New Stereoselective Reactions to Form Amido Alkyl C-N and Vinyl Triflate C-O

Home , Ritter reaction, Sulfonate, Triflate

NEW STEREOSELECTIVE REACTIONS TO FORM AMIDO ALKYL C-N AND

VINYL TRIFLATE C-O BONDS VIA CARBOCATION INTERMEDIATES &

ULTRAFAST SILICON FLUORINATION METHODOLOGIES FOR

APPLICATIONS IN PET IMAGING

Mohammed Alhuniti

A Dissertation Submitted to the Faculty of

The Charles E. Schmidt College of Science

in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

Florida Atlantic University

Boca Raton, Florida

December 2014

ACKNOWLEDGEMENTS

I would like to thank my advisor, Prof. Salvatore D. Lepore, for his excellent guidance throughout my doctoral studies. Prof. Lepore is truly an outstanding mentor who devotes his time to teach more than organic chemistry.

I want to thank my committee members Professors William Roush, Stéphane Roche, and

Lyndon West for their time and insightful advice on my research. I thank Susan Lepore for her efforts to facilitate my research in the Lepore group. I also thank all former and current members of the group.

Finally, I wish to express my gratitude to my family who have continually supported me.

I sincerely appreciate the endless encouragement of my mother, my brothers, and my lovely wife and daughters.

ABSTRACT

Author: Mohammed Alhuniti

Title: New Stereoselective Reactions to Form Amido Alkyl C-N and Vinyl Triflate C-O Bonds via Carbocation Intermediates & Ultrafast Silicon Fluorination Methodologies for Applications in PET Imaging.

Institution: Florida Atlantic University

Thesis Advisor: Dr. Salvatore D. Lepore

Degree: Doctor of Philosophy

Year: 2014

We report here the development of a Lewis acid catalyzed method for the dehydrative coupling of cyclic alcohols and nitriles to form amides with retention of configuration.

By contrast, the formation of amides by nitrile trapping of carbocations (Ritter reaction) usually affords racemic product. The present reaction was accomplished by first converting alcohol starting materials to their corresponding chlorosulfites in situ. Even after an extensive search, only copper (II) salts were able to produce the desired conversion of these chlorosulfites to amides though with low catalytic turnover.

Improving the turnover without deteriorating the stereochemical outcome was eventually accomplished by a careful selection of the reagent addition sequence and through the removal of gaseous byproducts. This Ritter-like coupling reaction proceeds in good yields with secondary cyclic alcohols under mild conditions. The stereochemical outcome

is likely due to fast nucleophilic capture of a non-planar carbocations (hyperconjomers) stabilized by ring hyperconjugation.

In a second project, we demonstrate that TMSOTf in the presence of several metal catalysts converts alkynes to vinyl triflates under mild conditions. Current methods for the formation of vinyl triflates directly from alkynes generally involve harsh conditions and are exclusively selective for the E-isomer. Further study and optimization revealed that internal alkynes are converted to the Z-vinyl triflate product though with modest selectivity. The reaction efficiently converts aliphatic and aromatic terminal alkynes as well as internal alkynes to their corresponding vinyl triflate products. A mechanism is put forward to explain the unique role of silicon in this system. In this mechanism, we propose a silyl vinyl triflate intermediate that undergoes protodesilylation to afford the vinyl triflate product. Importantly, we believe that silicon may play a similar role in other recently reported reactions

In a final project pursued in collaboration with NIH researchers, we describe our development of ultrafast silicon fluorination techniques for eventual applications in PET.

The demand for physiologically stable organosilicon 18F-fluorides required compounds with a high degree of steric hindrance at the silicon center. This in turn greatly slows the rate of fluorination using current methods which often entail high temperatures and very polar solvents. Our initial solution to this problem centered on the use of metal chelating units attached to silicon substrates to serve as leaving groups. We reasoned that such leaving groups would stabilize negative charge developed on the silicon center in the TS and thus lead to a faster fluorination. Using these leaving groups, we observed fast radiofluorination of bulky silicon substrates at room temperature in 15 minutes without

the need for the commonly used phase transfer reagents. Similar rate enhancements were

18 also observed with cyclotron-produced F-fluoride (t1/2 = 109.7 min, β+ = 97%). Based on our proposed mechanism for fluorination rate enhancement with chelating leaving groups, we reasoned that similar results could be achieved even without first attaching a chelating leaving group to a silicon center. As a result, we developed the concept of a

Crown Ether Nucleophilic Catalyst (CENC). The use of these new phase transfer agents allows for efficient sequestration and recovery of cyclotron-derived K18F. Using these

CENC/K18F complexes, we observed rapid radiofluorination of silicon substrates (5 min) which is significantly faster than currently reported methods.

vii

NEW STEREOSELECTIVE REACTIONS TO FORM AMIDO ALKYL C-N AND

VINYL TRIFLATE C-O BONDS VIA CARBOCATION INTERMEDIATES &

ULTRAFAST SILICON FLUORINATION METHODOLOGIES FOR

APPLICATIONS IN PET IMAGING

List of Figures……...... viii

List of Tables…...... ix

List of Scheme……...... x

CHAPTER ONE STEREORETENTIVE COPPER (II) CATALYZED RITTER REACTION OF SECONDARY CYCLOALKANOLS…..…. 1

1.1 Introduction …...... 1

1.2 Ritter reaction…………………………...... 3

1.3 Development of stereoretentive copper (II) catalyzed Ritter reactions of secondary cycloalkanols...... 14

1.4 Proposed catalytic cycle ….…...... 25

1.5 Experimental section...... 26

CHAPTER TWO ZINC (II) CATALYZED CONVERSION OF ALKYNES TO VINYL TRIFLATE IN THE PRESENCE OF SILYL TRIFLATE...... 32

2.1 Introduction………………………………...... 32

2.2 Synthesis of vinyl triflate …………………...... 36

2.3 Development of zinc (II) catalyzed conversion of alkynes to vinyl triflates in the presence of silyl triflates...... 40

viii

2.4 Mechanistic study ………………………..……...... 45

2.5 Experimental section…………………...... 51

CHAPTER THREE SILICON-NUCLEOPHILIC ASSISTED LEAVING GROUP (S-NALG)………………….………...... 57

3.1 Introduction…………………………………...... 57

3.2 18F-Organosilicon radiotracers…………...... 63

3.3 Enhanced nucleophilic fluorination and radiofluorination of organosilanes appended with potassium-chelating leaving groups...... 67

3.4 Mechanistic considerations...... 76

3.5 Enhanced radiofluorination of organosilanes using Nucleophilic crown ether phase transfer agents...... 76

3.6 Experimental section...... 81

Appendix……………………………………………….…………………………… 86

References……………………………………………….………………………….. 134

LIST OF FIGURES

Figure 1 Major NOE correlations observed Bach and coworkers in carbocation 22...... 6

Figure 2 Plot of LUMO orbital of three cyclohexyl carbocations...... 11

Figure 3 Stabilization modes for the hyperconjomers of cyclohexyl cation...... 12

Figure 4 Mechanistic rational for the formation of stereoretentive amide...... 14

Figure 5 Selected correlations from 2D-NOESY spectrum for 58 and 59...... 22

Figure 6 Proposed catalytic cycle for stereoretentive amidation…….………...... 26

Figure 7 Catalytic cycle for Heck coupling………..……...... 34

Figure 8 Structure of (+)-vernolepin...... 36

Figure 9 Comins reagent...... 38

Figure 10 Triflic acid addition to 113 in the presence and the absence of Lewis acid...... …..……….…………..….... 49

Figure 11 Common electrophilic fluorinating reagents...... 60

Figure 12 Structure of kryptofix 2.2.2 (K 2.2.2)...... 62

Figure 13 Structures of common SiFA building blocks...... 66

Figure 14 Radiofluorination of substrates 173−178...... 75

Figure 15 Radiofluorination of 173−178 in the presence of K2CO3 and K 2.2.2…...... 77

Figure 16 Radiofluorination of 178 to give [18F]180……………………………….… 82

LIST OF TABLES

Table 1 BrØnsted acid-catalyzed Ritter reaction………...... 4

Table 2 Leaving group study using TiF4………………………….…………...... 16

Table 3 Metal salt reactivity towards amidation………………….………………...... 17

Table 4 Study of catalyst loading and nitrile equivalents………….…………...... 19

Table 5 Nitrile generality study……………………………………...………...... 20

Table 6 Substrate generality study……………………………….….………...... 21

Table 7 Asymmetric Heck coupling using vinyl iodide…………….……………...... 35

Table 8 Screening of metal catalysts and silyl triflate reagents…….………………… 43

Table 9 Reactions with terminal alkynes...... 44

Table 10 Reactions with internal alkynes……………...... 45

Table 11 Effect of water on the formation of vinyl triflate…………..……………….. 57

Table 12 Control experiments involving TfOH………………………………………. 48

Table 13 Half-life time and average energy of commonly used positron-emitting nuclides………………………..……...…...……….. 59

Table 14 Fluorination of organosilanes with diverse alkoxide leaving groups…….…. 71

Table 15 Fluorination of 175 using 18-Cown-6 derivatives………………...………... 81

LIST OF SCHEMES

Scheme 1 Oxidative amidation...... 2

Scheme 2 Amidation using azides...... 3

Scheme 3 Mechanism for Ritter reaction...... 4

Scheme 4 Iron (III) catalyzes Ritter reactions...... 5

Scheme 5 Stereoselective synthesis of chiral amides...... 5

Scheme 6 Ritter type reaction through tertiary alcohol inversion...... 7

Scheme 7 Prins-Ritter reaction...... 8

Scheme 8 4-Tetrahydropyranyl cation...... 9

Scheme 9 Titanium (IV) fluoride amidation...... 9

Scheme 10 Dialkyl sulfite formation...... 18

Scheme 11 Formation of chlorosulfite from dialkylsulfite...... 19

Scheme 12 Iodine catalyzed amidation of borneol and isoborneol...... 22

Scheme 13 Reaction with cis-2-methylcyclohexanol...... 24

Scheme 14 Synthesis of 3-substituted carbacephem...... 36

Scheme 15 Enantioselective allene formation...... 36

Scheme 16 Synthesis of vinyl triflate from ketone...... 37

Scheme 17 Addition of triflic acid to alkyne...... 38

Scheme 18 Addition of triflic acid to propynoic acid derivatives...... 39

xii

Scheme 19 Gold-catalyzed addition of triflate to an alkyne...... 39

Scheme 20 Electrophilic carbon functionalization of alkyne...... 40

Scheme 21 Regioselective functionalization of internal alkynes...... 41

Scheme 22 Strategy for metal catalyzed alkyne conversion to vinyl triflate...... 42

Scheme 23 Suzuki coupling with vinyl triflate...... 45

Scheme 24 Tandem vinyl triflate/ Suzuki coupling...... 46

Scheme 25 Possible role of zinc (II) reagent involving zwitterionic intermediate..... 50

Scheme 26 Possible routes for the formation of vinyl triflate...... 50

Scheme 27 Formation of vinyl silicon from vinyl zinc reagent...... 51

Scheme 28 A Proposed mechanism for the observed Z-selectivity...... 52

Scheme 29 Electrophilic approach to synthesis 18F-FDOPA...... 61

Scheme 30 Electrophilic addition of 18F to dihy-2H-dropyran 154...... 62

18 Scheme 31 Synthesis of N-methyl-[ F]fluorospiperone using SNAr approach...... 63

Scheme 32 Synthesis of [18F]FDG by nucleophilic substitution...... 63

Scheme 33 Current radiofluorination methods...... 65

Scheme 34 Chelating leaving group...... 68

Scheme 35 Fluorination of silicon nucleophilic assisting group (S-NALG)...... 69

Scheme 36 Proposed radiofluorination mechanism involving stabilizing interactions with a 18-C-6-CH2O- leaving group……………...... 78

Scheme 37 Enhanced reactivity of pentacoordinated silicon...... 79

Scheme 38 Reactivity of tetra- and pentacoordinate silicon compounds...... 80

Scheme 39 Accelerated fluorination using nucleophilic crown ether...... 80

xiii

CHAPTER ONE STEREORETENTIVE COPPER (II) CATALYZED RITTER

REACTION OF SECONDARY CYCLOALKANOLS

1.1 Introduction

The amide functional group plays an important role in organic, polymer, and pharmaceutical chemistry.1 This functional group is considered to be the backbone of all natural peptides. In addition to its incorporation in various bioactive compounds, more than 25% of known drugs have the amide motif. 2 As a result of the importance of the amide functional group, numerous synthetic routes have been developed to access this motif. The commonly used method involves coupling of activated carboxylic acid derivatives (electrophilic) with amine nucleophiles.3 Though these coupling methods are widely used and largely successful, significant problems remain and thus various alternative strategies have emerged. As will be described in the paragraphs below, these new methods often involve creative changes to the traditional nucleophilic and electrophilic reactants to achieve higher degrees of efficiency and atom economy. While the scope of these new strategies is very large and involves many new developments, only a few recent examples are introduced here.

Oxidative amidation is an ideal alternative to the traditional synthesis of amides; here as well, the starting materials are often inexpensive and readily available.4 This approach entails the nucleophilic addition of an amine to aldehyde 1 leading to a hemiaminal intermediate (2), which furnishes the amide product (3) upon oxidation (Scheme 1). This method is not limited to aldehydes; when using an appropriate oxidant, alcohols can be

oxidized in situ to aldehydes that react with amines to give amide products after oxidation of the hemiaminal intermediate. Despite the best efforts to improve this reaction, some of the current methods require harsh conditions,5 toxic metals and excess amounts of base and oxidant.6

Scheme 1. Oxidative amidation

Other amidation methods that utilize azides as amine precursors have been developed.

One of these is the modified Staudinger reaction which utilizes alkyl azide 5 and carboxylic acid 4 to produce the amide product.7 This reaction proceeds by in situ activation of the carboxylic acid group followed by reaction with phosphazene intermediate 7 generated from the reaction between alkyl azide and a trialkylphosphine.

The Schmidt reaction, on the other hand, employs hydrogen azide and ketone 9 to access amide products (Scheme 2).8 This reaction is generally performed in acidic media to generate the protonated ketone species that react with hydrogen azide to give amide products after alkyl migration. Similar to Schmidt reaction, the Beckmann rearrangement, which exploits oximes as amide precursors, involves an alkyl migration step to produce the N-alkyl amide.8 Depending on the nature of the alkyl groups present in the reactant, the alkyl migration step in these reactions can result in the formation of two amide products. The Ritter reaction offers a useful strategy that avoids this potential problem with the Beckmann rearrangement utilizing nitriles as latent amides; as will be described in the next section, embedded in the nitrile group are the amine and carbonyl fragments

of traditional coupling approaches. Uniquely, this Ritter reaction yields an N-alkyl amide motif via the formation of a nitrogen to sp3-carbon bond.

Scheme 2. Amidation using azides

1.2 Ritter reaction

The Ritter reaction, a well-known amidation method that couples nitriles with alcohols or alkenes, has also received enormous attention due to its ability to prepare sterically hindered amides.9 This reaction has found wide application in industry as it can be performed on a large scale.10 The well-accepted mechanism involves the formation of carbocation species 12 that can be generated under acidic conditions from alcohols by dehydration or from alkenes by the addition of a proton. Carbocation 12 is then trapped with a nitrile to produce nitrilium ion intermediate 13, which furnishes amide 14 upon reaction with water (Scheme 3). This hydrolysis step consumes the water generated during the formation of carbocation from alcohol.9 This mechanism was supported by the isolation of nitrilium intermediate, and further supportive evidences were collected by spectroscopic techniques.11 Reactions involve alkenes as starting material requires the addition of water to hydrolyze the nitrilium ion intermediate. This amidation method is

limited to alcohols and alkenes that can yield a relatively stable carbocation intermediate, otherwise elimination and alkyl migration take place. Furthermore, the use of BrØnsted acids limits the application of this method to substrates containing acid-stable functional groups.

Scheme 3. Mechanism for Ritter reaction

This reaction usually performed in the presence of stoichiometric or near-stoichiometric amounts of BrØnsted acid even though the acid should be regenerated during this reaction. In fact, the use of catalytic amounts of BrØnsted acid was not reported till recently. Sanz and coworkers report a general procedure for BrØnsted acid catalyzed

12 Ritter reactions using 1-phenylethanol (15). Catalytic amounts of various BrØnsted acids were screened under the same conditions. In these reactions, the amide product was observed in good yields with the only difference being reaction time (Table 1).

Table 1. BrØnsted acid-catalyzed Ritter reaction

The Lewis acid catalyzed Ritter reactions were first reported in 1994 by Badparva using catalytic amounts of boron trifluoride.13 Since then, various Lewis acids were utilized to

catalyze this transformation. As an example, a catalytic amounts of the inexpensive and eco-friendly FeCl3•6H2O mediates Ritter reactions involving benzylic alcohols 17 to produce the amide products 18 (Scheme 4). Noteworthy, the methods reported using

Lewis acids suffer from low catalytic turnover and require high reaction temperatures

(>100 °C).14 More importantly, there are relatively few reports of stereoselective Ritter reactions and despite the potential utility of such reactions this area has been little explored.15 On the other hand, several methods for stereoselective Ritter-type reactions

(reactions that do not utilize alcohols or alkenes as starting material) have been reported.16

Scheme 4. Iron (III) catalyzed Ritter reactions

Bach and coworkers report a new diastereoselective synthesis of chiral benzylic amides

21 from the corresponding benzylic alcohol 19 using excess amounts of 2,4- dinitrobenzenesulfonic acid (DNBSA).17 This reaction proceeds by the formation of carbocation 20 that is trapped by nitrile to give amides 21 in high diastereoselectivity

(Scheme 5). According to Bach and coworkers, this transformation proceeds under kinetically controlled conditions and the product stereochemistry is independent of the relative configuration of starting material. Under the kinetically controlled condition, a nucleophilic attack is thought to occur at the less hindered face of the carbocation carbon in conformation 20 which the authors argue to be most energetically favorable.

Scheme 5. Stereoselective synthesis of chiral amides

This carbocationic center in 20 is stabilized by various conjugative interactions which taken together restrict the rotation around the σ-bond between the carbocationic center and its adjacent stereogenic carbon. Specifically, these are the conjugative interaction with the phenyl ring and the hyperconjugative interaction with the σ-(C-CMe3) bond. The stereochemical outcome of this reaction is in agreement with a previous study by the same authors that examined Friedel-Crafts type reactions using chiral benzylic alcohols as substrates.18 A low temperature 13C-NMR spectrum for structurally related carbocation

22 shows 12 carbon signals. Importantly, the carbocation carbon resonates at 262.2 ppm and the aromatic region encompasses six carbon signals. This observation suggests that the phenyl carbons are magnetically nonequivalent as a result of restricted rotation around the C+-phenyl bond. Furthermore, 1H-NMR NOE shows three important correlations summarized in Figure 1. These correlations suggest a conjugative stabilization with the phenyl ring; the p-orbitals of the aryl ring are parallel to the vacant carbocation p-orbital. Further stability is provided to the carbocationic center by hyperconjugation with the σ-(Cα-Cβ) bond (Figure 1). This stabilization would be

+ forfeited upon rotation of the σ-(C -Cα) bond. Consequently, the high energy barrier required for the rotation around this bond leads to a conformationally stable carbocation with nonequivalent faces.

