Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The use of dipyrromethene and polydipyrromethene ligands in supramolecular chemistry Zhang, Yongjun 2000

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

Item Metadata


831-ubc_2000-566544.pdf [ 7.67MB ]
JSON: 831-1.0061441.json
JSON-LD: 831-1.0061441-ld.json
RDF/XML (Pretty): 831-1.0061441-rdf.xml
RDF/JSON: 831-1.0061441-rdf.json
Turtle: 831-1.0061441-turtle.txt
N-Triples: 831-1.0061441-rdf-ntriples.txt
Original Record: 831-1.0061441-source.json
Full Text

Full Text

THE USE OF DIPYRROMETHENE AND POLYDIPYRROMETHENE LIGANDS IN SUPRAMOLECULAR CHEMISTRY By YONGJUN ZHANG B.Sc. Nankai University, China, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY 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 May 2000 © Yongjun Zhang, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract The main objective of this work was to study the use of dipynomethene and poly-dipyrromethene ligands in supramolecular chemistry. Numerous poly-dipyrromethene metal complexes were synthesized via self-assembly and characterized by mass and N M R spectrometry and X-ray crystallography. Some of their structures were simulated using Hyperchem release 5.01 MM+ force field. Methylene and directly linked bis(dipyrromethene) coordinated to metal ions to form complexes 60a, 60b and 59 which are dimer ((ligands^Mo). X-ray structures of these complexes showed them to have helical geometry. Ethylene-linked bis(dipyrromethene) once coordinated to metal ion, gave a mixture of monomer ((ligand)M) 61a and dimer 87 ll 61b ((ligand)2M2). Molecular modeling showed that simulated complex 61a possess a helical double-stranded geometry and 61b a distorted tetrahedral geometry. The methylene-linked tris(dipyrromethene) zinc complex 87 was also synthesized and shown to contain helical double-stranded structure by X-ray crystallography. Dipyrromethenes which contain a raeso-phenyl group and are free at a and (3 positions coordinated to trivalent metal ions forming octahedral complexes 131-135. The oxidation reaction of porphyrins with oxaziridines was studied and this leads to a new synthetic method towards the synthesis of chlorins which are potential photosensitizers in photodynamic therapy. Chlorins 179 and 180 were synthesized and characterized. The longest wavelength absorptions of these chlorins have an approximate M(lll) 131 R=H, M=Co 132 R = N 0 2 , M=Co 133 R = C 0 2 C H 3 , M=Co 134 R=H, M=Fe 135 R = N 0 2 , M=Fe 179 (±) 180 (±) in 20 nm bathochromic shift and a significant increase in intensity as compared with octaethyl porphyrin. Their structures were determined by H and C NMR spectrometry. A mechanistic proposal for this reaction will be presented. iv TABLE OF CONTENTS ABSTRACT ii T A B L E OF CONTENTS v LIST OF TABLES x LIST OF FIGURES xii LIST OF SCHEMES xvi LIST OF ABBREVIATIONS r xviii N O M E N C L A T U R E xxi ACKNOWLEDGEMENTS xxiii THE USE OF DIPYRROMETHENE LIGANDS IN SUPRAMOLECULAR CHEMISTRY Part 1: T H E USE O F D I P Y R R O M E T H E N E L I G A N D S IN S U P R A M O L E C U L A R C H E M I S T R Y 1. I N T R O D U C T I O N 2 1.1 Dipyrromethenes and Polydipyrromethenes as Ligands 2 1.1.1 Dipyrromethenes '. 2 1.1.2 Linear Polydipyrromethenes 3 1.2 Self-Assembly and Supramolecules 11 1.2.1 Definition of Self-Assembly Process 11 1.2.2 Self-Assembly in Nature 12 1.2.3 Self-Assembly in Synthetic Chemistry 14 1.2.4 Mechanism of Self-Assembly ". 16 1.2.5 Thermodynamic Issues of Self-Assembly 17 1.2.6 Features of Self-Assembly 19 1.2.7 Supramolecules, Nanostructures and Self-Assembly 20 1.2.8 Review of Synthesis and Properties of Self-Assembling Supramolecules 22 Hydrophobic interaction self-assembly 22 Hydrogen Bonded Self-Assembly 25 Inorganic Self-Assembly 32 2. R E S U L T S A N D DISCUSSION 43 2.1 Formation and Characterization of Biladiene Metal Complexes 60a and 60b 46 2.1.1 Syntheses of Complexes 60a and 60b 46 2.1.2 Spectroscopic Analysis of Complexes 60a and 60b Optical Spectra 48 2.1.3. Crystal Structure Analysis of Complex 60b 50 2.2 Formation and Characterization of 1,1 '-Bis-(dipyrrinyl) Zinc Complex 59 54 2.2.1 Synthesis of Complex 59 55 2.2.2 Crystal Structure Analysis of Complex 59 58 vi 2.3 Formation of l,2-Bis-(dipyrrin-r-yl)-ethane Cobalt Complex 61 60 2.4 Formation and Characterization of Trinuclear Helicate Complex 87 64 2.5 Formation and Characterization of bis-(l ,3,7,9-Tetramethyl-8-ethyl-dipyrrin-2-yl) Sulfide Zinc(II) Complex 92 70 2.6 Current Development 75 2.6 Summary and Future Work 77 3. E X P E R I M E N T A L ..81 3.1 Preparation 81 3.2 X-ray Crystallographic Analysis of 60b, 59, 87 and 92 90 4. R E F E R E N C E S 110 Part 2: SYNTHESIS , D E R I V A T I Z A T I O N A N D S T R U C T U R A L C H A R A C T E R I Z A T I O N O F O C T A H E D R A L TRIS(5-PHENYL-4,6-D I P Y R R I N A T O ) C O M P L E X E S O F COBALT(III ) A N D IRON(III) 1. I N T R O D U C T I O N 118 1.1 Dipyrromethenes And Their Properties 118 1.2 Preparation And Characterization Of meso-Phenyldipyrromethenes 121 vn 1.3 Formation and Characterization of Divalent Transition Metal Chelates of meso-Phenyldipyrromethenes 124 1.4. Goal of Project 130 2. R E S U L T S A N D DISCUSSION 132 2.1. Formation of meso-Phenyldipyrromethenes 132 2.2 Formation Of Trispyrrinato Complexes Of Co(III) and Fe(III) 132 2.3 Spectroscopic Properties of Tris-Dipyrrinato Complexes of Fe(III) and Co(III) 135 2.4 Crystal Structure of Tris[(5-Phenyl) Dipyrrinato] Co(III)-Acetone 137 2.5 Chemical Transformation of P-Phenyl Substituents of Tris[(5-Phenyl)Dipyrrinato] Co(III) Complexes 139 2.6 Future Perspective..... 142 3. E X P E R I M E N T A L S E C T I O N 145 3.1 Preparations 145 3.2 X-ray Crystallographic Analysis of 131-Acetone 157 4. R E F E R E N C E S 167 Part 3: O X I D A T I O N OF P O R P H Y R I N S W I T H N - S U L F O N Y L O X A Z I R I D I N E S 1. I N T R O D U C T I O N 172 vin 1.1 Introduction of Porphyrin-Related Macrocycles 172 1.2 Optical Absorption Spectra 174 1.3 Photodynamic Therapy (PDT) 177 1.3.1 General Introduction 177 1.3.2 Mechanisms of PDT 180 1.3.3 Desirable Properties For PDT Drug 182 1.4 Chlorins Used as PDT Drugs and Conversion of Porphyrins to Chlorins 184 1.4.1 Osmium Tetraoxide Oxidation of Porphyrins 185 1.4.2 Overview of Oxaziridines 187 1.6 Goal of Project 190 2. R E S U L T S A N D DISCUSSION 192 2.1 Reaction, Isolation and Characterization 192 2.2 Proposed Mechanism 199 2.3. Chemical Properties Of Compounds 179 And 180 201 2.4 Summary 202 3. E X P E R I M E N T A L 203 3.1 Preparations 203 3.2 Supplemental Analytical Data for Compound 179 208 4. R E F E R E N C E S 216 ix LIST OF T A B L E S Table 1-1. Crystallographic Data for 60b 90 Table 1-2. Atomic Coordinates and Beq for 60b 91 Table 1-3. Selected Bond Length (A) for Complex 60b 95 Table 1-4. Selected Bond Angles(°) for 60b 96 Table 1-5. Selected Structural Parameters of X-ray and Simulated Structures of Complex 60b , 52 Table 1-6. Crystal lographic Data for 59 97 Table 1-7. Selected Bond Length (A) for 59 98 Table 1-8. Selected Bond Angles (°) for 59 99 Table 1-9. Selected Structural Parameters of X-ray and Simulated Structures of Complexes 59 60 Table 1-10. Crystallographic Data for 87 100 Table 1-11. Atomic Coordinates and Bei/ for 87 101 Table 1-12. Bond Lengths (A) for 87 104 Table 1-13. Bond Angles (°) for 87 105 Table 1-14. Crystallographic Data for 92 106 Table 1-15. Selected Bond Lengths (A) for 92 107 Table 1-16. Selected Bond Angles (°) for 92 109 Table 2-1. Crystallographic Data for 131-Acetone 159 Table 2-2. Atom Coordinates and Beq. [A^] for Complex 131-Acetone 162 x Table 2-3. Selected Bond Length (A) for Compound 131 Acetone 164 Table 2-4. Selected Bond Angles (°) for Compound 131-Acetone 165 xi LIST OF FIGURES Figure 1-1. Geometric isomers of dipyrromethene 2 Figure 1-2. The helical structure of DNA 13 Figure 1-3. Schematic structure of A, C, D,F-tetra-6-(6-n-butyrylamino-n-hexyl-l-sulfenyl)-p-cyclodextrin 23 Figure 1-4. Self-assembly of a cyclodextrin-based ion channel 23 Figure 1-5. Self-assembly of an oligoether-based ion channel 25 Figure 1-6. Hydrogen bonded self-assembled cyclic aggregates 27 Figure 1-7. Self-assembly of C A M stack 29 Figure 1-8. Self-assembly of a peptide-based ion channel 31 Figure 1-9. Self-assembly of trinuclear homotopic double stranded helicate 39 34 Figure 1-10. Self-assembly of trinuclear homotopic triple stranded helicate 41 35 Figure 1-11. Self-assembly of ladder-like complexes 44 and 45 36 Figure 1-12. Self-assembly of grid-type complex 47 38 Figure 1-13. Self-assembly of bowl-shaped complex 49 39 Figure 1-14. Self-assembly of dendrimeric complex 51 41 Figure 1-15. M 2 + ions give charged species with bipyridines but give neutral species with dipyrromethenes 43 Figure 1-16. UV-visible spectrum of complexes 60a and 60b in CH 2 C1 2 49 xn Figure 1-17. ORTEP representation of complex 60b 51 Figure 1-18. ORTEP representative of complex 59 58 Figure 1-19. Two possible conformations of bis-(dipyrromethene) 60 Figure 1-20. Simulated Models for 61a and 61b 63 Figure 1-21. 