Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Templation studies of a tetramethylene-bridged hemicarceplex Pope, Douglas Jay 1997

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

Item Metadata

Download

Media
831-ubc_1998-0045.pdf [ 3.63MB ]
Metadata
JSON: 831-1.0061558.json
JSON-LD: 831-1.0061558-ld.json
RDF/XML (Pretty): 831-1.0061558-rdf.xml
RDF/JSON: 831-1.0061558-rdf.json
Turtle: 831-1.0061558-turtle.txt
N-Triples: 831-1.0061558-rdf-ntriples.txt
Original Record: 831-1.0061558-source.json
Full Text
831-1.0061558-fulltext.txt
Citation
831-1.0061558.ris

Full Text

Templation Studies of a Tetramethylene-Bridged Hemicarceplex by DOUGLAS JAY POPE B.Sc , The University of Manitoba, 1995 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A October, 1997 ©Douglas J. Pope, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference arid study. 1 further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C^rl^^OfTip The University of British Columbia Vancouver, Canada Date tin-a In DE-6 (2/88) Abstract This thesis presents a study of template effects on the formation of hemicarceplex 14«guest from two rigid bowl-shaped molecules (tetrol 7) and four tetramethylene bridges. The use of an appropriate template/guest molecule aids in the shell closure of hemicarceplex 14»guest through stabilizing van der Waals and electrostatic interactions during the transition state leading to formation of hemicarceplex 14»guest. By studying the effects of a template molecule on formation of a hemicarceplex, we gain valuable insight into the non-covalent forces that control self-assembling processes. Potential guest molecules varying in size, shape, polarity and symmetry were screened for templating abilities. The reaction was found to be moderately selective towards the template molecule used and competition experiments between successful guest molecules revealed a template ratio that spanned a factor of one thousand. This template effect is compared to those found previously in the formation of hemicarceplex 13 and carceplex 11. Br<CH2)4Br Base Guest/Template R = C H 2 C H 2 P h 14»guest R = C H 2 C H 2 P h u Table of Contents Abstract i i Table of Contents i i i List of Schemes vi List of Figures vi List of Charts and Tables vii List of Abbreviations viii Acknowledgements x Chapter 1: Templation Studies in Host-Guest Chemistry 1 1.1 Introduction 1 1.2 Carceplexes and Hemicarceplexes 4 (a) Background 4 (b) Cavitands 5 (c) Carceplexes 7 (d) Hemicarceplexes 9 i. Introduction 9 ii. The First Hemicarceplex 9 iii. Tetramethylene Bridged Hemicarceplex 11 iv. Other Hemicarceplexes .15 v. Conclusions 17 1.3 Templation 19 (a) Examples of Templation 19 (b) Template Effects of a Carceplex Reaction 20 (c) Template Effects of a Hemicarceplex Reaction 24 (d) Templation Aids in Preorganization of an o-Xylyl Hemicarcerand 25 iii 1.4 Project Goals and Thesis Objectives 28 1.5 References 30 Chapter 2: Template Effects of a Tetramethylene-Bridged Hemicarceplex 33 2.1 Synthesis of the Tetramethylene-Bridged Hemicarceplex 33 (a) Introduction 33 (b) Investigation of Linkers 35 (c) Investigation of Reaction Conditions 37 i. Time and Temperature 37 ii. Modification to the Base 38 iii. Modifications to Linker 39 iv. Summary 40 (d) Solvent Screening 40 2.2 Template Effect in a Tetramethylene-Bridged Hemicarceplex 46 (a) Guest Screening 46 i. Unsuccessful Guests 47 ii. Successful Guests Not Included in Templation Studies 49 (b) Competition Studies 50 (c) Control Experiments 51 (d) Template Ratios and Template Trends for Hemicarceplex 14»guest 53 i. Introduction 53 ii. Corrections to the Yield of Hemicarceplex 14«guest 54 iii. Template Trends-Size 54 iv. Template Trends-Aromatic Substitution 57 v. Template Trends-Other 57 vi. Conclusions 58 iv (e) Correlation with Previous Templation Studies 59 i. Template Ratios 59 ii. Yields 60 (f) Correlation between Guest Orientation and Changes in Chemical Shifts of Incarcerated and Free Guest Protons in *H N M R Spectra 61 2.3 Conclusions and Future Studies 66 2.4 Experimental 69 (a) General 69 (b) Tetramethylene-Bridged Hemicarceplex Synthesis 72 (c) Competition Experiments 84 (d) Crosscheck Experiments 86 (e) Control Experiments 86 2.5 References 89 v List of Schemes Scheme 1.1: D N A Replication Requires a Template 2 Scheme 1.2: Synthesis of a Cavitand 6 Scheme 1.3: Synthesis of the First Carceplex 7 Scheme 1.4: Synthesis of a Soluble Carceplex 8 Scheme 1.5: The First Hemicarceplex Synthesis 10 Scheme 1.6: Tetramethylene-Bridged Hemicarceplex Synthesis 12 Scheme 1.7: Example of Templation in Crown Ether Synthesis 20 Scheme 1.8: Competition Reaction between Two Different Guests 22 Scheme 1.9: Hydrogen Bonded Complex Formation 23 List of Figures Figure 1.1: CPK Drawing of Hemicarceplex 14 14 Figure 2.1: Cesium Ion Aids in Preorganization 38 Figure 2.2: Cesium Effect in Hemicarceplex 14«guest Synthesis 39 Figure 2.3: Tetramethylene-Bridged Hemicarceplex N M R Assignment Structure 71 vi List of Charts and Tables Table 1.1: Different Shell Closure Combinations and Yields of all Hemicarcerands 25 Chart 1.1: Cavitands and Hemicarcerands Referred to for Section 1.3 (d) 26 Table 2.1: Modifications to reaction Conditions for Hemicarceplex 14 Synthesis. .36 Table 2.2: Solvents Screened for the Hemicarceplex Reaction 42 Table 2.3: Successful Template Molecules 46 Table 2.4: Unsuccessful Template Molecules in the Synthesis of Hemicarceplex 14 48 Table 2.5: Results of Control Experiments Performed on Hemicarceplex 14» guest 52 Table 2.6: Template Ratios and Yields for Hemicarceplex 14»guest 56 Table 2.7: Chemical Shift (8) of Free and Incarcerated Guest Protons and Differences in the *H N M R Spectra of Hemicarceplex 14«guest in CDCI3 at Ambient Temperature 63 Table 2.8: Competition Results 85 vii List of Abbreviations A - angstroms A8 - change in chemical shift 8 - chemical shift CPK - Corey-Pauling-Koltun (models) COSY - correlation spectroscopy D M A - iV^V-dimethylacetamide D M F - TV^V-dimethylformamide D M P U - TVy/V'-dimethylpropyleneurea DMSO - dimethylsulfoxide D N A - deoxyribonucleic acid F A B - M S - fast atom bombardment mass spectrometry GDS - guest determining step J - coupling constant KI - potassium iodide m - meta M - parent mass (mass spectra); or molar (concentration) m/z - mass-to-charge ratio M A L D I - matrix assisted laser desorption ionization NBS - JV-bromosuccinimide NFP - Af-formylpiperidine N M P - iV-methylpyrrolidinone viii N M R - nuclear magnetic resonance (spectroscopy) o - ortho p - para Ph - phenyl (-C 6H 5) ppm - parts per million Ti - longitudinal relaxation time THF - tetrahydrofiiran ix Acknowledgments I sincerely wish to thank my research supervisor, Dr. John Sherman, for guiding and supporting me through these last two years. His addictive enthusiasm and encouragement has been essential to my education and enjoyment at UBC. I would like to thank all past members of the Sherman group, especially Bob Chapman for his extremely helpful suggestions. A special thanks to all present members of the Sherman group, especially Naveen Chopra, Adam Mezo, Ayub Jasat and Ashley Causton for being a great bunch of guys I will never forget. I am indebted to Ashley Causton and Ayub Jasat for proof-reading this thesis and also Naveen Chopra for his helpful suggestions and drawing expertise. I would like to also thank all my friends, especially Sundiep Tehara, Sharlene Faulkes and Robert Poe for making my life outside of the lab an enjoyable one. Special thanks goes to Kayla Feldman for her much appreciated friendship and loan of her computer. Finally, I wish to thank and dedicate this thesis to my mother, Myrna, and my late father, Edward, for their incredible support and encouragement throughout my undergraduate and graduate studies. Chapter 1: Templation Studies in Host-Guest Chemistry 1.1 Introduction The pursuit of larger molecules and complexes with traditional connectivity using the covalent bond is limited by certain structural and bonding rules. Chemists began organizing molecules using non-covalent forces to assemble "supermolecules" - creating a distinct area of science termed "supramolecular chemistry". The supermolecule has distinct structural and chemical properties from those exhibited by the original chemical building blocks.1 Synthesis of a supermolecule involves two main processes, molecular recognition and self-assembly. Molecular recognition is a process where a structurally defined pattern of molecular forces guides molecules to select and bind other molecules.2 This is readily apparent in the formation of double stranded DNA. If D N A is denatured, one single strand of DNA, with its structurally defined base sequence, will only bind another single strand of D N A having a base sequence with the required complementary bases. This is because only complementary bases can effectively hydrogen bond the bases on the first strand. The term self-assembly has been defined by Whitesides as "the spontaneous assembly of individual molecules into structured, non-covalently joined aggregates".3 Thus, it is the molecular recognition forces between the individual components that drive the construction of a self-assembling structure, or "supermolecule"4 An important factor in the self-assembly of some large molecules is the use of a template to assist in the self-assembling process. The Watson-Crick model of D N A does 1 more than explain base pairing, as briefly introduced above, it also demonstrates the ingenious way for D N A molecules to reproduce exact copies of themselves. The replication of D N A is used here to introduce the concept and importance of templation. Replication of D N A begins by a partial unwinding of the double helix, and as bases are exposed, new nucleotides (adenine (A), guanine (G), thymine (T), cytosine (C)) line up on each strand in a complementary manner, A to T and C to G. The enzyme, D N A polymerase, uses the newly separated strand as a template to add a new nucleotide triphosphate to the free hydroxyl group of the growing chain (Scheme 1.1).5 Scheme 1.1: D N A Replication Requires a Template T O H It is difficult to conceive the magnitude of the replication process, as the total D N A in a single human cell is estimated to be 3 billion base pairs. However, the replication of an Template DNA strand New DNA strand entire genome takes only minutes and a error is estimated to occur only once each 10 to 100 billion bases.5 If D N A replication did not evolve from use of a template to assist in replication, a small cut might take months to heal and the possibility for error would definitely be greater! A chemical template has been described by Busch as organizing an assembly of atoms with respect to one or more geometric loci, in order to achieve a particular linking of atoms.6 Templates are different from reagents in that they affect the geometry of the reaction and not the intrinsic chemistry. A template is a thermodynamic template if it shifts the equilibrium of the reaction toward a specific product by binding this product. A kinetic template favors a specific geometry of a reaction by binding to a specific transition state facilitating the formation of a single product.6 This thesis studies the self-assembly of seven molecules to form a hemicarceplex, in which a template molecule aids in the formation of this hemicarceplex. By studying the effects of a template molecule on formation of a supermolecule such as the hemicarceplex, we gain valuable insight into the non-covalent forces that control molecular recognition and therefore the self-assembly process. The first section of this chapter will give an overview of the recent literature focusing on self-assembling structures such as carceplexes and hemicarceplexes. The second part of this chapter will provide a few examples that demonstrate the use of templation in synthesis focusing on the importance of a template in the formation of supramolecular structures such as carceplexes and hemicarceplexes. 3 1.2 Carceplexes and Hemicarceplexes (a) Background In 1987, the Nobel Prize in chemistry was awarded to three researchers whose work created an explosion of growth in the field of supramolecular chemistry and molecular encapsulation. Pederson's discovery of crown ethers7, Lehn's work on cryptands8 and Cram's work with spherands9, provided the initial insight into the process of self-assembly. For example, spherand 1, a rigid macrocyclic molecule composed of a linked cyclic array of/?-anisyl units forming an enforced cavity, can bind and undergo guest exchange with alkali metal cations within its cavity.9 C H 3 R = C H 3 1 It is beyond the scope of this chapter to provide a comprehensive review of all self-assembling structures. The emphasis will be placed on closed or nearly closed surface supermolecules containing an enforced cavity. In 1983, Cram first proposed the concept of a carceplex, which is a closed surface molecule containing a molecule within its interior.10 The entrapped molecule (the "guest"), is held in the interior of the carceplex (the "host"; Latin career = prison, dungeon) in a non-covalent manner and guest escape 4 can only occur upon rupture of covalent bonds. Hemicarceplexes are similar to carceplexes, but differ in that they posses portals or holes in their shell through which guests may escape. Before these supramolecular structures are discussed in detail, this thesis must give an overview of the synthesis of the individual components, namely cavitands (bowl shaped molecules containing an enforced cavity), that form the building blocks of carceplexes and hemicarceplexes. (b) Cavitands Cram expanded on his development of spherands (see 1) in the synthesis of cavitands. Scheme 1.2 illustrates the synthesis of cavitand 7, 1 1 the building block used for the synthesis of the hemicarceplex studied in this thesis. As shown in Scheme 1.2, the first step of the reaction involves the acid catalyzed condensation reaction between four molecules of resorcinol (2) and four molecules of hydrocinnamaldehyde (3) to yield the resorcinarene or "octol" (4). Octol 4 is then brominated with N-bromosuccinimide (NBS) to give the "bromo-octol" (5). The next step is incorporation of intrahemispheric methylene groups ("spanners"), between the adjacent phenolic groups of bromo-octol 5 to yield cavitand 6. The aryl bromine at the rim of the cavitand is further modified by treatment of cavitand 6 with n-butyllithium followed by quenching with trimethylborate and subsequent oxidation using H 2 0 2 , to yield "tetrol" 7. 