Figure 1. Major NOE correlations observed Bach and coworkers in carbocation 22

Recently, Ritter-like reactions have been developed by Shenvi and coworkers for the conversion of tertiary alcohols to their corresponding isonitrile derivatives 24 with

19 inversion of configuration. In the presence of Sc(OTf)3 as a catalyst and TMSCN as a solvent, various trifluoroacetyl protected alcohols, formed in situ, were converted to their isonitrile derivatives. The reported inversion of configuration is suggested to be the result of a contact ion pair. In particular, Shenvi argues that the isonitrile nucleophile approaches the carbocation center preferably from the backside since the metal-leaving complex blocks the front face. However, this backside attack argument only explains a limited number of tertiary alcohol substrates. For example, Shenvi and coworkers show that the trifluoroacetyl esters of cis- and trans-4-t-butyl-1-methylcyclohexanol (23a and b) led to both inversion and retention products under the same scandium(III)-catalyzed conditions (Scheme 6). Noteworthy, the authors also state in their published work that this reaction only appears to function with the TMSCN nucleophile (as a solvent) and cannot be applied to different reactants. For example, acetonitrile under these conditions failed to give the expected Ritter-type product.

Scheme 6. Ritter type reaction through tertiary alcohol inversion

Yadav and coworkers report efficient diastereoselective Prins-Ritter reactions to produce

2-alkyl-4-aminotetrahydopyran derivatives (Scheme 7).20 The reaction between 3- phenylpropanal (25) and but-3-en-1-ol (26) in the presence of catalytic amounts of

Bi(OTf)3 gave product 27 with cis-geometry (Scheme 7). The authors rationalize the observed diastereoselective outcome as an equatorial attack of 4-tetrahydropyranyl cation

29 followed by aqueous workup. This intermediate is formed from the cyclization of oxocarbenium intermediate 28.

Scheme 7. Prins-Ritter reaction

Though Yadav provides very little additional analysis in his published work, a computational study by Alder and coworkers provides further insight. In their work, an analysis of conformational preferences of 4-tetrahydropyranyl cation 30 suggests an unusual global energy minimum for this carbocation.21 Carbocation 30 exhibits a unique geometry resulting from an extensive delocalization. According to Alder, the lone pair of the oxygen is delocalized with the carbocation center via σ-(CH2-CH2) bonds. As a

+ consequence of this interaction, the C-C and C-O bonds are shortened and the CH2-CH2 bonds are lengthened. This geometry places the C+-H bond in the pseudoaxial position which allows a nucleophile to be delivered from an equatorial trajectory. This analysis by

Alder seems to accord with the finding of Yadav. Compared to its acyclic isomer, carbocation 30 is 56 Kcal/mol more stable than 31 partly due to the charge delocalization

with the heteroatom (Scheme 8). Thus, the cyclization step leading to carbocation 30 is spontaneous when the appropriate conformation is achieved.

Scheme 8. 4-Tetrahydropyranyl cation

This importance of hyperconjugation in cyclic cations appears to extend beyond the chemistry of Yadav and Alder. In this regard, the Lepore group reported a new stereoretentive amidation method for cyclic alcohols.22 In this method, thionyl chloride was introduced for the first time for in situ generation of a leaving group leading to non- chlorinated products. Specifically, cyclic alcohols are treated with thionyl chloride to give chlorosulfite derivative 32, which was then allowed to react with nitrile (40 equivalents) in the presence of titanium (IV) fluoride (10 equivalents) to yield amide 34 after aqueous work-up with retention of configuration (Scheme 9). Under these identical conditions, acyclic alcohols reacted to give racemic mixtures in contrast to cyclic substrates. The stereoselectivity observed with cyclic alcohols is believed to be the result of titanium (IV) fluoride chelating both nitrile and the chlorosulfite oxygen. The metal chelation to chlorosulfite increases its leaving group ability leading to the formation of a cyclohexyl cation. As will be described in subsequent paragraphs, the authors argue that this cyclic carbocation does not immediately lose its configuration allowing the nitrile to be delivered from the same face of the leaving group (33, Scheme 9).

Scheme 9. Titanium (IV) fluoride amidation

The Lepore group has also developed a stereoretentive chlorination method of cyclic alcohols; a method that is thought to involve similar cyclic carbocation intermediates.23

The mechanistic argument for this stereoretentive chlorination was further supported by a computational study that suggests non-planar cyclic carbocation stabilized by hyperconjugation.

It is important to clarify that the stereochemical outcome of the method reported by Bach

(Scheme 5) is a result of hyperconjugative stabilization with the C-C bond of an existing stereocenter adjacent to the carbocation directing the approach of an incoming nucleophile. While the selectivities observed by Yadav and Lepore are different, in both cases, hyperconjugative stabilization plays an important role in the geometry of the carbocation intermediates formed, ultimately leading to one stereoisomer predominantly.

This unique type of stabilization of cyclic carbocations was first proposed by Sorensen and Schleyer.24

Sorensen and Schleyer began their studies of cyclohexyl cation stabilization by examining the 1-methyl-1-cyclohexyl cation using low temperature 1H and 13C-NMR techniques. These techniques show the presence of two interconverting species which were assigned as the chair and the twist boat conformers with the latter being the lower energy structure.24a However, further studies performed by the same group, showed that

1-methylcyclohexyl cations have two distinctive chair conformational minima. These conformers have different patterns of hyperconjugative interactions as a result of the

spatial orientation of the vacant p-orbital (Figure 2). The p-orbital occupying the axial position as in structure 35, is perfectly aligned to allow hyperconjugation with the axial

C-H σ-bonds, hence, 35 is referred to as the C-H hyperconjomer. The equatorial position of the p-orbital in 36 allows its interaction with C-C σ-bonds and therefore 36 is called the C-C hyperconjomer. Compared to the planar 1-methylcyclohexyl cation (37) the C-C and the C-H hyperconjomers are slightly more stable and the energy barrier for their interconversion is 4-5 Kcal/mol based on low temperature 13C-NMR experiments.

Noteworthy, the planar carbocation is considered to be the transition state for the interconversion between C-H and C-C hyperconjomers.24b This concept is summarized and expanded in subsequent paragraphs (see Figure 4).

Figure 2. Plot of LUMO orbital of three cyclohexyl carbocations24b

Additional and more recent computational studies on secondary cyclohexyl cations were performed by Alabugin and Manoharan. In their work, several new modes of hyperconjugation were explored relative to the previously described work of Sorensen and Schleyer. For example, stabilizing effects of the C-H or C-C bonds of the γ-carbon are included in the Alabugin study (Figure 3).25 Interestingly, very similar hyperconjomers to Sorensen and Schleyer’s work (38 and 39) were determined to be the low energy minima of the system. These results generally parallel the conclusions of

Sorensen and Schleyer, namely that the C-C hyperconjomer is more stable than the C-H hyperconjomer (2.2 Kcal/mol in this case).

Figure 3. Stabilization modes for the hyperconjomers of cyclohexyl cation

Sorensen and Schleyer discussed in detail a theory of non-classical cyclohexyl cations and how the nature of the C-C and the C-H hyperconjomers can affect the stereochemical outcome of reactions involving them.22d They proposed that the position of the leaving group (axial or equatorial) determines which conjomer is initially formed in an ionizing reaction. Cyclohexyl derivatives with an axial leaving group will initially give the C-H hyperconjomer and those with an equatorial leaving group will give the C-C hyperconjomer.

Most importantly for the work described in this chapter, Sorensen and Schleyer predicted that these two hyperconjomers could be trapped with nucleophiles to give products with opposite configurations at the reacting carbon center if the rate of interconversion between C-C and C-H hyperconjomers is slower than the rate of nucleophilic attack. On the contrary, rapid interconversion to the more stable hyperconjomer before a nucleophilic attack will lead to a predominant retention or inversion depending on the initial stereochemistry assuming the rate of the nucleophile attack of these hyperconjomers is the same. These predictions of Sorensen and Schleyer provided an

important mechanistic starting point in our exploration of a stereoretentive Ritter reaction.

To summarize, a cyclohexyl cation has two stable isomers each in the chair conformation as a result of hyperconjugative stabilization. These isomers are called C-C and C-H hyperconjomers and the carbocation center in these hyperconjomers adopt a uniquely non-planar geometry. These hyperconjomers, 35 and 36, are more stable than a traditional planar cyclohexyl cation (37, Figure 4). The calculated energy barrier for their interconversion is 4-5 Kcal/mol. However, we expect that this barrier will be higher for secondary cyclohexyl cations due to the absence of some hyperconjugative interactions present in tertiary cations (e.g. CH3 in structure 37, Figure 4) though computational verification of this assertion is required. Sorensen and Schleyer predict that, depending on the configuration of the reactant, one hyperconjomer will form initially. If this cation is rapidly trapped with a nucleophile, retentive product will result (Figure 4). We believe that this prediction has been verified by Lepore and coworkers in which Ritter reactions proceed with retention of configuration under TiF4 conditions.

Figure 4. Mechanistic rationale for the formation of stereoretentive amide

1.3 Development of Stereoretentive Copper (II) Catalyzed Ritter Reactions of

Secondary Cycloalkanols

As mentioned earlier, the main drawback with the titanium (IV)-mediated amidation method is the use of large excesses of reagents; the optimal conditions require 10 equivalents of TiF4 and 40 equivalents of nitrile. Therefore, we undertook to improve the conditions of this method by carefully examining the key components of this reaction. As previously mentioned, the originally reported method calls for the formation of a chlorosulfite from a cycloalkanol followed by the addition of TiF4 and a nitrile. With 2- methyl cyclohexanol, these conditions led to amide product in 88% yield with a 10:1 selectivity for stereoretentive product 49 (Table 2, entry 1). We sought to evaluate other leaving groups using the same substrate and reaction conditions (Table 2). Thus variety

of substrates containing mainly sulfonate leaving groups (41 – 48) were prepared and then subjected to TiF4 amidation conditions. Interestingly, the reaction with compound 41 containing a chelating leaving group gave a diastereoselectivity similar to the outcome with chlorosulfite 40 but in a poorer yield. Based on previous related work in the Lepore group, we selected compound 41 due to its diethylene oxide chelating arm with the expectation that this would provide high stereoselectivity in nitrile addition.26 Similar chelation was also expected with 8-sulfonylquinoline 42 which gave amide product in good yield upon reaction with TiF4 and nitrile but with inferior diastereoselectivity.

However, reactions involve non-chelating arylsulfonate leaving groups afforded similar results (entries 4-8). Surprisingly, substrate 48, containing the triflate leaving group, reacted to give mainly inverted amide product 50 (entry 9). Although the leaving group study did not provide a clear understanding of the relationship between the leaving group ability and selectivity, the results clearly suggested that in situ formed chlorosulfites were optimal substrates for further study. Specifically, we endeavored to develop a catalytic form of this stereoretentive Ritter reaction.

Table 2. Leaving group study using TiF4

We then proceeded to examine numerous non-nucleophilic metal salts in an attempt to find the best metal to catalyze this transformation (Table 3). With substrate 40, the readily available triflate salts of copper (II) and scandium (III) react to afford amide products 49/50, albeit with a high catalyst loading of 50%. Using copper (II) carbonate, the reaction gave a significantly lower yield than copper (II) triflate. Thus, to rule out the role of triflate as anything other than a spectator ion, KOTf was used (entry 6); however, no amide product was observed. Product formation has also been observed with reactions involving other metal salts including those of Co(II) and Ag(I) though in modest yields

(entries 4 and 5). Due to its lower toxicity and cost relative to the scandium reagent, copper (II) triflate was used to further study the present amidation reaction.

Table 3. Metal salt reactivity towards amidation

Before going further in this study, the reactivity of compound 42, bearing 8-quinoline sulfonate leaving group, was compared with chlorosulfite substrate 40 using copper (II) triflate as a catalyst. Here as well, the substrate bearing a chlorosulfite leaving group afforded a higher reaction yield and stereoselectivity. Then, various solvents were examined in an attempt to improve the yield of this reaction. Reactions performed in coordinating solvents such as diethyl ether deteriorated the stereoselectivity. The low selectivity might be due to solvent-metal chelation, which interferes with metal-nitrile (or

chlorosulfite) coordinating interactions. Despite these extensive optimization efforts, the reaction suffered from low catalytic turnover. The breakthrough occurred when a literature search uncovered an article suggesting that copper (II) is unstable in the presence of sulfur dioxide, a byproduct in our reaction.27 To circumvent the catalyst decomposition pathway and the solvent issue, we performed the reaction under solvent- free conditions. In this system, the nitrile coupling partner would serve as both reagent and solvent. This approach also allowed for easier removal of SO2 and HCl, generated during the course of the reaction, by flushing the head-space in the reaction vessel with argon. Under these conditions the catalyst load was reduced to 20%. Experiments involving l-menthol revealed another problem with this reaction, the formation of dialkyl sulfite 53 as a side product in the amidation reaction (Scheme 10). The menthol substrate turned out to be useful for further study since it led to 53, a uniquely stable and isolable compound. This product is expected to arise from the addition of another molecule of l- menthol chlorosulfite 52.28 In separate experiments, we observed that copper (II) triflate catalyzed the reaction between alcohol 51 and chlorosulfite 52 to give dialkylsulfite 53

(quantitative yield was observed within 5 min). In addition, dialkylsulfite 53 was separately prepared and subjected to our reaction conditions leading to amide product 54a only after long reaction times (> 24 h).

Scheme 10. Dialkyl sulfite formation

Generally, dialkyl sulfites (55) can be converted to chlorosulfite 56 in the presence of thionyl or chloride ions (Scheme 11).29 However, under our conditions the use of excess amounts of thionyl chloride or metal chlorides did not minimize the formation of dialkyl sulfite. Consequently, we sought to minimize its formation by altering the addition sequence. To accomplish this, a CH2Cl2 solution of nitrile, copper (II) triflate, and alcohol was added slowly to thionyl chloride (neat). The concentration of this solution was based on the nitrile (1 M). This is in contrast to the previous conditions which involved the pre-formation of chlorosulfite by reacting alcohol with thionyl chloride followed by reaction with a copper (II)/ nitrile complex. Using the new addition sequence, the yield of amide formation was improved 20 to 25 %.

Scheme 11. Formation of chlorosulfite from dialkylsulfite

Using the SO2/HCl removal technique and the new addition strategy, catalyst loading and nitrile stoichiometry studies were then performed with 2-methylcyclohexanol (Table 4).

Not surprisingly, after 24 h and in the absence of metal catalyst no amide product was observed (entry 1); however, increasing the reaction time to 7 days led to a 25% yield

though with virtually no selectivity (1.25:1 ratio of 49:50). In terms of product yield, a catalyst loading of 20% appears to be optimal, especially with 4 to 5 equivalents of nitrile

(entries 6 and 7). However, these additional equivalents of nitrile lowered the ratio of retention to inversion. Taking into account the yield and stereoselectivity, 20% copper

(II) triflate and 1.9 or 2.0 equivalents of nitrile appeared to be the optimal conditions for our generality studies.

Table 4. Study of catalyst loading and nitrile equivalents

Previous reports indicate that l-menthol is a problematic substrate in Ritter reactions owing to its propensity to undergo cation rearrangements.30 Thus, we proceeded to examine the nitrile generality with this substrate to further probe the limits of the current catalytic reaction. This study revealed that both aliphatic and aromatic nitriles react to afford amides 54a-g with complete retention of configuration in moderate to good yields with no detectable hydride shift products (Table 5). Bulky pivalonitrile reacted under these conditions to afford the amide product with retention of configuration (entry 5).

Interestingly, the amide product was also observed in the presence of a phenolic hydroxyl group (entry 4). The primary limitation involves reactions with electron deficient nitriles such as p-chlorobenzonitrile (entry 3) and especially trichloroacetonitrile (entry 8). For

optimal yield, 1.9 equivalents of nitrile were used; however, in most cases, reasonable yields were also obtained with as little as 1.2 equivalents.

Table 5. Nitrile generality study

We next examined a range of saturated cyclic alcohols to explore the substrate generality

(Table 6). Very similar to the previously mentioned l-menthol example, we observed the formation of amide products with retention of configuration with a number of cyclic chiral alcohols (entries 1 – 4).

When isoborneol (entry 3, Table 6) was subjected to our copper catalyzed reaction

(copper (II) triflate and thionyl chloride) the stereoretentive amide (exo-product) product was observed though in low yield (21%). The major product was isobornyl chloride

(retention of configuration) formed preferentially over the inverted isomer (5:1).

Interestingly, when borneol (entry 2) was reacted under the same conditions, stereoretentive amide 58 was isolated. Even with these bridged bicyclic systems, it

appears that the present amidation reaction proceeds by a mechanism favoring retention of configuration.

Table 6. Substrate generality study

The borneol and isoborneol substrates have been investigated by several groups in Ritter reactions.31 In each case, the exo-amide product (59) was isolated from both starting materials. For example, using molecular iodine as a catalyst, both 64 and 65 gave the exo amide product (Scheme 12).31b However, as just discussed, our copper (II) catalyzed reaction led to retention of configuration with both alcohols.

Scheme 12. Iodine catalyzed amidation of borneol and isoborneol

The stereochemistry of these isobornyl and bornyl products in our reaction was confirmed by comparison to spectra in previously published works.31 In case these previous reports had misassigned the relative stereochemistry of their products, we further confirmed our observation stereoretention using NOESY NMR spectroscopy

(Figure 5). Characteristic correlations were concluded from the C-2 methine proton. The

C2-methine proton in 58 shows a correlation with one of the C7 methyl groups and one of the methylene protons at C3. On the other hand, the C2 methine proton in compound

59 shows no correlations with a C7 methyl groups; instead, it correlates with methylene protons at C3 and C6.

Figure 5. Selected correlations from 2D-NOESY spectrum for 58 and 59

Cyclic alcohols with variety of ring sizes also reacted under the copper (II) conditions to afford amide products in good yields (Table 6, entries 5 – 8). Poor yields were observed in the case of cyclododecanol which gave almost exclusive elimination product (entry 9).

Medium-sized cyclic carbocations, C8-C11, are known to adopt a special type of hyperconjugation involving a transannular hydrido bridge. We suspect that this type of stabilizing factor was present in our Ritter reaction with a cyclooctyl substrate leading to good yields (entry 8). However, this transannular hydrido bridge is not likely accessible in the cyclododecyl case (entry 9). Thus, a possible explanation for the low yield of the

amide from cyclododecanol is that upon carbocation formation the elimination is faster than the addition of a nitrile. Eliminations were also observed with cis-2- methylcyclohexanol (66) under the copper (II) conditions along with hydride shift product 67 (Scheme 13). It is expected that the hydrogen at C2, trans to chlorosulfite, participates in the formation of a carbocation at position C1 thus facilitating a hydride shift to a thermodynamically more stable tertiary carbocation. From this common intermediate, products 67 and 68 are likely to arise.