'H NMR spectrum (CDC13) of complex 87 67 Figure 1-22. UV-visible spectrum of complex 87 in methylene chloride 68 Figure 1-23. ORTEP representation of complex 87 69 Figure 1-24. ORTEP representation of complex 92 73 Figure 1-25. (3-Linked multi-dipyrromethene ligands 75 Figure 1-26. Stereo view for the X-ray structure of complex 99 76 Figure 2-1. Structure, atom numbering scheme and formal nomenclature of a,a'-dipyrromethene 1 18 Figure 2-2. UV-visible spectra of complexes 124 and 125 123 Figure 2-3. X-ray structure of complex 130 126 Figure 2-4. Limiting resonance forms of dipyrrinato ligands 127 Figure 2-5. UV-visible spectra of complex 127 and 128 128 Figure 2-6. UV-visible spectra of complexes 132 and 135 135 Figure 2-7. X-ray structure of 131 137 Figure 2-8. Model structure for 2 n d generation dendrimer 144 Figure 3-1. Chromophores of macrocyclic tetrapyrroles 172 xiii Figure 3-2. The most important naturally occurring porphinoids 173 Figure 3-3. Four types of porphyrin Q-band absorptions 176 Figure 3-4. Typical optical spectra of chlorins and metallochlorins 177 Figure 3-5. Structures of selected second-generation photosenseitizers 179 Figure 3-6. Modified Jablonski diagram for generation of excited porphyrin states and reactive singlet molecular oxygen 180 Figure 3-7. N-sulfonyl oxazirdines 188 Figure 3-8. Suggested structures of compounds 179 and 180 193 Figure 3-9. UV-visible spectrum of 179, 180 and 181 194 Figure 3-10. l H NMR spectra of compounds 179 and 180 196 Figure 3-11. H M Q C NMR spectrum of compound 179 197 Figure 3-12. 'H COSY spectrum of compound 179 198 Figure 3-13. Proposed mechanism for reaction of OEP with 2-benzylsulfonyl-3-phenyloxaziridine 200 Figure 3-14. The steric interactions within TPP 201 Figure 3-15. I 3 C NMR spectrum of compound 179 208 Figure 3-16. Detailed H Q M C spectra (1) of compound 179 209 Figure 3-17. Detailed HQMC spectra (2) of compound 179 210 Figure 3-18. Detailed H Q M C spectra (3) of compound 179 211 Figure 3-19. Detailed H Q M C spectra (4) of compound 179 212 Figure 3-20. Detailed H Q M C spectra (5) of compound 179 213 xiv Figure 3-21. Detailed H Q M C spectra (6) of compound 179 Figure 3-22. Detailed H Q M C spectra (7) of compound 179 LIST OF SCHEMES Scheme 1-1. Deprotonation and Metallation of Dipyrromethene 3 Scheme 1-2. Preparation of biladienes-ac (Method 1) 4 Scheme 1-3. Preparation of biladienes-ac (Method 2) 5 Scheme 1-4. Preparation of biladienes-ac (Method 3) 6 Scheme 1-5. Products available from biladiene-ac deravitives 7 Scheme 1-6. Synthesis of bis-(dipyrrinyl) dihydrobromide 16. 8 Scheme 1-7. Synthesis of 1,19 dideoxy-b-norbilenes-a 21 9 Schemel-8. Preparation of ammonium carboxylate 26. 24 Scheme 1-9. Self-assembly of peptide rings 30 Scheme 1-10. Syntheses of dinuclear poly dipyrromethene complexes 45 Scheme 1-11. Synthesis of Zn(II) complex 60a and Co(II) Complex 60b 47 Scheme 1-12. H-D exchange at the meso-position of 1,19-dideoxybiladiene-ac 50 Scheme 1-13. Complexation of 1,19-dideoxy-b-norbilane-a 21 with Cu(II) 54 Scheme 1-14. Synthesis of 1,1 '-bis-(dipyrrinyl) zincll complex 59 56 Scheme 1-15. Synthesis of 1,2-bis-(dipyrrin-l '-yl)-ethane Co (II) complex 61 62 Scheme 1-16. Synthesis of trinuclear duplex helicate 87 ..66 Scheme 1-17. Synthetic Route to Bis(l ,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-yl) Sulfide Zinc Complex 92 72 Scheme 2-1. Valence-tautomeric structure of dipyrromethene 119 Scheme 2-2. Preparation of dipyrromethene (Method 1) , 120 xvi Scheme 2-3. Preparation of dipyrromethene (Method 2) 120 Scheme 2-4. Preparation of dipyrromethene (Method 3) 121 Scheme 2-5. Preparation of dipyrromethene (Method 4) 121 Scheme 2-6. Preparation of dipyrromethene (Method 5) 121 Scheme 2-7. Synthesis of meso-phenyldipyrranes and meso-phenyldipynines 122 Scheme 2-8. Formation of>-phenyldipyrrinato complexes from 125 Scheme 2-9. Formation of tris[me.s,o-phenyldipyrronato] complexes 133 Scheme 2-10. Chemical reactions of p-phenyl substituents of dipyrrinato metal complexes 139 Scheme 2-11. Reaction of 137 with benzoic acid under catalytic conditions using D M A P / D C C 140 Scheme 2-12. Reaction pathways in carbodiimide condensation 142 Scheme 3-1. Biosynthesis of Uroporphyrinogen III 174 Scheme 3-2. Reactions between singlet oxygen with some biomolecules 181 Scheme 3-3. Oxidation of OEP by OsOVpyridine 185 Scheme 3-4. Oxidation of TPP with osmium tetraoxide 186 Scheme 3-5. Synthesis of 2-Benzylsulfonyl-3-phenyloxaziridine 189 Scheme 3-6. Oxygen-Transfer Reactions of N-Sulfonyloxazidines 189 Scheme 3-7. Oxidation of OEP with oxaziridine 192 xvii LIST OF ABBREVIATIONS A angstrong um 10"9 m A L A 5-aminolevulinic acid aq. aqueous br broad BPDMA benzoporphyrin derivative monoacid ring A CA cyanuric acid COSY homonuclear correlation spectroscopy d doublet DCC dicyclohexylcarbodiimide DDQ 2,3-dichloro-5,6-dicyano-1,4-benzoquinone DMAP 4-(dimethylamino)-pyridine DMF N,N-dimethylformamide DMSO dimethylsulfoxide DPTS 4-(dimethylamino)-pyridinium /?-toluenesulfonate DNA deoxyribonucleic acid EI electron impact eq. equivalent ESI electrospray ionization FAB fast atom bombardment xviii HOMO highest occupied molecular orbital HMQC heteronuclear multiple quantum correlation spectroscopy HpD haematoporphyrin derivative HR high resolution hr hour Hz hertz ISC intersystem crossing LR low resolution M melamine m multiplet ra-CPBA 3-chloroperoxybenzoic acid min minute MS mass spectrometry NMR nuclear magnetic resonance NOE nuclear Overhauser effect OEP 2,3,7,8,12,13,17,18-octaethylporphyrin ORTEP Oak Ridge thermal ellipsoid plot PBG porphyrinogen PDT photodynamic therapy py pyridine q quartet s singlet Sens sensitizer xix t triplet TFA trifluoroacetic acid THF tetrahydrofuran T L C thin layer chromatography TsOH p-toluenesulfonic acid TPP 5,10,15,20-tetraphenylporphyrin UV ultraviolet xx NOMENCLATURE MONOPYRROLIC SYSTEMS Pyrrole (1) is numbered as shown at right. The Greek letters a and [3 are 3 4 used to distinguish the two types of carbon positions in pyrrole. Thus, 2 |\f 5 H 1 carbons 2 and 5 are referred as oc-carbons and carbons 3 and 4 as (3 carbons. -NH HN-1 10 11 9 2 4,6-Dipyrromethane DIPYRROLIC SYSTEMS The most important dipyrrolic molecules are those compromised of two pyrrole moieties joined by a single carbon bridge. Their numbering scheme is shown as 2 at right. In the same way, the dehydro-4,6-dipyrromethane 3 is numbered. This pigment is commonly named as dipyrromethene. In both systems, 1 1 0 1 1 9 4,6-Dipyrromethene carbons l and 9 are also referred to as a-carbons, 2-(2-H-Pyrrol-2-ylidenemethyl)pyrrole positions 2, 3, 7 and 8 as B-positions and position 5 as the mew-position.In this work, two other dipyrrolic systems are used. Their skeletal structures, numbering schemes and names are shown as following. 4 3 3' 4' w N x — ' N H 1 2 H 2,2'-Bipyrrole 2,2'-Bipyrrolyl 1,2-bis-pyrrolylethane xxi T E T R A P Y R R O L I C S Y S T E M S There are four types of tetrapyrrolic molecules used in this study. Their skeletal structures, numbering schemes and names are shown as following. 1,19-Dideoxybiladiene-ac 7 1,1 '-Bis-(dipyrrinyl) 1,2-Bis-(dipyrrin-1 '-yl)ethane Bis-(dipyrrin-2-yl) sulfide xxn Acknowledgement I wish to thank those people who have been so helpful to the development of my research project and writing of this thesis. Firstly, my supervisor, Dr. David Dolphin: I am wholeheartedly grateful for the guidance, academic freedom and financial support he has offered. Secondly, Dolphin lab group members: During these years, they have given me. so much help, advice and friendship. Special thanks go to Dr. Alison Thompson, Dr. Qingqi Chen, Mr. Jack Chow and Dr. Ethan Sternberg. Dr. Thompson has suffered longest and most in helping me proof-read the manuscript. Without her, this thesis might never have been transformed from manuscript form. Needless to say, any remaining errors are my responsibility alone. Dr. Chen and Mr. Chow have performed tedious pyrrole chemistry and provided me with extremely valuable pyrrole intermediates with no selfishness. Their collaboration has made this work more productive and interesting. Dr. Sternberg has given invaluable help and advice when I desperately needed it. Thirdly, those who provided technical support for this work: Without their help, this work would not have been possible. Sincere thanks go to Drs. Steve Rettig and Brian Patrick in the X-ray crystallography lab, Mr. Peter Boda in the Microanalysis lab, Ms. Marietta Austria and Ms. Liane Darge in the NMR lab and the crew in the Mass Spectroscopy Lab. Finally, and most importantly, my parents: I owe so much gratitude to them for their enduring love, support and encouragement. xxm PART 1 THE USE OF DIPYRROMETHENE LIGANDS IN SUPRAMOLECULAR CHEMISTRY i 1. INTRODUCTION 1.1 D I P Y R R O M E T H E N E S A N D P O L Y D I P Y R R O M E T H E N E S AS L I G A N D S 1.1.1 Dipyrromethenes Dipyrromethenes (1 in Figure 1-1) consist of two pyrrolic rings linked at a p -positions with a methine bridge. To achieve maximum conjugation of the n system, the two pyrrolic rings and the methine bridge are coplanar, and thus the molecule as a whole is planar. It has been pointed out that dipyrromethenes may exist as geometric isomers 1 and 2. ' However, the intramolecular hydrogen bond N-H---N enhances the formation of Figure 1-1. Geometric isomers of dipyrromethene isomer 1. Dipyrromethenes, unlike pyrroles, are relatively strong bases and very reactive. They are best handled in their more stable protonated form, dipyrromethenium salts, of which the hydrobromide and hydrochloride salts 3 (Scheme 1-1) have been by far the most popular. The propensity of dipyrromethenes to strongly chelate transition metal ions has long been recognized. Bis(dipyrromethene) complexes of bivalent transition metal ions including cobalt, nickel, palladium, copper, zinc, cadmium and mercury have been 2 Scheme 1-1. Protonation and Metallation of Dipyrromethene X = Br, C l ; M = Co, Ni , Pd, Cu, Zn, Cd, Hg. prepared with various geometries around the metal ion including tetrahedral, distorted tetrahedral and even planar (Scheme l - l ) . 