1 1 Cavitands such as 6 and 7 are highly rigid molecular bowls capable of binding either ions or neutral guests!10 5 Scheme 1.2: Synthesis of a Cavitand; R = CH 2 CH 2 Ph. HH HH 4 NBS/ 2-butanone 50% I. nBuLI ii. B(OMe)3 iii. H202/"OH 53% The pendent groups, or "feet" at the bottom of the cavitand can be varied by using different aldehydes, and a variety of spanners have been recently introduced between the adjacent phenols.11"14 These possible modifications, coupled with the easily functionalized aryl bromide allow a wide variety of cavitands to be synthesized. 6 (c) Carceplexes Shortly after proposing the encapsulation of molecules within a carceplex, Cram reacted tetrabenzylthiol 8 and tetrabenzylchloro 9 cavitands to synthesize the first carceplex (Scheme 1.3).15 Scheme 1.3: Synthesis of the First Carceplex; R = CH3. 9 8 10«guest Carceplex 10 was characterized by mass spectrometry and elemental analysis as containing an assortment of species, such as dimethylformamide (DMF), tetrahydrofuran (THF) and Cs + , found in the reaction mixture. Unfortunately, characterization was limited to F A B mass spectrometry, elemental analysis and 1 3 C solid state cross polarization magic angle spin N M R spectroscopy, as carceplex 10 was virtually insoluble. The use of the more flexible, lipophilic phenethyl groups in the feet was explored 7 to impart enhanced solubility to the carceplex. Thus reaction of tetrol 7 with bromochloromethane yielded the first soluble carceplex (11) (Scheme 1.4).16 The shell closure reaction was performed in neat DMF, dimethylacetamide (DMA) and dimethyl sulfoxide (DMSO) solvents, and produced carceplex 11«DMF, carceplex 11«DMA and carceplex l l»DMSO in 49%, 54% and 61% yields respectively.16 For a presumed entropically disfavored assembly of seven molecules (two tetrols, four bridges and one guest), the yields are remarkably high. This thesis will explore this surprising result and its relevance to our study in more detail in section 1.3 (b). Scheme 1.4: Synthesis of a Soluble Carceplex; R = CH2CH2PI1. 2 ll»guest The *H N M R chemical shifts of entrapped guests were solvent dependent and carceplex 11«DMA was separable from carceplex 11«DMF via silica gel chromatography, demonstrating the entrapped species were capable of interacting with the external environment. Corey-Pauling-Koltun (CPK) molecular model examination of 8 carceplex 11 shows that the largest holes or portals, located at the bottom of each cavitand (or poles of the carceplex) were only about 2 A in diameter.16 Therefore, entrapped guests cannot escape the carceplex, even with prolonged heating in solution. (d) Hemicarceplexes /'. Introduction Hemicarceplexes differ from carceplexes in that the portals between the cavitand units are large enough for the guests to escape the interior with sufficient heat treatment. When greater than 350 °C is required to liberate guests, the two bridged bowl complex is referred to as a carceplex, and when temperatures less than 350 °C is required, the complex is a hemicarceplex.17 Therefore, a hemicarceplex can be isolated and characterized in solution without loss of a guest. There are two general strategies for creating a hemicarceplex: (1) Instead of bridging the two cavitands, such as tetrol 7, with four interhemispheric groups ("linkers"), three bridging groups can be incorporated, thus creating a portal where the fourth bridge would normally be found and; (2) Enlarging the size and/or length of the bridging group so guest escape could occur through the portals between the bridges. /'/'. The First Hemicarceplex Cram utilized the side product, triol 12, formed during the synthesis of tetrol 7, to synthesize the first hemicarceplex. Hemicarceplexes 13»DMA, D M F and DMSO were prepared in 42%, 20% and 51% yields respectively by bridging two molecules of triol 12 9 with bromochloromethane (Scheme 1.5).18 Since hemicarceplex 13«guest possesses a portal through which a guest can potentially escape, heating hemicarceplex 13»guest in a solvent too large to be incarcerated, forces the guest to exit the interior, yielding "empty" hemicarceplex 13 (referred to as hemicarcerand 13). 1 8 Scheme 1.5: The First Hemicarceplex Synthesis; R = CH2CH2PI1 2 13»guest Hemicarceplex 13«guest has lower free energy relative to the noncomplexed hemicarcerand 13, thus helping to explain its thermodynamic stability. Hemicarceplex 13«guest also possesses high kinetic stability. The terms intrinsic and constrictive binding were introduced to explain the kinetic stability. Intrinsic binding is the free energy of complexation of a guest by the hemicarceplex and is based on stabilizing non-covalent interactions between host and guest. Constrictive binding is explicitly defined as the free energy of activation for that complexation; which is equal to the free energy of decomplexation plus the thermodynamic free energy of the hemicarceplex»guest 10 (intrinsic binding). The host itself mechanically imposes the constrictive binding on the guest. For example, the half-life for decomplexation of D M F from hemicarceplex 13 was calculated to be 14 hours at 140 °C. The crystal structure of hemicarceplex 13«DMF showed that a 13° twist of one hemisphere with respect to the other compresses the cavity and increases the number of stabilizing host to guest interactions. Thus, constrictive binding is the energetic barrier to be overcome in order for guests to escape.18 Hemicarcerand 13 is made by forcing the initial guest out of hemicarceplex 13«guest with sufficient heat. Other hemicarceplexes can be made by forcing guests into the hemicarcerand 13. Complexation of a guest with a hemicarcerand is dependent not only on the interactions between that host and guest, but on entropic effects. For instance, free guests in solution are solvated, but upon complexation, the solvent molecules become more disordered providing a solvophobic driving force.17 Additionally, upon complexation, the guest occupies empty space that was in the hemicarcerand, and distributes this empty space throughout the medium ("nature abhors a vacuum").17 ;;'/. Tetramethylene Bridged Hemicarceplex A hemicarceplex can also originate from expanding the size of the four bridging groups between the two cavitand parts. Hemicarceplex 14»DMA, containing four aliphatic bridges was synthesized in 30-40% yields by reacting two tetrol 7 molecules with TsO(CH2)40Ts in D M A solvent (Scheme 1.6).19 Cram has incarcerated 44 different guest molecules into hemicarceplex 14 by complexation of the desired guest with hemicarcerand 14 and by shell closure around the solvent guest molecule. 1 9 ' 2 0 Cram obtained hemicarcerand 14 from hemicarceplex 14»DMA by heating at 205 °C in phenyl 11 ether as solvent. Decomplexation of hemicarceplex 14«DMA in C6D5NO2 occurred at 170 °C with an activation energy barrier of 23.5 kcal/mol. This corresponds to a half-life of decomplexation of 223 minutes at 140 °C, thus demonstrating strong binding despite the large 26-membered-ring portals.19 Scheme 1.6. Tetramethylene Bridged Hemicarceplex Synthesis; R = CH2CH2PI1. 14»guest Although no kinetics of complexation were examined, smaller guests complexed more rapidly than larger guests did and by CPK model examination, the larger guests could only be complexed by difficult, synchronous conformational changes of the hemicarceplex. It was also found that the rates of complexation of disubstituted benzene isomers were in the order: /?-disubstituted > zw-disubstituted > o-disubstituted. Generally, one or two hydroxyl groups attached to benzene also made complexation easier. These results exemplify the high kinetic steric recognition hemicarceplex 14 demonstrates toward constitutional isomers having identical molecular parts, but widely different shapes. 12 Six crystal structures were determined for complexes of hemicarceplex 14 containing either; six H 20,/?-I 2C6H4,/MCH 3) 2C6H4, C 6 H 5 N 0 2 , o-BrCeHtOH or D M A as guests. These crystal structures indicated that there were a number of conformations available to hemicarceplex 14. One conformation has the four (CH2)4 interhemispheric bridges lying inside the volume of the cavity, defined by drawing imaginary lines between the eight phenol oxygens of the host. However, to provide this type of geometry, the phenol oxygens' unshared electron pairs must face inward toward the cavity in the same direction as the adjacent oxygen electron pairs of the O C H 2 0 spanners. The other conformation has the four (CH 2 ) 4 interhemispheric bridges lying outside the volume of the cavity. This allows an alternate out-in-out arrangement of the three adjacent oxygens, thus providing the greatest compensation of dipoles and lowest energy. In this configuration, the portals are widest in the horizontal dimension and shortest in the vertical dimension. Another feature of the four bridges "out" conformation is that a maximum twist of 15° of one hemisphere with respect to the other is possible. This brings the rims of the hemispheres in contact with each other, reducing the cavity size and increasing the number of stabilizing atom to atom (H-H and H-O) van der Waals contacts. The cavity volume ranged from 6895 A3 for hemicarceplex 14»o-I2C6H4 to 6799 A3 for hemicarceplex 14»DMA. The cavity shape is not uniform, as it is longer along the vertical polar axis than along the horizontal equatorial axis (see Figure 1.1). The length of the polar axis ranged from 11.81 A in hemicarceplex 14»/?-I2C6H-4 to 10.55 A in hemicarceplex 14»6H20. Contrary to the polar axis, the size of the equatorial axis ranged from only 6.18 A for hemicarceplex 14»6H20 to 5.84 A for hemicarceplex 14«/?-I2C6H4. 13 Figure 1.1: CPK Drawing of Hemicarceplex 14. Hemicarceplex 14 possesses D 4 symmetry with one C 4 long polar axis and four C 2 short equatorial axes. (Drawing is not accurate representation of actual crystal structure as it has been altered to improve visibility of portals) A C 4 "polar" axis C2 "equatorial" axis Hemicarceplex 14 14 By comparison, the cavity dimensions of carceplex 11«DMA were 9.0 A along the polar axis and 6.2 A along the equatorial axis. The size of the equatorial axis is similar for carceplex 11 and hemicarceplex 14 due to the rigidity of the identical tetrol 7 cavitands. The length of the polar axis is longer in hemicarceplex 14 due to the tetramethylene linker compared to the methylene linker of carceplex 11, demonstrating an elongation of the cavity in hemicarceplex 14. iv. Other Hemicarceplexes Similar hemicarceplexes have also been synthesized with bridges of the form 0(CH2)nO, where n= 2,3,5 and 6. 1 3 ' 2 1 ' 2 2 However, hemicarceplexes are not limited to having aliphatic bridges as there have been numerous other types of bridges used to join two cavitands. For example, Cram et al. have shell closed two tetrol molecules with o-xylyl dibromide in D M A to yield hemicarceplex 15«DMA in 23% yield. 2 3 R R R R R= C H 2 C H 2 P h 15«guest R=CH 2 CH 2 Ph or ( C H 2 ) 4 C H 3 16»guest 15 Hemicarceplexes have been prepared with even larger cavities, such as hemicarceplex 16»guest and hemicarceplex 17. The octaacetylenic hemicarcerand 16 was synthesized in 5-8% yields by oxidation of a tetraacetylenic cavitand with pyridine-CV Cu(OAc)2 2 2 Large molecules such as [2.2] paracyclophane and [3.3] paracyclophane have been successfully incarcerated within hemicarceplex 16. The largest hemicarceplex cavity synthesized to date (17) contains diphenyl o-xylyl bridges providing 42 membered ring portals.24 Hemicarceplex 17 was obtained free of guest, as the solvents used in its shell closure were poorly bound by the large cavity. Complexation of hemicarceplex 17 with Ceo failed even though model examination demonstrated the cavity was of sufficient size. The 16 cyclophanes illustrated above remain the largest neutral molecules encapsulated within a hemicarceplex. The first water soluble hemicarceplex was prepared recently by Cram by reacting tetrol 7 with dibromide 18 in N-methylpyrrolidinone (NMP) solvent and CS2CO3 as base to yield hemicarceplex 19a«NMP in 15% yield. 2 5 Hemicarceplex 19a»NMP was then hydrolyzed to give hemicarceplex 19b«NMP in a 90% yield. Complexation of hemicarceplex 19b with various guests in D 2 0 at 25 °C was complete in a few minutes with the exception of naphthalene (12 hours) whose limited solubility in D2O was the rate-limiting step rather than complexation. v. Conclusions The few examples outlined above demonstrate that hemicarceplexes have enormous potential in medical applications as slow release drug delivery systems. Great strides have been undertaken in synthesizing hemicarceplexes with larger cavities to accommodate large molecular mass antibiotics and medicines and to create a hemicarceplex soluble in water to be physiologically compatible.25 We also require a deeper understanding of the interactions between the guest and host molecules so one day we could design a hemicarceplex to selectively bind a toxin or transport specific drugs. 17 Why are certain shapes of guests preferred by the hemicarceplex? What are the driving forces for the high yield eight-bond shell closure? We shall see in the next section that a template molecule is required to preorganize the self-assembly of seven separate components (two cavitands, four bridges and a guest/template) into a hemicarceplex. The size, shape and functionality of the template molecule can also have a tremendous impact on its ability to template hemicarceplex (and carceplex) formation. 18 1.3 Templation (a) Examples of Templation One of the more widely recognized examples of templation is the use of a metal ion to template the formation of crown ethers. These compounds have the property of binding metallic ions and each individual crown ether can bind a different ion, depending on the size of the cavity. For example, 12-crown-4 (20) binds L i + but not K + , 2 6 while dicyclohexano-18-crown-6 (21) binds K + but not L i + . 2 7 . c T " " V / ^ / O o c r v o 9 20 21 However, the cavity lacks structure until addition of the appropriate cation; the cation aids in preorganizing the crown ether. The Williamson reaction, which is widely used for crown ether synthesis, clearly shows a template effect.28 The K + ion acts as a kinetic template for the formation of 18-crown-6 (22) in a 93% yield, by binding the triethylene glycol dialkoxide, and preorganizing the reactive ends in a conformation favorable for ring closure (Scheme 1.7).28'29 The yield was drastically reduced when the potassium cation was replaced with tetrabutylammonium. 