Scheme 13. Reaction with cis-2-methylcyclohexanol

Finally, tertiary and primary alcohols under the copper (II) conditions gave little or no yields of amide products except in the case of 1-adamantanol (entry 11, Table 6). The lack of reactivity with primary alcohols is due to the instability of the resultant carbocation. For tertiary alcohols the lack of reactivity is possibly due to the reactivity of their chlorosulfite derivatives. These derivatives react rapidly to form the chloride product.

1.4 Proposed Catalytic Cycle

Using our copper (II) conditions, the reactivity of acyclic alcohols was investigated.

When (R)-2-octanol was subjected to the present reaction, the inverted amide product was isolated in 47% yield in 65% enantiomeric excess (ee). We conclude that the present

method only leads to retention of configuration with cyclic alcohol substrates especially cyclohexyl derivatives.

As previously discussed, 2-methylcyclohexanol will undergo a Ritter reaction in the absence of the Cu(OTf)2 though the reaction proceeds very slowly (25% yield in 7 days).

Notably, the amide product is formed with virtually no selectivity (see footnote, entry 1,

Table 4). In the presence of the copper catalyst this observed selectivity was 15:1 in favor of stereoretention. This difference in selectivity can be explained in terms of a mechanistic pathway proceeding through non-planar carbocation intermediates. Without a catalyst, the secondary carbocation forms very slowly as one hyperconjomer and has an opportunity, under these conditions, to interconvert to a hyperconjomer of the opposite configuration. These two hyperconjomers are then captured by nitrile (possibly at different rates) leading to the observed selectivity. By contrast, we interpret the high selectivity in the presence of copper (II) to be the result of a fast capture of the hyperconjomer formed. Upon ionization, a tight ion pair is possibly formed. The anion

(leaving group) of this tight ion pair contains a metal part that chelates the nitrile coupling partner. An ion-pair return process in such scenario leads to formation of nitrilium ion intermediate instead of reforming the reactant. This might be due to the poor nucleophilicity of the chlorosulfite oxygen or the rapid formation of SO2 gas upon ionization. As mentioned in the introduction, we argue that this initially formed species is a non-planar cyclohexyl cation that maintains some ―memory‖ of its original configuration and is preferentially attacked from the same face as the leaving group.

In conclusion, we suggest the following mechanism to explain both the stereochemical outcome of the present reaction and the catalytic role played by the copper agent.

Chlorosulfite is formed in situ and is then chelated by the copper (II) catalyst.

Computational studies have shown that Ti(IV) prefers to chelate chlorosulfites at the sulfonyl oxygen.23 It seems reasonable to assume that Cu(II) will bind chlorosulfites in the same manner. Furthermore, we expect that the nitrile coupling partner will also complex with the copper catalyst32 ultimately leading to complex 69 (Figure 6). Metal chelation to chlorosulfite likely increases its leaving group ability leading to rapid formation of carbocation. This cyclic carbocation, stabilized by hyperconjugative interactions discussed earlier, is then rapidly trapped by nitrile delivered from the front face in tight ion pair 70 to form a nitrilium salt 71. The catalyst is then regenerated upon its dissociation from the chlorosulfite anion (to give sulfur dioxide) (Figure 6). Efforts are made to rid the reaction of this sulfur dioxide since, as mentioned earlier, it is known to destroy the copper catalyst. Finally, nitrilium salt 71 is converted to the stereoretentive amide product upon aqueous work-up.

Figure 6. Proposed catalytic cycle for stereoretentive amidation

Based on the proposed mechanism, the degree of stereoretention should depend on the rate of formation of nitrilium ion (nucleophile capturing a non-planar carbocation). As eluded earlier, if the rate of interconversion between C-C and C-H hyperconjomers is slower than the rate of nucleophilic attack complete stereoretention will be observed.

Therefore, experiments or computational studies that help determine the rate of nucleophile capturing a cyclohexyl carbocation might be very useful to further understand the mechanism of this reaction.33

1.5 Experimental section

General Information: All reactions were purified using flash silica gel 40-63μ. Analytical thin layer chromatography was performed on 0.25mm silica gel 60-F plates.

Visualization was accomplished with UV light and aqueous potassium permanganate solution staining. 1H-NMR spectra were recorded on a Varian Mercury 400 (400 MHz) spectrometer and are reported in ppm using solvent as an internal standard (CDCl3 at 7.26 ppm). Coupling constants were reported in Hz and 13C-NMR spectra were recorded on a

Varian Mercury 400 (100 MHz) spectrometer. Chemical shifts are reported in ppm, with solvent resonance employed as the internal standard. High-resolution mass spectra were obtained from University of Florida Mass Spectrometry Laboratory. All reagents were used without any further purification. Solvents were dried with activated molecular sieves.

General titanium (IV) fluoride procedure: To a solution of alcohol (1.0 eq) in dichloromethane (1.0 M) at 0°C was added thionyl chloride (1.5 eq) followed by stirring for 1 h to form the chlorosulfite. In a separate reaction vessel, nitrile (40 eq) was added to a TiF4 (10 eq) suspension in dichloromethane (4.0 M) and allowed to stir at room

temperature until complete dissolution (~15 min). Since TiF4 is fairly moisture sensitive, it was quickly transferred to a reaction vessel under argon and then weighed. The amount of each remaining reagent was then based on the weight of the TiF4. The titanium/nitrile solution was then cooled to 0 °C and to it was added the previously prepared chlorosulfite transferring by cannula under argon pressure. The chlorosulfite containing vessel was further washed with an amount of dichloromethane necessary to bring the final concentration of TiF4 in the other vessel to the desired concentration (2.5 M). After stirring for 2 h, the reaction was quenched with deionized water water and stirred (~30 min) until the organic layer became clear. The organic layer was removed and the aqueous layer was extracted twice with dichloromethane. All organic layers were combined, dried over anhydrous sodium sulfate, and concentrated in vacuo.

General copper (II) triflate procedure: An oven-dried vial under argon atmosphere was charged with nitrile (1.9 mmol), copper (II) triflate (0.2 mmol), and anhydrous dichloromethane alcohol (1.0 mmol). The mixture was stirred at room temperature until a homogenous mixture was observed. This solution was then transferred, over >45 min, through cannula to an oven-dried flask that contained thionyl chloride (1.5 mmol) under argon at 15-22°C. Headspace gases were replaced with argon several times in the first 1 h. The reaction was monitored by TLC. After completion, the reaction was diluted and poured into aqueous cooled KOH solution (20% w:v; 25 mL) and stirred for 20 min. The aqueous layer was then extracted several times, dried over anhydrous sodium sulfate, and concentrated in vacuo.

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]benzamid (54a).34 1H-NMR (400

MHz, CDCl3): δ 7.77-7.75 (m, 2H), 7.51-7.46 (m, 1H), 7.44-7.40 (m, 2H), 5.88 (d, J =

9.28 Hz, 1H), 4.04-3.95 (m, 1H), 2.10-2.05 (m, 1H), 2.01-1.94 (m, 1H), 1.76-1.69 (m,

2H), 1.59-1.48 (m, 1H), 1.22-1.09 (m, 2H), 0.91(d, J = 7.38 Hz, 3H), 0.90-0.89 (m, 1H),

0.89 (d, J = 6.94 Hz, 3H), 0.86-0.85 (m, 1H), 0.84 (d, J = 6.88 Hz, 3H); 13C-NMR (100

MHz, CDCl3) δ 166.6, 135.0, 131.2, 128.5, 126.7, 50.3, 48.3, 43.1, 34.5, 31.8, 26.9, 23.8,

22.1, 21.2, 16.2; -57.8 (c 4.0 CHCl3), HRMS (ESI+) Calc. For C17H26NO [M +

H]+: 260.2009. Found: 260.2000.

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]-4-methoxybenzamide (54b).34 1H-

NMR (400 MHz, CDCl3): δ 7.74-7.71 (m, 2H), 6.94-6.90 (m, 2H), 5.73 (br d, J = 7.61

Hz, 1H), 4.02-3.94 (m, 1H), 3.84 (s, 3H), 2.10-2.05 (m, 1H), 2.00-1.93 (m, 1H), 1.76-

1.69 (m, 2H), 1.57-1.50 (m, 1H), 1.17-1.13 (m, 2H), 0.98-0.89 (m, 1H), 0.91 (d, J = 6.84

Hz, 3H), 0.89 (d, J = 6.33 Hz, 3H), 0.88-0.85 (m, 1H), 0.83 (d, J = 6.94 Hz, 3H); 13C

NMR (100 MHz, CDCl3) δ 166.1, 161.9, 128.5, 127.3, 113.6, 55.4, 50.2, 48.4, 43.2, 34.5,

31.9, 26.9, 23.8, 22.1, 21.2, 16.2; -60.0 (c 4.3 CHCl3), HRMS (ESI+) C18H28NO2

[M + H]+: 290.2115 . Found: 290.2108.

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]-4-chlorobenzamide (54c).22 1H-

NMR (400 MHz, CDCl3): δ 7.71-7.68 (m, 2H), 7.40-7.37(m, 2H), 5.88 (br s, 1H), 4.02-

3.93 (m, 1H), 2.07-2.03 (m, 1H), 1.98-1.90 (m, 1H), 1.76-1.69 (m, 2H), 1.57-1.49 (m,

1H), 1.19-1.11(m, 2H), 0.91 (d, J = 7.18 Hz, 3H), 0.98-0.90 (m, 1H), 0.90 (d, J = 6.74

13 Hz, 3H), 0.86-0.80 (m, 1H), 0.83(d, J = 6.88 Hz, 3H); C NMR (100 MHz, CDCl3) δ

165.6, 137.3, 133.3, 128.6, 128.2, 50.47, 48.2, 43.0, 34.4, 31.8, 27.0 23.8, 22.1, 21.1,

+ 16.2; -51.8 (c 1.2 CHCl3), HRMS (ESI+) calc. for C17H25ClNO [M + H] : 294.1619.

Found: 294.1614.

2-Hydroxy-N-((1S,2R,5S)-2-isopropyl-5-methylcyclohexyl)benzamide (54d). 1H-

NMR (400 MHz, CDCl3): δ 12.53 (s, 1H) , 7.41-7.34 (m, 2H), 6.98 (dd, 1H, J = 8.4, 0.8

Hz,1H), 6.86-6.82 (m, 1H), 6.0 (d, J = 0.8 Hz, 1H), 4.04-4.95 (m, 1H), 2.07-2.02 (m,

1H) 1.95-1.89 (m, 1H), 1.77-1.71 (m, 2H), 1.57-1.47 (m, 1H), 1.25-1.09 (m, 2H), 0.92-

13 0.89 (m, 6H), 0.82 (d, J = 7.2 Hz, 3H); C-NMR (100 MHz, CDCl3) δ 170.5, 162.9,

135.4, 126.5, 120.0, 119.9, 115.8, 51.5, 49.5, 44.5, 35.8, 33.3, 28.4, 25.2, 23.5, 22.5,

+ 17.5; -58.6 (c 3.1 CHCl3), HRMS calc. for C17H25NO2 [M + H] : 276.1958 Found

[M+Na]+: 276.1954.

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]pivalamide (54e). 1H-NMR (400

MHz, CDCl3): δ 5.26 (d, J = 8.28 Hz, 1H), 3.80-3.71(m, 1H), 1.95-1.92 (m, 1H), 1.85-

1.81 (m, 1H), 1.72-1.65 (m, 2H), 1.52-1.44 (m, 1H), 1.19 (s, 9H), 1.12-1.02 (m, 2H), 0.89

(d, J = 7.75 Hz, 3H), 0.87 (d, J = 6.75 Hz, 3H), 0.84-0.73 (m, 2H), 0.78 (d, J = 6.89 Hz,

13 3H); C-NMR (100 MHz, CDCl3) δ 177.4, 49.5, 48.2, 43.1, 38.6, 34.5, 31.8, 27.6, 26.9,

+ 23.8, 22.1, 21.0, 16.1; -66.2 (c 3.0 CHCl3), HRMS calc. for C15H30NO [M + H] :

240.2322. Found: 240.2317

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]-3-phenylpropanamide (54f). 1H-

NMR (400 MHz, CDCl3): δ 7.29-7.25 (m, 2H), 7.21-7.19 (m, 3H), 5.04 (s,1H), 3.77-3.68

(m, 1H), 3.02-3.89 (m, 2H), 2.45 (t, J = 7.54 Hz, 2H), 1.89-1.85 (m, 1H), 1.70-1.59 (m,

3H), 1.48-1.41 (m, 1H), 1.09-0.90 (m, 2H), 0.85 (d, J = 6.52 Hz, 3H), 0.81 (d, J = 7.05

13 Hz, 3H), 0.76-0.63 (m, 2H), 0.72 (d, J = 6.86 Hz, 3H); C-NMR (100 MHz, CDCl3) δ

171.0, 140.8, 128.5, 128.3, 126.1, 49.7, 47.9, 43.0, 38.8, 34.4, 31.8, 31.7, 26.4, 23.6, 22.1,

+ 21.1, 16.0; -48.8 (c 7.3 CHCl3), HRMS calc. for C19H30NO [M + H] : 288.2322.

Found: 288.2321.

N-[(1R, 2S, 5R)-2-isopropyl-5-methylcyclohexyl]acetamide (54g). 1H-NMR (400

MHz, CDCl3): δ 5.23 (s, 1H), 3.81-3.72 (m, 1H), 2.03-1.87 (m, 2H), 1.97 (s, 3H), 1.73-

1.65 (m, 2H), 1.51-1.44 (m, 1H), 1.10-0.95 (m, 2H), 0.89 (d, J = 7.24 Hz, 3H), 0.88 (d, J

= 7.28 Hz, 3H), 0.79 (d, J = 6.90 Hz, 3H), 0.84-0.72 (m, 2H); 13C-NMR (100 MHz,

CDCl3) δ 169.1, 49.8, 48.0, 43.1, 34.4, 31.7, 26.7, 23.6, 23.5, 22.1, 21.1, 16.1; -64.3

+ (c 1.10 CHCl3), HRMS calc. For C12H23NO [M + H] : 198.1852. Found: 198.1852.

N-[(1R, 2R)-2-methylcyclohexyl]benzamide (racemic) (49).2,35 1H-NMR (400 MHz,

CDCl3): δ 7.78-7.75 (m, 2H), 7.51-7.47 (m, 1H), 7.44-7.41 (m, 2H), 5.90 (br s, 1H), 3.75-

3.66 (m , 1H), 2.07-2.04 (m, 1H), 1.81-1.68 (m, 3H), 1.42-1.31 (m, 2H), 1.28-1.12 (m,

13 3H), 0.99 (d, J = 6.52 Hz, 3H); C-NMR (100 MHz, CDCl3) δ 166.9, 135.1, 131.2,

128.5, 126.7, 54.4, 38.7, 34.3, 33.7, 25.7, 25.4, 19.1; HRMS calc. for C14H20NO [M +

H]+: 218.1539. Found: 218.1539.

Endo-N-(1,7,7-trimethylbicyclo[2.2.1]heptan-2-yl)benzamide (racemic) (58).2 1H-

NMR (400 MHz, CDCl3): δ 7.72-7.70(m, 2H), 7.51-7.47(m, 1H), 7.45-7.41 (m, 2H), 6.08

(d, J = 7.73 Hz, 1H), 4.14-4.09 (m, 1H),1.99-1.93 (m, 1H),1.81-1.58 (m, 4H), 1.40-1.1.34

(m, 1H), 1.24-1.18 (m, 1H), 1.01 (s, 3H), 0.92 (s, 3H), 0.87 (s, 3H); 13C-NMR (100 MHz,

CDCl3) δ 166.9, 135.3, 131.4, 128.8, 126.8, 57.2, 48.9, 47.3, 45.0, 39.4, 36.0, 27.2, 20.5,

+ + 20.5, 12.00. HRMS calc. for C14H20NO [M] :257.178 Found [M + H] :258.1855.

Exo-N-(1,7,7-trimethylbicyclo[2.2.1]heptan-2-yl)benzamide (59).29b 1H-NMR (400

MHz, CDCl3): δ 7.72-7.70 (m, 2H), 7.47-7.38 (m, 3H), 6.16 (d, J = 7.6 Hz, 1H), 4.13-

4.08 (m, 1H),1.96-1.90 (m, 1H),1.80-1.57 (m, 4H), 1.38-1.34 (m, 1H), 1.23-1.16 (m, 1H),

13 1.01 (s, 3H), 0.91 (s, 3H), 0.86 (s, 3H); C-NMR (100 MHz, CDCl3) δ 166.7, 135.1,

131.2, 128.5, 126.7, 57.1, 48.9, 47.1, 44.9, 39.0, 35.9, 27.0, 20.3, 20.3, 11.8. HRMS calc.

+ for C14H20NO [M] :257.178 Found:258.1855.

1 N-cyclopentylbenzamide (61a). H-NMR (400 MHz, CDCl3): δ 7.77-7.74 (m, 2H),

7.48- 7.37 (m, 3H), 6.33 (bs, 1H), 4.41-4.36 (m, 1H), 2.09-2.02 (m, 2H), 1.73-1.70 (m,

13 4H), 1.53-1.45 (m, 2H); C-NMR (100 MHz, CDCl3) δ 167.3, 134.9, 131.3, 128.5,

+ 126.9, 51.7, 33.2, 23.9; HRMS calc. for C12H15NO [M + H] : 190.1226 Found :

190.1227.

1 N-Cyclohexylbenzamide (61b). H-NMR (400 MHz, CDCl3) δ 7.76-7.74 (m, 2H), 7.49-

7.40 (m, 3H), 6.01 (bs, 1H) 4.02-3.94 m, 1H), 2.05-2.01 (m, 2H), 1.78- 1.63 (m, 4H),

13 1.44-1.371 (m, 2H), 1.28-1.18 (m, 4H). C-NMR (100 MHz, CDCl3) δ 166.9, 135.3,

+ 131.5, 128.7, 127.0, 48.90, 33.5, 25.8, 25.1. HRMS calc. for C13H17NO [M + H] :

204.1383 Found: 204.1384.

1 N-cyclooctylbenzamide (61d). H-NMR (400 MHz, CDCl3) δ 7.77-7.74 (m, 2H), 7.55 –

7.35 (m, 3H), 4.20 (m,1H), 1.93 (m, 2H), 1.76 – 1.46 (m, 13H). 13C-NMR (100 MHz,

CDCl3) δ 166.4, 135.22, 131.3, 128., 126.9, 49.9, 32.4, 27.3, 25.6, 23.8. HRMS calc. for

+ + C15H21NO [M+H] : 232.1696 Found [M + H] :232.1696

1 N-cyclododecylbenzamide (61e). H-NMR (400 MHz, CDCl3): δ7.77- 7.74 (m, 3H).

7.49- 7.41 (m, 3H), 5.93 (d, J = 8 Hz, 1H), 4.31-7.25 (m, 1H) 1.76-1.68 (m, 2H), 1.53-

13 1.37 (m, 24H). C-NMR (100 MHz, CDCl3) δ 166.9, 135.2, 131.5, 128.7, 127.0, 46.7,

+ 30.5, 24.22, 23.9, 23.7, 23.5, 21.6. HRMS calc. for C19H29NO [M + H] : 288.2322.

Found: 288.2329.