4 More details of the properties and preparations of dipyrromethenes and their transition-metal ion complexes will be discussed in Section 1.1 in Part 2. 1.1.2 Linear Polydipyrromethenes Linear polypyrrolic compounds have been extensively investigated, largely because they can be employed as intermediates for the synthesis of porphyrins and related macrocycles.5 To this end, most studies on linear polypyrrolic compounds have focused on those containing four pyrrole rings or less. Hence, polypyrrolic systems of this kind with more than four pyrrole rings are relatively rare, with only a few examples including some five- , six-"' and eight- membered compounds reported to date. Many synthetic routes that lead to the formation of linear polypyrrolic compounds have been developed. Generally, these polypyrrolic compounds contain a certain number of pyrrole rings linked at the a-positions either directly or with -CH2- or -CH= bridges. Owing to the large number of these compounds reported and varying strategies to synthesize them, it is impractical to review them as a whole within the limited length of this introduction. Consequently, only reviews on the properties and syntheses of linear polypyrrolic compounds constituting dipyrromethene units, used in our study of the use of the polydipyrromethenes in self-assembly, are reported herein. Biladienes-ac and their metal complexes Biladienes-ac, such as 7 (Scheme 1-2) can be considered as two dipyrromethenes linked by a methylene group at the a,a'-positions. They are the most useful linear tetrapyrrolic intermediates for the preparation of porphyrins and related macrocycles. Like dipyrromethenes, biladienes-ac are usually obtained as the corresponding salts of the mineral acid present in the reaction mixture. They have been synthesized using the following methods: 1) condensation of a a,a'-unsubstituted dipyrromethane or a corresponding a ,a'-dicarboxylic acid with two equivalents of a a-formylpyrrole in the presence of H H H H 7 Scheme 1-2. Preparation of biladienes-ac (Method 1) 4 hydrogen bromide or hydrogen chloride (Scheme 1-2).9'10 Alternatively, condensation of a,a'-diformyldipyrromethanes with two equivalents of a-unsubstituted pyrroles or a-pyrrole carboxylic acids11 also gives biladienes-ac as product. These methods introduce two identical pyrroles to both ends of the biladiene-ac structure and as such are generally used to synthesize symmetrical biladienes-ac. 2) condensation of an a-unsubstituted dipyrromethene and a oc-bromo-a'-bromomethyldipyrromethene in the presence of tin tetrachloride in methylene chloride, and subsequent demetallation by treatment with hydrobromic acid in 12 13 methanol (Scheme 1-3). ' This reaction allows the synthesis of biladiene-ac derivatives bearing different substituents on each of the pyrrole rings. H H H H 10 Scheme 1-3. Preparation of biladienes-ac (Method 2) 3) condensation of an a-unsubstituted tripyrrene with a a-formylpyrrole (Scheme 1-4). 1 4 ' 1 5 This method is used to prepare symmetrical or asymmetrical biladienes-ac. However, since it requires additional steps and extra efforts to synthesize intermediate-tripyrrene, its application is relatively rare in practice. 13 Scheme 1-4. Preparation of biladienes-ac (Method 3) Biladienes-ac have very rich chemical reactivity. As Scheme 1-5 shows, several kinds of biladiene-ac derivatives can be transformed into several different macrocyclic tetrapyrroles such as porphyrins,16 metalloporphyrins,16 corroles,17 and tetradehydrocorrins metal chelates.18'19 For example, l-bromo-19-alkyl-l,19-dideoxybiladiene-ac can be directly converted to the corresponding porphyrins upon heating in o-dichlorobenzene (route I in Scheme 1-5). This method is very useful to prepare unsymmetrical porphyrins. In refluxing methanol, l,19-dimethyl-l,19-dideoxybiladiene-ac undergoes a facile oxidative cyclization in the presence of copper acetate to the corresponding porphyrin copper complexes (route II, Scheme 1-5). Biladiene-ac 14 bearing two methyl groups at C-10 , which are, therefore, unable to loose a proton from the methylene bridge, do not cyclize to form metal porphyrins.21 6 T3 CD m x 1 1 tl, CM CM OC DC m x" II _M o o cr Q-r I 5 D D CM II n II nr CM 'A, CM CE DC DC O O : m DC DC DC K > 5 ^ ^ X II CM DC DC ? a> <u >_ II II II II, CM CM CM <M DC DC DC DC CD <D CD <U !L 'J_ !L 'J-DC CCCE DC . _ : = > E o c CD CD . a CD > CD ••—• W _C0 o C/J CD > CO > CD •D O CO I CD CD T3 ca E o i_ _Q) X ! _C0 'CO > CO w -t—' o T3 O up T— CD E CD O It has been reported that treatment of a l,19-dimethyl-l,19-dideoxybiladiene-ac with zinc acetate in refluxing methanol gave a helical dimer.21 In this dimer, each zinc ion is bound to the two nitrogen atoms of both ligands and the coordination sphere is thus a distorted tetrahedron. Bis-(dipyrrinyl) dihydrobromides Bis-(dipyrrinyl) dihydrobromides can be synthesized from 5-unsubstituted dipyrromethene hydrobromides. ' " a-Unsubstituted dipyrromethene hydrobromide 15 was oxidized smoothly to the palladium complex of a,a'-bis-(dipyrrinyl) by the action of 3% palladium oxide on strontium carbonate in boiling methanol. After treating with hydrobromic acid to remove the metal, a,a'-bis-(dipyrrinyl) dihydrobromide 16 was obtained (Scheme 1-6). Br 2Br 15 16 Scheme 1-6. Synthesis of bi-(dipyrrinyl) dihydrobromide 16 8 Other attempts to synthesize these compounds have been reported by Dolphin et al. in 1965 (Scheme 1-7).21 5,5'-Unsubstituted 2,2'-bipyrrole 17 condensed with 2-formylpyrrole 18 in the presence of hydrobromic acid to give 5-(pyrrol-2-yl)dipyrromethene hydrobromide 19. However, a second mole of 2-formylpyrrole 18 did not react further to give a 5,5'-bis-(dipyrrinyl) dihydrobromide due to the general deactivation of 19 towards further electrophilic attack.21 In order to circumvent this lack of reactivity, 5-(pyrrol-2-yl)dipyrromethene 19 was reduced to 5-(pyrrolyl)dipyrromethane 20, which was then condensed with a 2-formylpyrrole to yield tetrapyrrolic compounds of the type exemplified by 21, the so-called 1,19-dideoxy-b-norbilenes-a. 20 21 Scheme 1-7. Synthesis of 1,19 dideoxy-b-norbilenes-a 21 Reaction conditions: i) HBr/MeOH; ii)Mg, Pd/C; iii) 1 eq. 18, HBr/MOH. In an attempt to bring about cyclization of bis-(dipyrrinyls), l,19-dideoxy-l,19-dimethyl-b-norbiladiene-a 21 was treated with copper acetate in boiling ethanol.21 This reaction failed to give cyclization products and instead gave a mixture of two open-chain copper tetrapyrrolic complexes. On the basis of molecular weight determination by mass 9 spectrometry, the major product, was regarded to be monomer 22, and the minor product as dimer 23. However, no further structural analysis was available on these two complexes at that time. As mentioned previously, there are very few reports relating to the synthesis of longer linear polypyrrolic chains. In order to perform our study of dipyrromethene-metal ion complexation, we have synthesized this kind of compound in a convenient and efficient way. It will be discussed in Chapter 2 of this part. 10 1.2 S E L F - A S S E M B L Y A N D S U P R A M O L E C U L E S 1.2.1 Definition of Self-Assembly Process Self-assembly is the spontaneous association of two or many moieties under equilibrium conditions into stable, structurally well-defined aggregates through either covalent or non-covalent binding.2 4 In terms of molecular magnitude and interaction, self-assembly can be divided into two types: molecular self-assembly (at the molecular, covalent level) and supramolecular self-assembly (at the supramolecular, non-covalent level). 