19 Scheme 1.7: Example of Metal Templation in Crown Ether Synthesis Extensive examples of metal templation are available in the literature (including 31 the synthesis of complex molecules such as rotaxanes and catenanes ). But all templation effects are not dependent on metal ions. Examples of templation in synthesis using organic molecules include; Rebek's self-replicating system32, formation of specific zeolites from use of specific organic templates33, and template effects in the synthesis of cyclic porphyrins.34 A common thread for all studies involving templates is that the template serves to preorganize the reactants in order to create a specific product from a number of possible products, or to achieve a higher yield of the product. (b) Template Effects of a Carceplex Reaction Carceplex ll»guest was the first soluble carceplex. The remarkably high yields obtained for carceplex 11»DMA, carceplex 1UDMF and carceplex l l«DMSO (when run in neat solvents), was not the only interesting result interpreted from study of carceplex ll«guest. When neat N-formylpiperidine (NFP), a molecule too large to fit the interior of 20 the carceplex, was used as a solvent, no carceplex was formed. But when the NFP solvent was doped with 0.5 mol % D M A , carceplex 11»DMA was obtained in a 10% yield. Additionally, when the reaction was conducted in 1:1 mixture (v/v) of D M A and D M F as solvent a 5:1 ratio of carceplex 11»DMA to carceplex 11»DMF was obtained.16 These results suggest that a template molecule is required to form carceplex ll«guest, and that the carceplex can demonstrate selectivity toward one template over another. Our group has further investigated the template effects upon carceplex ll»guest formation.35 By performing the carceplex 11 reaction in N-methylpyrrolidinone (NMP) solvent, (which is poor template due to its large size), in presence of two different guest molecules, a mixture of two carceplex ll»guests were formed (Scheme 1.8). 21 Scheme 1.8: Competition Reaction between Two Different Guests 2 11 "guest Carceplex 11 can encapsulate either guest molecule preferentially during its formation, and the product ratios (interpreted as a template ratio) were determined by lH N M R integration of the separate guest peaks. Thirty-four successful template molecules were competed against each other and the ratios for all template molecules were referenced to the poorest template, N M P . 3 5 It was shown that the reaction to form carceplex ll»guest was subject to a million-fold template effect from the best guest (pyrazine) to the poorest guest (NMP). The differences in templating power for the thirty four guests were attributed to the optimal van der Waals interactions between carceplex 11 and guest in the interior of the developing carceplex. Thus, the template ratios reflect the relative rates of what is known as the guest determining step (GDS) for carceplex 22 ll»guest synthesis for the various guest molecules. The GDS is the step in the reaction pathway when no further guest exchange with the medium can occur. The GDS for carceplex ll»guest formation was found to be the formation of the second bridge, either adjacent to or across from the first bridge. This study also demonstrated that 2 molecules of tetrol 7 form a charged hydrogen-bonded complex 23 in the presence of a template molecule and base (Scheme 1.9).35'36 Scheme 1.9: Hydrogen Bonded Complex Formation The binding affinity of complex 23 toward guests correlated with the templating effect of those guests, indicating that formation of complex 23 played an integral role in carceplex ll»guest formation. The template dependent tetrol 7 dimerization preorganizes the subsequent formation of the eight bonds leading to carceplex ll»guest, explaining the high yields obtained. 23 (c) Template Effects of a Hemicarceplex Reaction Previously in this chapter we reported that the yields for hemicarceplex 13»guest formation were greater than statistically calculated. The shell closure involving two asymmetric triol 12 molecules allows the possibilities for misalignment and a statistical yield of 28% is expected. However, yields ranging between 20% and 51% have been reported for hemicarceplex 13«guest.18 This can be attributed to a template effect on hemicarceplex 13»guest formation. Would the presence of a portal in hemicarceplex 13 significantly affect the template effect on formation of the hemicarceplex? An ensuing study to investigate the template effects upon hemicarceplex 13«guest attempted to answer this question.37 Nine representative guests were selected from the previous templation study that spanned the entire template range of 1 to 1 million (best guest, poorest guest and seven in between), and were subjected to competition experiments in NFP as the solvent. The reaction to form hemicarceplex 13«guest was subject to a template effect that spanned 170,000 between the best guest (pyrazine) and the worst guest (NMP) with the same ordering of guests in terms of their templating abilities. Indeed, N M P was approximately a 10 times better template in hemicarceplex 13 than in carceplex 11 under identical reaction conditions. Therefore, the template effects on hemicarceplex 13«guest formation were very similar to the effects on carceplex ll»guest formation.37 Thus, similar van der Waals interactions and steric strain are present in the transition state of the GDS for formation of carceplex ll«guest and hemicarceplex 13«guest. It is also believed that the preorganization of two triol 12 molecules, where the 24 triol molecules align their six hydroxyls to form three hydrogen bonds, prior to hemicarceplex 13»guest synthesis, accounts for the greater than statistical yields observed. (d) Templation aids in Preorganization of an o-Xylyl Hemicarcerand A recent study by Cram demonstrated that hemicarcerand yields improve when the shell closures involve starting materials that are preorganized for hemicarcerand formation.14 Six different hemicarcerands with identical o-xylyl bridges were synthesized by coupling different or similar pairs of cavitands differing in their intrahemispheric or spanner groups (see Chart 1.1).14 The yields for all shell closure combinations are summarized in Table 1.1. Table 1.1: Different Shell Closure Combinations and Yields of all Hemicarcerands8 Cavitand 1 Cavitand 2 Hemicarcerand Yield 24a 24a 27 50% 24a 24b 27 2.2% 25a 25a 28 8.0% 25a 25b 28 43% 26a 26a 29 0% 26a 26b 29 6%b 24b 25a 30 21% 24a 25b 30 2.0% 24b 26a 31 1.8%b 24a 26b 31 0% 25b 26a 32 30% 25a 26b 32 20% a All reactions were performed under identical conditions as described in the text. b 1,2,3-(CH30)3C6H4 had to be present in NMP solvent for shell closure to occur. 25 Chart 1.1: Cavitands and Hemicarcerands Referred to for Section 1.3 (d) Cram et al. hypothesized that yields would improve by reacting tetrol's 24a, 25a and 26a with tetrachloride cavitands 24b, 25b and 26b rather than reacting tetrol's 24a, 25a and 26a with themselves in presence of 1,3-(C1CH2)2C6H4 bridging reagent, as less 26 side reactions would be encountered in the making of four bonds than in making of eight new bonds leading to the shell closures. But inspection of the yields, under the identical shell closure reaction conditions, N M P as solvent, 25-60 °C, and CS2CO3 as the base, revealed some interesting results. The eight bond closure of hemicarcerand 27 prepared from 2 mol of tetrol 24a gave a 50% yield, which was 23 times greater than the four bond shell closure when tetrol 24a was reacted with tetrachloride cavitand 24b (2.2% yield). As previously shown in section 1.3 (b) of this chapter, two tetrol molecules 7 can form a hydrogen bonded complex prior to shell closure, and this dimerization was correlated to template binding power and yields of carceplexes.35'36 An explanation presented believed that the N M P solvent templated the formation of the first two bridges (but exited upon isolation, yielding an empty hemicarceplex—a hemicarcerand), and the hydrogen bonding between the remaining four hydroxyls keeps the two hemispheres preorganized for subsequent shell closure.14 Therefore in the reaction of 24a with tetrachloride cavitand 24b, hydrogen bonding of 24a with itself inhibits shell closure, decreasing the yield. Hemicarcerands 28 and 29 prepared from 2 mol each of cavitands 25a and 26a were lower in yield (8% and 0% respectively) than the same hemicarcerands prepared from cavitands 25a and 26a and tetrachloride cavitands 25b and 26b (43% and 6% yields respectively). This result was as expected as cavitands 25a and 26a cannot form a preorganized complex due to the increased steric hindrance of the ethylene and propylene spanners. For the synthesis of hemicarceplexes 30 and 31 with mismatched hemispheres, reactions with tetrol 24a with tetrachloride cavitands 25b and 26b gave lower yields than cavitand 25a and 26a with tetrachloride 24b, as disruption of the complex had to occur prior to shell closure. 27 The order of shell-closure yields leading to these six hemicarcerands was 27 > 28 > 32 > 30 > 29 > 31. Interestingly, the power of an appropriate template also affected the yields of two hemicarcerands. In the synthesis of 29 from cavitand 26a and tetrachloride cavitand 26b and of 31 from cavitand 26a and tetrachloride cavitand 24b, no shell closures occurred unless the N M P solvent was doped with 5 mol % 1,2,3 trimethoxybenzene. This suggests that for the two largest hemicarcerands, a larger molecule than N M P , such as 1,2,3 trimethoxybenzene, was required to template the shell closure. Hemicarcerand 31 was actually isolated containing 1,2,3 trimethoxybenzene (therefore a hemicarceplex). For hemicarcerand 29, it is believed 1,2,3 trimethoxybenzene exited the interior by mass-law driven exchange with N M P after the final bridge is in place, due to the large 32-membered ring portals. This study illustrates the power of preorganization in two aspects: (1) that a template can effectively preorganize two molecules to properly shell close and; (2) the preorganization of two tetrol units can have both a positive and negative effect on the yields of desired hemicarcerands. 1.4 Project Goals and Thesis Objectives The incredible power and selectivity of template effects have been demonstrated upon a completely closed surface carceplex 11, and on a similarly structured hemicarceplex 13 having a small portal and a slightly increased cavity. The presence of this portal had no significant effect on the templating power of identical guests as the cavity showed similar selectivity toward the same guests as carceplex 11. We also saw the templates role in 28 forming a preorganized complex in both reactions to account for the remarkably high yields. Would increasing the size of the cavity and the introduction of four portals significantly affect the template effect? To implement these changes the bridging reagent must be larger, and i f so, would this diminish the positive effect of the preorganized complex in obtaining high yields? Would this larger cavity exhibit identical selectivity toward the same guests used in the templation study of carceplex ll»guest an hemicarceplex 13»guest? These questions were addressed by study of the template effects upon the large tetramethylene bridged hemicarceplex 14 (described in section 1.2 (d) iii.) and the results are discussed in the following Chapter. 29 1.5 References 1. Desiraju, G. R. Angew. Chem. Int. Ed Engl. 1995, 34, 2311. 2. Lehn, J. M . Angew. Chem. Int. Ed. Engl. 1988, 27, 89. 3. Whitesides, G. M . ; Mathias, J. P.; Seto, C. T. Science 1991, 254, 1312. 4. Bradley, D. Chem. Soc. Rev. 1995, 379. 5. (a) Voet, D.; Voet, J. G. Biochemistry, John Wiley and Sons: New York, 1990; pp 948-973. (b) McMurry, J. Organic Chemistry, Brooks/Cole Publishing: Pacific Grove, 1992; pp 1121-1125. 6. Busch, D.H.J. Inclusion Phenom. Mol. Recognit. Chem. 1992,12, 389. 7. Pederson, C. J. Angew. Chem. Int. Ed Engl. 1988, 27, 1021. 8. Lehn, J. M . Angew. Chem. Int. Ed. Engl. 1988, 27, 89. 9. Cram, D. J. Angew. Chem. Int. Ed. Engl. 1988, 27, 1009. 10. Cram, D. J. Science 1983, 219, 1177. 11. (a) Cram, D. J.; Karbach, S.; Kim, H . E.; Knobler, C. B. ; Maverick, E. F.; Ericson, J. L . ; Helgeson, R. C. J. Am. Chem. Soc. 1988,110, 2229. (b) Tunstad, L . M . ; Tucker, J. A. ; Dalcanale, E. ; Weiser, J.; Bryant, J. A . , Sherman, J. C ; Helgeson, R. C ; Knobler, C. B. ; Cram, D. J. J. Org. Chem. 1989, 54, 1305. 12. (a) Gibb, B. C ; Chapman, R. G ; Sherman, J. C. J. Org. Chem. 1996, 61, 1505. (b) Timmerman, P.; Verboom, W.; Reinhoudt, D. N . Tetrahedron 1996, 52, 2663. 13. Helgeson, R. C ; Paek, K ; Knobler, C. B ; Maverick, E. F.; Cram, D. J. J. Am. Chem. Soc. 1996,118, 5590. 14. Helgeson, R. C ; Knobler, C. B. ; Cram, D. J. J. Am. Chem. Soc. 1997,119, 3229. 15. Cram, D. J.; Karbacck, S.; Kim, Y . H . ; Baczynskyj, L , Marti, K . ; Sampson, R. M . ; Kalleymeyn, G. W. J. Am. Chem. Soc. 1988,110, 2554. 16. Sherman, J. C ; Knobler, C. B. ; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2194. 17. Maverick, E. ; Cram, D. J. Comprehensive Supramolecular Chemistry; Pergamon: New York, 1996; Vol. 2, pp 367-388. 30 18. (a) Tanner, M . E ; Knobler, C. B.; Cram, D. J. J. Am. Chem. Soc. 1990,112, 1659. (b) Cram, D. J.; Tanner, M . E.; Knobler, C. B. J. Am. Chem. Soc. 1991,113, 7717. 19. Robbins, T. A. ; Knobler, C. B . , Bellow, D. R.; Cram, D. J. J. Am. Chem. Soc. 1994,116, 111. 20. Kurdistani, S. K ; Helgeson, R. C ; Cram, D. J. J. Am. Chem. Soc. 1995,117, 1659. 21. Byun, Y . S.; Robbins, T. A. ; Knobler, C. B. ; Cram, D. J. J. Chem. Soc, Chem. Commun. 1995, 1947. 22. Cram, D. J.; Jaeger, R.; Deshayes, K. J. Am. Chem. Soc. 1993,115, 10111. 23. Cram, D. J.; Blanda, M . T.; Paek, K. ; Knobler, C. B. J. Am. Chem. Soc. 1992, 114,1165. 24. Von dem Bussche-Hunnefeld, C ; Buhring, D ; Knobler, C. B. ; Cram, D. J. J. Chem. Soc, Chem. Commun. 1995, 1085. 25. Yoon, J.; Cram, D. J. J. Chem. Soc, Chem. Commun. 1997, 497. 26. Bartsch, R. H . ; Czech, B. P.; Kang, S. I ; Stewart, L . E. ; Walkowiak, W.; Charewicz, W. A , Heo, G. S.; Son, B. J. Am. Chem. Soc 1985, 707, 4997. 27. Izatt, R. M ; Nelson, D. P.; Rytting, J. H . , Haymore, B. L . ; Christensen, J. J. J. Am. Chem. Soc. 1995, 93, 1619. 28. Greene, R. N . Tetrahedron Lett. 1972, 1793. 29. Cacciapaglia, R.; Mandolini, L . Chem. Soc. Rev. 1993,22,221. 30. (a) McMurry, T. J.; Raymond, K. N ; Smith, P. H. Science 1989, 244, 938. (b) Busch, D. L . ; Vance, A. L . ; Kolchinski, A. G. Comprehensive Supramolecular Chemistry, Pergamon: New York, 1996; Vol. 9, pp 1-40. 31. Schill, G. Catenaries, Rotaxanes and Knots, Academic Press: New York, 1971. 32. Wintner, E. A. ; Conn, M . M . ; Rebek, J. Acc. Chem. Res. 1994, 27, 198. 33. Zones, S. I ; Nakagawa, Y . ; Harris, T. V. J. Am. Chem. Soc. 1996,118, 7558. 34. Anderson, S.; Anderson, H . L . ; Sanders, J. K. M . J. Chem. Soc, Perkin Trans. 1 1995, 18, 2255. 31 35. (a) Chapman, R. G.; Chopra, N . ; Cochien, E. D.; Sherman, J. C. J. Am. Chem. Soc. 1994,116, 369. (b) Chapman, R. G ; Sherman, J. C. J. Am. Chem. Soc. 1995, 777,9081. 