Bis((2S,5R)-2-isopropyl-5-methylcyclohexyl) sulfite (mixture of diastereomers) (53).

1 H-NMR (400 MHz, CDCl3) δ 4.34-4.18 (m, 2H), 2.17-2.01(m, 4H), 1.67-1.58 (m, 4H),

1.43-1.29 (m, 4H), 1.32-1.14 (m, 2H), 1.06-0.97 (m, 2H), (m, 12H), 0.8-0.77 (m, 6H).

13 C-NMR (100 MHz, CDCl3) δ 76.9, 75.2, 76.9, 48.1, 48.0, 43.6, 43.0, 34.2, 32.0, 31.9,

25.7, 25.5, 23.3, 23.2, 22.2, 21.2, 21.1, 15.8. colorles oil. HRMS cal. for C20H38O3S [M +

+ NH4] : 376.2880 Found: 376.2879

CHAPTER TWO ZINC (II) CATALYZED CONVERSION OF ALKYNES TO VINYL

TRIFLATES IN THE PRESENCE OF SILYL TRIFLATES

2.1 Introduction

Vinyl triflates continue to play an important role in organic synthesis most especially in

C-C bond formation using palladium-catalyzed coupling reactions. These coupling reactions have been exploited in the total synthesis of various natural products.36 Of these palladium-catalyzed reactions, the Heck reaction has found many exciting applications in this field,37,38 including methods to introduce new stereocenters as will be discussed below. Generally, this reaction allows the coupling of aryl or vinyl halides with alkenes in the presence of mild bases. With cyclic alkenes, Heck reactions can produce new stereogenic centers. Similar to other palladium-mediated coupling reactions, the Heck reaction proceeds by oxidative addition of palladium to a vinyl halide or triflate 72. To better understand the advantages of vinyl triflates over vinyl halides in palladium coupling reactions, we briefly examine the initial step of the Heck reaction as a representative example. The initial step in this reaction affords σ-complex 73 (Figure 7).

The rate of this step depends on the nature of the leaving group. Here palladium insertion is fastest with vinyl iodides followed by vinyl triflates. Vinyl bromides and chlorides react slower in this insertion step.

Figure 7. Catalytic cycle for Heck coupling

The mechanism and outcome of the second step in this coupling reaction depends on the leaving group (X) present in complex 73. The ligand exchange on palladium leading to π- complex 74 follows a cationic pathway with triflate leaving group (X = OTf). On the other hand, a similar exchange leading to 75 goes by a neutral route (X = halide, Figure

7) in the absence of a halide scavenger.39 The cationic pathway has important implications in asymmetric versions of the Heck coupling. Importantly, the triflate anion can more easily dissociate from the palladium center than halides (including iodide) to allow the formation of cationic species 74. It is this cationic intermediate that is most receptive to interactions with chiral ligands allowing the development of asymmetric

Heck reactions.

The utility of vinyl triflates over other vinyl halides is nicely highlighted in a report by

Shibasaki and coworkers. In their paper, an asymmetric intramolecular Heck-coupling is described and applied to the enantioselective synthesis of cis-decalins. Starting from prochiral vinyl iodide 76 and using (R)-(2,2'-bis(diphenylphosphino)-1,1'-binaphthyl)

(BINAP) as a chiral ligand, enantio-enriched cis-decalin 82 is produced (Table 7).40 Other vinyl triflate derivatives gave 82 but with lower enantioselectivities (36 - 44% ee).

However, replacing the iodide leaving group with triflate led to high optical purities (91% ee) though in lower yields relative to vinyl iodides (Table 7).

Table 7. Asymmetric Heck coupling using vinyl iodide

Further optimization studies performed by the same group led to the formation of cis- decalin from vinyl iodide with higher enantioselectivity (80% ee) by changing the counter anion of the base; nevertheless, this was achieved at the expense of lower reaction yields.41 When vinyl triflates were coupled in the presence of a tertiary amine base the reaction furnished a cis-decalin product in both high yield and enantioselectivity

(95% ee).42 The results of these optimization studies underscore the continuing need for efficient methods to access vinyl triflates. The positive outcome with vinyl triflates certainly encouraged Shibasaki and coworkers to attempt the total synthsis of (+)- vernolepin (83, Figure 8) from a cis-decalin intermediate that was prepared from a vinyl triflate in high enantiomeric excess.43

Figure 8. Structure of (+)-vernolepin

Vinyl triflates have found significant applications even outside of palladium-catalyzed reactions. As an example, the coupling of triflate 84 with various alkyl and aryl cuprate reagents provides a quick access to 3-substituted carbacephem 85 (Scheme 14).44

Scheme 14. Synthesis of 3-substituted carbacephem

Even outside of coupling reactions, vinyl triflates have proved valuable. Recently, Frantz and coworkers developed a method to prepare difficult-to-access chiral allenes from vinyl triflates 86 using palladium as a catalyst in the presence of a chiral phosphite ligand

(Scheme 15).45 Under these conditions, oxidative addition followed by β-hydride elimination gives the allene product with high enantiomeric excesses. This method utilizes β-hydride elimination which has often been considered an undesired pathway in other reactions.

Scheme 15. Enantioselective allene formation

2.2 Synthesis of Vinyl Triflate

Despite the utility of vinyl triflates in synthesis, relatively few methods are available to prepare this motif and these often involve the use of costly reagents.46 Though this area has not been recently reviewed,36a most methods continue to use carbonyl compounds, especially ketones, as starting materials. One typical approach would be the enolization of ketone 88 followed by trapping with a trifluoromethylsulfonylating agent leading to the formation of vinyl triflates 90 and 91 (route a, Scheme 16).47,48 An alternative approach has been reported in which an initial enolization step is not required. In this method the direct addition of triflic anhydride to ketones takes place to give gem- bis(triflate) intermediate 92; this intermediate is then converted to vinyl triflate products upon treatment with a bulky base (route b, Scheme 16).49 One of the important drawbacks of these methods (routes a and b) is that often products are formed with little regio- and/or diastereoselectivity. Even starting from aldehydes where regioselectivity is not an issue, the terminal vinyl triflate product is often formed as a mixture of Z/E isomers.

Scheme 16. Synthesis of vinyl triflate from ketone

An improved synthesis of vinyl triflates from metal enolates has been reported by

Comins using N-(2-pyridyl)triflimide (93, Figure 9). The Comins reagent 93, a common trifluoromethylsulfonylating agent, is used to prepare vinyl triflates from ketones, aldehydes, and lactams.50 N-acyl lactams can be easily converted to vinyl triflates using

the Comins reagent.51 This transformation requires the use of a base to form enolate intermediates which react with N-(2-pyridyl)triflimide to form vinyl triflates.

Figure 9. Comins reagent

Important to the topic of this chapter, alkynes serve as the only alternative to carbonyl compounds as precursors to acyclic vinyl triflates. The addition of trifluoromethanesulfonic acid (triflic acid) to alkynes was first reported by Stang

(Scheme 17).52 Studies have shown that, at low temperatures, protonation of alkyne 94 gives vinyl cation 95 that is subsequently trapped by triflate anion to furnish both stereoisomers of 96.53 As with the enolate approaches just described, conversions of internal alkynes using triflic acid always produce a mixture of vinyl triflate isomer. As expected, the regioselectivity of the addition is dependent on the ability of R or R1 to stabilize vinyl carbocation intermediate 95.

Scheme 17. Addition of triflic acid to alkyne

As a partial solution to the problem of stereoselectivity, Sommer and coworkers recently reported regio- and diastereoselective additions of triflic acid to propynoic acid derivatives 97; this approach allows the formation of E-vinyl triflate 99 (Scheme 18).54

Vinyl cation 98 which formed upon protonation of the precursor alkyne is trapped by the triflate anion from the face that is anti to the alkyl carboxylate group. The authors suggested that this facial selectivity is a result of an electronic repulsion (not steric) between the incoming triflate anion and the alkoxy group of the ester (Scheme 18).

Scheme 18. Addition of triflic acid to propynoic acid derivatives

The addition of triflic acid to alkynes has also been reported in the presence of Lewis acids. Zhang and coworkers report a gold-catalyzed addition of sulfonic acids to alkynes

(Scheme 19).55 The addition of sulfonic acid to an alkyne-Au complex gives intermediate

101 which isomerize to intermediate 103 via cationic gold carbene species 102. Vinyl triflate with E-geometry is then formed by protodemetallation. Similar stereoselectivity has been reported for the addition of sulfonic acids to alkynes in the presence of a Rh catalyst.56

Scheme 19. Gold-catalyzed addition of triflate to an alkyne

While ketone and triflic acid based approaches have long been used to produce valuable vinyl triflate intermediates, these methods have numerous limitations. The triflic acid strategy in particular suffers from several drawbacks including harsh acidic conditions and the requirement to use an exacting work-up protocol to obtain the modest diastereoselectivities reported. Thus several groups have recently strived to develop new methods capable of efficiently producing vinyl triflates in higher selectivity and appropriate for use in the synthesis of more complex targets.

In an attempt to meet the challenge of a more mild and selective method, Gaunt reported a new approach to the synthesis of highly substituted vinyl triflates using diaryl iodonium triflate reagents.57 This copper (I) catalyzed method starts with alkyne 94 and entails the transfer of an aryl (and vinyl) ligand from iodonium reagent via complex 106 (Scheme

20). Vinyl triflate 105 was then isolated as the major product in exceptional diastereoselectivity by a reductive elimination of intermediate 107 (Scheme 20). This is an excellent approach that allows the synthesis of highly substituted vinyl triflates.

Scheme 20. Electrophilic carbon functionalization of alkynes

Though not formally a vinyl triflate synthesis method, Liu reported a regioselective functionalization of internal alkyne 108 also employing hypervalent iodine reagents in the presence of a base (Scheme 21).58 This reaction mainly afforded hydrolysis product

109 though small amounts of vinyl triflate were also observed.

Scheme 21. Regioselective functionalization of internal alkynes

2.3 Development of Zinc (II) Catalyzed Conversion of Alkynes to Vinyl Triflates in

the Presence of Silyl Triflates

Partially as a result of serendipity, we discovered that alkynes react with Cu(OTf)2 to form vinyl triflates. In our subsequent literature search to better understand the significance of this discovery, we found a preliminary report suggesting that the synthesis of vinyl triflates could be accomplished with aryl alkynes using trimethylsilyl triflate

(TMSOTf) over long reaction times.59 Though not a vinyl triflate-forming method, we also noted that Shaw and coworkers reported the formation of TMS-alkyne (112, Scheme

22) under rather similar conditions. Specifically, starting from terminal alkyne 94,

TMSOTf was used in the presence of a Zn(II) catalyst and an amine base to form 112

(Scheme 22).60 Based on mechanistic insight gained from studies of our new reaction, we envisioned that the transformation reported by Shaw to form the TMS-alkyne proceeds via a vinyl triflate intermediate (111, R2 = H). From these reports we envisioned the

possibility of forming vinyl triflates from alkyne using catalytic transition metals with

TMSOTf as the stoichiometric triflate source.

Scheme 22. Strategy for metal catalyzed alkyne conversion to vinyl triflate

We began our study by examining the conversion of phenyl acetylene (113) to vinyl triflate 114 in the presence of TMSOTf (1 equiv). While this transformation required 48 hours at room temperature to achieve a 90% conversion, the same reaction in the presence of 10% Cu(OTf)2 afforded 114 in a 75% conversion in 1 hour (Table 8, entry 1).

Further rate enhancement was observed when excess TMSOTf (1.5 equiv) was used

(Table 8, entry 2). We then examined the reactivity of other Lewis acids; to our surprise, we found that copper (I) (entries 3 – 6) and a variety of other metals also expedite this transformation. The fact that so many metals can catalyze this reaction led us to suspect that these metals were merely a source of water in this reaction. As will be discussed later in this chapter, our studies of the role of water on this transformation demonstrate that the metal plays a key role. Of the metals examined, Zn(OTf)2 was found to be the optimal metal salt to catalyze this transformation (entry 9). Then we proceed to examine the effect of other silyl triflates; bulkier silyl triflate reagents appeared to slow the rate of the reaction considerably (entries 10-12). Thus, while reactions with TMSOTf leads to high conversions in 20 min, reactions of 113 with triethylsilyl triflate required 210 min to give a similar conversion to vinyl triflate 114. With still bulkier silyl triflates high conversions

to 114 were only achieved after extended reaction times (48 hours). Ultimately, 1.5 equivalents of TMSOTf and 10% Zn(OTf)2 were found to be the optimal conditions for this reaction.

Table 8. Screening of metal catalysts and silyl triflate reagents

With the optimized conditions in hand, the generality of this method with various terminal and internal alkynes was examined (Tables 9 and 10). Terminal aryl alkynes were rapidly converted to their corresponding vinyl triflate derivatives, especially with an electron-donating group on the aromatic ring (Table 9, entry 3). On the other hand, reactions involving electron withdrawing groups on the aryl ring required longer times

(Table 9, entries 2 and 4). Similar reaction times were required for reactions involving terminal alkynes bearing alkyl groups (entries 5 – 9). These reactions required substantially longer times (16 – 20 h) to produce the corresponding triflates in good yields.

Table 9. Reactions with terminal alkynes

Internal alkynes also reacted slowly under these conditions and appeared to favor products of Z-geometry especially with increasing the bulkiness of alkyl substituents. We observed an increase in Z-selectivity from smaller groups such as methyl (Table 10, entry

1) to isopropyl alkyne 133 which gave nearly a 5:1 ratio of Z-vinyl triflate 137 (Table 10, entry 4). However, when the bulky tertiary butyl group was used, various side products were observed. We presumed that some triflate derivatives were present in this mixture of products. Thus the mixture was allowed to react with diluted aqueous KOH in 1,4- dioxane in an attempt to characterize any potential ketone derivatives. This approach did not yield any useful information. Otherwise, these hydrolysis conditions effectively converted vinyl triflates 114 and 123-129 to their corresponding ketone derivatives in good yields.

Table 10. Reactions with internal alkynes

Due to the relatively poor stability of these vinyl triflates, we were not able to perform the usual rigorous level of characterization. We relied primarily on 1H and 13C-NMR reported spectra for confirmation of the identity of our compounds. Furthermore, as just discussed, vinyl triflates were hydrolyzed and their corresponding ketones were characterized. To further confirm the identity of these products, compound 114 was prepared and then was treated with phenylboronic acid derivatives under palladium coupling conditions (Scheme 23). These reactions led to the formation of products 114a and 114b in a good two-step yield. Importantly, these coupling products were rigorously characterized thus providing additional evidence for the existence of vinyl triflate 114.

Scheme 23. Suzuki coupling with vinyl triflate

Though originally used to confirm the identity of triflate product, this palladium coupling reaction (Scheme 23) led us to consider the possibility of a direct formation of disubstituted alkenes from alkyne. We envisioned that palladium coupling (Suzuki

reaction in this case) could be made to occur in tandem with the in situ formation of vinyl triflate for a convenient access to germinal disubstituted alkenes. In fact, this tandem reaction led to formation of product 114a but with a low yield (17%, Scheme 24). We suspected that this low yield is due to poor solubility of certain reactant(s) in chloroform.

To address this concern, the same reaction was repeated using THF, a common solvent for Suzuki reaction, but no coupling product was observed. Further control studies showed that THF was a very poor solvent for the formation of vinyl triflates. We suspect that more appropriate conditions can be identified for a more efficient tandem reaction; however, this was not pursued partly due to the presence of very similar methods in literature.61

Scheme 24. Tandem vinyl triflate/Suzuki coupling

2.4 Mechanistic studies

As alluded to earlier, the metal salt study showed that several metals can catalyze the formation of vinyl triflate and this observation led us to investigate the possibility of water (of hydration) playing an important role in the present reaction. Indeed, in our studies of this reaction, we noticed that the product yield diminished substantially as greater efforts were made to rigorously exclude water. When zinc (II) triflate was subjected to high-vacuum (0.01 torr) overnight to remove trace moisture, the conversion of 113 to 114 decreased to 39% (Table 11, entry 1). The inclusion of 0.25 equivalents of

water led to similar rates of formation of vinyl triflate as with the unmodified catalyst; however, further increase in water content did not change the reaction rate (Table 11).

When the same study was performed using internal alkynes, water appears to accelerate the formation of vinyl triflate but deteriorate the Z-selectivity; 0.25 equivalents of water gave 91% conversion but with little selectivity (Table 11).

Table 11. Effect of water on the formation of vinyl triflate

In light of these results, we suspected that the direct addition of triflic acid to the alkyne substrate (presumably formed in situ from TMSOTf and water) was a possible mechanism for this transformation. Thus, a comparison study was performed using phenyl acetylene and triflic acid as source of triflate (Table 12, entry 2). This reaction resulted in the formation of several side products that are best explained from the intermediacy of a vinyl cation. By contrast, the same reaction performed in the presence of a catalytic amount of zinc (II) triflate resulted in the formation of vinyl triflate 114 with very little side products (Table 12, entry 4). A comparison of the crude reaction H-

NMR spectra for these two reactions (i.e. with and without Zn(OTf)2) clearly demonstrates the impact of the metal on product purity (Figure 10).

When we examined the present reaction with an internal alkyne in the presence and in the absence of zinc (II), no selectivity was observed (Table 12, entries 3 and 5). Once again, we examined the possibility that the metals merely served to produce minute quantities of triflic acid throughout the course of the reaction. To evaluate this possibility, a dilute triflic acid solution was added slowly to a reaction mixture containing internal alkyne 129 to mimic the in situ formation of the acid (Table 12, entry 1). This reaction gave vinyl triflate 135 as a major product with predominant E-geometry (Z/E 1:4). This is a complete reversal of selectivity relative to our zinc (II) triflate catalyzed reaction. These control experiments argue against a mechanism involving the direct addition of triflic acid to alkyne.

Table 12. Control experiments involving TfOH

Figure 10. Triflic acid addition to 113 in the presence and the absence of metal salt

In a separate study, we examined the reaction between phenylacetylene and TMSCl in the presence of 10% Zn(OTf)2 expecting the formation of vinyl chloride product. Under these conditions, ~20% conversion to vinyl triflate 114 was observed and vinyl chloride was not detected. This observation along with the previous result, suggests that the zinc

(II) reagent plays a unique role, likely by delivering its triflate ligand to the carbocationic center in zwitterionic intermediate 139 (Scheme 25). Evidence for similar intermediates has been put forward by Sudo and coworkers to explain the regio- and stereoisomers observed from the reaction between alkynes and trialkylsilanes in the presence of a Lewis acid (Scheme 25).62 Sudo and coworkers proposed metal coordination to alkynes to produce zwitterionic intermediate 140. This intermediate is then trapped by a hydride delivered from triethylsilane reagent to produce aluminium ate complex 141. Coupling between this complex and the electrophilic silicon takes place to give vinyl silane as a product 142.