2 5 Molecular self-assembly is a special type of synthetic procedure where several reactions between the components occur in one experimental operation to generate the well-defined covalent molecules. This type of self-assembly is controlled by the intramolecular conformational features and stereochemistry of the reaction(s). In order to gain efficiency of reaction, molecular self-assembly requires that covalent bond formation be reversible so as to allow searching for the final structure. There are many successful applications of molecular self-assembly in the generation of macromolecules26'27 and probably the most recent exciting achievement in this area is the formation of the curved carbon structures of the spheroidal fullerences Cgo, C70, ere.28 and the formation of carbon nanotubes29 in the high-temperature regime of carbon vapor that permits a certain structure fluidity. In contrast to molecular self-assembly, supramolecular self-assembly is referred to as the spontaneous association of a few or many components by intermolecular non-covalent interactions such as intermolecular hydrogen bond formation, metal-ligand 11 coordination, and electrostatic or hydrophobic interactions. These interactions, coupled with spatial confinement, direct and control the construction of a giant architecture. Supramolecular self-assembly is a reversible process. The end products are non-covalent, structurally organized, self-assembled structures. They might be either discrete oligomolecular supramolecules or extended polymolecular assemblies such as molecular layer, films, membranes, etc. 1.2.2 Self-Assembly in Nature In living systems, tremendously large biological molecules exist, for example, DNA, proteins etc. Nature "magically" constructs these giant molecules from small, simple molecules.' More amazingly, nature achieved the construction of large biomolecules even during protogenesis. To explain how primordial cells were created from relatively simple building blocks, when these "cells" did not contain the necessary machinery, catalytic, genetic, or otherwise, to direct their own synthesis, in 1954, Wald 3 1 first proposed that the building block molecules contained all the necessary information to recognize and interact with other appropriate molecules. Directed by this inherently "stored" information, the building blocks could thus assemble into larger molecules. This process, known as self-assembly, has now been recognized to be both a crucial component in the molecular events that comprise the evolution of life and an essential participant in the biosynthesis of contemporary biological systems. Indeed, a wide variety of cellular constituents, such as ribosomes, mitochondria, and the multitude of smaller multicomponent enzyme complexes, are synthesized from comparatively simple sub-units via non-covalent interactions. Among them, probably the best known of nature's 12 assemblies is the double helix of D N A (Figure 1-2). Here, the two polynucleotide chains are held together by hydrogen bonding between the complimentary base pairs, by hydrophobic interactions (the hydrophobic residues are hidden from water), by n,n interactions between parallel aromatic residues and by electrostatic and hydrophilic interactions of the charged phosphate groups in contact with water. Figure 1-2. The helical structure of DNA. The dashed lines shown hydrogen bonding between complementary nucleotide bases and the double-headed arrows indicate TZ-TX interactions. Self-assembly in biological system offers several advantages relative to the alternate approach, linear synthesis. These advantages includes: 1) a reduction of structural errors by rejection of defective sub-units during self-assembly, 2) facile formation of large biomolecules by rapidly established non-covalent interactions, and 3) economical synthesis of the complex end products from simple sub-units. 4) fewer steps required compared to the linear synthesis because the process is highly convergent. 13 1.2.3 Self-Assembly in Synthetic Chemistry In modern synthetic chemistry, there are four main strategies used to synthesize larger, more complicated molecules from small simple molecules.32 The first strategy is sequential covalent synthesis, which assembles molecules based on the sequential formation, cleavage and rearrangement of covalent bonds, usually one or a few at a time. This strategy is usually limited to molecules with small molecular weight (usually lower than 1000) due to its lack of efficiency in the synthesis of large molecules and the difficulty in preparing the required starting materials. The second synthetic strategy, called covalent polymerization, is used for preparing molecules with high molecular weights.33 In this strategy a relatively simple, reactive low molecular weight substance (a monomer) reacts repetitively with itself to produce a molecule (a polymer) comprising many covalently connected monomers. The prototype of this synthetic strategy was the conversion of ethylene to polyethylene. The molecular weight of polyethylene can be as high as 106, but it is easily prepared by polymerization of ethylene. The disadvantage of this strategy is that molecular structure of the end product is simple and the process offers only limited opportunity for controlled variation in this structure or for control of its three-dimensional shape. The third strategy abandons the covalent bond as a required connection between atoms and relies instead on weaker and less directional bonds, such as ionic bonds, hydrogen bonds, and van der Waals interactions. It organizes atoms, ions, or molecules by adjusting their own positions to reach a thermodynamic minimum and introduces 14 high-level organized structures. This strategy is referred to by Whitesides as "self-^9 organizing syntheses" because it has distinguishing features of self-organization. Representative structures prepared via these techniques include molecular crystals34, liquid crystals,35 colloids, 3 6 micelles,37 emulsions,38 phase-separated polymers,39 Langmuir-Blodgett f i lm, 4 0 and self-assembled monolayers.41 Supramolecular self-assembly, the protocol exclusively driven by non-covalent interactions,27 is the fourth but the most recently developed strategy. This method is most relevant to the preparation of nanostructures. It combines features of each of the preceding strategies to prepare large, stable and structurally well-defined aggregation products through the spontaneous assembly of molecules. In some discussions, the concept of self-assembly overlaps partially or entirely with that of self-organization.30'46 Compared to alternative approaches, the strategy of supramolecular self-assembly has several advantages: 1. Supramolecular self-assembly is directed by information stored in the sub-units to form a non-covalent interaction. This reduces the structural errors in the final product by rejection of defective sub-units during self-assembly. 2. Molecules in self-assembled aggregates are joined by non-covalent interactions (hydrogen bonds and hydrophobic interactions and n-n interactions). These interactions are usually weak (such as less than 10 kcal/mol) relative to covalent bonds (-100 kcal/mol) and comparable to thermal energies. (RT ~ 0.6 kcal/mol at 300 K). Self-assembly uses networks of such interactions to offset the unfavorable entropic terms. 15 Thus by delicate design, self-assembly facilitates the formation of suprastructures by rapidly established non-covalent interactions. 3. The self-assembly process is a highly convergent synthetic protocol and therefore requires fewer steps to convert starting material to final product compared to that of a linear synthesis encountered in traditional synthetic methods. This approach enables the rapid construction of chemical systems displaying a level of structural complexity which would be inaccessible by utilization of conventional sequential covalent synthetic methodology. It provides a method to construct complicated chemical systems with power of design and control through various intermolecular interactions.27 1.2.4 Mechanism of Self-Assembly How does the self-assembly process generate structures of the size and complexity of biological structures, but without using biological catalysts or information coded in genes? The answer is that self-assembly uses self-programmed information to direct the formation of giant products. Jean-Marie Lehn describes the self-assembly process as "recognition-directed self-processes". To better understand the self-assembly process, Lawrence utilized an analogy: think of the assembly line construction of a car -the process is performed in a stepwise manner following pre-set programs.30 The parameters (e.g. time and environmental conditions) associated with each discrete step are unique in order to maximize efficiency. Finally, a completed car is released from the 16 assembly line after the parts are assembled and the construction strategy is employed. The self-assembly process bears this resemblance, with the products being counterparts to cars and the molecular sub-units being counterparts to automobile parts. Although this example simplifies self-assembly, it clearly gives an insight into this process. In the components (molecules in this case) is stored the information necessary for a process to take place. This information is based on the steric conformation of the sub-unit structures and on the bonding geometry (e.g. carbon-metal ion and hydrogen bonds). Three levels of informational input may be distinguished in self-assembly:42 1. Molecular recognition for the selective binding of complementary components; 2. Orientation, to allow growth through binding of components in the correct relative disposition; 3. Termination of the process, requiring a built-in feature (a stop signal) that specifies the end point thus signifying that the process has reach completion. The self-assembly process "reads" this information stored in the structure of the precursor that participates in self-assembly and thus defines an algorithm through interactions between the components. Self-assembly follows the algorithm to form supramolecules. The program is molecular, as the information is contained in the covalent framework; its operation through non-covalent recognition algorithms is supramolecular. The processing of molecular information via molecular recognition events constitutes the route from molecular to the supramolecular level. 