36. Chapman, R. G. Ph. D. Thesis, The University of British Columbia, 1997, 137-289. 37. Chopra, N . ; Sherman, J. C. Supramolecular Chemistry 1995, 5, 31. 32 Chapter 2: Template Effects of a Tetramethylene-Bridged Hemicarceplex 2.1 Synthesis of the Tetramethylene-Bridged Hemicarceplex (a) Introduction A wide variety of hemicarceplexes can be synthesized from two cavitands ("bowls"); these can arise by altering the spanners or links between the arene units within the bowls2 or by varying the bridges in length and structure joining the bowls to change the size of the cavity.3'4 Modifying the "feet" attached to the underside of the "bowl" to impart different physical properties, such as solubility, to the hemicarceplex is another avenue open to the researcher. Cram synthesized the tetramethylene bridged hemicarceplex 14 in 30-40% yields by bridging tetrol 7 with TsO(CH2)40Ts in D M A solvent (also the incarcerated guest) with CS2CO3 as the base.4 The crystal structures of hemicarceplex 14 which contained a variety of incarcerated guests, demonstrated that hemicarceplex 14 can adjust its cavity size and shape to optimize the van der Waals interaction between host and guest.4 Therefore, hemicarceplex 14 has a strong binding affinity for its guests as shown by the high energy barrier of 23.5 kcal/mol for decomplexation of D M A and relatively large M«guest signals in the F A B mass spectra, whose harsh conditions usually prevent non-covalent species from being observed. According to Cram, "this host is the strongest and most versatile hemicarcerand yet prepared" 4 This hemicarceplex is a logical choice to study the template effects of a large cavity, owing to its strong binding properties and synthesis in high yield from two molecules of tetrol 7. 33 Before we could study template effects in this hemicarceplex 14's formation we decided to first investigate some of the variables involved in the reaction. We had to have a standard set of reaction conditions that would produce the hemicarceplex 14 in good yield and be suitable for competition reactions. An obstacle to achieving a high yield was the side reaction leading to polymer formation, that would be encountered in the sequential making of eight bonds to form hemicarceplex 14. An increase in temperature and/or time of reaction might be enough to overcome the high energy barrier of the eight-bond shell closure required. However, it would not be advisable to increase the temperature due to the risk of decomplexation of guest from host. This leads to the second objective of finding a set of reaction conditions that would be appropriate for our competition experiments. We required a time and temperature of reaction that would not allow this decomplexation to occur, and we also ideally needed a solvent that would not become incarcerated during the shell closure. The solvent chosen must be a poor template for hemicarceplex 14 formation, in order to allow the hemicarceplex to preferentially encapsulate the competing guests. Therefore, the linker reagent, solvent, time, temperature, base and other reaction conditions were examined to satisfy our requirements of a high yield synthesis and still be suitable for our competition studies involving hemicarceplex 14. 34 (b) Investigation of Linkers Cram et al reported the synthesis of hemicarceplex 14»DMA by treating tetrol 7 with Br(CH2)4Br, without high dilution, using D M A as the solvent in a 15% yield. 4 Cram also explored using TsO(CH2)40Ts under high dilution conditions as the bridging reagent. By using the ditosylate linker and D M A as the solvent, the yield of hemicarceplex 14«DMA was increased to 30-40%.4 Initial attempts to reproduce hemicarceplex 14»DMA using the ditosylate linker under the identical reaction conditions outlined by Cram resulted in no greater than a 20% yield in our hands (see condition N , Table 2.1). We then explored 1,4 dibromobutane as a potential bridging reagent. The best yield obtained for hemicarceplex 14«DMA was 29% using 10 equivalents of linker per molecule tetrol, two days at 80 °C in presence of KI using D M A as the solvent and CS2CO3 as the base (condition A, Table 2.1). KI was used because the I* species would displace the bromine in the linker via halide exchange and since I" is a better leaving group, the linker should now have a higher lever of reactivity. Other modifications (summarized in Table 2.1) either decreased or did not affect the yield of hemicarceplex 14»DMA. Therefore, since it appeared that Br(CH.2)4Br bridged the two tetrol 7 molecules in a better yield than observed for the TsO(CH2)40Ts linker, without a need for high dilution conditions, 1,4 dibromobutane was our obvious linker of choice. 35 Table 2.1: Modifications to Reaction Conditions for Hemicarceplex 14 Synthesis8 Rxn. Condition Time/Temo. Linker (eauiv./tetrol) Add. Method of linker1* Base KI (equiv/linker) Dilution of tetrol/solvent Yield A 2d/80°C Br(CH2)4Br (10) Dump C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 22-29% B 3d/80°C Br(CH2)4Br (60) 20 equiv./day (dump) C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 20-22% C 2d/80°C Br(CH2)4Br (20) 10 equiv./day (dump) C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 25% D 2.5d/80°C Br(CH2)4Br(10) Slow addit. (48hrs) C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 3% E 3.5d/80°C Br(CH2)4Br (10) Slow addit. (48hrs) C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 5% F 2d/80°C Br(CH2)4Br(10) Dump C s 2 C 0 3 None 50 mg / 20 ml 20% G 2d/80°C Br(CH2)4Br(10) Dump C s 2 C 0 3 Yes (6.0) 50 mg /20 ml 26% H 2d/80°C Br(CH2)4Br(10) Dump K 2 C 0 3 Yes (0.8) 50 mg / 20 ml 4% I 2d/rt-2d/80°C Br(CH2)4Br(10) Dump C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 17% J 2d/80°C Br(CH2)4Br (10) Dump C s 2 C 0 3 Yes (0.8) 25 mg/30 ml 19% K 2d/50°C Br(CH2)4Br (10) Dump C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 5% L 4d/60°C Br(CH2)4Br(10) Dump C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 15% M 7d/rt Br(CH2)4Br(10) Dump C s 2 C 0 3 Yes (0.8) 50 mg / 20 ml 2% N 4d/60°C TsO(CH2)4OTs (8) Slow addit. (48hrs) C s 2 C 0 3 N / A 50 mg/ 20 ml 20% All reactions in neat DMA solvent b Dump refers to addition of linker in one portion. No high dilution conditions. 36 (c) Investigation of Reaction Conditions Once we had our linker of choice we were left with acquiring a standard procedure that would be used for all future competition reactions between different guests. This standard procedure had to meet two goals: (1) To improve the yield of 29% which represents the yield in the best solvent (DMA). Our competition reactions were not going to be performed in a templating solvent such as D M A , but rather in a non-templating solvent (see section 2.1 (d)). (2) Attempt to decrease the harshness of the reaction conditions previously employed by Cram (2 days, 80 °C). Even though the energy barrier for decomplexation of D M A was high, hemicarceplex 14 contains four 26-membered ring portals through which guests can potentially escape. For example, Cram reported that refluxing hemicarcerand 14 in acetone for 3 days at 70 °C was all that was required to form hemicarceplex 14«acetone.4 If one or more of the competing guests we were to employ in our future competition reactions were able to escape the hemicarceplex 14 under the reaction conditions, the template ratios reported in Table 2.6 would be invalid. i. Time and Temperature As shown in Table 2.1, the extension of the initial reaction time (2 days at 80 °C) by one full day at 80 °C (condition B) or two full days at ambient temperature (condition I) produced hemicarceplex 14«DMA in yields of 22% and 17% respectively. We still wished to decrease the temperature of the reaction conditions and still have a yield 37 comparable to those observed for 2 days at 80 °C. Unfortunately, reaction temperatures and times of 2 days at 50 °C (condition K), 4 days at 60 °C (condition L) and 7 days at room temperature (condition M) decreased the yields to 5%, 15% and 2% respectively. It appeared that at least 80 °C was required to overcome the energy barrier to hemicarceplex 14»DMA formation. 77. Modification to the Base Carbonate base is an effective base in a wide variety of carceplex and hemicarceplex shell closure reactions,2"4 but would a change of the counter ion have an impact on the overall yield of hemicarceplex 14»guest? In order to investigate this, we decreased the counter ion size from Cs + to K + , and the yield decreased substantially to 4% (condition H , Table 2.1). This observation may be attributed to what is known as the "cesium effect".5 Cesium forms the cation with the largest ionic radius and follows T l + with respect to the largest polarizability, in the first main group of elements. The large cesium ion, in contrast to the smaller alkali metal cations, has the capability of preorganizing the 6 7 reactants. ' A study by Vogtle and Meurer demonstrated favored intramolecular ring closure in the synthesis of cyclophanes using cesium thiolates (Figure 2.1).7 Figure 2.1: Cesium ion aids in preorganization (a) Preorganization of an 11-membered intermediate during cyclization by cesium cations, (b) With the smaller Na+ cation the cyclic intermediate cannot be preorganized due to steric repulsion between the two H atoms. 38 The Cs ion could behave in a similar fashion in our system by coordination of the oxygen anion and the linker at its surface as depicted in Figure 2.2. Figure 2.2: Cesium effect in hemicarceplex 14*guest synthesis. Comparison of the coordination of the tetrol complex and linker to the large cesium and small potassium cation A larger cation reduces steric repulsion in formation of the first bridge. Size of cations is relative to each other and are not exact. R = C H 2 C H 2 P h Cesium's large size and polarizability reduces steric interference between the linker and the two tetrol units 7 in the formation of the first bridge. Thus, the synthesis of hemicarceplex 14»guest is another example of the many cyclizing and self assembling systems that demonstrate a cesium effect.8 iii. Modifications to Linker i We next looked at increasing the equivalents of 1,4 dibromobutane linker per molecule of tetrol. A two-fold increase from 10 to 20 linker equivalents by addition of 10 equivalents per reaction day for a total of two days (condition C, Table 2.1) achieved a similar yield to our initial conditions. Addition of 20 equivalents per day for three 39 reaction days (i.e.: a six-fold increase) did not significantly affect the yield obtained (condition B). The next logical step was then to alter the method of linker addition. By using high dilution conditions we hoped to eliminate polymer formation. Unfortunately, as observed in Table 2.1, slow addition of 10 equivalents of linker over a 48 hour period and an additional reaction time of 12 or 36 hours decreased the yields of hemicarceplex 14«DMA to 3% and 5% respectively (conditions D and E). Other attempts to decrease polymer production were unsuccessful and are summarized in Table 2.1 (conditions F, G and J). iv. Summary For our templation studies we concluded that the initial reaction conditions of 2 days at 80 °C, 10 equivalents of 1,4 dibromobutane linker per tetrol, CS2CO3 as the base, 0.8 equivalents KI per linker molecule and a dilution of 2.5 x 10"3 M were the best possible conditions to achieve a balance between modest yield and prevention of guest decomplexation. As we will see in section 2.2 (c), we devised a set of control experiments to test for guest decomplexation and/or exchange at these chosen reaction conditions. (c) Solvent screening The highest yield of hemicarceplex 14 formation was found to be in D M A solvent, but D M A also templated the formation of hemicarceplex 14«DMA. Therefore, 40 D M A was considered to be too good of a guest to be used as a solvent for our competition reactions where we want to study a number of different guests. As shown earlier in Chapter 1, dipolar, aprotic solvents such as D M A , D M F , DMSO have been used extensively in the formation of carceplexes and hemicarceplexes, but when used as a neat solvent, became incarcerated in the hosts. Therefore, our goal was to find a solvent that would not template the formation of hemicarceplex Insolvent; such a solvent could then be doped with a number of competing guests but would still drive hemicarceplex 14«guest formation. The solvents N-methylpyrrolidinone (NMP) and N-formylpiperidine (NFP) have been used in previous templation studies, as they depicted the polar and aprotic characteristics that seem necessary for carceplex and hemicarceplex formation,9'10 and are too large to become incarcerated within the host. Previous studies done by Chapman demonstrated that other polar and non-polar solvents can also lead to carceplex ll»guest formation.11 For example, when THF, toluene and benzene were employed as solvents, carceplex 11 was synthesized in yields of 20%, 25%, and 30% respectively. Unfortunately, the reaction to form carceplexes proceeds much slower in these solvents, as it required times of between 5 and 10 days at 60 °C to achieve the reported yields. Chapman produced carceplex 11 in 70% yield using nitrobenzene as a solvent in only 2 days at 60 °C. Suprisingly, the highly apolar solvent cyclohexane (6 days, 60 °C) led to a 60% yield of carceplex. The experiments by Chapman opened up a wider number of possible solvents that could be tested for our competition studies and we screened 14 solvents in the presence of various templating guests (summarized in Table 2.2). 