Scheme 25. Possible role of zinc(II) reagent involving zwitterionic intermediate

By analogy, we suggest that trapping the vinyl cation in 139 by a triflate attached to zinc

(II) leads to the formation of vinyl intermediate 143 which may lead to vinyl triflate product directly by reaction with water (pathway a, Scheme 26). This pathway should yield the E-isomer of vinyl triflate (Scheme 26). However, our reaction leads to a predominance of the Z-isomer, thus a possible pathway for the formation of vinyl triflates with Z-geometry might involve the E-isomer of intermediate 143 (assuming vinyl zinc

143 is configurationally stable). An alternative approach involves the reaction between

143 and an electrophilic silyl reagent to furnish vinyl silane intermediate 144 (pathway b,

Scheme 26). Furthermore, the formation of vinyl silicon 144 is expected to be slow with hindered silyl triflates. Thus, this pathway might explain the slow formation of vinyl triflates with bulky silicon reagents (Table 8, entries 10 – 12).

Scheme 26. Possible routes for the formation of vinyl triflate

The reaction between a similar vinyl zinc intermediate and an electrophilic silicon reagent has been reported by Knochel who demonstrates the reaction between intermediate 145 and TMSCl to give product 146 in a good yield (Scheme 27).63

Scheme 27. Formation of vinyl silicon from vinyl zinc reagent

Based on the aforementioned control studies and literature precedents, we argue that the formation of a vinyl silane intermediate under our Zn(II) conditions requires electrophilic silicon reagents. However, control studies showed that water is required for this transformation. We do not fully understand the role of water in this reaction. We suspect that, instead of producing triflic acid, water reacts with TMSOTf to give protonated

+ silanol (TMSOH2 ). Studies suggest that silanols are relatively basic and their protonated form possesses a silicon geometry resembling silicon cation. Specifically, these crystal

+ structure studies reveal that the Si-OH2 bond distance is 1.779 Å which is 0.1-0.2 Å longer than Si-OH and Si-OR bonds; furthermore, the C-Si-C bond angel is 116° that is very similar to trialkylsilicon cation with coordinated acetonitrile (115.6°).64 Judging for this report, we argue that silicon (and not a proton) serves as the primary electrophile for the reactions with metal vinyl intermediate. This reaction leads to the formation of vinyl silane intermediate 144 and provides a proton source for its conversion to oxonium ion54

147 (Scheme 28). Then this intermediate undergoes protodesilylation once the silyl group in 147 is orthogonal to the carbonyl unit. This can occur in two conformations 147a and

147b. However, conformer 147b is less favored due to an eclipsing interaction between

the R and R1 groups (Scheme 28). Elimination of the silyl group from the more favored conformer then affords vinyl triflate product with the Z-geometry as the kinetic product.

Scheme 28. Proposed mechanism for the observed Z-selectivity

In summary, we described a mild reaction for the preparation of vinyl triflates from alkynes catalyzed by several metal triflates. The method is general to both aliphatic and aromatic terminal alkynes as well as internal alkynes bearing an aryl substituent. For Z- diastereoselectivity, the reaction requires silyl triflates pointing to a unique role for silicon in the mechanism. Further studies suggest that silicon initially serves as a target for the nucleophilic vinyl metal intermediate allowing the catalyst to be regenerated.

Subsequent removal of silicon from the substrate in the presence of limited amounts of water leads to the observed triflate product.

2.5 Experimental section:

General Information: Reactions were carried out under an argon atmosphere (unless otherwise stated) in oven dried glassware with magnetic stirring. Purification of reaction products was carried out using flash silica gel 40-63μ. Analytical thin layer chromatography was performed on 0.25mm silica gel 60-F plates. Visualization was accomplished with UV light and aqueous potassium permanganate solution staining followed by air drying. 1H-NMR was recorded on a Varian Mercury 400 (400 MHz)

spectrometer and are reported in ppm using solvent as an internal standard (CDCl3 at 7.26 ppm). Data are reported as: (b = broad, s = singlet, d = doublet, t = triplet, q = quartet, p = pentet, m = multiplet; coupling constant(s) in Hz, integration). 13C-NMR were recorded on Varian Mercury400 (100 MHz) spectrometer. Chemical shifts are reported in ppm, with solvent resonance employed as the internal standard (CDCl3 at 77.16 ppm). High- resolution mass spectra were obtained from University of Florida Mass Spectrometry

Laboratory.

Materials: All reagents were purchased form commercially available sources and were used without further purification.

General procedure for Sonogashira Coupling:65 A mixture of aryl halide 1 (0.5 mmol), alkyne 2 (0.6 mmol), PdCl2(PPh3)2 (3 mol %), and TBAF.3H2O (3 equiv) was stirred at rt for the desired time until complete consumption of starting material as monitored by TLC.

After the mixture was washed with water, extracted with ether, and evaporated, the residue was purified by flash column chromatography (10 cm silica, hexane) to afford the corresponding coupled products

General procedure for the formation of vinyl triflate: In an oven dried vial charged with stirrer, alkyne (1mmol) and zinc triflate (0.1mmol) were added then dissolved in 3mL

CDCl3. To that solution TMSOTf (1.5 mmol) was added and the reaction was monitored by

H-NMR. After completion 200 mg silica was added, to hydrolyze excess TMSOTf, and then the mixture was filtered through cotton. The resultant solution was evaporated under a stream of argon then under high vacuum for 1h.

Spectral data:

66 1 (3-methylbut-1-yn-1-yl)benzene (133). H-NMR (400 MHz, CDCl3): δ 7.42-7.40 (m,

13 2H), 7.29-7.26 (m, 3H), 2.8 (m, 1H), 1.39 (d, 9H); C-NMR (100 MHz, CDCl3): δ 131.8,

+ 128.4. 127.7, 124.3, 96.0, 80.0, 23.3, 21.4. HRMS calc. for C11H13 [M+H] : 145.1017

Found: 145.1020

67 1 (3,3-dimethylbut-1-yn-1-yl)benzene (134). H-NMR (400 MHz, CDCl3): δ 7.39 -7. 27

13 (m, 2H), 7.27 -7. 25 (m, 3H), 1.32 (s, 9H); C-NMR (100 MHz, CDCl3): δ 131.7, 128.3,

127.6, 124.3, 98.7, 79.2, 31.3, 28.1.

1-phenylvinyl trifluoromethanesulfonate (114).21 1H-NMR (400 MHz, C6D6): δ 7.56

(d, J = 8 Hz, 2H), 7.41 (d, J = 8 Hz, 2H), 4.89 (d, J = 4Hz, 1H), 4.85 (d, J = 4Hz, 1H),

13 1 C-NMR (100 MHz, C6D6): δ 153.7, 130.2, 128.9, 128.3, 125.5, 105.0, 118.6 (q, J C-F =

318 Hz).

1-(4-bromophenyl)vinyl trifluoromethanesulfonate (123).68 1H-NMR (400 MHz,

CDCl3): δ 7.56 (d, J = 8Hz, 2H), 7.41 (d, J = 8 Hz, 2H), 5.62 (d, J = 4 Hz, 1H), 4.42 (d, J

13 = 4Hz, 1H); C-NMR (100 MHz, CDCl3): δ 152.6, 105.0, 132.3, 131.1, 126.9, 124.9,

1 118.6 (q, J C-F = 318 Hz).

1-(4-methoxyphenyl)vinyl trifluoromethanesulfonate (124). 1H-NMR (400 MHz,

CDCl3): δ 7.48 (d, J = 12 Hz, 2H), 6.92 (d, J = 12 Hz, 2H), 5.46 (d, J = 4 Hz, 1H), 5.25

(d, J = 4 Hz, 1H), 3.02 (s, 3H). This compound was hydrolyzed and 4ʹ-

1 methoxyacetopheneone was isolated; H-NMR (400MHz, CDCl3): δ 7.93 (d, J = 8 Hz,

13 2H), 6.91 (d, J = 8 Hz, 2H), 3.85 (s, 3H), 2.54 (s, 3H), C-NMR (100 MHz, CDCl3): δ

196.9, 163.5, 130.7, 127.9, 113.7, 55.5, 26.42. methyl 4-(1-(((trifluoromethyl)sulfonyl)oxy)vinyl)benzoate (125). 1H-NMR (400

MHz, CDCl3): δ 8.09 (d, J = 8 Hz, 2H), 7.61 (d, J = 8 Hz, 2H), 7.16 (d, J = 8Hz, 1H),

5.49 (d, J = 8Hz, 1H), 4.39 (t, J = 7.2 Hz, 2H), 1.04 (t, J = 7.2 Hz, 3H). 13C-NMR (100

1 MHz, CDCl3): δ 165.8, 152.6, 136.0, 132.2, 130.2, 125.4, 118.6 (q, J C-F = 318 Hz),

106.4, 61.5, 14.8.

70 1 hex-1-en-2-yl trifluoromethanesulfonate (126). H-NMR (400 MHz, CDCl3): δ 5.09

(d, J = 4Hz, 1H), 4.93 (d, J = 4Hz, 1H), 2.34 (t, J = 8 Hz, 2H), 1.51 (m, 2H), 1.39 (m,

13 1 2H), 0.93 (t, J = 8 Hz, 3H); C-NMR (100 MHz, CDCl3): δ 157.2, 118.6 (q, J C-F = 318

Hz), 104.1, 33.7, 28.2, 21.9, 13.8.

1 4-methylpent-1-en-2-yl trifluoromethanesulfonate (127). H-NMR (400 MHz, CDCl3):

δ 5.12 (d, J = 3.2 Hz, 1H), 4.92 (dd, J = 3.4, 0.4 Hz, 1H), 2.2 (d, J = 7.2 Hz, 2H), 1.95 -

13 1. 85 (m, 1H), 0.96 (d, J = 6.8 Hz, 6H); C-NMR (100 MHz, CDCl3): δ 156.2, 118.5 (q,

1 J C-F = 315Hz), 105.3, 43.3, 25.6, 21.9.

3,3-dimethylbut-1-en-2-yl trifluoromethanesulfonate (128).70 1H-NMR (400 MHz,

13 CDCl3): δ 5.06 (d, J = 4 Hz, 1H), 4.96 (d, J = 4 Hz, 1H), 1.18 (s, 9H), , C-NMR (100

1 MHz, CDCl3): δ 146.6, 118.5 (q, J C-F = 318 Hz), 99.9, 36.6, 27.6.

3-methylbut-1-en-2-yl trifluoromethanesulfonate (129).18 1H-NMR (400 MHz,

CDCl3): δ 5.07 (d, J = 3.6 Hz, 1H), 4.93 (dd, J = 1.12, 3.6 Hz, 1H), 2.56 (m, 1H), 1.16

13 1 (d, J = 8 Hz, 1H), C-NMR (100 MHz, CDCl3) δ 162.3, 118.6 (q, J C-F = 318 Hz), 101.7,

33.1, 19.9.

69 1 1-cyclohexylvinyl trifluoromethanesulfonate (130). H-NMR (400 MHz, CDCl3): δ

5.05 (d, J = 4 Hz, 1H), 4.88 (dd, J = 4, 2.8 Hz, 1H), 2.29-2.15 (m, 1H), 1.31-1.17 (m,

13 6H), 1.18-1.17 (m, 4H), C-NMR (100 MHz, CDCl3) δ 161.6, 117.1, 101.9, 42.4, 30.4,

25.9, 25.7.

1-phenylprop-1-en-1-yl trifluoromethanesulfonate (135).70 Two isomers were

1 observed 68:32 Z:E, for Z isomer (major) H-NMR (400 MHz, CDCl3): δ 7.46 -7. 37 (m,

5H), 5.95 (q, J = 8 Hz, 1H), 1.85 (d, J = 8 Hz, 1H), For E isomer (minor) 1H-NMR (400

MHz, CDCl3): δ 7.46 -7. 37 (m, 5H), 5.94 (q, J = 8 Hz, 1H), 1.95 (d, J = 8Hz, 1H).

1-phenylbut-1-en-1-yl trifluoromethanesulfonate (136). two isomers were observed

75:25 Z:E, for Z isomer (major), 1H-NMR (400 MHz, CDCl3): δ 7.46 -7. 37 (m, 5H),

5.85 (t, J = 8 Hz, 1H), 2.23 (p, J = 8 Hz, 2H), 1.09 (t, J = 8 Hz, 3H); 13C-NMR (100

1 MHz, CDCl3): δ 146.7, 131.8, 130.0, 128.7, 128.6, 125.9, 118.6 (q, J C-F = 318 Hz), 21.1,

1 14.1. For E isomer (minor), H-NMR (400 MHz, CDCl3): 7.37-7.46 (m, 5H), 5.83 (t, J =

13 8 Hz, 2H), 2.39 (p, J = 8 Hz, 2H), 1.14 (t, J = 8 Hz, 3H); C-NMR (100 MHz, CDCl3): δ

1 146.7, 131.7, 129.5, 128.8, 128.7, 125.8, 118.6 (q, J C-F = 318 Hz), 20.4, 13.4. Hydrolysis

71 1 of compoud 11a gives Butyrophenone : H-NMR (400 MHz, CDCl3): δ 7.97 -7. 95 (m,

2H), 7.56-7.53 (m, 1H), 7.53-7.43 (m, 2H), 2.94 (t, J = 7.4 Hz, 2H), 1.77 (m, 2H), 1.00

13 (t, J = 7.4 Hz, 3H); C-NMR (100 MHz, CDCl3): δ 200.5, 137.1, 132.9, 128.6, 128.1,

40.5, 17.8, 14.0.

3-methyl-1-phenylbut-1-en-1-yl trifluoromethanesulfonate (137). 1H-NMR (400

MHz, CDCl3): δ 7.36-7.44 (m, 5H), 5.67 (d, 1H), 2.54 (m, 1H), 1.09 (d, J = 4Hz, 6H),

13 1 C-NMR (100 MHz, CDCl3): δ 145.6, 131.2, 130.0, 128.9, 128.7, 126.0, 118.6 (q, J C-F =

318 Hz), 27.5, 23.0.

Spectral data for Suzuki coupling products:

1-methoxy-4-(1-phenylvinyl)benzene (114a).72 (74% two step yield); 1H NMR (400

MHz, CDCl3) δ 7.32-7.29 (m, 5H), 7.26 (d, J = 8.8 Hz, 2H), 6.84 (d, J = 8 Hz, 2H), 5.38

(d, J = 1.2 Hz, 1H), 5.34 (d, J = 1.2 Hz, 1H), 3.78 (s, 3H); 13C NMR (100 MHz, CDCl3):

δ 159.4, 55.36, 149.6, 141.9, 134.0, 128.4, 129.5, 128.2, 127.8, 113.6, 113.0. HRMS calc.

+ for C15H15O [M + H] : 211.1117. Found: 211.1118

1-chloro-4-(1-phenylvinyl)benzene (114b).73 (71% two-step yield); 1H-NMR (400

13 MHz, CDCl3): δ 7.28-7.35 (m, 8H), 5.49 (s, 1H), 5.47 (s, 1H); C-NMR (100 MHz,

CDCl3): δ 149.1, 141.1, 140.1, 133.7, 129.7, 128.5, 128.4, 128.3, 128.0, 114.8.

CHAPTER THREE SILICON-NUCLEOPHILIC ASSISTED LEAVING GROUP (S-

NALG)

3.1 Introduction

Positron Emission Tomography (PET) is a non-invasive imaging method that allows monitoring of physiological, biochemical, and pharmacological functions at the molecular level.74 PET is emerging as an important diagnostic tool in modern medicine due to its ability to identify some diseases in the early stages. This technique requires patients to be injected with small quantities of radiotracers prior to imaging. These radiotracers are small bioactive molecules labeled with radionuclides which emit positrons. Each of these positrons yields two photons generated in a process known as annihilation.75 Annihilation occurs when a positron collides with an electron; this process emits two gamma ray photons that are 180° apart from each other. Ultimately, these photons are detected and the resulting signal is used to accurately determine the location of the radiotracer. Using computer software, a three dimensional image of the region where the radiotracer accumulates in an organism is then reconstructed.

Table 13. Half-life time of commonly used positron-emitting nuclides

Several positron-emitting radionuclides can be used to label bioactive compounds for

PET imaging (Table 13).76 However, the half-life of these nuclides limits their practical application. For example, the very short half-life of oxygen-15 (2.03 min) makes it impractical for PET usage. Nitrogen-15 and carbon-11 have somewhat longer half-lives that allow their incorporation through chemical reaction into bioactive molecules but their preparation must be carried out with extreme rapidity. In practice this requires on- site preparation meaning that a medical imaging facility must be equipped with a cyclotron to enable the preparation of these radiotracers which can then be used immediate (no travel time required). Conversely, iodine-124 has a relatively long half-life which allows its incorporation into bioactive compounds; however, this long half-life subjects patients to unnecessary radiations after PET scan.77 In light of these considerations, many have considered fluorine-18 to possess optimal chemical and nuclear properties for PET applications. The half-life of 18F (109.8 min) enables its incorporation into bioactive molecules without the need for an on-site cyclotron facility and without exposing patients to a useless radiation after PET imaging.78 Furthermore, positrons emitted from 18F nuclides have relatively low energy and high abundance which makes a picomolar quantity of 18F-radiotracers sufficient for a good quality PET image.79

From a radiochemical viewpoint, the preparation of radioactive fluorine can be achieved with different purities using one of the following techniques.80 A carrier free (c.f.) method refers to a preparation in which the radioactive isotope is undiluted by its nonradioactive isotope (19F in this case). In a no carrier added (n.c.a.) approach, radionuclide is produced along with its non-radionuclide contaminant (19F) in roughly equal amounts. Finally, in a carrier added (c.a.) approach, the stable isotope (19F) of the radionuclide is added during the preparation.

Furthermore, depending on the synthesis possibilities of the starting material, radiofluorination can be achieved electrophilically or nucleophilically.81 While nucleophilic 18F-fluoride reagents are initially prepared from 18O-enriched water as will be described later, electrophilic 18F-fluoride reagents are prepared from high pressure Ne

18 gas or O2 enriched oxygen.

18 18 Originally, [ F]F2 gas, an electrophilic source of fluorine-18, was used to prepare F- radiotracers.82 However, due to its high reactivity and low selectivity several other electrophilic fluorinating reagents have emerged. Examples include acetyl hypofluorite

(148, Figure 11)83 and N-[18F]-fluoropyridinium triflate (149, Figure 11).84 These are

18 prepared from F2 and employed in electrophilic radiofluorination of bioactive molecules.85

Figure 11. Common electrophilic fluorinating reagents

Though many examples can be found in the literature for the electrophilic radiofluorination of bioactive compounds, two examples that exemplify the most

common approaches will be discussed here. [18F]-6-fluoro-dihydroxyphenylalanine (6-

[18F]-FDOPA) (151) is a currently used radiotracer for patients with Parkinson’s disease.86 This radiotracer has been synthesized with low radiochemical yield (RCY) by electrophilic aromatic substitution approach (SEAr). Starting from L- dihydroxyphenylalanine (150), product 151 is obtained along with other side products

18 87,88 upon reaction with either F2 or acetyl hypofluorite (148) (Scheme 29).