17 1.2.5 Thermodynamic Issues of Self-Assembly Self-assembled structures are formed by reversible association of a number of individual molecules through non-covalent bonds. Generally, this kind of interaction holds individual subunits together more weakly than covalent bonds. Hence, the interplay of enthalpy and entropy changes (AH and AS) in their formation is more important than in syntheses based on the formation of covalent bonds. The values of A H for the interactions that hold together self-assembled structures vary widely, from 2-20 kcal/mol for hydrogen bonds, electrostatic and hydrophobic interactions and K-TZ interactions43 to -20-80 kcal/mol of metal-ligand coordination. Entropy of reaction is usually secondary in importance in reactions that irreversibly form a covalent bond. However, it is more important in equilibrium reactions. The approximate loss in translational entropy upon bringing together two particles originally at millimolar concentration contributes approximately -TAS =+5.5 kcal/mol to A G , 3 2 and the loss in conformational entropy in freezing a freely rotating bond with three equally populated conformations in one conformation contributes approximately -TAS = 0.7 kcal/mol. If there are a number of particles associating, and if a number of conformationally mobile sections of the participating molecules are frozen on aggregation, the sum of these unfavorable entropic terms can be significant. These considerations lay the foundations for self-assembly design suggesting that the final products should be rigid and the area of contacting molecular surface be large. Assemblies using networks of hydrogen bonds in non-aqueous solvents represent 18 examples of such designs to meet the criteria of rigidity and multipoint contact, and these systems have, in consequence, been extensively examined as models for self-assembly. 1.2.6 Features of Self-Assembly Self-assembly, as a newly emerged synthetic strategy, has a lot of intriguing features, such as high efficiency, selectivity and cooperativity. These features give self-assembly many advantages over alternative strategies. An interesting experiment was carried out by Jean-Marie Lehn and his coworkers.44 When a mixture of the different tris-bipyridine ligands A , B, C and D was allowed to react with copper(I) ions, four corresponding double helicates CuA2, CuB2 CuC2 and CuD2 were spontaneously formed without crossover. In this experiment, the desired helicates are generated from a mixture of starting compounds. Self-assembly preferentially selects the like ligand strands in the mixture and binds them in a certain conformation to afford the corresponding helicates. This characteristic of self-assembly is called "self-recognition"- the recognition of like from unlike, or self from non-self. It is embodied in spontaneous selection and preferential assembly of like components in a mixture. Self-recognition is a complicated process that combines various effective factors.45 It involves three structure factors and two thermodynamic factors. The former comprises: 19 (1) The structural features of the ligands (nature, number and arrangement of the binding subunits; nature and position of spacers); (2) The coordination geometry of the metal ions; (3) The steric and conformational effects within different assembled species resulting from the various possible combinations of ligands and metal ions in a given mixture. The two thermodynamic factors are: (1) The energy-related principle of maximal site occupancy, which implies that the system evolves towards the species or a mixture of species that presents highest occupancy of the binding sites available on both the ligand and the ions. It corresponds to the formation of the highest number of coordination bonds and therefore to more stable state of the system. Full site occupancy (site saturation) is achieved in "closed" architectures; (2) The entropy factor, which favors the state of the system with the largest number of product species. These factors not only apply to self-assembly utilizing metal coordination, but also apply to other interactions, such as hydrogen bonding or hydrophobic interactions. A typical example is the chiral selection occurring in the course of the self-assembly of homochiral helical strands and ribbons through hydrogen bonding. 4 6 1.2.7 Supramolecules, Nanostructures and Self-Assembly Supramolecular chemistry and nanochemistry have rapidly become concepts of major importance in the last decade. Supramolecules are molecules that cannot be made 20 by conventional synthetic strategies due to their large size and complexity. Thus, supramolecular chemistry is also called "chemistry beyond the molecule, the designed chemistry of intermolecular (non-covalent) binding interactions." 2 4 Nanostructures are assemblies of bonded atoms that have dimensions in the range of 1 to 102 nanometers (1 nm = IO"9 m = 10 A), and that contain a large number of atoms (103 to 109) and have molecular weights of 104 - 10 1 0 daltons. Self-assembly is the main strategy used in constructing supramolecular and nanomolecular structures from small molecules. Although investigation in this field is still generally considered as being in its infancy, many pioneering approaches have been successfully developed and a wide range of complex supramolecular structures (usually nanostructures) have been built from small and simple molecules. For example, J .-M. Lehn and his colleagues have used polybipyridine ligands to generate helicates,47"49 grids, 5 0" 5 2 ladder, cages,4 5'5 3 and rings. 4 2 ' 5 4 ' 5 5 Whitesides et al. have prepared doubly and triply stacked cyclic arrays through hydrogen-bond networks.56 Suprastructures and nanostructures pose many intriguing futuristic possibilities, as their broad application possibilities are being realized. Ultimately suprastructures may be utilized in many areas: interface and colloid science,57 molecular recognition,58 electronics microfabrication,59 polymer science,60 electrochemistry,61 zeolites and clay chemistry and scanning probe microscopy. Other potential applications include the design of assembly inhibitors which are of interest for interfering with the association of components of multiprotein complexes.64 21 In particular, assembly inhibition could represent a fruitful approach to drug designs; for instance, the inhibition of insulin aggregation may facilitate its absorption.65 1.2.8 Review of Synthesis and Properties of Self-Assembling Supramolecules Hydrophobic interaction self-assembly Hydrophobic interaction is one of the most common phenomena in nature and plays a fundamental role in the living system (lipids, proteins, nucleic acids, etc.).30 Since such interaction is weak, it usually acts as an auxiliary force in forming self-assembling aggregates. For example, in the assembly system of the double helix of D N A (Figure 1-2), coexist hydrogen bonding, hydrophobic interaction (the hydrophobic residues are hidden from water) and TC-TT interactions. Among them, the hydrogen bond is the strongest (5-10 kcal/mol). Although hydrophobic interactions are relatively weaker and non-directional, they are still important factors effecting conformation. Due to the weakness of the interaction, studies on synthetic assembly by using hydrophobic interaction as the driving force are more difficult (compared to that of hydrogen bonds) and its practical application are few. However, several tube-like cylindrical complexes have recently been designed based upon hydrophobic interaction-induced self-assembly. One of the earliest examples of artificial self-assembled ion channels was reported by Tabushi and his colleagues in 1982.66 The basic strategy of this approach was to prepare molecules of appropriate hydrophilic and hydrophobic part, of a length comparable to a lipid molecule, and of appropriate numbers of ionophilic sites. Based on this concept, A , C, D, F -tetra-6-(6-n-butyrylamino-n-hexyl-l-sulfenyl)-p-cyclodextrin 24 22 (this species is illustrated in a schematic fashion in Figure 1-3) was chosen as a promising artificial channel candidate ("half channel"). After "planting" of this compound into an egg lecitin-based liposome, the rate of metal ion transport from the outside to the interior aqueous solution across the liposome membrane was enhanced to such an extent that it is much faster than that of ion-carrier 18-azacrown-6. C o 2 + transport exhibited second-order kinetics with respect to the membrane concentration of 24. Consequently, a model was proposed in which two sub-units of 24 dimerized to produce a species that can span the length of the lecithin bilayer (Figure 1-4). 