41 Table 2.2 Solvents Screened for the Hemicarceplex Reaction8 Solvent Guest(s) Time % Yield hemicarceplex 14«guest DMF b DMA 2 days 4.7% Diethylacetamideb DMA 2 days 0% NMP b DMA 2 days 16% NFP b DMA 2 days 13% l-acetyl-3-methylpiperidineb DMA 2 days 4.5% Tetramethlene sulfoneb NMP & 2-butanol 2 days 0% DMPU b NMP & 2-butanol 2 days 0% Nitrobenzeneb DMA & NMP 2 days 0% Ethyl acetate5 EtOAc 2 days 0% Cyclohexanec DMA & NMP 2 days <1% Cyclohexane0 DMA & NMP 4 days <1% Acetonitrile" DMA & NMP 2 days 0% Acetone0 NMP & 2-butanol 2 days 0% 2-butanone° NMP & 2-butanol 2 days 2.5% 2-butanonec NMP & 2-butanol 4 days 2.4% THF° NMP & 2-butanol 2 days 0% a Using 1,4 dibromobutane linker, KI, and CS2CO3. b Reaction was stirred at 80 °C. 0 Reaction was refluxed. d Guest was present as 1 mol % based on solvent screened, except EtOAc. Diethylacetamide has similar polar, aprotic characteristics to D M A but since it was larger than D M A we thought it would be a suitable solvent, but unfortunately, when screened in presence of D M A , no hemicarceplex 14»DMA product was isolated (Table 2.2). N M P solvent had the best potential as Chapman had successfully used it previously 42 in his templation experiments. While we achieved a respectable yield of 16% for hemicarceplex 14«guest(s) with N M P as solvent, N M P itself was incarcerated in the hemicarceplex. In fact the ratio of hemicarceplex 14«NMP to hemicarceplex 14»DMA products was approximately 1.4 : 1 demonstrating that N M P was a better template for hemicarceplex 14 formation than D M A . When we employed NFP as a solvent (1 mol % D M A ) we isolated a 20 : 1 mixture of hemicarceplex 14«DMA : hemicarceplex 14«NFP in a 13% yield. This promising result led us to test l-acetyl-3-methylpiperidine, a similar structure to NFP, but with two additional methyl groups, one altering the formyl group into an acetyl and one at the 3 position. A yield of 4.5% was obtained and no l-acetyl-3-methylpiperidine was incarcerated; but the lH N M R for the product showed an incomplete reaction, as the linker protons did not correlate to 32 protons. This could mean a mixture of two and/or three bridged species was recovered. Substantiating this theory was the absence of incarcerated D M A by lH N M R , as D M A could more easily escape an incompletely bridged hemicarceplex. We investigated a number of other solvents such as tetramethylene sulfone, dimethylpropyleneurea (DMPU), nitrobenzene, cyclohexane, ethyl acetate, acetone and 2-butanone. Of these solvents, only cyclohexane and 2-butanone showed any promise with yields of <1% and 2.5% respectively. Extension of reaction time by 2 days for these two solvents did not improve hemicarceplex 14 yields. Our initial approach was to screen solvents that were too large to become incarcerated during the competition reactions, but what if we found a solvent that was small enough to enter and depart the interior rapidly, thus avoiding permanent 43 encapsulation? We therefore tried D M F in the presence of D M A as guest, and obtained a 4.7% yield of hemicarceplex 14»DMA and hemicarcerand 14. *H N M R characterization of the product showed that most of the hemicarceplex created was empty and very little hemicarceplex 14»DMA was isolated. It therefore appears that D M F was indeed templating the formation of hemicarceplex 14, but then escaped the interior under the reaction conditions. D M A , being larger than DMF, was not able to enter the interior without sufficient heating. D M F would then not be a suitable solvent, as the competition reactions would not be based on kinetic templating factors, but on complexation rates of the competing guest molecules (thermodynamic). Additional screening of small solvent molecules such as, acetone, acetonitrile and THF did not yield any noticeable amount of hemicarceplex 14»guest (Table 2.2). The solvent that best met our criteria of being a poor guest and producing hemicarceplex 14»guest in a decent yield was NFP. Attempts to separate hemicarceplex 14»NFP from the competing guests by column chromatography were unsuccessful, therefore 10 out of 17 guests used in our templation studies were characterized as hemicarceplex 14»guest: hemicarceplex 14VNFP mixtures. These 10 guests were: 2-butanol, chlorobenzene, benzene, toluene, 1,4 thioxane, ethyl acetate, cyclohexane, THF, 2-propanol and isopropylacetate. NFP solvent had additional drawbacks: by being a modest template itself it automatically eliminated any potential guests that: (a) are worse templates than NFP, and (b) when used in 1 mol % or 5 mol %, based on NFP solvent, they might not become incarcerated due to the 20-100 times excess of NFP, even though they have template ratios above that of NFP. In spite of these disadvantages, NFP became our solvent of choice for our competition reactions. 44 2.2 Template Effect in a Tetramethylene-Bridged Hemicarceplex (a) Guest Screening Having established suitable reaction conditions, we now set out to investigate the templating effects of the tetramethylene bridged hemicarceplex. 54 potential guest molecules were screened. The guests chosen varied in size, shape, symmetry and polarity; and were screened by incorporating them as a co-solvent (1 mol % - 5 mol % based on NFP) into the reaction mixture. Of these 54 molecules, 22 were found to template the formation of hemicarceplex 14»guest (Table 2.3) while 32 molecules failed to become incarcerated (Table 2.4). Table 2.3: Successful Template Molecules. Molecules boxed were successful, but were not used in templation studies. o •t A H3C CH3 A * H3C CH3 O H C H 3 H 3 C O A .CH3 H 3 C N C H 3 O H 3 C O C H 3 O CH3 H 3 C ^ O \ H 3 0 o CH3 o s o CH3 CI C I — V 7—CI Br—V 7—Br H 3C—<f y—CH3 45 /. Unsuccessful guests The potential guests that were unsuitable as templates were possibly too small or too large to template hemicarceplex 14 formation. Another likely reason is that some unsuccessful molecules might have the potential to act as templates, but are possibly worse templates than the solvent NFP, which is in large excess. A final possibility is that some of these guests may indeed template the formation of hemicarceplex 14, but exit the interior during the course of the reaction or work up of the final product. Hemicarceplexes 14 with guests that were found to be ineffective as templates, such as acetone, nitrobenzene, hexafluorobenzene, o-xylene and w-xylene, have been reported to form complexes, hemicarceplex 14»guest, when mixed with hemicarcerand 14.4 It is possible that these molecules can behave as templates but are worse templates than NFP. Alternatively, it is likely that while these molecules can be forced into hemicarceplex 14's interior with sufficient heating, they do not lower the transition state energy of the guest determining step (GDS) prior to hemicarceplex 14»guest formation, thus exhibiting no template effect. The configuration of the transition state could differ enough from the final hemicarceplex to influence the templating ability of the guest. /7-Dihydroxybenzene was another unsuccessful guest screened that too has been characterized within hemicarceplex 14.4'12 However, this guest was incarcerated in that study, as with the aforementioned guests, by forcing it into hemicarcerand 14 with sufficient heat4, and not by templation of hemicarceplex 14 shell closure. From screening experiments with this molecule we isolated no hemicarceplexes, even in presence of other guests, leading us to believe that /?-dihydroxybenzene was being alkylated by the linker under our reaction conditions. 46 Table 2.4. Unsuccessful Template Molecules in the Synthesis of Hemicarceplex 14 C H 3 C N CH3OH cH 3 CH 2 OH CH 3CH 2CH 2OH O 11 H3C CH3 o II H 3 C C H 2 C H 3 HqC O -CH3 u CH 3 O 3 H A N ^ C H 2 C H 3 I C H 2 C H 3 O H 3 C ^ N ' C H 2 C H 3 C H 2 C H 3 C H 2 C H 3 H 3 C H 2 C C H 2 C H 3 o O \ o \ I H ' x // O N—\ \ / H P O CH 3 H3C CH 3 /—N Oo CH 3 \\ H CH 2CH 3 HO—(\ // 0 H / = \ O H 3 C ^ / ) \ N i — J H CH, H3C \ / ^ C H a CH 3 -OCH, OCH3 H3CO <(J>-OCH 3 H3CO 47 ii. Successful guests not included in templation studies Five molecules that were outlined in Table 2.3 as successful guests were not included in the competition studies. Cyclohexanone, 1,4-dioxane and thiophene demonstrated approximate template abilities only slightly greater than that of NFP and will be pursued in future studies by our group. Hemicarceplex 14»pyrazine, on the other hand, was observed only by M A L D I mass spectrometry, as it is likely that from *H N M R , the protons of incarcerated pyrazine are buried under the host peaks. This would cause difficulty in integration and therefore hemicarceplex 14»pyrazine was left out of our competition experiments. Incarcerated />-dimethoxybenzene was observed by both M A L D I mass spectrometry and *H N M R but was not included in the template studies due to an anomaly observed only by this guest. For instance, at 1 mol % /?-dimethoxybenzene we either (a) did not observe any incarcerated /?-dimethoxybenzene or (b) the template ratio ranged from 8 to 54 relative to NFP. At 5 mol %/>dimethoxybenzene (in NFP) we observed a template ratio range of 155 to 2400. The first problem was the 300-fold variation in the template ratio observed for this guest, and the second was the general trend that as we increased the concentration of /?-dimethoxybenzene, the template ratio seemed to increase. These preliminary competition reactions were carried out in the presence of other different guest molecules, with similar results, dispelling the hypothesis that another competing guest was interfering with or enhancing/7-dimethoxybenzene's ability to template hemicarceplex 14 formation. If the 1,4 dimethoxybenzene were forming aggregates in solution, thus tying up individual /?-dimethoxybenzene molecules, this would show a general decrease in templating ability as the concentration increased. If the solubility of 48 /j-dimethoxybenzene in NFP were a factor, again we should be observing a general decrease in its templating effects as the concentration increased. Therefore, p-dimethoxybenzene was not included in our competition experiments. (b) Competition Studies Competition experiments were performed on 17 of the successful template molecules to directly study the templation of hemicarceplex 14«guest formation. This initially entailed running the reaction under the standard conditions chosen previously but in the presence of two competing guest molecules (1 mol % guests based on NFP). This gives the hemicarceplex a choice of entrapping either guest molecule preferentially during its formation. After completion of the reaction, the mixture of two hemicarceplex products was purified and the product ratios for each pair were determined from integration of the unique guest peaks in the ' H N M R spectra. Incarceration of guest molecules in hemicarceplex 14 had a dramatic effect upon the N M R chemical shift of the guest species' protons; as much as 4.40 ppm upfield from their normal position. The upfield shift also occurs in other related systems, such as carceplex 11 and hemicarceplex 13.9'10 This provides a clear window for ' H N M R analysis. The product ratios obtained for each mixture are then interpreted as template ratios, providing us with a hierarchical listing of the template molecules studied. Once we had our initial ordering of template ratios, we performed further competition experiments between guests that had similar templation abilities to obtain a more precise ordering of the templates; occasionally unequal amounts of guest were used to give an equal mixture of two hemicarceplexes, thus improving the 49 accuracy of the integration. Each of the template ratios were then referenced to the poorest template (NFP), which was arbitrarily assigned a value of 1. The template ratios obtained are summarized in Table 2.6 and the results are discussed in sections 2.2 (d) & (e) of this chapter. (c) Control Experiments A series of control experiments were performed to ensure the validity of the template ratios outlined in Table 2.6. By definition, a hemicarceplex possesses a portal by which guest exchange may take place. If such guest exchange were possible under our reaction conditions (2 days, 80 °C), the template ratios would not be dependent upon the guest lowering the energy of the transition state of the hemicarceplex reaction, but upon the relative decomplexation and recomplexation rates. This would lead to a thermodynamic ratio, rather than a kinetic ratio of the products. Hemicarceplex 14»guest was therefore subjected to conditions used for the competition reactions, but in the presence of various other guest molecules. The *H N M R integration of guest and host peak(s) for hemicarceplex 14»guest was examined for exchange in and out of the hemicarceplexes' interior before and after the control experiment. Percent in was based on appearance of new guest peaks, while % out was calculated based on integration of guest peaks and comparing to integration of host. Only the highest value of exchange was reported and it was found that no more than 9% guest exchange had occurred (see Table 2.5).' This is within our estimated experimental error of ±10% in the *H N M R integration. 1 see experimental for detailed procedure of control experiments. 50 Table 2.5: Results of Control Experiments Performed on Hemicarceplex 14*guest Hemicarceplex 14* Procedure3 % Guest Exchange /7-Xylene E 1% /7-Dibromobenzene E 5% /7-Dichlorobenzene E 0% 2-Butanol E 0% Chlorobenzene E 8% Benzene E 9% Toluene D 6% Thioxane D 7% EtOAc D 9% Cyclohexane D 0% N M P D 0% D M A D 2% THF D 0% 2-Propanol D 0% DMSO D 6% Isopropyl acetate D 0% NFP D 0% see experimental for detailed procedure for control experiments. The first templation study by Chapman on carceplex 11 was done in presence of a 2.0 second relaxation delay during acquisition of the ! H N M R spectra since the Ti relaxation rates for incarcerated guests were calculated to be between 1 and 2 seconds.9 , 1 1 A subsequent templation study with hemicarceplex 13 by Chopra was done in the absence of a relaxation delay, as it was determined that no appreciable difference was observed in the template ratios with or without a 2.0 second delay.10 The host to guest integration of hemicarceplex 14»DMA without a relaxation delay showed 83% hemicarceplex 14«DMA (the remainder being hemicarcerand 14 impurity). The X H N M R spectrum was acquired with a 2.0 second relaxation delay and yielded 81% hemicarceplex 14»DMA. 51 The difference is minimal and within error; therefore our template ratios were obtained in the absence of a relaxation delay. (d) Template Ratios and Template Trends for Hemicarceplex 14*guest /'. Introduction The competition experiments revealed that the formation of hemicarceplex 14«guest is subject to a template effect such that the best guest (p-xylene) is 950 times more effective as a template than the poorest guest, NFP. The 17 successful template molecules, their template ratios and yields are listed in Table 2.6. The template ratios presented reflect the relative rates of the guest determining step (GDS) in hemicarceplex 14»guest formation. To reiterate from Chapter 1, the GDS is the step in the hemicarceplex reaction where the guest can no longer exchange under the reaction conditions. Therefore, the transition state for the GDS is lower in energy in the presence of /7-xylene than in the presence of NFP, such that that step is 950 times faster in the presence of p-xylene. The template ratios in Table 2.6 and the comparisons between successful and unsuccessful guests also demonstrate interesting trends for template effects upon hemicarceplex 14 formation. 