Scheme 29. Electrophilic approach to synthesis 18F-FDOPA

Another example is [18F]-2-fluoro-2-deoxyglucose ([18F]FDG), one of the few 18F- labelled radiopharmaceutical agents in general usage. This compound has been

89 18 synthesized with low RCY (8%) from dihydro-2H-pyran 154. Using [ F]F2, 154 reacts to give a mixture of [18F]FDG (155) and its epimer [18F]-2-fluoro-2-deoxymannose

([18F]FDM, 156) (Scheme 30). To minimize the formation of side products in the electrophilic radiofluorination approach, transition metal-mediated methods have been reported. However, due to the issues with metal toxicity, these approaches are currently used to synthesize tracers for use with non-human primate (NHP) PET imaging.90

Scheme 30. Electrophilic addition of 18F to dihy-2H-dropyran 154

Important to the topic of this chapter, a nucleophilic radiofluorination approach provides an alternative avenue to the previously discussed electrophilic radiofluorination techniques. Nucleophilic [18F]-fluoride ions can be prepared from an aqueous solution of

[18F]HF which can be generated form 18O-enriched water.91 This acidic solution is then

18 treated with K2CO3 to produce an aqueous [ F]KF solution. However, the hydrated fluoride exhibits poor nucleophilicity. Even isolated as [18F]KF, problems of solubility in typical organic reaction solvents arise. Thus preparations involving the azeotropic removal of water and simultaneous sequestering of [18F]KF with phase transfer agents are essential for effective nucleophilic radiofluorination methods. One of the most commonly used reagents in these preparations is kryptofix 2.2.2 (K 2.2.2, Figure 12).

Figure 12. Structure of kryptofix 2.2.2 (K 2.2.2)

Nucleophilic aromatic substitution (SNAr), a complimentary approach to SEAr, is a common route that allows activated aryl rings to be functionalized with radioactive fluorine. An example of SNAr is the direct radiofluorination of 157 to produce N-methyl-

[18F]fluorospiperone (158) using [18F]KF and K 2.2.2 (Scheme 31).92

Scheme 31. Synthesis of N-methyl-[18F]fluorospiperone using SNAr approach

Aliphatic substitution reactions are also used to introduce the radioactive fluoride into bioactive molecules. A well-known example is the synthesis of 18F-FDG from 159

(Scheme 32).93 The triflate leaving group at C-2 is displaced by a 18F-fluoride. This conversion is highly optimized for that monosaccharide system (> 50% yield using automated procedures); however, the reaction conditions have not translated well to the labeling of other compounds especially those that are more complex.

Scheme 32. Synthesis of [18F]FDG by nucleophilic substitution

As alluded to earlier, both electrophilic and nucleophilic approaches find important applications in the preparation of radiotracers. However, although fluorine-18 is perhaps the optimal nuclide for PET imaging, it still presents great challenges to synthetic chemist. Specifically, the half-life of the 18F-radionuclide (109.8 min) requires that synthesis of 18F-lableled radiotracers be accomplished in very brief amount of time.

Furthermore, these radiotracer products then need to be purified rapidly and transported to the medical imaging site so that sufficient time is available for imaging to take place with a usable radioactivity of radiotracer.94

This synthetic challenge limits the widespread application of PET imaging. While many radiotracers have been reported in the literature, only few (such as FDG) have found wide-spread medical application. The methods found in literature for labeling bioactive molecules with 18F-nuclide generally suffer from drawbacks including poor chemoselectivity and harsh conditions.

Some have attempted to circumvent these drawbacks by preparing simple 18F-containing fragments which are then coupled to bioactive molecules for subsequent use as imaging agents. However, this approach has its drawbacks as well since the fragments often extensive purification and still must be coupled to a bioactive molecule.95 The ideal synthetic approach to an imaging agent should have a rapid 18F-labeling occur as the last step with high chemoselectivity to avoid side product formation allowing for a rapid and trivial purification of the radiotracer. This goal has been difficult to achieve in 18F- labeling involving C-F bond formation leading a few researchers to seek out other targets for nucleophilic 18F-fluorine.

3.2 18F-Organosilicon radiotracers

More recently, organosilicon compounds have been identified as excellent receptors for radioactive fluorine due to the strength of the silicon-fluorine bond. This approach was first reported by Rosenthal and coworkers in 1985.96 However, in vivo studies showed that simple 18F-labelled silicon based radiotracers are unstable under physiological conditions.97 This is mainly due to the hydrolysis of Si-18F bond under these conditions ultimately causing an accumulation of fluoride ions in bones and hence useless exposure to radiation. Several stability studies were conducted to examine the lifetime of organofluorosilane derivatives under physiological conditions.98 Not surprisingly, these

studies showed that the stability of these derivatives increases as the steric hindrance around the silicon atom increases. Some sterically congested organofluorosilane derivatives are stable up to 300 hours under physiological conditions. However, one negative consequence of increasing the size of the groups attached to silicon is that radiofluorination becomes more difficult and therefore requires the use of harsh conditions.

Two main methods are currently employed in the radiofluorination of organosilicon compounds (Scheme 33).99 The first of these is the radiofluorination of silyl ethers 162 which gives the 18F-labled radiotracer and an alkoxy leaving group. Due in part to the poor leaving group property of alkoxides, this approach requires the use of dilute acetic acid in polar aprotic solvent such as DMSO or DMF and also requires, in some cases, high temperatures. In addition, this method still requires long purification times to remove reaction side products. The 18F-19F exchange is another currently employed method for radiofluorination. This method depends on the ability of silicon to exchange its fluorine ligand in the presence of a high concentration of [18F]-fluoride ions. Under these conditions, compound 163 undergoes 18F-19F exchange leading to the formation of

18F-labeled organosilicon compounds and the 19F ion as a leaving group (Scheme 33).

Scheme 33. Current radiofluorination methods

With the increasing demand for reducing purification steps, 18F-19F exchange method attracts enormous attention. The exchange process can label silicon compounds leaving

[19F/18F]KF/ K 2.2.2 complex as the only byproduct. This isotopic exchange method was utilized in Silicon-Fluoride-Acceptor (SiFA) developed by Wangler and coworkers.100

Several useful SiFA building blocks have been prepared and successfully labeled (Figure

13).101 Compound 165 bearing an aldehyde group can be easily coupled with N-terminal peptides, and then the resultant coupling product can be radiolabeled using the 18F-19F exchange method. The azide analog (167) has been utilized in azide-alkyne cycloaddition reactions (click chemistry) allowing the synthesis of triazole-based silicon radiotracers.

Figure 13. Structures of common SiFA building blocks

Despite of these efforts, the main drawback in 18F-19F exchange radiofluorination is the inability to separate the radiolabeled compound from starting material, which is essentially the same compound. Thus, this method generally leads to the production of

18F-silicon radiotracers with low specific activity (SA); in the isolated product, the ratio between 18F-labeled to non-18F-labeled tracers is low.102 From a practical perspective, tracers with low SA lead to a poor quality PET imaging.

In light of these shortcomings with the current organosilicon radiofluorination methods

(and other problems as well), silicon radiotracers have seen only limited use in PET

imaging. Thus, the development of improved silicon fluorination methods is expected to significantly expand the use of 18F-organosilicon radiotracers. We thought to approach this problem from a more mechanistic perspective by designing leaving groups on silicon capable of stabilizing intermediates (or transition states) in the fluorination mechanism to facilitate radiolabeling.

Recently, the Lepore group has developed a new class of leaving groups termed nucleophilic assisted leaving group (NALG) and utilizes them for the formation of 18F-C bond.103 This class of leaving group has several advantages over others in literature. The sulfonate esters of these leaving groups are quite stable and easily isolated (unlike trifluoromethanesulfonate esters). Importantly, their reactivity is triggered by nucleophilic metal salts. This increased reactivity is proposed to be the result of chelation. The chelation arm, when hosting a metal cation can stabilize the partial negative charge developed during the departure of leaving group 169 (Scheme 34). Thus, we envisioned a possible increase in the rate of radiofluorination by incorporating similar chelating leaving groups onto organosilicon compounds. We hypothesized that such a leaving group would enhance the rate of fluorination (and radiofluorination) of silicon in the presence of metal fluorides. Specifically, the metal cation was expected to be hosted by the chelating leaving group ultimately providing a stabilizing interaction with the growing negative charge developed during the formation of pentacoordinated silicon species. Furthermore, we considered that chelating leaving groups might help solubilize the [18F]KF salt without the need for K 2.2.2 for easier purification.

Scheme 34. Chelating leaving group

3.3 Enhanced nucleophilic fluorination and radiofluorination of organosilanes

appended with potassium-chelating leaving groups

Our initial efforts to explore the effect of chelating leaving groups on the rate of fluorination of organosilanes involved compounds 170 and 171 which were prepared and then subjected to fluorination conditions (4 equivalents of KF in anhydrous acetonitrile).

Fortunately, reactions involving compound 170 with a triethylene glycol chelating leaving group showed an enhanced rate of fluorination compared to 171 containing a leaving group of comparable size (Scheme 35). Furthermore, in the presence of equimolar amounts 18-crown-6 (18-C-6), substrate 171 reacted to give 60% of 172 after

11 hours. This result suggested that the rate enhancement observed in 170 may not have been entirely due to a phase transfer effect (i.e. solubilizing KF). With this preliminary result, we chose to further examine the reactivity of organosilicon compounds with similar chelating leaving groups to better understand the nature of the observed rate enhancement.

Scheme 35. Fluorination of silicon nucleophilic assisting group (S-NALG)

3.3.1 Results of fluorination of organosilicon compounds in the absence of phase

transfer agent

In this study, several silicon compounds with different leaving groups have been examined (Table 14). All reactions were conducted with 0.1 M organosilane substrate at room temperature in acetonitrile-d3 (1.0 mL) with potassium fluoride as the limiting reagent (0.5 eq) in the absence of any phase transfer agent. This use of limiting KF was chosen to better mimic conditions generally found in radiofluorinating protocols. Under these conditions, potassium fluoride was initially present in the reaction vessel as a suspension; however, the reaction turned into a homogenous clear solution after 10 to 15 minutes. All reactions examined were allowed to proceed for only 30 min. In this brief time, substrate 170, containing a linear oligoether moiety and consequently a potential potassium-chelating leaving group, failed to give detectable amounts of the desired organofluorosilane product 172 upon its reaction with KF. Similarly, reactions involving substrate 171, containing a linear non chelating leaving group yielded no product and remained inert even after 6 hours. We next examined the reactivity of substrates 173−175 and 176−178 which are more useful for PET applications due to their stability under physiological conditions. The silicon atoms in these substrates are variously congested with alkyl and phenyl groups and carry one of three types of alkoxide leaving group.

Gratifyingly, substrate 173, containing a metal-chelating hydroxymethyl-18-crown-6 leaving group, reacted with potassium fluoride in anhydrous acetonitrile-d3 to give tert- butyldiphenylsilyl fluoride (TBDPSF, 179) in 59% yield after 30 min and in quantitative yield after 2.3 hours. By contrast, reactions involving compound 174, containing a linear methoxy triethylene oxide leaving group, failed to give any of the organofluorosilane 179 after 30 minutes. Not surprisingly, substrate 175, possessing a simple methoxy leaving

group, also gave no measurable amounts of organofluorosilane under reaction conditions.

Similar reactivity was observed with reactions involving substrates 176−178, substrate

176 reacted to give 180 in 79% yields in 30 minutes, but reactions involving 177 and 178 did not give any measurable amounts organofluorosilane 180 under the same conditions.

Table 14. Fluorination of organosilanes with diverse alkoxide leaving groups

3.3.2 Results of fluorination of organosilicon compounds in the presence of phase

transfer agent

The goal of the present project was to develop an ultrafast radiofluorinating procedure without the need to employ phase transfer agents (e.g. K 2.2.2). Nevertheless, we

examined the reactivity of substrates 173−178 towards KF in the presence of a phase transfer agent 18-crown-6 (18-C-6) in an attempt to better understand the role of leaving group chelation in this system. Reactions involving substrates 173 and 176 bearing a cyclic crown ether leaving group gave no yield (in 30 min) of their corresponding fluorinated products (179 and 180), in stark contrast to the high yield obtained in the absence of 18-C-6 described in the previous section. It is possible that the added 18-C-6 preferentially chelated the available potassium ion (limiting reagent). This bulky 18-C-

6/KF complex was probably inhibited in its approach to the sterically crowded silicon center. This may explain the poor reaction kinetics. From another perspective, the added

18-C-6 took away available potassium needed to activate the crown ether leaving group in 173 and 176. Without these cations (needed for the NALG effect) the chelating leaving groups (bulky) in these substrates then merely acted as impediments to the fluorination mechanism.

Substrates 174 and 175 reacted to give organofluorosilane 179 in low (10%) and moderate (33%) yields respectively with the assistance of 18-C-6. Similar 18-C-6 assistance was also required for reactions involving substrates 177 and 178 which produce organofluorosilane 180 in good yields (72 and 77%, respectively).

To summarize the previous studies, in the presence of phase transfer agent, reactions involving substrates containing the methoxy leaving group showed fast formation of fluorinated products. Under the same conditions, reactions involving the triethylene glycol (monomethyl ether) leaving group also led to the formation of fluorination products, but these reactions were slower than those involving the small methoxy leaving group. The sterically hindered crown ether leaving group seems to impede the formation

of product in the presence of phase transfer agent. This reactivity was completely reversed in reactions performed in the absence of phase transfer agent (where potassium salt was available to activate the 18-C-6 leaving group).

3.3.3 Results of radiofluorination of organosilicon compounds in the absence of phase

transfer agent

As described in the introduction, the primary goal of this project was to develop silicon fluorination methods for applications in PET imaging. To validate some of the key findings presented in the preceding sections in a radiochemical laboratory, we collaborated with Dr. Pike (PET Section Chief) and Dr. Lu (Researcher in the Pike Lab) at the National Institutes of Health (NIH). In this work, compounds were prepared at

FAU and then shipped to the NIH PET lab which has access to a cyclotron on-site. In addition, optimal silicon fluorination conditions and other data were shared with the NIH radiochemists to serve as a starting point in their validation studies. As will be described in the next two sections, the radiolabeling results generated by the Pike lab (using K18F) closely paralleled our findings.

For the radiofluorination of substrates 173−178 in the absence of K 2.2.2, substrates were dissolved in anhydrous acetonitrile and were added to a V-vial containing no carrier added (n.c.a) relatively anhydrous potassium [18F]fluoride. This salt was prepared by drying cyclotron-produced aqueous [18F]fluoride ion in the presence of potassium carbonate through conventional azeotropic distillation of acetonitrile. In the absence of the K 2.2.2 phase transfer agent, 173 gave a 70−80% decay-corrected radiochemical yield. This yield was calculated based on the radioactivity that entered liquid phase. It is worth noting that in the absence of the phase transfer agent, less than 5% of the

[18F]fluoride salt was soluble in acetonitrile (low recovery of radioactivity). The [18F]KF salt, under these phase transfer-free conditions, gave only trace amounts of radioactive product when allowed to react with 174 and 175. Again, in both cases, only very low proportions of [18F]fluoride ions (~1%) were recovered. Recovery of radioactivity is the percent of radioactivity dissolved in solvent in this case acetonitrile. The poor recovery of

[18F]fluoride ions is due to several reasons in which the solubility in acetonitrile solvent is among. In radiofluorination laboratory reactions are performed in microgram scale which makes the reactions outcome sensitive to the adsorption of [18F]KF on the inner surface of reaction vessel. This lose of KF on the inner surface of the reaction vessel was not observed in the fluorination reactions (not radiofluorination) performed earlier at larger scale (Table 13). For these reasons, the well-accepted procedure to improve the recovery of [18F]KF is to use a phase transfer agent. As will be discussed later in this chapter, another approach was developed to overcome this low availability of 18F-fluoride ion in reaction solvents.

As a potential solution to the problem of fluoride recovery, our collaborators chose to include a low concentration of water (0.5% v/v) in the reaction solvent. This dramatically increased the recovery of [18F]fluoride ion in acetonitrile from 4 to 31%. However, further increase in the water content (up to 5% v/v) reduced the overall RCY. With 0.5% water as an optimal hydration level for the reaction, the reactivities toward radiofluorination of a series of organosilanes were examined in the presence of 1.8 mM

K2CO3 in acetonitrile at room temperature (Figure 14).

6 0 0 .3 6 m M K 2 C O 3 1 7 3

) 9 .0 m M K C O

4 0 2 3 %

( 1 7 6

Y 1 7 7

C 1 7 8

R 2 0 1 .8 m M K 2 C O 3 1 7 3 1 7 4 0 1 7 5

0 1 0 2 0 3 0 4 0 5 0 T im e (m in )

Figure 14. Radiofluorination of substrates 173−178

The radiofluorination of 173 in just 2 minutes gave [18F]179 in a moderate yield (35%) from the solubilized [18F]fluoride ion though the yield gradually decreased showing that product was degrading. Substrates 174 and 175 were not so reactive under these conditions; each gave [18F]179 in less than 5% yield after 45 minutes. Hence, the organosilicon-NALG having a cyclic crown ether unit (173), rather than a linear oligoether component (174), was clearly most effective for promoting radiofluorination.

This finding accords with the results performed using non-radioactive fluoride.

To circumvent the decomposition of [18F]179, these reactions were repeated with a lower concentrations of potassium carbonate (0.36 mM instead of 1.8 mM). Using these conditions, the yields of [18F]179 from reactions involving substrates 174 and 175 were improved slightly to near 10% when the concentration of potassium carbonate was reduced (not shown). On the other hand, significantly less decomposition was observed in the radiofluorination of 173 giving a 50% yield in 20 minutes at room temperature

(Figure 14). To put this into context, to our knowledge, this is the fastest silicon radiofluorination ever reported without the use of a phase transfer agent, such as K 2.2.2.

The effect of base concentration on the outcome of the radiofluorination of diisopropyl biphenyl silane substrates (176-178) was somewhat different. Literature reports establish that the fluoride product resulting from 176 - 178 (i.e. 180) is relatively stable at higher aqueous base concentrations. Thus, radiofluorination of these substrates in the absence of

K 2.2.2 was performed using an optimal K2CO3 concentration for the efficient production of K18F (9 mM, Figure 14). The highest yield (45%) for this substrate group was obtained from the reaction of 176 containing a cyclic crown ether. Yields for 177 and 178 were

33% and 30%, respectively. Overall, these radiofluorination results were in keeping with our non-radioactive findings; in both cases substrates containing a cyclic crown ether leaving group were clearly the most reactive towards fluorination (in the absence of a phase transfer catalyst).

3.3.4 Results of radiofluorination of organosilicon compounds in the presence of phase

transfer agent

As a complement to our non-radioactive studies, our NIH collaborators also performed radiofluorination of substrates 173-178 in the presence of excess amounts a phase transfer agent (K 2.2.2). Here again, we sought to better understand the role of potassium- chelating leaving groups on the reactivity of these compounds (Figure 15). Desired product [18F]179 was obtained from reactions involving substrates 173−175 with yields peaking and leveling off after 20 minutes. Slower rates of radiofluorination were observed with reactions leading to the formation of [18F]180 (from substrates 176−178,

Figure 15). Yields from reactions involving 177 and 178 reached plateaus after about 45

minutes whereas the reaction yield from 176 gradually increased but failed to achieve more than 25% even after 2 hours. In both series the highest yields were obtained from reactions involving substrates containing a methoxy leaving group (i.e. 175 and 178) rather than a metal-chelating unit (Figure 14). These results mirror those described above involving non-radioactive potassium fluoride. The reduced reactivity of substrates containing chelating leaving groups may be due to the fact that K 2.2.2 more strongly chelates K+. Without access to this cation to trigger the reactivity of these chelating leaving groups, the substrates containing them reacted more slowly. Also, without this cation activation, the large metal-chelating units might also impart some steric hindrance under these conditions further slowing down the reaction.