24 Figure 1-3. Schematic structure of A, C, D,F-tetra-6-(6-n-butyrylamino-n-hexyl-l-sulfenyl)-p-cyclodextrin Figure 1-4. Self-assembly of a cyclodextrin-based ion channel. 23 (CH 3 ) 2 N + (CH 2 ) 1 7 CH 3 (CH 2)mCH 3 "OOC (CH 3 ) 2 N + (CH 2 ) 1 7 CH 3 (CH 2)mCH 3 26 Kobuke et al. have designed and synthesized amphiphilic molecules that enter into the planar bilayer membrane and assemble into a channel-containing species.67 The polyether 25 was converted to the corresponding ammonium carboxylate 26 (Scheme 1-8). The additional long hydrophobic alkyl chains on the amine are present in order to promote incorporation of 26 into the lipid membrane. Upon the insertion of 26 into lipid bilayer membrane, several v 'n intrinsic single ion channel characteristic were observed, including stable and constant conductance levels and cation selectivity over anion. It is believed that the tendency of the hydrophilic oligoether chain to separate from the hydrophobic membrane components is the driving force for assembly formation and consequent connection of two half-channels located in different lipid layers. The possible mechanisms are postulated as four key steps (Figure 1-5). Step 1: Incorporation of amphiphile into bilayer membrane. Step 2: Molecular recognition of the polar oligoether chains by surrounding hydrophobic lipid molecules to include the assembly. Step 3: The connection of polar domains in the different lipid layers via discrimination of the polar molecular domain from the hydrophobic domain. Step 4: Transfer of ions from the aqueous phase into the channel via ion-dipolar stabilization and movement through the polar pore. It should be pointed out that this artificial channel is Schemel-8. Preparation of ammonium carboxylate 26 24 characterized by extreme simplicity of the structure and a function remarkably similar to that of natural single ion channels. Figure 1-5. Self-assembly of an oligoether-based ion channel Hydrogen Bonded Self-Assembled Structures The hydrogen bond, which is stronger than hydrophobic forces but weaker than the covalent bond and metal-carbon coordination, has a moderate bond energy between 5-10 kcal/mol. According to approximate calculations, to form a self-assembled structure, the loss in translational entropy and the loss in the conformational entropy contribute approximately -TAS = +5.5 kcal/mol and +0.7 kcal/mol to A G respectively (discussed in 1.2.5). These values are in the range of enthalpy of hydrogen bond formation. Thus by using networks of hydrogen bonds in non-aqueous solvents, thermodynamically 25 reversible self-assembly can easily be achieved. This consideration is consistent with the fact that a large number of H-bonded self-assembled structures widely exist in biological systems. However, understanding the thermodynamics of self-assembly in biological systems is made difficult by several factors. Firstly, water is a complicated solvent, and the thermodynamic origins of the hydrophobic effect remain a matter of debate. The entropically favorable release of structured water on association of hydrophobic regions of aggregating molecules is an important contribution to overcoming the unfavorable loss of translational entropy in this aggregation. Secondly, many intermolecular interfaces in aggregated biological systems involve supramolecules and are large (1 to 5 nm2). It is difficult to disentangle the contributions of individual organic groups (with an area of 0.05 to 0.5 nm2) to these interfaces. Finally, changes in conformation upon self-assembly are common but may be distributed as small changes in a large number of bonds. The enthalpic sum of these changes is again difficult to estimate. Recently, chemists have made great progress in the study of hydrogen bond oriented self-assembly by defining the reacting systems and using non-aqueous solvents. Many intriguing macromolecular structures have been generated and many promising applications had been discovered or are being proposed. This has become the most extensively examined area in self-assembly study groups and has been seen as a model for self-assembly. Here we choose several typical hydrogen bonded systems to demonstrate the fascination of this area. 26 1. Cyclic array Generally, a set of aromatic rings with hydrogen-bond donor or hydrogen-receptor functionalities are employed to assemble into cyclic aggregates, such as cyclic tetramers 27 and 28 (Figure 1-6).68 This synthetic strategy risks forming linear end products because the individual subunits have the opportunity to assemble into either a cyclic or linear array. Entropically the formation of the latter would be favored because cyclic structures generally exhibit more order than their linear counterparts. In contrast, since there will be a greater number of hydrogen bond interactions per sub-unit in a cyclic species, compared to that of a linear complex, the former will be favored. Consider a hypothetical species that self-associates to form a cyclic complex composed of three identical subunits, one in which each sub-unit is hydrogen bonded to the other two subunits. In this case, there is an average of 1.0 hydrogen bonds per monomer. However this drops to 0.67 for the corresponding linear trimer. Nevertheless, the enthalpic advantage is dramatically diminished in large multicomponent aggregates (e.g. A 20-mer R 27 28 Figure 1-6. Hydrogen bonded self-assembled cyclic aggregates 27 composed of one hydrogen bond between monomers: cyclic (1.0 bonds/monomer), linear (0.95 bonds/monomer). It has been noticed that derivatives of barbituric acid or the closely related cyanuric acid 29 and melamine 30 (Figure 1-7) will assemble into astonishing variety of structural motifs and minor variations in monomer structure can have a profound effect on the architecture of self-assembling systems.69 By delicate design, Whitesides and his colleagues56 have constructed a molecular aggregate with two parallel planes of cyanuric acid (CA)-melamine (M) lattice, each containing one hexagonal array of three C A units and three M units. To bring together 12 molecules into one is an unfavorable process entropically; moreover, even if hydrogen-bonded arrays were strong enough to allow assembly, there is every reason to expect them to assemble as one sheet, not two parallel sheets. In order to overcome these disadvantages, the investigators preorganized C A and M units by connecting them covalently with a benzene ring as a central hub, with "spokes" (33 and 34) designed to position the C A and M units in approximately the correct position. This helped to effect a balance between entropy and enthalpy in the system and led to the self-assembly of two components into an aggregate via the interaction of hydrogen bonds. The final aggregate (shown as a schematic representation at the bottom of Figure 1-7) forms quantitatively in chloroform solution. This structure is very stable (it can be heated to 450° without change) as a result of the network of hydrogen bonds that hold it together. It is roughly spherical with diameter 2.5 nm. This approach is a modest start along a pathway leading to functional nanostructures, but nevertheless illustrates the basic components of this synthetic 2 8 Figure 1-7. Self-assembly of C A - M stack aggregate The self-assembled C A - M stack aggregate is shown at the bottom of the figure. Structure 32 in the middle represents the C A - M lattice. 29 strategy: the use of reversible interactions (hydrogen bonds in this case) to bind the participating molecules in the aggregate; preorganization of the interacting groups through a network of covalent bonds to control the entropy of association and to determine the shape of the aggregate; choice of components so that recognition with high selectivity occurs; and design of the system to effect positive cooperativity. 