52 ii. Corrections to the hemicarceplex 14 •guest yields Due to the presence of impurities such as hemicarceplex 14«NFP for the 10 hemicarceplex 14»guest: hemicarceplex 14VNFP mixtures outlined previously, and hemicarcerand 14 for all hemicarceplexes 14«guest except hemicarceplex 14«DMSO, the recorded yields had to be corrected." So, for example, comparing the integration of unique guest to host peaks showed 95% hemicarceplex 14«benzene. Multiplying this value by the yield recorded (11.5%) resulted in the 10.9% yield outlined in the experimental and Table 2.6. Thus, these yields are not of isolated products. An identical procedure was utilized for all the remaining hemicarceplex 14«guest yields shown in Table 2.6 and for all hemicarceplex 14»guest mixtures in Tables 2.1 and 2.2. /'//. Template trends-Size From looking at the unsuccessful guests in Table 2.4 we see that the very small molecules such as acetonitrile, methanol, and ethanol all contain two or three non-hydrogen atoms. The smallest molecules to successfully template hemicarceplex 14 formation were DMSO and 2-propanol, each containing 4 non-hydrogen atoms. While these small molecules can escape the interior of the hemicarceplex with ease on the basis of CPK model examination, they were screened in the hope that a combination of two or more of these molecules would have an observable template effect. Unfortunately, we did not observe incarceration of two or more molecules of C H 3 C N , C H 3 O H or CH3CH2OH. a see experimental for exact % hemicarceplex 14«guest produced. 53 The sensitive nature of hemicarceplex 14 toward guest size can also be observed in its selectivity toward D M A over D M F and ethyl acetate over methyl acetate. The addition of one methyl or methylene group turned unsuccessful molecules (i.e. D M F and MeOAc) into moderately good templates (i.e. D M A and EtOAc). However, increasing the size of D M A and D M F by two methylene groups each to give diethylacetamide and diethylformamide as templates, failed. Increasing the size of ethyl acetate by one non-hydrogen atom using isopropyl acetate demonstrated a 6-fold decrease in template effect and attempts to template the formation of hemicarceplex 14 using l-acetyl-3-methylpiperidine, a larger version of NFP, also failed. These examples show that even though large molecules can fit the cavity by C P K modeling, the use of a larger template does not necessarily mean a parallel increase in template effects, even with more potential van der Waals interactions between host and guest in the transition state of the GDS of the hemicarceplex. The remarkable sensitivity of the cavity can also be observed in the preference of the hemicarceplex toward benzene over hexafluorobenzene. The hemicarceplex is possibly sensitive to the substitution of a hydrogen (van der Waals radius = 1.20 A) for a fluorine (van der Waals radius = 1.47 A).13 Pyridine was also an unsuccessful template apparently due to the substitution of a nitrogen for a C-H, and furan was unsuccessful whereas tetrahydrofuran (THF) was a modest template. The extra hydrogen atoms may allow THF and benzene to have more favorable van der Waals interactions with the developing cavity of the hemicarceplex. The template variations observed for the above two examples could also be electronic in nature (CH-TC host-guest interactions). 54 Table 2.6: Template Ratios and Yields for Hemicarceplex 14*guest Guest Structure Template ratio % Yield3 /^ -Xylene H 3 C — ^ ^—CH3 947 5.2%b />-Dibromobenzene Br~jryBr 799 3.9%b /7-Dichlorobenzene C I—^ ^—-CI 728 2.6%b 2-Butanol OH A ,CH} H3C ^ 240 3.6%b Chlorobenzene 205 1.5%b Benzene 0 194 10.9%b Toluene 133 2.7%c Thioxane \> 98 5.2%c Ethyl acetate 0 58 2.5%b Cyclohexane O 36 1.4%° NMP 26 8.8%d DMA 0 A , C H , H 3 C N C H j 22 29.0%d THF (!) 19 1.6%° 2-Propanol A H3C C H 3 16 2.3%° DMSO 0 A H3C CH3 12 4.4%d Isopropyl acetate O CH, X X H 3 C ^ O - ^ C H 3 9 1.3%° NFP 0 1 2.8%d Yield of hemicarceplex in presence of only one guest template. See section 2.2 (d) for more information. 1 mol % guest, 2 days at 80 °C. c 5 mol % guest, 2 days at 80 °C. d Reaction was run in neat guest as solvent. 55 iv. Template trends-Aromatic substitution Aromatic molecules generally behave as better templates than non-aromatics as the best three templates discovered are all para di-substituted benzene species. The more symmetric substitution of aromatics leads to a better template as shown by the fact that /^-xylene is a 7.5-fold better template than toluene and ethylbenzene was completely unsuccessful. Benzene is highly symmetric and it is a slightly better template than toluene, despite its smaller size. Chlorobenzene is slightly more effective than benzene, but the improved templating characteristics of chlorobenzene could be electronic in origin. From Table 2.4 no ortho, meta or tri-substituted benzene rings screened as guests were incarcerated by this templation procedure. The crystal structure for hemicarceplex 14«/?-xylene has been determined by Cram, and it was discovered that each methyl group occupies the highly shielded axial bowl regions while the aromatic ring is centered in the equatorial region of the hemicarceplex (see section 2.2 (f) for further discussion)4 This helps explain that even though 1,3,5 trimethoxybenzene can fit inside the hemicarceplex's interior according to CPK model examination, the introduction of an additional methoxy group causes enough steric strain in the transition state of the GDS such that it cannot template hemicarceplex formation. v. Template trends-Other Successful guests varied greatly in their polarity. Polar molecules such as DMSO and highly apolar molecules such as cyclohexane were found to template the reaction. Guests possessing a variety of functional groups were also successful templates. 56 However, it is impossible to identify trends by comparing template effects between alcohols and esters or amides and esters for example, as the template difference is minimal and in fact could be due to differences in size and shape of guests. vi. Conclusions Hemicarceplex 14«guest formation by using the templating procedure is dependent on the guest effectively lowering the transition state energy of the GDS. This dependency is relatively insensitive to the polarity of the guests screened as both highly polar and apolar guests behaved as effective templates. Size of the guest(s) present had a moderate impression on the template effect, as there appeared to be an optimal size requirement to lowering the transition state energy. The transition state has been shown to be very sensitive to the configuration and symmetry of the potential guest molecules, especially in the aromatic guests, and it was found that para di-substituted benzene rings were the most effective templates. 57 (e) Correlation with Previous Templation Studies i. Template ratios As previously discussed in Chapter 1, for carceplex 11 formation there was a million fold template difference between the best guest (pyrazine) and the worst guest ( N M P ) 9 The slightly more capacious hemicarceplex 13 had a range of template abilities that spanned a factor of 200,000 between the best guest (pyrazine) and the worst guest (NMP). 1 0 These two previous results correlated well, when we take into effect that N M P (arbitrarily assigned a value of 1) is a better template (~10 times) for hemicarceplex 13»guest than for carceplex ll«guest formation. A plot of log hemicarceplex 13 template ratios vs. log carceplex 11 template ratios yielded a straight line correlation coefficient (R) of 0.97, thereby demonstrating that the effects of template molecules on formation of these species is the same.10 These effects include minimizing steric strain and optimizing of the van der Waals interactions in the transition state of the host-guest complexes' guest determining step. Do the template molecules have a similar effect upon the transition state of the GDS of the larger cavity present in the tetramethylene bridged hemicarceplex? An immense difference exists in the guests that act as templates and in the range of template ratios obtained with tetramethylene verses methylene bridges. The range of template abilities was approximately 1000-fold between the best guest (p-xylene) and the worst guest (NFP). Increasing the length of the bridges or linkers between the two bowls from one methylene to four methylene units not only increases the size of the cavity but should also impart more flexibility between the two bowls. 58 However, because a large component of the stabilization of the transition state of the GDS is due to stabilizing van der Waals and electrostatic interactions between host and guest, the increased cavity size of hemicarceplex 14»guest should exhibit similar template effects to those observed in carceplex ll«guest and hemicarceplex 13»guest. In fact, due to the increased size and therefore increased potential van der Waals contacts, the template effects should be even greater. It is believed that the 1000-fold template effect for hemicarceplex 14 is simply because the solvent, NFP, used in our study is in itself a fair guest. For example, i f a solvent were discovered that was 10-fold worse than NFP, a 10 000-fold template effect would be observed for the same range of guests. A solvent that is a very poor template, as in previous templation studies, might demonstrate the million fold template effect expected for formation of hemicarceplex 14«guest. ii. Yields In the previous templation studies, the yields of carceplex ll«guest and hemicarceplex 13«guest generally increased as the template ratio increased, while in our system, the yields were very similar and very poor for all hemicarceplexes 14«guest. Carceplex 13»pyrazine was synthesized in a remarkable 87% yield, while only a 5.2% yield was observed for hemicarceplex 14»/>xylene. Chapman et al. have shown that two molecules of tetrol 7 formed a hydrogen bonded tetrol dimer complex (23) in the presence of base and a suitable template.11 This hydrogen bonded complex effectively preorganizes the guest within the tetrol molecules and the single methylene linker used to form carceplex 11 and hemicarceplex 13 in those 59 previous templation studies easily bridges this preorganized complex. The guests that were most efficient at forming the complex had the highest yields.1 1 Thus, the reason for the low yields in the present system is that due to our larger tetramethylene linker, the formation of a hydrogen bonded complex is not required to form hemicarceplex 14«guest. The complex would have to disassemble before formation of the bridges, so the formation of such a complex inhibits formation of hemicarceplex 14»guest. The yields for hemicarceplex 14«guest did not increase as the template ratio increased because, while good templates have been shown to best form the preorganized complex in carceplex 11, this result is inconsequential as the GDS occurs after any complex formation. Therefore, the yield of any hemicarceplex 14*guest is independent of the guests templating abilities. For example, i f the GDS is formation of the fourth bridge, one can envisage the mono-tetramethylene-bridged intermediate to bind the template molecules poorly and polymerize. Only those few entities that make it to the tris-bridged state will bind the template and seal off to yield hemicarceplex 14»guest. (f) Correlation between Guest Orientation and Changes in Chemical Shifts of Incarcerated and Free Guest Protons in ' H NMR Spectra Table 2.7 lists the chemical shifts of the free and incarcerated guest protons in the *H N M R spectra used in the templation experiments. The difference in these shifts (A8) ranged from 0.23 ppm to 4.40 ppm. A l l protons of the incarcerated guests experienced an upfield shift due to the higher level of shielding supplied by the hemicarceplex. The two hemispheric, or axial regions, of the hemicarceplex's interior provide the highest level of shielding due to the presence of four aromatic rings per hemisphere. The lowest shielding 60 is found in the equatorial region between the bowls. Therefore, it is likely the A8 values should provide us with an estimate to the guests' orientation within hemicarceplex 14's interior. For example, the large A8 of 4.40 ppm for the methyl groups of /^-xylene and the A8 of only 1.17 ppm for the aromatic protons indicate that the methyl groups are thrust deep into the shielding faces of the axial region, while the aromatic ring resides in the equatorial region. This correlates well with the crystal structure of hemicarceplex 14»/?-xylene, previously solved by Cram, 4 which shows each methyl substituent occupying each hemisphere. The rest of this section will focus on the orientation of the guests not previously incarcerated in hemicarceplex 14. Difference in chemical shift values for chlorobenzene and toluene demonstrates that they are in a similar orientation to p-xylene, where the single substituent for each of these guests resides in the axial region. The highest A8 values of the aromatic protons (3.82 for chlorobenzene and 3.73 for toluene) were observed for the proton para to the substituent indicating it is situated in the hemispheric region. This means the substituent would be sitting in the opposite axial region and this is manifested by the high A8 of 4.09 ppm for the methyl group of toluene. The smaller size of the proton para to the methyl substituent allows the methyl to be situated not so deep in the axial region as the methyl groups of/>-xylene. 