1 0 0

8 0 1 7 5

) 1 7 3

% 6 0 (

1 7 4 Y

C 4 0 1 7 8 R

2 0 1 7 7 1 7 6 0

0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 T im e (m in )

Figure 15. Radiofluorination of 173−178 in the presence of K2CO3 and K 2.2.2

3.4 Mechanistic considerations

We hypothesize that chelating leaving groups, especially 18-C-6-CH2O-, enhance the rate silicon fluorination, at least in part, by positioning the fluoride ion near to its silicon target (entropic advantage). The reaction then proceeds through the well-established pentacoordinate silicon anion intermediate (182, Scheme 36). This anionic intermediate is likely stabilized by the potassium cation positioned nearby due to chelating interactions with the leaving group. As will be discussed later in this chapter, others have shown that an 18-C-6/K+ counter ion is capable of stabilizing a pentacoordinate silicon anion.104,105

Scheme 36. Proposed radiofluorination mechanism involving stabilizing interactions with a 18-C-6-CH2O- leaving group

3.5 Enhanced Radiofluorination of Organosilanes Using Nucleophilic Crown Ether

Phase Transfer Agents.

Though we were able to report a fast radiofluorination method in the absence of added phase transfer agent, the low recovery of [18F]KF ions in acetonitrile, highlighted earlier,

remains a drawback to this approach. Therefore, we envisioned an alternative approach based on hypervalent silicon chemistry.

The ability of silicon to increase its valency to penta- and hexacoordination has been exploited in various organic transformations.106 Pentacoordinate silicon attracted our attention due to its enhanced reactivity towards nucleophiles, allowing for faster reactions. Over the course of a decade of study, Corriu and coworkers demonstrated in a series of publications that pentacoordinate silicon complexes such as 184 are formed in the presence of nucleophilic catalysts (Scheme 37).107 In this negatively charged complex, the silicon center acts as a Lewis acid which allows rapid reactions with nucleophiles to form a hexacoordinated complex (185). This highly reactive hexacoordinated complex then collapses rapidly to form product 186 (Scheme 37).

Scheme 37. Enhanced reactivity of pentacoordinated silicon

Furthermore, Shiro and coworkers reported that an 18-C-6/K+ counter cation stabilizes pentacoordinated silicon species.104 Using this finding, Corriu and coworkers compared the reactivity of pentacoordinated complex 187 to the thermodynamically stable tetracoordinate analog (188) towards carbon-nucleophiles. Reactions involving intermediate 187 show rapid formation of 189 (94% yield 2 h); conversely, the tetracoordinate analog (188) reacts far more slowly to give only a 17% yield after 17 h

(Scheme 38).108,109 As described in the next section, we envisioned a strategy to alleviate

the problem of low recovery of [18F]KF salt and, at the same time, enhance silicon fluorination rates. This would be accomplished by employing a crown ether nucleophilic phase transfer agent capable of accessing the hypervalent silicon chemistry of Corriu.

Scheme 38. Reactivity of tetra- and pentacoordinate silicon compounds

3.5.1 Enhanced Fluorination Using Nucleophilic Crown Ether.

We hypothesized that the nucleophilic hydroxyl group in hydroxymethyl-18-C-6/KF complex 190 would attack the silicon center of the silyl ether leading to the formation of pentacoordinated species 191. Then complex 191 will expand its valency by the addition of fluoride ion to hexacoordinated complex 192 that eventually collapse rapidly to give the tetracoordinated organofluorosilane derivatives (193, Scheme 39).

Scheme 39. Accelerated fluorination using a nucleophilic crown ether

To examine our hypothesis, 18-C-6-CH2OH/KF complex 190 and 18-Crown-6/KF complex 194 were prepared and allowed to react with silyl ether 175 (Table 15). Using two mole equivalents of compound 175, complex 194 reacted to yield 51% of fluorinated product 179. The same conditions but using complex 194 gave a significantly lower yield

(27%). Both of these reactions were then repeated with rigorous exclusion of water to determine if the rate enhancement was related to a disruption in fluoride ion hydration but these conditions did not change the reaction outcomes. Furthermore, in order to understand the role of the hydroxyl group on the reaction mechanism, 18-C-6-

CH2OMe/KF complex 195 was prepared and allowed to react with 175 under the same conditions. This reaction was monitored by 1H-NMR but no product formation was observed after 30 min.

Table 15. Fluorination of 175 using 18-crown-6 derivatives

3.5.2 Results for the radiofluorination of 160 using Nucleophilic Crown Ether. (Performed by Dr. Pike and Dr. Lu National Institutesof Health)

18 Similar to the previous studies, 18-C-6-CH2OH/K F was the most effective agent for radiofluorination. Specifically, substrate 178 was treated with limiting amount of three different phase transfer agents pre-complexed with [18F]KF. Overall, 18-C-6-

18 CH2OH/K F gave superior results leading to a 70% radiochemical yield in 5 min (Figure

16). To our knowledge this is the fastest and most mild silicon radiofluorination reaction ever reported. Using a traditional phase transfer agent complex, 18-C-6/K18F, substrate

178 was fluorinated in a much lower yield in the same 5 min reaction time (35% RCY).

When the same reaction was performed using the most common radiofluorination phase transfer agent (K 2.2.2/K18F) the results were also poor after 5 min (<30% RCY) although reasonable yields were obtained after 45 min (60% RCY) (Figure 16). This poor result with K 2.2.2/K18F is not surprising since substrates similar to 178 have been radiofluorinated by others though under significantly more forcing conditions typically in

DMSO with 1% AcOH at a higher temperatures (up to 90 °C) for 15 min.101

w ith 1 8 -C -6 -C H 2 O H w ith 1 8 -C - 6 1 0 0 w ith K 2 .2 .2

8 0 )

6 0

( Y

C 4 0 R

2 0

0 1 0 2 0 3 0 4 0 5 0 T im e (m in )

Figure 16. Radiofluorination of 178 to give [18F]180.

These studies reveal that silicon substrates containing crown ether leaving group is fluorinated at nearly the same rate as simple silyl methoxy substrates in the presence of

18 18-C-6-CH2OH/KF. However, as alluded earlier, due to the small amounts of K F typically used in radiolabeling procedures (μg scale), the radioactive salt must be

sequestered by a phase transfer agent during the drying process improve its solubility

(recovery of [18F]KF). In this regard, the radioactivity recovery (percent of radioactivity dissolved in MeCN) was the highest when K 2.2.2 was used (89%). Importantly, radioactivity recovery was also high and very consistent using 18-C-6-CH2OH (74%) though relatively poor using 18-C-6 (44%). Ultimately, radiofluorination using nucleophilic phase transfer agent takes advantage of silicon hypervalent chemistry allowing for a rapid silicon fluorination (radiofluorintation). More importantly, it provides a solution for the low recovery problem [18F]KF salt.

3.6 Experimental section:

General procedure for silylation of alcohols: In an oven-dried vial under argon atmosphere, alcohol (1.0 mmol) was dissolved in dichloromethane (5 mL) to which was added imidazole (2.0 mmol). This solution was then stirred in an ice-bath for 10 min.

Silyl chloride (2.0 mmol) was then added and the reaction was allowed to warm to rt.

After completion, volatiles were removed under reduced pressure and the crude mixture was purified with silica gel chromatography using hexanes/EtOAc as eluent.

General procedure for fluorination: In an oven-dried vial purged with argon, silyl ether

(0.10 mmol) was dissolved in anhydrous acetonitrile-d3 (0.1 M solution, 1.0 mL) and stirred at rt for 5 min. Finely powdered anhydrous potassium fluoride (0.5 mmol) was then added to the stirred solution resulting initially in a suspension and then a homogeneous mixture after 10 - 15 min. After KF addition, the reaction was stirred at room temperature for 30 min (overall time) and then directly analyzed by 1H-NMR.

Percent conversion to silyl fluoride product was calculated based on the limiting reagent

KF using the proton integration of an alkyl (tBu and iPr) group present on silicon in both starting material and product.

2-Methyl-2-phenyl-3,6,9,12-tetraoxa-2-silatridecane (170). Yield = 61%. 1H-NMR

(400 MHz, CDCl3): δ 7.58−7.56 (m, 2H), 7.37−7.34 (m, 3H), 3.74 (t, J = 5.2 Hz, 2H),

3.63−3.61 (m, 6H), 3.55−3.52 (m, 4H), 3.36 (s, 3H), 0.38 (s, 6H); 13C NMR (100 MHz,

CDCl3): δ 138.0, 133.7, 129.8, 128.0, 72.6, 72.1, 70.8, 70.8, 70.7, 62.6, 59.2, −1.5. MS

+ calc. for C15H26O4Si [M + H] : 299.17. Found: 299.61.

1 (Heptyloxy)dimethyl(phenyl)silane (171). Yield = 71%. H-NMR (400 MHz, CDCl3): δ

7.62−7.61 (m, 2H), 7.42−7.40 (m, 3H), 3.62 (t, J = 6.6, 2H), 1.56 (m, 2H), 1.29 (m, 8H),

13 0.91 (t, J = 6.7, 3H), 0.41 (s, 6H); C-NMR (100 MHz, CDCl3) δ 138.3, 133.7, 129.8,

128.0, 63.4, 32.9, 32.1, 29.4, 26.0, 22.9, 14.4, −1.5. APCI_TOF-MS calc. for C15H26OSi

[M + H]+: 251.1826. Found: 251.1825.

((1,4,7,10,13,16-Hexaoxacyclooctadecan-2-yl)methoxy)-(tert-butyl)diphenylsilane

1 (173). Yield = 97%. H-NMR (400 MHz, CDCl3): δ 7.67 (m, 4H) 7.39−7.32 (m, 6H),

13 3.65 (m, 25H), 1.04 (s, 9H); C-NMR (100 MHz, CDCl3): δ 135.5, 135.4, 133.3, 133.3,

129.5, 127.5, 127.5, 79.9, 71.0, 70.7, 70.5, 70.6, 70.6, 70.5, 70.5, 70.5, 70.4, 69.9, 63.3,

+ 26.6, 19.1. ESI-HRMS calc. for C29H44O7Si [M + Na] : 555.2749. Found: 555.2754.

13,13-Dimethyl-12,12-diphenyl-2,5,8,11-tetraoxa-12-silatetradecane (174). Yield =

1 74%. H-NMR (400 MHz, CDCl3): δ ppm 7.70−7.67 (m, 4H), 7.42−7.36 (m, 6H) 3.81 (t,

J = 4 Hz, 2H), 3.67−3.52 (m, 10H), 3.37 (s, 3H), 1.05 (s, 9H); 13C-NMR (100 MHz,

CDCl3): δ 135.8, 133.9, 129.8, 127.9, 72.7 72.2, 71.0, 70.9, 70.8, 63.6, 59.3, 27.0, 19.4.

+ ESI-HRMS Calc. for C23H34O4Si [M + Na] : 425.2119. Found: 425.2126.

1 tert-Butyl(methoxy)diphenylsilane (175). Yield = 84%. H-NMR (400 MHz, CDCl3): δ

13 7.74-7.73- (m, 4H), 7.47-7.43 (m, 6H), 3.57 (s, 3H), 1.08 (s, 9H); C-NMR (CDCl3, 100

MHz ): δ 135.7, 133.7, 129.8, 127.9, 52.4, 27.0, 19.4. CI-HRMS Calc. for C17H22OSi [(M

+ + H) - CH3OH] : 239.1256. Found: 239.1251.

((1,4,7,10,13,16-Hexaoxacyclooctadecan-2-yl)methoxy)-([1,1'-biphenyl]-4-

1 yl)diisopropylsilane (176). Yield = 68%. H-NMR (400 MHz, CDCl3): δ 7.61−7.27 (m,

6H), 7.47−7.43 (m, 2H), 7.37−7.35 (m, 1H), 3.86−3.63 (m, 25H), 1.33−1.27 (m, 2H),

13 1.08−1.03 (m, 12H); C-NMR (100 MHz, CDCl3): δ 141.8, 141.0, 135.1, 132.9, 128.7,

127.3, 127.1, 126.3, 80.2, 71.6, 70.9, 70.8, 70.7, 70.7, 70.6, 70.6, 70.6, 70.5, 70.1, 63.7,

+ 17.4, 17.3, 12.0. ESI-HRMS calc. for C31H48O7Si [M + Na] : 583.3062. Found:

583.3078.

12-([1,1'-Biphenyl]-4-yl)-12-isopropyl-13-methyl-2,5,8,11-tetraoxa-12-

1 silatetradecane (177). Yield = 91%. H-NMR (400 MHz,CDCl3): δ 7.65−7.58 (m, 6H),

7.47−7.43 (m, 2H), 7.37−7.35 (m, 1H), 3.94 (t, J = 5.6 Hz, 2H), 3.69 (m, 8H), 3.35 (m,

2H), 3.37 (s, 3H), 1.34−1.26 (m, 2H), 1.09 (d, J = 7.6 Hz, 6H), 1.04 (m, J = 7.6 Hz, 6H);

13 C-NMR (100 MHz, CDCl3): δ 141.8, 141.0, 135.1, 128.2, 127.3, 127.1, 126.2, 72.6,

71.9, 70.8, 707.0, 70.6, 63.3, 59.0, 17.4, 17.3, 12.02. ESI-HRMS calc. for C25H38O4Si

[M + Na]: 453.2432. Found: 453.2447.

[1,1'-Biphenyl]-4-yldiisopropyl(methoxy)silane (178). Yield = 90%. 1H-NMR (400

MHz, CDCl3): δ 7.64−7.60 (m, 6H), 7.46−7.43 (m, 2H), 7.37−7.35 (m, 1H), 3.64 (s, 3H),

1.37−1.29 (m, 2H), 1.10 (d, J = 7.2 Hz, 6H), 1.05 (d, J = 7.2 Hz, 6H); 13C-NMR (100

MHz, CDCl3): δ 141.6, 141.0, 135.1, 132.7, 128.7, 127.3, 127.1, 126.3, 52.1, 17.5, 17.3,

+ 11.95. CI-HRMS calc. for C19H26Osi [M + H] : 299.1831. Found: 299.1844.

110 1 tert-Butylfluorodiphenylsilane (172). H-NMR (400 MHz, C6D6): δ 7.70–7.77 (m, 4H),

4 7.12–7.20 (m, 6H, ), 1.08 (d, JH–F = 1.1 Hz, 9H).

111 1 Fluorodimethyl(phenyl)silane (179). H NMR (400 MHz, in CDCl3): δ 0.49 (d, JH-F = 7.3

Hz, 6H), 7.42-7.65 (m, 5H ).

[1,1'-Biphenyl]-4-ylfluorodiisopropylsilane (180). Yield = 81%. 1H-NMR (400 MHz,

CDCl3): δ 7.67−7.56 (m, 6H), 7.48−7.41 (m, 2H), 7.39−7.32 (m, 1H), 1.37-1.25 (m, 2H),

13 1.12−1.05 (m, 12H); C-NMR (100 MHz, CDCl3): δ 141.5, 134.9, 134.3, 134.3, 129.0,

128.8, 127.7, 127.5, 127.3, 127.1, 127.0, 126.5, 126.1, 17.8, 17.6, 16.7 (d, JC-F = 2 Hz),

+ 16.5, 13.8, 12.2 (d, JC-F = 14 Hz). CI-HRMS calc. for C18H23FSi [M + H] : 287.1631.

Found: 287.1635.

2-(Methoxymethyl)-1,4,7,10,13,16-hexaoxacyclooctadecane (195). Yield = 56%. 1H

NMR (400 MHz, CDCl3): δ 3.93-3.88 (m, 1H), 3.76—3.60 (m, 24H), 3.38 (s, 1H); 13C-

NMR (100 MHz, CDCl3): δ 76.7, 70.8, 70.7, 69.6, 69.6, 69.3, 69.3, 69.2, 69.2, 69.1, 67.7,

+ + 58.6. EI-LRMS calc. for C14H28O7: [M + K] : 347.1467 Found: [M + K] :347.1473.

Radiochemistry. Cyclotron-produced NCA [18F]fluoride ion (30−100 mCi) in

18 [ O]water (60−250 μL) was mixed with stock K2CO3/K 2.2.2 (0.7 mmol and 2.6 mmol in 9:1 MeCN:H2O mixture) or aqueous K2CO3 (0.1–3.6 µmol), then dried by four cycles of azeotropic evaporation with acetonitrile (0.65 mL for each addition) at 110 °C using a robot-based automation module. All reactions were performed at room temperature. For radiochemical studies with K 2.2.2, substrate (1 mmol) and anhydrous 18F-/K+- K 2.2.2 in

MeCN (200 mL) were added to a glass vial to form a clear solution. For radiochemical studies without K 2.2.2, substrate (1 mmol) was dissolved in the MeCN containing H2O

(0.5−5%, v/v, 200 mL), and the solution transferred to the V-vial that contained dried

K18F. For both types of experiment, an aliquot (10 mL) of reaction mixture was sampled at a designated time and quenched in a mixture of H2O:MeCN (1:1, v/v, 500 mL). A portion of the quenched aqueous solution (20 mL) was injected onto a reverse phase

HPLC (Luna, C18, 10 mm, 250 × 4 mm) for analysis with UV-absorbance (254 nm) and radioactivity detectors. The mobile phase was a mixture of 25 mM aqueous ammonium formate (A) and MeCN (B), initially with B at 40% for 2 min, and then increased to 90% over 1 min. Flow rate was 2 mL/min. Product identity was confirmed by coelution with authentic non-radioactive compound. RCY (%) was calculated as the ratio of the peak area of the product to the sum of peak areas of all radio-peaks. No decay correction was performed because the time lapse between the two radio-peaks was small (around 10 min).

Appendix

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

References

1. Humphrey J. M., Chamberlin A. R. Chem. Rev. 1997, 97, 2243-2266.

2. Ghose A. K., Viswanadhan V. N., Wendoloski J. J. J. Comb.Chem.1999, 1, 55- 68. 3. Montalbetti C. A. G. N., Falque V. Tetrahedron 2005, 61, 10827-10852.