2. Ion Channels Another interesting example is the construction of ion channels via hydrogen-bonding self-assembly. M . R. Ghadiri and his colleagues at The Scripps Research Institute have reported the synthesis of self-assembling peptide nanotubes that may find i PCy 3 Cy=cyclohexyl Scheme 1-9. Self-assembly of peptide rings 30 applications ranging from medicine to materials science. This supramolecular architecture is based on building blocks that consist of cyclic peptides. Since rings composed entirely of L-amino acids tend to form intramolecular hydrogen bonds between the carbonyl oxygen of one amide and the hydrogen attached to the amide nitrogen of another, alternating D- and L- amino acids were used to built peptide rings that have flat conformations with amide carbonyl and N H groups pointing up and down, perpendicular to the plane of the ring 35 (Scheme 1-9). These rings self-assemble into nanotubes by stacking one on top the other, linked by intermolecular hydrogen bonds in a p-pleated sheet motif. To hold the self-assembly supramolecular structure tightly together (37 in Scheme 1-9), Ghadiri et al. have devised a method to tie the individual rings together by covalent bonds thus increasing the kinetic stability of the cyclic peptide stack.71 The peptides precipitated out of acid solution as rod shaped crystalline objects 200 to 300 um long, composed of hundreds of tightly packed parallel tubular structures (shown as a schematic representation in Figure 1-8). The tube diameters vary with the Figure 1-8. Self-assembly of a peptide-based ion channel 31 number of amino acids in the ring. The largest reported so far has a diameter of 13 A. These nanotubular supramolecules may have significant applications. The interiors of the peptide tubes are very hydrophilic so that ions or molecules can be placed inside. The outside surface of the tubes can also be varied by changing the amino acid side chain. Ghadiri and his group are now working to design nanotubes with lipophilic surfaces, which self-assemble into channels that transport ions and even glucose through lipid bilayers. In addition, these chemists have engineered a system in which carboxylic acid side-chains on the outside surface of the tube bind copper ions. Such assemblies may find use as heterogeneous catalysts and could lead to highly ordered superlattices known as quantum dots. I n o r g a n i c Self-Assembly One of the most fascinating areas of self-assembly is the spontaneous generation of well-defined metallo-supramolecular architectures from specific interactions between the covalent organic ligands and metal ions. 7 2 Metal ions are systematically used since they possess useful properties for the assembly: (i) a set of coordination numbers and stereochemistry preferences depending on their size, charge and electronic structure, (ii) a large variation in binding strength and kinetic stability, (iii) variable affinities for different binding units, and (iv) specific magnetic, electronic, and spectroscopic properties expressed in the final complex. Ligands should posses (i) several binding units allowing the recognition and coordination of the various metal ions; (ii) judicious factors such as rigid spacers or steric hindrance which orientates the conformation of the metal complex. During the last decade, an enormous body of research activity has been focused on inorganic self-assembly which generates metal-containing aggregates. Jean-Marie Lehn, undoubtedly the foremost practitioner in the exploration of supramolecular architectures, and his colleagues have successfully synthesized a wide range of polynuclear metal clusters by multi-component self-assembly using polybipyridine ligands. These assemblies have such wondrous molecular structures as helicates,47"49 grids, 5 0 - 5 2 cages,4 5'5 3 cylinder, ladders 7 2and rings. 4 2 ' 5 4 ' 5 5 1. Helicates One of the most rapidly expanding families of metallo-assembled architectures is that of the helicates, which consist of linear arrays of metal atoms, held together by bridging ligands, helically arranged about a central axis. Although evidence for the existence of such structures was reported as early as 1958, most of the work in this area have been accomplished since the mid-1980s.73 Among these initial pioneering investigations is the preparation of complex 39.7 4 This complex was formed by treatment of tris-bipyridine ligand 38 with a slight stoichiometric excess of a Cu 1 salt (Figure 1-9). It was characterized as a trinuclear double-stranded helicate by X-ray crystallography. Lehn further extended this work to include tetra- and penta-nuclear arrays of metals held together by helically arranged polytopic oxomethylene-bridged bipyridines.75 Their total lengths are estimated to be about 2,200 pm and 2,700 pm, respectively, and 33 can therefore be considered as examples of self-assembled structurally well-defined nanoarchitectures. The mechanism of their formation almost certainly takes place with positive cooperativity and self-recognition. 39 X= H, CHgCHgCO^Bu Figure 1-9. Self-assembly of trinuclear homotopic double stranded helicate 39 Tris-bipyridine 40 connected via (3 positions with ethylene groups leads to complex 41 ([Ni 3 (40) 3 ] 6 + ) , when reacted with N i 2 + (Figure 1-10).76 The complex was characterized by X-ray crystallography and found to exhibit the trinuclear triple-helical structure with three stands coordinated to three packed pseudo-octahedral Ni(II) ions. The length per helical pitch (rate of axially linear to angular properties) 4 7 is about 41 A, a value significantly larger than the 12 A found in the double-stranded cuprohelicate 39, and attributed to the transoid conformation of the ethylene spacer. 34 41 Figure 1-10. Self-assembly of trinuclear homotopic triple stranded helicate 41 2. Ladders One fundamentally interesting topology is that of a molecular ladder, which could in principle be constructed by arranging linear chains of metals parallel to one another and connecting them together by bridging units. A possible way in which molecules of this type may be assembled is to allow two linear wrack-like ligands (ladder sides) with n metal binding sites to react with 2n metal ions in the presence of bridging ligands containing "back-to-back" coordinating sites (ladder steps). The shape of the resulting architecture could then be described as a stepladder. Molecules of this type may be expected to exhibit unusual and interesting physicochemical properties. Thus, a mixture of quaterpyidine 42a, tetraphenylbipyrimidine 43 and [Cu(MeCN) 4]X (X = CIO4, PF 6 ) in a 1:1:2 stoichiometric ratio reacted to give dark-brown product. *H N M R and fast atom bombardment (FAB) and electrospray ionization 35 (ESI) mass spectrometric data of the product complexes supported the product formulation [Cu(42a)2(43)2]X4. Although not characterized by X-ray crystallography, the tetranuclear product 44 is almost certainly a metallo-macrocyclic ring of four copper(I) ions and four ligands or, in the other words, a "two-step" stepladder (Figure 1-11).77 45 Figure 1-11. Self-assembly of ladder-like complexes 44 and 45 Further extension of this building principle led to the synthesis of larger stepladders with n > 2 steps, by using of the linear tropic ligands sexipyridine 42b in 36 place of 42a. Thus, from a mixture of 42b, 43 and [Cu(MeCN)4]X in a 1:1.5:3 stoichiometric ratio in nitromethane as solvent was isolated a complex 45 formulated as [Cu(42b)2(43)3]X6. The structural formulation was determined on the basis of 'H NMR, FAB and ESI mass spectrometric data, and is consistent with a "three-rung" stepladder-type architecture (Figure 1-11). The 'Ft NMR spectrum was particularly informative in that the ortho- and meto-phenyl ring protons of the two outer and one inner ligands 43 (outer and inner "rungs") were clearly distinguishable as a pair of doublets and a pair of triplets, respectively, each in the expected 2:1 ratio. This latter system represents an impressive example of a programmed molecular system, in which 11 "informed" components spontaneously and correctly assemble to a single architecture of precisely defined composition. When 42a, 42b, 43 and [Cu(MeCN)4]PF6 were mixed in a 1:2.5:1:5 stoichiometric ratio in nitromethane, only two products were found to exist in the reaction mixture after 72h. 'H NMR and ESI mass spectrometric measurements confirmed the identity of both products as being associate, without error, to give the two structurally complex and precisely defined nanoarchitectures 44 and 4578 in a self-recognition process, as described previously. 