61 Table 2.7: Chemical Shift (5) of Free and Incarcerated Guest Protons and Differences in the 1 H NMR Spectra of Hemicarceplex 14*guest in CDCI3 At Ambient Temperature Guest 8 free (ppm) 8 bound (ppm) A8 (ppm) /^-xylene H a 2.32 -2.08 4.40 b H b 7.07 5.90 1.17 H3CH^J^-CH3 ^-dibromobenzene H a 7.29 6.40 0.89 B r ^ V -Br /7-dichlorobenzene H a 7.25 6.20 1.05 a CI—<f y—CI 2-butanol(±) Ha 0.86 -2.76 3.62 Hq H b l 1.40 -0.52 1.92 \ /**o H b 2 1.40 -0.61 2.01 A H 3 C X Y C X 3 65 1.71 1.94 / \ C H 3 E 1 .94 1.00 0.94 BIH ''HB2 He 1.11 -2.35 3.46 Chlorobenzene H a 7.30 6.01 1.29 b_a H b 7.25 5.66 1.59 He 7.18 3.36 3.82 0-a Benzene H a 7.37 4.70 2.67 a Toluene H a 2.30 -1.79 4.09 c b H b 7.07 5.83 1.24 /T\ a He 7.15 5.65 1.50 d <' V - C H 3 1 3 Ha 7.05 3.32 3.73 Thioxane H a 2.60 0.14 2.46 a b H b 3.88 1.54 2.34 62 Ethyl acetate A H3C O CH 3 Cyclohexane H a H b He Ha 1.25 4.12 2.04 1.44 -1.97 2.28 -2.19 -0.57 3.22 1.84 4.23 2.01 N M P cm D H a H b He Hd 2.22 1.89 3.25 2.70 1.99 -0.77 -0.59 -0.88 0.23 2.66 3.84 3.58 D M A 0 a H 3C N C H 3 b T H F 2-propanol (±) d HO H b c A . a H 3C CH 3 H a H b He H a H b H a Hb He H , 3.02 2.94 2.08 1.85 3.75 1.13 3.94 1.13 2.14 -0.42 1.61 -1.64 -0.23 1.38 -1.63 2.10 -1.63 3.44 1.33 3.72 2.08 2.37 2.76 1.84 2.76 DMSO o A a H 3 C C H3 Ha 2.46 -0.49 2.95 Isopropylacetate a O C H 3 dH r A A b H 3C O c Cr+3 H a Hb He Hd 1.20 1.20 4.95 1.98 -0.42 -0.42 3.31 -2.38 1.62 1.62 1.64 4.36 1.33 1.38 1.49 3.27 3.12 7.80 -1.22 -0.03 0.18 2.09 2.04 4.47 2.55 1.41 1.31 1.18 1.08 3.33 63 Isopropyl acetate, a large guest with 7 non-hydrogen atoms, can be situated in two different orientations within hemicarceplex 14 by CPK model examination. This long guest could lie in the equatorial region perpendicular to the polar axis of the host, with the methyl adjacent to the carbonyl protruding through the bridging groups. However, the high A5 of 4.36 ppm for that same methyl group indicates it is thrust deep into the hemispheric region, and CPK model examination shows that the propyl methyl groups reside in the less shielded equatorial region. Supporting this interpretation is that the A8 for the propyl methyl's is only 1.62 ppm, indicating low shielding, and the propyl's two methyl groups have identical chemical shift meaning they are rotating rapidly on the J H N M R time scale. From model examination, free rotation of the two methyl groups cannot occur if they were pushed into the hemispheric region instead of the methyl adjacent to the carbonyl. The J H N M R spectra of hemicarceplex 14»guest containing linear like-ended guests such as 2-propanol and DMSO show the presence of only one proton signal which mean that at ambient temperature, these guests can rotate freely on the N M R time scale and are not impeded by the host's interior. Six-membered ring guests like thioxane and cyclohexane, when free in solution, can ring flip rapidly at 25 °C, so the axial and equatorial protons are equivalent on the N M R time scale. Incarceration of these two guests in hemicarceplex 14 did not allow differentiation between axial and equatorial protons, as only one singlet for cyclohexane and only two singlets for the two types of protons in 1,4 thioxane, were observed in the *H NMR. This shows that these molecules can freely ring flip and tumble about the interior of hemicarceplex 14. 64 One linear guest whose rotation may be influenced upon incarceration was 2-butanol. The high shielding of the HA'S and HE'S of the methyl groups, shown by the high A8 values of 3.62 ppm and 3.46 ppm respectively, indicates that 2-butanol is orientated lengthwise, parallel to the polar axis of hemicarceplex 14. CPK model examination of this configuration shows that the hydroxyl of 2-butanol is in a position near an unshared electron pair of an oxygen of the [0(CH2)40]4 bridges. This may allow formation of a hydrogen bond, restraining the motion of 2-butanol within the interior. A hydrogen bond during the transition state of the GDS would have a positive stabilizing effect and may help in preorganizing the shell closure. This would account for the high templating effects of this guest, as it is our best acyclic guest to date, and even surpasses aromatics (generally better templates) like benzene and toluene in its templating ability. 2.3 Conclusions and Future Studies In Chapter 2, we investigated the effect of a template on formation of hemicarceplex 14«guest and the results were compared to the previously reported carceplex ll»guest and hemicarceplex 13«guest systems. The template effect for hemicarceplex 14 was on the order of 1000-fold from the worst guest (NFP) to the best guest (^-xylene). This effect is a combination of optimized van der Waals interactions and minimized steric strain in the host-guest transition state of the GDS. The transition state of the GDS for hemicarceplex 14 demonstrated a sensitivity to guest size and symmetry of the aromatic substitution. Para di-substituted benzene's had the most profound effect on the energy of 65 the transition state due to their high complementarity to the interior. We found that the increased size of the interior of hemicarceplex 14, relative to carceplex 11 and hemicarceplex 13, yielded a difference in guest selectivity. An approximate 1000 times decrease in template effect was also observed, but is likely due to the fair templating ability of the solvent used in the template studies. Since the formation of the preorganized tetrol complex had no implications on the shell closure of hemicarceplex 14»guest, the yields for all guests studied were very low compared to the high yields observed for formation of carceplex ll«guest and hemicarceplex 13»guest. Immediate goals for the future of this project include the continued screening of guest molecules as there still might be a certain structure or configuration of guest that might yield a template effect greater than the 1000-fold effect observed in this study. Ongoing research of a suitable solvent, other than NFP, that will provide modest yields and yet be an unsuccessful template for the formation of hemicarceplex 14 is currently underway. If such a solvent were discovered, and for example, it was 1000 times worse at templating the reaction than NFP, this new solvent would be arbitrarily given a value of 1 and the template ratios listed in Table 2.6 would conversely increase 1000-fold. A longer term goal would be to determine the GDS for hemicarceplex 14«guest formation. Previous studies on carceplex ll«guest determined that the GDS for this system was the formation of the second bridge, either adjacent to or directly across from the first connected bridge. This could be achieved by isolating or directly synthesizing the intermediates involved in hemicarceplex 14 formation, such as the mono-bridged, the two possible bis-bridged and tris bridged species. These species could then be subjected to binding studies with a template such asp-xylene and analyzed for guest release. The 66 species in which no guest escape is observed represents the GDS for the hemicarceplex 14 formation. The template effect observed is powerful enough to preorganize the entropicaly disfavored assembly of seven molecules, two tetrols, four linkers and one guest, to form hemicarceplex 14»guest. This power of preorganization is dependent on optimal van der Waals interactions between host and guest in the rate determining step and one must keep this knowledge in hand when attempting to synthesize larger and more complex self-assembling systems. 67 2.4 Experimental (a) General N-Formylpiperidine (NFP), N-methylpyrrolidinone (NMP), and dimethylacetamide (DMA) were stirred over BaO for 24 hours, distilled under reduced pressure and stored over 4 A molecular sieves under N 2 prior to use. Dimethyl sulfoxide (DMSO) and chlorobenzene were distilled under reduced pressure and stored over 4 A molecular sieves under N 2 prior to use. Isopropyl acetate was distilled and stored over 4 A molecular sieves, and benzene was stored over Na wire under N 2 prior to use. A l l other reagents were commercially available at >98% purity and were used without further purification. Silica gel (BDH, 230-400 mesh) was used for column chromatography. Silica gel thin-layer chromatography was performed on glass-backed plates (Aldrich, silica gel 60, F254, 0.25mm). Matrix assisted laser desorption ionization (MALDI) mass spectra were recorded on a Bruker Biflex II in reflectron mode. M A L D I mass spectrometry samples were prepared by mixing 5ul of a 0.85 u M sample in C H C I 3 with 5ul of a 50mM dihydroxybenzoic acid (DHB) matrix in THF. 0.5 u.1 of this mixture was applied to a M A L D I target disc and allowed to dry prior to running the mass spectrum. Bovine insulin protein standard (molecular weight = 5733.55 g/mol) was used to calibrate the spectrometer prior to use. For the molecular weight range of the samples reported here (2250-2500 g/mol) the errors in M A L D I are assumed to be ± 2 m/z of the calibrated mass. The molecular weights are averaged and not exact due to limited resolution of the M A L D I mass spectrometry. ! H N M R spectra were run on a Bruker WH-400 68 spectrometer at ambient temperature in CDCI3 using the residual signal (8 = 7.24 ppm) as a reference. The N M R spectra of the following hemicarceplexes 14»guest were assigned based on COSY spectroscopy: hemicarceplex 14«THF, hemicarceplex 14»(+) 2-butanol, hemicarceplex 14»thioxane, hemicarceplex 14«chlorobenzene, hemicarceplex 14«2-propanol and hemicarceplex 14»toluene. A l l new hemicarceplexes 14«guest were characterized as mixtures due to the presence of hemicarceplex 14«NFP and/or hemicarcerand 14 in the sample. Therefore, elemental analyses and melting points were not performed on the new hemicarceplex 14»guest compounds synthesized in this study. Further work is underway to fully characterize these compounds. The percentage of the desired hemicarceplex 14»guest obtained is outlined in the following experimental and the yields have been corrected as discussed in section 2.2 (d) of this thesis. 69 Figure 2.3: Tetramethylene bridged hemicarceplex NMR assignment structure 70 (b) Tetramethylene Bridged Hemicarceplex Synthesis Hemicarceplex 14«/>-xylene Procedure " A " : To 40 rnL (361mmol) of N-formylpiperidine (NFP) were added 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g (10.4 mmol) of Cs 2 C0 3 ,126 mg (0.76 mmol) of KI , and 442 uL (3.6 mmol) of /^-xylene. The reaction was stirred at 80 °C under N 2 for 2 days. The solvent was removed in vacuo and 20 mL of 2 M HC1 was added to acidify the crude mixture and dissolve the remaining Cs2CC>3. This mixture was extracted with CHCI3 (2 x 20 mL) and the combined organic extracts were washed with saturated NaHC03 (40 mL), saturated NaCl (40 mL) and dried over anhydrous MgSC*4. The crude oil was dissolved in CH 2 C1 2 , silica gel was added, and the solvent was removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column and eluted with CH 2 C1 2 . Recrystallization from CH 2 C1 2 -hexanes and drying at 80 °C (0.1 mm Hg) for 24 hours gave 80%111 hemicarceplex 14»/>-xylene (6.0 mg, 5.2%) as a white solid. This compound gave an identical *H N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2379 (M«p-xylene + Na + ; 100). Calcd for Ci44Hi36024»C6H4(CH3)2 + N a + : 2380. m The % listed before hemicarceplex 14«guest is the % of that hemicarceplex out of the total hemicarceplexes formed in the reaction, determined by 'H NMR integration. Remaining % is hemicarcerand 14 and hemicarceplex 14»NFP. The % listed in brackets is the corrected yield of hemicarceplex 14»guest (see section 2.2 (d)). 71 Hemicarceplex 14*p-dibromobenzene Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of CS2CO3 (10.4 mmol), 126 mg of KI (0.76 mmol), 0.86 g (3.6 mmol) of /?-dibromobenzene in 40 mL of NFP yielding 98% hemicarceplex 14«/?-dibromobenzene (4.8 mg, 3.9%) as a white solid. This compound gave an identical ' H N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2507 (M«/>dibromobenzene + Na + ; 100). Calcd for Ci44Hi36024«C6H4Br2 +Na + : 2509. Hemicarceplex 14*p-dichlorobenzene Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (l.OOmmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 0.54 g (3.6 mmol) of/>dichlorobenzene in 40 mL of NFP yielding 91% hemicarceplex 14«/>dichlorobenzene (3.1 mg, 2.6%) as a white solid. 72 *H NMR (CDCb, 400 MHz): 5 7.20 (m, 24H, C H j C H z C ^ ) , 7.16 (s, 16H, C H J C H J C ^ ) , 6.84 (s, 8H, A r H of hemisphere), 6.20 (s, 4H, C ^ C 1 2 ) , 5.67 (d, J= 6.8 Hz, 8H, outer OCH 2 0) , 4.84 (t,J= 7.8 Hz, 8H, methine), 4.06 (d, J = 6.8 Hz, 8H, inner OCH 2 0) , 3.90 (br s, 16H, OCH2CH2), 2.70 (m, 16H, CH 2 C# 2 C6H 5 ) , 2.50 (m, 16H, C//2CH2C6H5), 1.87 (br s, 16H, OCH 2C#2). MS (MALDI) m/z (rel intensity): 2419 (M»/?-dichlorobenzene + Na + , 100). Calcd for Ci44Hi36024»C6H4Cl2 + N a + : 2421. Hemicarceplex 14«(±) 2-butanol Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 pL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 330 uL (3.6 mmol) of 2-butanol in 40 mL of NFP yielding 67% hemicarceplex 14»2-butanol (4.2 mg, 3.6%) as a white solid. *H NMR (CDCI3,400 MHz): 6 7.20 (m, 24H, CH 2 CH 2 C6 / /5), 7.14 (s, 16H, CH 2 CH 2 C6#5), 6.81 (s, 8H, A r H of hemisphere), 5.81 (d, J= 7.1 Hz, 8H, outer OCH 2 0) , 4.81 (t, J= 7.4 Hz, 8H, methine), 4.19 (d, J= 7.1 Hz, 8H, inner OCH 2 0) , 3.91 (br s, 16H, OC# 2 CH 2 ) , 2.67 (m, 16H, CH 2 C#2C 6 H 5 ) , 2.47 (m, 16H, C/ / 2 CH 2 C 6 H5) , 1.99 (br s, 16H, O C H 2 C # 2 ) , 1.71 (br m, 1H, H c ) , 1.00 (br, 1H, H D ) , -0.52 (br m, 1H, H B i ) , -0.61 (br m, 1H, H B 2 ) , -2.35 (d, J= 5.8 Hz, 3H, H E ) , -2.76 (t, J - 6.9 Hz, 3H, H A ) . 73 MS (MALDI) m/z (rel intensity): 2348 (M»2-butanol + Na + ; 100). Calcd for Ci44Hi36O24«C4H10O + N a + : 2348. Hemicarceplex 14*chlorobenzene Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg K I (0.76 mmol), 330 uL (3.6 mmol) of chlorobenzene in 40 mL of NFP yielding 69% hemicarceplex 14«chlorobenzene (1.8 mg, 1.5%) as a white solid. *H NMR (CDC13,400 MHz): 8 7.21 (m, 24H, C H j C H j C ^ ) , 7.16 (s, 16H, CH 2 CH 2 C6#5), 6.90 (s, 8H, A r H of hemisphere), 6.01 (d, J= 7.8 Hz, 2H, H A ) , 5.66 (m, 10H, outer O C H 2 0 and H B ) , 4.83 (t, J= 7.1 Hz, 8H, methine), 4.11 (d, J = 7.0 Hz, 8H, inner OCH 2 0) , 3.87 (br s, 16H, OC# 2 CH 2 ) , 3.36 (t, J = 7.4 Hz, 1H, H c ) , 2.68 (m, 16H, CH 2C#2C6H 5), 2.51 (m, 16H, C# 2 CH 2 C6H 5 ) , 1.90 (br s, 16H, OCH 2 C# 2 ) . MS (MALDI) m/z (rel intensity): 2363 (M»chlorobenzene; 100). Calcd for Ci44Hi36024«C6H5Cl: 2363. A H 3C ,C. CH 3 E M B2 CI 74 Hemicarceplex 14* benzene Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 322 uL (3.6 mmol) of benzene in 40 mL of NFP yielding 95% hemicarceplex 14«benzene (12.4 mg, 10.9 %) as a white solid. This compound gave an identical lH N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2350 (M»benzene + Na + ; 100). Calcd for Ci44Hi36024»C6H6 + N a + : 2352. Hemicarceplex 14*toIuene Procedure "B": This procedure is similar to Procedure " A " , but 5 mol % guest to solvent was used. To 40 mL (361 mmol) of N-formylpiperidine (NFP) were added 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g (10.4 mmol) of C s 2 C 0 3 ) 126 mg (0.76 mmol) of KI and 1.92 mL (18 mmol) of toluene. Hemicarceplex 14«toluene (83%) was recovered as a white solid (3.1 mg, 2.7 %). *H NMR (CDC13,400 MHz): 8 7.21 (m, 24H, CH 2 CH 2 C6#5), 7.16 (s, 16H, CH 2 CH 2 C6#5), 6.91 (s, 8H, A r H of hemisphere), 5.83 (d, J= 6.4 Hz, 2H, HB), 5.65 (m, 10H, outer O C H 2 0 and He), 4.84 (t, J= 7.4 Hz, 8H, methine), 4.12 (d, J= 7.2 Hz, 8H, inner OCH 2 0) , 3.32 (br, 1H, H D ) , 3.85 (br s, 16H, OCi7 2 CH 2 ) , 2.69 (m, 16H, 75 CH 2 C# 2 C 6 H 5 ) , 2.51 (m, 16H, C#2CH 2 C 6 H 5 ) , 1.90 (br s, 16H, OCH 2 C# 2 ) , -1.79 (s, 3H, C6H5C//?). MS (MALDI) m/z (rel intensity): 2387 (M«NFP + Na + ; 100), 2365 (M«toluene + Na + ; 93). Calcd for Ci44Hi36024»C6H5CH3 + N a + : 2366. Hemicarceplex 14*1, 4 thioxane Procedure " B " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 1.68 mL (18 mmol) 1,4 thioxane in 40 mL of NFP yielding 75% hemicarceplex 14*1,4-thioxane (3.0 mg, 5.2 %) as a white solid. *H NMR (CDC13,400 MHz): 8 7.20 (m, 24H, CH 2 CH 2 C6# 5 ) , 7.14 (s, 16H, CH 2 CH 2 C6# 5 ) , 682 (s, 8H, A r H of hemisphere), 5.79 (d, J= 7.1 Hz, 8H, outer OCH 2 0) , 4.81 (t, J = 7.9 Hz, 8H, methine), 4.33 (d, J= 7.1 Hz, 8H, inner OCH 2 0) , 3.92 (br s, 16H, OC# 2 CH 2 ) , 2.66 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.47 (m, 16H, C#2CH 2 C 6 H 5 ) , 1.98 (br s, 16H, OCH 2 C/f 2 ) , 1.54 (br s, 4H, QZ/sOS), 0.14 (br s, 4H, C4#sOS). CH 76 MS (MALDI) m/z (rel intensity): 2353 (M»thioxane, 100). Calcd for Ci44Hi36024*C4H80S '. 