181 (a) Y. Tu, Z-X. Wang, Y. Shi, J. Am. Chem. Soc., 1996, 118, 9806. (b) H. Tian, X. She, L. Shu, H. Yu, Y. Shi, J. Am. Chem. Soc., 2000, 122, 11551.

182 G. Procter, J. Leonard, B. Lygo, in Advanced Practical Organic Chemistry, Chapman and Hall/Stanley Thornes, England, 1995.

183 H. M. C. Ferraz, R. M. Muzzi, T. de O Vieira, H. Viertler. Tetrahedron Lett., 2000, 41, 5021.

184 D. Chen, R. J. Motekaitis, I. Murase, A. E. Martell, Tetrahedron, 1995, 51, 77.

185 S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B. Hansen, A. E. Gould, M. E. Furrow, E. N. Jacobsen, J. Am. Chem. Soc., 2002, 124, 1307.

186 O. Silberrad, H. A. Phillips, J. Chem. Soc., 1908, 93, 474.

159