4. Woo-Jin Y., Chao-Jun L. J. Am. Chem. Soc. 2006, 128, 13064-13065.

5. Ekoue-Kovi K., Wolf. C. Chem. Eur. J. 2008, 14, 6302 – 6315

6. (a) Qian C., Zhang X., Zhang Y., Shen Q. J. Organomet. Chem. 2010, 695, 747- 752; (b) Zhang L. J., Su S. P., Wu H. P., Wang S. W. Tetrahedron 2009, 65,10022-10024; (c) Seo S. Y., Marks T. J. Org. Lett. 2008, 10, 317-319; (d) Wang J. F., Li J. M., Xu F., Shen Q. Adv. Synth. Catal. 2009, 351, 1363-1370; (e) Wu Y. J., Wang S.W., Zhang L. J., Yang G. S., Zhu X. C., Zhou Z. H., Zhu H., Wu S. H. Eur. J. Org. Chem. 2010, 326-332. (f) Annegret T., Ivo R., Matthias B. Eur. J. Org. Chem. 2001, 523-528; (g) Shie J., Fang J. J. J. Org. Chem. 2003, 68, 1158-1160; (h) Marko. I. E., Mekhalfia A. Tetrahedron Lett. 1990, 31, 7237-7240; i) Gao J., Wang G. W. J. Org. Chem. 2008, 73, 2955-2958. 7. a) Reddy K. L. Tetrahedron Lett. 2003, 44, 1453-1455; b) Bures J., Martin M., Urpi F., Vilarrasa J. J. Org. Chem. 2009, 74, 2203-2206; c) Aube J., Milligan G. L. J. Am. Chem. Soc. 1991, 113, 8965-8966. 8. a) Saxon E., Bertozzi C. R. Science 2000, 287, 2007-2010; b) Hashimoto M., Obora Y., Sakaguchi S., Ishii Y. J. Org. Chem. 2008, 73, 2894-2897

9. Guerinot A., Reymond S., Cossy J. Eur. J. Org. Chem. 2012, 19-28.

10. Vardanyan R., Hruby V. 2006 Synthesis of essential drugs, 1st Ed. Amsterdam: Elsevier 137.

11. Gerasimova N. P., Nozhnin N. A., Ermolaeva V. V., Ovchinnikova A. V., Moskvichev Y. A. Mendeleev Commun. 2003, 13, 28-83.

Eugeni M. Alov and Angelina S. Danilova 12. Sanz R., Martinez A., Guilarte V., Alvarez-Gutierrez J. M., Rodriguez F. Eur. J. Org. Chem. 2007, 4642-4645.

134

13 Firouzabadi H., Sardarian A. R., Badparva H. Synth. Commun. 1994, 24, 601- 607 14. a) Barton D. H. R., Magnus P. D., Garbarino J. A., Young R. N. J. Chem. Soc. Perkin Trans.1 1974, 2101-2107; b) Top S., Jaouen G. J. Org. Chem. 1981, 46, 78-82; c) Mukhopadhyay M., Reddy M. M., Maikap G. C., Iqbal J. J. Org. Chem. 1995, 60, 2670-2676; d) Callens E., Burton A. J., Barrett A. G. M. Tetrahedron Lett. 2006, 47, 8699-8701.

15. a) Sohar P., Mathre J. D., Blacklock J. T. Eur. Pat. Appl. EP XXDW EP 617037 A1 19940928, 1994; b) Popova A. L., Kozlov G. N., Makhnach A. S., Vestsi Akad Navuk BSSR, Ser Khim Navuk 1990, 3, 49–54.

16. a) Indukuri K., Unnava R., Deka M. K., Saika A. K. J. Org. Chem. 2013, 78, 10629-10641; b) Martinez A. G., Vilar E. T., Moreno-Jimenez F., Garcia A. M. A., Tetrahedron: Asymmetry 2006, 17, 2970-275; c) Senanayak C. H., Roberts F. E., DiMechele L. M., Ryan K., M., Liu J., Fredenburgh L. E., Foster B. S., Douglas A. W., Laesen R. D., Verhoeven T. V., Reider P. J. Tetrahedron Lett. 1995, 36, 3993-3996.

17. Rubenbauer P., Bach T. Chem. Commun. 2009, 2130-2132.

18. Muhlthau F., Stadler D., Goeppert A., Olah G. A., Prakash G. K. S., Bach T. J. Am. Chem. Soc. 2006, 128, 9668-9678.

19. Pronin S. V., Reiher C. A., Shenvi R. A. Nature, 2013, 501, 195-199.

20. Yadav J. S., Reddy S. B. V., Pandurangam T., Reddy J., Gupta M. K. Catal. Commun. 2008, 9, 1297-1301.

21. Alder R. W., Harvey J. N., Oakley M. T. J. Am. Chem. Soc. 2002, 124, 4960- 4961.

22. Mondal D., Bellucci L., Lepore S. D. Eur. J. Org. Chem. 2011, 7057–7061.

23. Mondal D., Li S. Y., Bellucci L., Laino T., Tafi A., Guccione S., Lepore S. D. J. Org. Chem. 2013, 78, 2118-2127.

24. a) Kirchen R. P., Sorensen T. S. J. Am. Chem. Soc. 1978, 100, 1487-1494; b) Rauk A., Sorensen T. S., Maerker C., Carneiro J. W., De M., Sieber S., Schleyer P. V. R. J. Am. Chem. Soc. 1996, 118, 3761-3762; c) Kirchen R. P., Ranganayakulu K., Sorensen T. S. J. Am. Chem. Soc. 1987, 109, 7811-7816; d) Rauk A., Sorensen T. S., Schleyer P. V. R. J. Chem. Soc. Perkin Trans. 2 2001, 869-874

25. Alabugin I. V., Manoharan M. J. Org. Chem. 2004, 69, 9011-9024.

135

26. Lepore S. D., Mondal D., Li S.Y., Bhunia A. K. Angew. Chem. Int. Ed. 2008, 47, 7511-7514.

27. Mann R. S., Yao K. C. Ind. Eng. Chem. Prod. Res. Dev. 1969, 8, 331-333

28. Van-Woerden H. F. Chem. Rev. 1963, 63, 557-571.

29. Bartlett P. D., Herbrandson H. F. J. Am. Chem. Soc. 1952, 74, 5972-5974.

30. Shrestha-Dawadi P. B., Jochims J. Synthesis 1993, 426-432.

31. a) Carman R. M., Greenfield K. L. Aust. J. Chem. 1984, 37, 1785-1790; b) Kasashima Y., Uzawa A., Hashimoto K., Yokoyama Y., Mino T., Sakamoto M., Fujita T. J. Oleo Sci. 2010, 59, 607-613; c) Kitagawa N., Nojima M., Tokura N. J. Chem. Soc. Perkin Trans. 1 1975, 2369-2371.

32. Michelin R. A., Mozzon M., Bertani R. Coord. Chem. Rev. 1996, 147, 299-338.

33. a) Richard J. P., Rothenberg M. E., Jencks W. P., J. Am. Chem. Soc. 1984, 106, 1361-1372; b) Richard J. P., Toteva M. M., J. Am. Chem. Soc. 1996, 118, 11434- 11445.

34. Kulisch J., Nieger M., Stecker F., Fischer A., Waldvogel S. R. Angew. Chem. Int. Ed. 2011, 50, 5564-5567.

35. Bexrud J. A., Eisenberger P., Leitch D. C., Payne P. R., Schafer L. L. J. Am. Chem. Soc. 2009, 131, 2116-2118.

36. a) Ritter K. Synthesis 1993, 735-762; b) Dounay A. B., Overman L. E. Chem. Rev. 2003, 103, 2945-2963.

37. Cartney D. M., Guiry P. J. Chem. Soc. Rev. 2011, 40, 5122-5150.

38. Nicolaou K. C., Frederick M. O., Burtuloso A., Denton R. M., Rivas F., Cole K. P., Aversa R. J., Gibe R., Umezawa T., Susuki T. J. Am. Chem. Soc. 2008, 130, 7466-7476.

39. Crisp G. T. Chem. Soc. Rev. 1998, 27, 427-434.

40. Sato Y., Sodeoka M., Shibasaki M. J. Org. Chem. 1989, 54, 4738-4739.

41. Sato Y., Watanabe S., Shibasaki M. Tetrahedron Lett. 1992, 33, 2589-2592.

42. Kondo K., Sodeoka M., Mori M., Shibasaki M. Tetrahedron Lett. 1993, 4219- 4222. 43. Ohrai K., Kondo K., Sodeoka M., Shibasaki M. J. Am. Chem. Soc. 1994, 116, 11737-11748.

136

44. a) Kant J., Sapino Jr. C., Baker S. R. Tetrahedron Lett. 1990, 31, 3389-3392; b) Cook G. K., Hornback W. J., Jordan C. L., McDonald J. H., Monroe J. E. J. Org. Chem. 1989, 54, 5828-5830.

45. Crouch I. T., Neff R. K., Frantz D. E. J. Am. Chem. Soc. 2013, 135, 4970-4973. 46. Suzuki A. J. Organomet. Chem.1999, 576, 147-168.

47. Stang P. J., Mangum M. G., Fox D. P., Haak P. J. Am. Chem. Soc. 1974, 96, 4562-4569.

48. Subramanian L. R., Bentz H., Hanack M. Synthesis 1973, 293-294,

49. Wright M. E., Pulley S. R. J. Org. Chem. 1989, 54, 2886-2889.

50. Comins D. L., Dehghani A. Tetrahedron Lett. 1992, 33, 6299–6302. 51. Foti C. J., Comins D. L. J. Org. Chem. 1995, 60, 2656-2657.

52. Stang P. J.; Summerville R. J. Am. Chem. Soc. 1969, 91, 4600-4601.

53. Summerville R. H., Senkler C. A., Schleyer P. V. R, Dueber T. E., Stang P. J. J. Am. Chem. Soc. 1974, 96, 1100-1110.

54. Vasilyev A. V., Walspurger S., Chassaing S., Pale P., Sommer J. Eur. J. Org. Chem. 2007, 5740–5748.

55. Cui D.-M., Meng Q., Zheng, J.-Z., Zhang C. Chem. Commun. 2009, 1577-1579.

56. Yang Y., Moschetta E. G., Rioux R. M. ChemCatChem. 2013, 5, 3005-3013.

57. Suero M. G., Bayle E. D., Collins B. S. L., Gaunt M. J. J. Am. Chem. Soc. 2013, 135, 5332-5335

58. Xu Z.-F. ; Cai C.-X. ; Liu J.-T. Org. Lett. 2013, 15, 2096-2099

59. Rivers J. The ANU Undergraduate Research Journal, 47-56

60. Rahaim R. J., Shaw J. T. J. Org. Chem. 2008, 73, 2912-2915.

61. a) Oh C. H., Jung H. H., Kim K. S., Kim N. Angew. Chem. Int. Ed. 2003, 42, 805- 808; b) Bhilare S. V., Darvatkar N. B., Deorukhkar A. R., Raut D. G., Trivedi G. K., Salunkhe M. M. Tetrahedron Lett. 2009, 50, 893–896

62. Sudo T., Asao N., Gevorgyan V., Yamamoto Y. J. Org. Chem. 1999, 64, 2494- 2499.

137

63. Knochel P. Singer R. D., Chem. Rev. 1993, 6, 2117-2188.

64. Xie Z., Bau R., Reed C. A. J. Chem. Soc. Chem. Commun. 1994, 2519-2520.

65. Liang Y., Xie Y-X., Li J.-H. J. Org. Chem., 2006, 71, 379-381.

66. Egi M., Kawai T., Umemura M., Akai S. J. Org. Chem. 2012, 77, 7092-7097.

67. Cahiez G., Gager O., Buendia J. Angew. Chem. Int. Ed. 2010, 49, 1278-1281.

68. Saulnier M. G., Kadow J. F., Tun M. M., Langley D. R., Vyas D. M. J Am. Chem Soc, 1998, 111, 8320-8321.

69. Matsubara S., Hibino J., Morizawa Y., Oshima K., Nozaki H. J. Organomet. Chem. 1985, 285, 163-72.

70. Luan L., Song J-S., Bullock R. M. J. Org. Chem. 1995, 60, 7170-7176.

71. Meng L., Su J., Zha Z., Zhang L., Zhang Z., Wang Z. Chem. Eur. J. 2013, 19, 5542-5545

72. Alacid E., Najera C. J. Org. Chem. 2008, 73, 2315-2322.

73. Zou Y., Qin L., Ren X., Lu Y., Li Y., Zhou J. Chem. Eur. J. 2013, 3504-3511.

74. Lee E., Kamlet A. S., Powers D. C., Neumann C. N., Boursalian G. B., Furuya T., Choi D. C., Hooker J. M., Ritter T. Science 2011, 334, 639-642.

75. Ganguly B. N., Mondal N. N., Nandy M., Roesch F. J. Radioanal. Nucl. Chem. 2009, 279, 685-698.

76. Stoll H. P., Hutchins G. D, Winkle W. L., Nguyen A. T., Appledorn C. R., Janzen I., Seifert H., Rube C., Schieffer H., March K. L. J. Nucl. Med. 2001, 42, 1375-1383.

77. a) Dolle F., Curr. Pharm. Des. 2005, 11, 3211-3235; b) Okarvi S. M. Eur. J. Nucl. Med. Mol. Imaging. 2001, 28, 929-938; c) Wilbur D. S. Bioconjugate Chem. 1992, 3, 433-470; d) Bouvet V. R., Wuest M., Wiebe L. I., Wuest F. Nucl. Med. Biol. 2011, 38, 235-245.

78. a) Smith G. E., Salden H. L., Biagini S. C., Blower P. J. Dalton Trans., 2011, 40, 6196-6205; b) McBride W. J., D’souza C. A., Sharkey R. M., Karacay H., Rossi E. A., Chang C. H., Goldenberg D. M. Bioconjugate Chem. 2010, 21, 1331-1340; c) Pike V. W. Trends in Pharmacol. Sci. 2009, 30, 431-440.

138

79. Kilbourn M. R., Dence C. S., Welch M. J., Mathias C. J. J. Nucl Med. 1987, 28, 462-470.

80. De Goeij J. J. M., Bonardi M. L. J. Radioanal. Nucl. Chem. 2005, 263, 13-18.

81. Miller P. W., Long N. J., Vilar R. Gee A. D. Angew. Chem. Int. Ed. 2008, 47, 8998 – 9033.

82. Visser G. W. M., Gorree G. C. M., Braakhuis B. J. M., Herscheid J. D. M. Eur. J. Nucl. Med. 1989, 15, 225-229.

83. Neirinckx R. D., Lambrecht R. M., Wolf A. P. Intl. J. Appl. Radiat. Isotopes 1978, 29, 323-327.

84. Adam M. J., Ruth T. J., Grieson J. R., Abeysekera B., Pate B. D. J. Nucl. Med. 1986, 27, 1462-1466.

85. Teare H., Robins E. G., Arstad E., Luthra S. K., Gouverneur V. Chem. Commun. 2007, 21, 2330-2332.

86. Wahl L, Nahmias C. J Nucl Med. 1996, 37, 432–437.

87. Adam M. J., Jivan S. Appl. Radiat. Isot. 1988, 39, 1203-1210.

88. Luxen A., Gutllaume M., Melega W. P., Pike V. W., Solin O., Wagner R. Review. Nucl. Med. Biol. 1992, 19, 149-158.

89. Fowler J.S., Ido T. Semin. Nucl. Med. 2002, 32, 6-12.

90. Kamlet A. S., Neumann C. N., Lee E., Carlin S. M., Moseley C. K., Stephenson N., Hooker J. M., Ritter T. PLOS one 2013, 8, e59187.

91. Okarvi S. M. Eur. J. Nucl. Med.2001, 28, 929–938.

92. Hamacher K., Hamkens W. Appl. Radiat. Isot. 1995, 46, 911-916.

93. Hamacher K., Coenen H. H., Stocklin G. J. Nucl. Med.1986, 27,235–238.

94. Hohne A., Mu L., Honer M., Schubiger P. A., Ametamey S. M., Graham K., Stellfeld T., Borkowski S., Berndorff D., Klar U., Voigtmann U., Cyr J. E., Friebe M., Dinkelborg L., Srinivasan A. Bioconjugate Chem. 2008, 19, 1871-1879.

95. Mu L., Hohne A., Schubiger P. A., Ametamey S. M., Graham K., Cyr J. E., Dinkelborg L., Stellfeld T., Srinivasan A., Voigtmann U., Klar U. Angew. Chem. Int. Ed. 2008, 47, 4922-4925.

139

96. Rosenthal M. S., Bosch A. L., Nickles R. J., Gatley S. J. Int. J. Appl, Radiat. Isot. 1985, 36, 318-319. 97. a) Hohne A., Yu L., Mu L., Reiher M., Voigtmann U., Klar U., Graham K., Schubiger P. A., Ametamey S. M. Chem. Eur. J. 2009, 15, 3736-3743; b) Bohn P., Deyine A., Azzouz R., Bailly L., Fiol-Petit C., Bischoff L., Fruit C., Marsais F., Vera P. Nuclear Medicine and Biology 2009, 36, 895-905.

98. Schirrmacher E., Wangler B., Cypryk M., Bradtmoller G., Schafer M., Eisenhut M., Jurkschat K., Schirrmacher R. Bioconjugate Chem., 2007, 18, 20852089.

99. Kilbourn M. R., Subramanian R. J. Labelled Compd. Radiopharm 1990, 28, 1355-1361.

100. Schirrmacher R, Bradtmoller G., Schirrmacher E., Thews O., Tillmanns J., Siessmeier T., Buchholz H. G., Bartenstein P., Wangler B., Niemeyer C. M., Jurkschat K. Angew. Chem. Int. Ed. 2006, 45, 6047-6050.

101. Wangler C., Kostikov A., Zhu J., Chin J., Wängler B., Schirrmacher R. Appl. Sci. 2012, 2, 277-302.

102. Wangler C., Waser B., Alke A., Iovkova L., Buchholz H-G., Niedermoser S., Jurkschat K., Fottner C., Bartenstein P., Schirrmacher R., Reubi J-C., Wester H- J., Wangler B. Bioconjugate Chem. 2010, 21, 2289-2296.

103. Lu S., Lepore S. D., Li S. Y., Mondal D., Cohn P. C., Bhunia A. K., Pike V. W. J. Org. Chem., 2009, 74, 5290-5296.

104. Damrauer R., Danahey S. E. Organometallics 1986, 5, 1490−1494.

105. a) Tamao K., Hayashi T., Ito Y. Organometallics 1992, 11, 182-191; b) Deiters J. A., Holmes R. R. J. Am. Chem. Soc. 1990, 112, 7197–7202; c) Harland J. J., Payne J. S., Day R. O., Holmes R. R. Inorg. Chem. 1987, 26, 760-765; d) Johnson S. E., Payne J. S., Day R. O., Holmes J. M., Holmes R. R. Inorg. Chem. 1989, 28, 3190-3198.

106. For recent reviews on hypercoodritated silicon : Chuit C., Corriu R. J. P., Reye C., Young J. C. Chem. Rev. 1993, 93, 1371−1448; (b) Kost D., Kalikhman I. Acc. Chem. Res. 2009, 42, 303−314.

107. Boudin A., Cerveau G., Chuit C., Corriu R. J. P., Reye C. Angew. Chem. Int. Ed. 2003, 25, 473−474.

108. Corriu R. J. P., Guerin C., Henner B. J. L., Man W.C. Organometallics 1988, 7, 237-238.

140

109. Lennon P., Mack D. P., Thompson Q. E. Organometallics 1998, 8, 1121−1122.

107. Damrauer R., Simon R. A., Kanner B. Organometallics, 1988, 7, 1161-1164.

108. Kunai A., Sakurai T., Toyoda E., Ishikawa M. Organometallics 1996, 15, 2478- 2482.

141