2355. Hemicarceplex 14«ethyl acetate Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 354 uL (3.6 mmol) ethyl acetate in 40 mL of NFP yielding 18 % hemicarceplex 14«ethyl acetate (2.9 mg, 2.5 %) as a white solid. This compound gave an identical J H N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2362 (M«ethyl acetate + Na + ; 100). Calcd for Ci44H136024»C4H802 + N a + : 2362. Hemicarceplex 14*cyclohexane Procedure " A " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 pL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 388 uL (3.6 mmol) cyclohexane in 40 mL of NFP yielding 39 % hemicarceplex 14»cyclohexane (1.6 mg, 1.4 %) as a white solid. 77 *H NMR (CDCb, 400 MHz): 8 7.20 (m, 24H, CH 2 CH 2 C6 / /5), 7.15 (s, 16H, CH 2CH 2C6#j), 6.84 (s, 8H, A r H of hemisphere), 5.79 (d,J= 6.9 Hz, 8H, outer OCH 2 0) , 4.81 (t, J= 7.9 Hz, 8H, methine), 4.17 (d, J= 6.9 Hz, 8H, inner OCH 2 0) , 3.90 (br s, 16H, OC# 2 CH 2 ) , 2.67 (m, 16H, CH 2 C# 2 C 6 H 5 ) , 2.48 (m, 16H, C # 2 C H 2 C 6 H 5 ) , 1.92 (br s, 16H, OCH 2 C# 2 ) , -0.57 (br, 12H, CsH6). MS (MALDI) m/z (rel intensity): 2387 (M«NFP + Na + ; 100), 2335 (M« cyclohexane; 74). Calcd for Ci44H ] 36024»C6Hi2 : 2335. Hemicarceplex 14»NMP Procedure "C": This procedure is similar to Procedure " A " , but where the reaction was run in neat guest as solvent. To 40 mL (417 mmol) of N-methylpyrrolidinone (NMP) were added 100 mg (0.098 mmol) of tetrol 7, 120 uL (100 mmol) of 1,4 dibromobutane, 3.4 g (10.4 mmol) of C s 2 C 0 3 and 126 mg (0.76 mmol) of KI. Hemicarceplex 14»NMP (80 %) was recovered as a white solid (9.8 mg, 8.8 %). X H NMR (CDCI3,400 MHz): 8 7.20 (m, 24H, CH 2CH 2C6r7 5), 7.14 (s, 16H, CH 2 CH 2 C6#5), 6.83 (s, 8H, A r H of hemisphere), 5.78 (d,J= 7.4 Hz, 8H, outer OCH 2 0) , 4.82 (t, / = 7.9 Hz, 8H, methine), 4.30 (d, J = 7.4 Hz, 8H, inner OCH 2 0) , 3.94 (br s, 16H, OCrY 2CH 2), 2.68 (m, 16H, C H 2 C / / 2 C 6 H 5 ) , 2.48 (m, 16H, C # 2 C H 2 C 6 H 5 ) , 1.99 (br, 2H, H A ) , 1.97 (br s, 16H, OCH 2 C# 2 ) , -0.59 (br m, 2H, H c ) -0.77 (br m, 2H, HB), -0.88 (s, 3H, H D ) . 78 MS (MALDI) m/z (rel intensity): 2372 (MVNMP + Na + ; 100). Calcd for C ^ o e O j ^ C s H g O N + N a + : 2373. Hemicarceplex 14«DMA Procedure " C " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol) in 40 mL D M A (430 mmol) yielding 100% hemicarceplex 14-DMA (33.3 mg, 29 %) as a white solid. This compound gave an identical X H N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2362 (M»DMA + Na + ; 100). Calcd for Ci44Hi36024»C4H9ON + N a + : 2361. 79 Hemicarceplex 14«THF Procedure " B " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 u.L (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 1.46 mL (18 mmol) of THF in 40 mL of NFP yielding 27 % hemicarceplex 14»THF (1.8 mg, 1.6 %) as a white solid. *H NMR (CDCI3,400 MHz): 8 7.20 (m, 24H, CH 2 CH 2 C6#5), 7.15 (s, 16H, CH 2 CH 2 C6#5), 6.83 (s, 8H, A r H of hemisphere), 5.82 (d,J= 7.4 Hz, 8H, outer OCH 2 0) , 4.81 (t, J = 7.9 Hz, 8H, methine), 4.10 (d, J = 7.4 Hz, 8H, inner OCH 2 0) , 3.88 (br s, 16H, OC# 2 CH 2 ) , 2.67 (m, 16H, CH 2 Cr7 2 C 6 H 5 ) , 2.48 (m, 16H, C # 2 C H 2 C 6 H 5 ) , 1.97 (br s, 16H, OCH 2 C# 2 ) , 138 (br s, 4H, H A ) , -0.23 (br s, 4H, H B ) . MS (MALDI) m/z (rel intensity): 2387 (M«NFP + Na + ; 100), 2346 (M»THF + Na + ; 55) Calcd for Ci44Hi3602 4»C4H80 + N a + : 2346. B Hemicarceplex 14*2-propanol Procedure " B " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 80 1.38 mL (18 mmol) of 2-propanol in 40 mL of NFP yielding 29 % hemicarceplex 14*2-propanol (2.6 mg, 2.3 %) as a white solid. *H NMR (CDCb, 400 MHz): 8 7.20 (m, 24H, C H J C H J C ^ X 7.15 (s, 16H, CH 2 CH 2 C6#5), 6.83 (s, 8H, A r H of hemisphere), 5.82 (d, J = 6.9 Hz, 8H, outer OCH 2 0) , 4.81 (t, J= 7.9 Hz, 8H, methine), 4.17 (d, J= 6.9 Hz, 8H, inner OCH 2 0) , 3.94 (br s, 16H, OC# 2 CH 2 ) , 2.68 (m, 16H, CH 2 C# 2 C 6 H 5 ) , 2.48 (m, 16H, C / / 2 C H 2 C 6 H 5 ) , 1.98 (br s, 16H, OCH 2 C# 2 ) , -1.63 (d,J= 5.9 Hz, 6H, C T ^ C H O H C ^ ) . MS (MALDI) m/z (rel intensity): 2385 (M«NFP + Na + ; 100), 2331 (M»2-propanol + Na + ; 51). Calcd for Ci 44Hi360 24»C 3H 80 + N a + : 2334. Hemicarceplex 14«DMSO Procedure " C " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol) in 40 mL of DMSO (565 mmol) yielding 100% hemicarceplex 14»DMSO (5.0 mg, 4.4 %) as a white solid. This compound gave an identical *H N M R spectrum to that previously reported.4 MS (MALDI) m/z (rel intensity): 2352 (M-DMSO + Na + ; 100). Calcd for Ci44Hi36024»(CH3)2SO + N a + : 2352. 81 Hemicarceplex 14«isopropylacetate Procedure " B " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of C s 2 C 0 3 (10.4 mmol), 126 mg KI (0.76 mmol), 2.11 mL (18 mmol) of isopropylacetate in 40 mL of NFP yielding 10 % hemicarceplex 14»isopropylacetate (1.5 mg, 1.3 %) as a white solid. *H NMR (CDCfe, 400 MHz): 8 7.20 (m, 24H, CH2CH2C6//5), 7.15 (s, 16H, CH2CH2C6//5), 6.81 (s, 8H, A r H of hemisphere), 5.68 (d, J= 6.5 Hz, 8H, outer OCH 2 0) , 4.89 (t, J= 8.1 Hz, 8H, methine), 4.17 (d, J= 6.5 Hz, 8H, inner OCH 2 0) , 3.99 (br s, 16H, OCH2CR2), 3.31 (m, 1H, CH 3COOC Jr7(CH 3) 2), 2.68 (m, 16H, CH 2 C//2C 6 H 5 ) , 2.48 (m, 16H, C# 2 CH 2 C 6 H 5 ) , 1.92 (br s, 16H, OCH 2 C/ / 2 ) , -0.42 (d, J = 6.0 Hz, 6H, CH 3COOCH(C//3)2), -2.38 (s, 3H, C// iCOOCH(CH 3 ) 2 ) . MS (MALDI) m/z (rel intensity): 2388 (MVNFP + Na + ; 100), 2374 (M«isopropylacetate + Na + ; 42). Calcd for C ^ n s C ^ C j H w C ^ + N a + : 2376. Hemicarceplex 14»NFP Procedure " C " was applied using 100 mg (0.098 mmol) of tetrol 7, 120 uL (1.00 mmol) of 1,4 dibromobutane, 3.4 g of Cs2C0 3 (10.4 mmol), 126 mg KI (0.76 mmol) in 40 mL of NFP (361 mmol) yielding 84 % hemicarceplex 14»NFP (3.2 mg, 2.8 %) as a white solid. 82 *H NMR (CDCb, 400 MHz): 8 7.20 (m, 24H, CH 2 CH 2 C6#5), 7.15 (s, 16H, C H 2 C H 2 C 6 ^ J ) , 6.84 (s, 8H, A r H of hemisphere), 5.76 (d, J = 6.8 Hz, 8H, outer OCH 2 0) , 4.81 (t, J= 7.9 Hz, 8H, methine), 4.47 (s, 1H, H F ) , 4.17 (d, J= 6.8 Hz, 8H, inner OCH 2 0) , 3.93 (br s, 16H, OC# 2 CH 2 ) , 2.67 (m, 16H, CH 2 C# 2 C 6 H 5 ) , 2.48 (m, 16H, C i / 2 C H 2 C 6 H 5 ) , 2.09 (m, 2H, H D ) , 2.04 (m, 2H, H E ) , 1.92 (br s, 16H, OCH 2 C# 2 ) , 0.18 (br, 2H, He), -0.03 (br, 2H, H B ) , -1.22 (br, 2H, H A ) . MS (MALDI) m/z (rel intensity): 2387 (MVNFP + Na + ; 100). Calcd for Ci44Hi360 24*C 6Hi 1ON + N a + : 2387. (c) Competition Experiments Table 2.6 reports the template ratios obtained directly from competition experiments. The typical competition reaction was run according to the following procedure: To 20 mL of NFP (180 mmol) were added 50 mg (0.049 mmol) tetrol 7, 60 ul 1,4 dibromobutane (0.50 mmol), 1.7 g Cs 2 C03 (5.2 mmol), 63 mg KI (0.38 mmol) and guest 1 and guest 2. The relative ratios of the competing guests were chosen as to reflect a 1:1 ratio in the N M R spectrum. The product ratios for the hemicarceplex»guest mixture were calculated from the *H N M R spectrum by integration and comparison of the individual guest peaks. The template ratios obtained directly from competition reactions are listed in Table 2.8. 83 Table 2.8: Competition Results Template Ratio Hemicarceplex 14»Guest 1 Hemicarceplex 14*Guest 2 Guest 1 Guest 2 p-xylene p-dibromobenzene 1.23 1.00 p-xylene p-dichlorobenzene 1.26 1.00 p-xylene Benzene 4.90 1.00 p-dibromobenzene Benzene 4.30 1.00 p-dichlorobenzene Benzene 3.61 1.00 2-butanol Chlorobenzene 1.19 1.00 2-butanol Benzene 1.28 1.00 chlorobenzene Toluene 1.70 1.00 chlorobenzene ethyl acetate 4.46 1.00 toluene Thioxane 1.30 1.00 toluene ethyl acetate 2.60 1.00 toluene Cyclohexane 3.80 1.00 ethyl acetate Cyclohexane 1.80 1.00 N M P Isopropyl acetate 3.00 1.00 cyclohexane D M A 1.75 1.00 N M P D M A 1.20 1.00 D M A THF 1.12 1.00 D M A DMSO 2.10 1.00 THF 2-propanol 1.07 1.00 2-propanol NFP 16.0 1.00 DMSO isopropyl acetate 1.37 1.00 DMSO NFP 11.5 1.00 isopropyl acetate NFP 8.60 1.00 A sample calculation on how Table 2.6 was constructed follows: Template ratio of DMSO : isopropyl acetate = 1.37:1.00 Template ratio of isopropyl acetate : NFP = 8.60:1.00 The template ratio of isopropyl acetate : NFP is then calculated: 1.37/1.00X8.60/1.00= 11.8 84 (d) Crosscheck Experiments Crosscheck experiments were carried out to double-check the accuracy of the template ratios outlined in Table 2.6. The crosschecks were performed by competing non-adjacent guests from Table 2.6. For instance, a competition experiment was run starting with a 10:1 ratio of thioxane: /^-xylene, which, after adjusting for the starting ratio, gave a 1:9.5 ratio of hemicarceplex 14«thioxane : hemicarceplex 14»/?-xylene respectively. This is in excellent agreement with a 1.9.1 ratio obtained from the template ratios in Table 2.6. Continuing, a similar competition between thioxane and D M A resulted in a 1:3.4 ratio of hemicarceplex 14»DMA : hemicarceplex 14»thioxane respectively. This result is in good accord with a 1:4.4 template ratio derived from Table 2.6. Finally, a competition experiment between guests D M A and NFP resulted in a 1:21 ratio for hemicarceplex 14«NFP : hemicarceplex 14»DMA, which again is in good agreement with a 1:22 ratio obtained from Table 2.6. (e) Control Experiments Each hemicarceplex 14«guest was subjected to a control experiment to check for possible guest exchange. The samples used for control experiments were those as prepared in the experimental. A typical control experiment was run according to the following procedures: 85 Procedure "D": The hemicarceplex 14»guests and hemicarceplex 14»guest/NFP mixtures below and including hemicarceplex 14»toluene in Table 2.6 were subjected to competition reaction conditions and 1 mol % each of 2-butanol and benzene. For example: To 4 mL (36 mmol) of NFP were added 10 mg of hemicarceplex 14»DMA, 0.34g (1.04 mmol) C s 2 C 0 3 , 12.6 mg (0.076 mmol) KI , 47.6 mg (0.4 mmol) KBr , 12 uL (0.10 mmol) of 1,4 dibromobutane, 33 uL (0.36 mmol) of 2-butanol and 32 uL (0.36 mmol) of benzene. The reaction was stirred at 80 °C under N 2 for 2 days. The solvent was removed in vacuo and 2 mL of 2 M HC1 was added to acidify the crude mixture and dissolve the remaining CS2CO3. This mixture was extracted with C H C I 3 ( 2 x 4 mL) and the combined organic extracts were dried over anhydrous MgSO/t. The crude oil was dissolved in CH 2 C1 2 , silica gel was added, and the solvent removed in vacuo. The silica gel-absorbed sample was dry loaded onto a silica gel gravity column and eluted with CH 2 C1 2 . Recrystallization from CH2Cl2-hexanes and drying at 80 °C (0.1 mm Hg) for 24 hours gave hemicarceplex 14«DMA (8.8 mg, 88 % recovery) as a white solid. Examination of the lH N M R spectrum showed that only 2.3 % guest exchange had occurred. Procedure "E": This procedure is similar to Procedure " D " but the hemicarceplex 14«guests and hemicarceplex 14«guest/NFP mixtures above and including hemicarceplex 14»benzene in Table 2.6 were subjected to competition reaction conditions and 1 mol % each of the other guests in the top 5 of Table 2.6. For example: To 2 mL (18 mmol) of NFP were added 5 mg of hemicarceplex 14«2-butanol, 0.17 g (0.52 mmol) of C s 2 C 0 3 , 6.3 mg (0.038 mmol) KI, 24 mg (0.20 mmol) KBr , 6 pL (0.050 mmol) of 1,4 86 dibromobutane, 16 uL (0.18 mmol) of benzene, 18 uL (0.18 mmol) of chlorobenzene, 43 mg (0.18 mmol) of/?-dibromobenzene, 27 mg (0.18 mg) of/7-dichlorobenzene and 22 uL (0.18 mmol) of /^-xylene. Purification yielded hemicarceplex 14«2-butanol (4 mg, 80 % recovery) as a white solid. Examination of the ! H N M R spectrum showed that no guest exchange had occurred. The results of the control experiments are summarized in Table 2.5. Comparison of the host to guest integration before and after control experiment, showed that the guest exchange is within the +10 % error for the *H N M R integration. 87 2.5 References 1. Cram, D. J.; Karback, S.; Kim, Y . H . ; Baczynskyj, L . ; Kallemeyn, G. W. J. Am. Chem. Soc. 1985,107, 2575. 2. Helgeson, R. C ; Paek, K. ; Knobler, C. B. ; Maverick, E. F.; Cram, D. J. J. Am. Chem. Soc. 1996, 118, 5590. 3. (a) Quan, M . C ; Knobler, C. B. ; Cram, D. J. J. Chem. Soc. Chem. Commun. 1991, 660. (b) Quan, M . C ; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2754. (c) Judice, J. K . ; Cram, D. J. J. Am. Chem. Soc. 1991,113, 2790. (d) Cram, D. J.; Tanner, M . E.; Knobler, C. B. J. Am. Chem. Soc. 1991,113, 7717. (e) Choi, H . J., Buhring, D.; Quan, M . C ; Knobler, C. B. ; Cram, D.J.J. Chem. Soc. Chem. Commun. 1992, 1733. (f) Cram, D. J.; Blanda, M . T.; Paek, K . ; Knobler, C. B. J. Am. Chem. Soc. 1992,114, 7765. (g) Cram, D. J.; Jaegor, R ; Deshayes, K. J. Am. Chem. Soc. 1993, 115, 10111. (h) Eid, C. N . ; Knobler, C. B ; Gronbeck, D. A. ; Cram, D. J. J. Am. Chem. Soc. 1994,116, 8506. (i) Byun, Y.-S. ; Knobler, C. B ; Robbins, T. A . ; Cram, D. J J. Chem. Soc. Chem. Commun. 1995, 1947. 4. Robbins, T. A. ; Knobler, C. B. ; Bellow, D. R.; Cram, D. J. J. Am. Chem. Soc. 1994, 116, 111. 5. (a) Illuminati, J.; Mandolini, L. Acc. Chem. Res. 1981,14, 95. (b) Illuminati, J.; Mandolini, L . ; Masci, B. J. Am. Chem. Soc. 1981,103, 4142. 6. (a) Kruizinga, W. H ; Kellogg, R. M . J. Am. Chem. Soc. 1981,103, 5183. (b) Piepers, O ; Kellogg, R. M . J. Chem. Soc. Chem. Commun. 1978, 383. 7. Meurer, K ; Luppertz, F.; Vogtle, F. Chem. Ber. 1985,118, 4433. 8. For further examples see Ostrowicki, A. ; Koepp, E. ; Vogtle, F. Top. Curr. Chem. 1991,161, 38 9. (a) Chapman, R. G.; Chopra, N . ; Cochien, E. D.; Sherman, J. C. J. Am. Chem. Soc. 1994,116, 369. (b) Chapman, R. J.; Sherman, J. C. J. Am. Chem. Soc. 1995,117, 9081. 10. Chopra, N . ; Sherman, J. C. Supramol. Chem. 1995, 5, 31. 11. Chapman, R. G. Ph.D. Thesis, The University of British Columbia, 1997, pp. 85. 12. Robbins, T. A. ; Cram, D. J. J. Am. Chem. Soc. 1993,115, 12199. 13. Cary, F. A. ; Sundberg, R. J. Advanced Organic Chemistry, 3rd ed.; Plenum Press: New York, 1990. 88 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0061558/manifest

Comment

Related Items