3. Grids Another interesting inorganic self-assembling structure involves structurally well-defined two-dimensional arrays formed from metal ions and organic ligands. This architecture is of great interest in that it offers intriguing prospects for the future 37 development of electronic devices for information storage. One way to achieve this would be to use rigid, linear, polytopic ligands containing bidentate coordinating sites that may be able to assemble into chessboard-type arrays upon mixing with metal ions of tetrahedral coordination geometry. Combination of the tritopic ligand 6,6'-bis[2-(6-methylpyridyl)]-3,3'-bipyridazine (Me2-bpbpz) 46 and silver(I) trifluoromethanesulfonate in a 1:1.5 stoichiometric ratio in nitromethane solution resulted in the spontaneous assembly of a 3 X 3 square grid 47 consisting of nine silver(I) ions and six ligands (Figure 1-12).80 X-ray structural analysis of this crystal shows that the structure of the cation is that of a 3 X 3 grid of nine Ag(I) ions. The grid is distorted into a rhomboid shape, with an angle of about 72(3)° between the mean planes through the ligands and an average Ag-Ag distance of approximately 372(3) pm. A l l the silver ions experience a distorted tetrahedral environment, and collectively together sit on a slightly warped "saddle-back"-type surface. This is mainly due to the curved nature of each ligand, which, in turn, arises from the fact that 9+ 46 47 @ = Ag Figure 1-12. Self-assembly of grid-type complex 47 38 pyridazine rings are contracted about the N=N bond and therefore not regular hexagons. The average distance between the mean planes through adjacent ligands is 374 pm and therefore just outside that expected for van der Waals contact (-340 pm). This molecule is remarkable in that its formation involves the spontaneous and correct association of 15 particles (nine metal ions and six ligand components), and as such fulfils the three basic levels of operation of programmed supramolecular system, that is, recognition, orientation and termination. 4. Circular Complexes (Rings) Lehn et al. used compound 48, which combines two different subroutine binding components A and B , as ligands which undergo specific and distinct self-assembly processes with Cu 1 (Figure 1-13) E r r o r ! B o o k m a r k n o t d e t i n e d - The resulting complex, which was 112+ Figure 1-13. Self-assembly of bowl-shaped complex 49 39 made up from four ligand molecules and twelve metal ions, was formed in almost quantitative yield. Characterized by X-ray crystallography, complex 49 was a bowl-shaped macrocycle and contained both double-helical and grid-like motifs. It consists of macrocycles of nanometer dimensions with an external diameter of 2.8A. The central cavity has a diameter of 11 A, and contains four anions (PF6~) as well as solvent molecules. The spontaneous formation of 49 results from a self-assembly process based on a "program" combining two assembly subroutines, in that subroutine A gives a double-helical structure with Cu 1 and B a grid-like structure, each specific to one of the ligand subunits. This example again demonstrates a complicated self-assembly system which is comprised of a large number of components. In summary, self-assembly through a combination of double or, more generally, multiple assembly subroutines can be used to generate a wide variety of highly complex supramolecular architectures. 5. D e n d r i m e r Metal ion-induced self-assembly can also be used in the preparation of dendric macromolecules. The usual approach to dendrite synthesis involves the consecutive construction of one generation at a time (divergent method), or preparing sections of molecule and connecting together the segments in the final stages of the synthesis (convergent method).81 Both strategies employ sequential covalent bond forming methodology and also require additional protection/deprotection steps at each stage of growth. In 1995, a single step generation of a dendrite prepared via self-assembly was reported by Huck. 8 2 Dendrite 51 was constructed from precursor complex 50, which combined two kinetically labile MeCN ligands with one potentially kinetically inert 40 cyanomethylene donor. Displacement of the ligated MeCN by warming a solution of the complex in nitromethane resulted in coordination of each palladium ion by the cyanomethylene groups. In this way, the monomeric complexes were able to assemble in three dimensions to give a macromolecular aggregate (Figure 1-14). The final shape and size would be determined solely by maximal steric crowding effects experienced in the last generation. Such effects would collectively prevent further growth, and together constitute the termination step. Figure 1-14. Self-assembly of dendrimeric complex 51 41 In conclusion, inorganic self-assembly, by using metal ions as the "cement" which collectively bonds together the final supramolecular ensemble, is an exciting and rapidly expanding field. Many preexisting organic molecules such as crown ethers and crypates must be synthesized using laborious consecutive sequential steps and are obtained in poor to moderate overall yields. Structural analogues of these molecules which incorporate metal ions as integral architectural units can now be prepared under equilibrium conditions using a more simple set of components. The complexes are usually isolated in high to quantitative yields. It appears that very large and complex spatial organizations, which would be almost impossible to achieve using conventional organic covalent synthetic methods, can be built in this way. Self-assembly has brought us direct access to nanostructural molecules that lead to many applications such as light harvesting and biomolecular transport/delivery system, and also information transfer and transduction via molecular photonics, ionics, electronics and so on. Further developments in this field are likely to unearth more molecules that exhibit particularly fascinating and unexpected physicochemical properties. 42 2. R E S U L T S A N D DISCUSSION Bipyridines 52, the core binding units in the ligands used in Lehns' work 8 3 on supramolecular self-assembly, are neutral ligands and form charged complexes when coordinated to metals at any oxidation level (>M°) (Figure 1-15). Thus, counterions are needed to generate a neutral species. Unfortunately, such counterions may give rise to Bipyridine (52) //~\ --H neutral ligand Dipyrromethene (53) VN N= mono-anionic ligand M 2 + dicationic complex neutral complex Figure 1-15. M 2 + ions give charged species with bipyridines but give neutral species with dipyrromethenes. disorder in the solid state. In addition, salts are often difficult to purify using traditional techniques such as chromatography. Dipyrromethenes 53, on the other hand, generate mono-anionic resonance stabilized ligands (Figure 1-15) which readily coordinate metals to give neutral square planar, tetrahedral and octahedral complexes which do not require counterions. 84 43 Dipyrromethenes, which are also called dipyrrins, are basic, fully conjugated and flat bipyrrolic pigments. Their propensity to strongly chelate transition metals has long or been recognized and will be discussed in detail in Part 2. Multi-dipyrromethenes bridged with various alkyl chains have also been found to be good ligands for chelation to transition metals. A mixture of zinc:octaethyl formyl-biliverdinate complexes 1:1 and 2:2 was reported by Fuhrhop et al in 1976.86 X-ray crystallography showed that the 2:2 complex exists as a helix. Similarly, in 1965, Dolphin 8 7 ' 8 8 observed a 2:2 (ligand 54:Co") complex and suggested that it existed in a helical conformation, and a year later he also observed the existence of the 2:2 (55:Cun) complex (Scheme 1-10).21 In 1980, the helical structure of an analogous zinc complex of 1,2,3,7,8,12,13,17,18,19- decamethyl-biladiene-ac was confirmed by X-ray crystallography.89 The above studies show us that poly-dipyrromethenes usually coordinate with transition metal ions on the basis of one individual dipyrromethene unit and generate well-organized structures. Encouraged by J.-M. Lehn's pioneering work on the self-assembly of poly-bipyridines with transition metal ions, we tried to use poly-dipyrromethenes as the ligands to generate various structurally organized supramolecules with neutral charge. Additionally, we wanted to examine the impact of transition metal and spacer group on the final structures. On the basis of these project goals, we re-examined the reactions discussed above90 and explored the use of a spacer between dipyrromethene units by reacting the hydrobromide salts 56-58 with zinc or cobalt acetates (Scheme 1-10). Dimers 59-61 were obtained respectively. 44 (54) n=1, R1=R2=R3=CH3, R 4=CH 2CH 3 (55) n=0 R 1 =C0 2 CH 2 CH 3 , R2=R3=CH3, R4=H (56) n=0, R1=R2=R4=CH3, R 3=CH 2CH 3 (59) n=0, M=Zn", R1=R2=R4=CH3, R 3=CH 2CH 3 (57) n=1, R1 =R3=R4=CH2CH3, R 2=CH 3 (60a) n=1, M=Zn", R1 =R3=R4=CH2CH3, R 2=CH 3 (58) n=2, R1=R2=R3=CH3, R 4=CH 2CH 3 (60b) n=1, M=Co", R1=R3=R4=CH2CH3, R 2=CH 3 (61) n=2, M=CoM, R1=R2=R3=CH3, R 4=CH 2CH 3 Scheme 1-10. Syntheses of dinuclear polydipyrromethene complexes In addition, we further extended our work to the preparation of larger architectures; namely tri-nuclear arrays of metals, held together by helically arranged methylene-bridged dipyrromethenes. A trinulear complex was prepared by refluxing the corresponding ligand and Zn(OAc) 2 in methanol, followed by chromatography. EI mass spectrometry showed the product to have molecular mass corresponding to the trinuclear bis-ligand complex, with Zmligand ratio of 3:2. X-ray crystallography confirmed that it also exists as a helical double-stranded structure. 45 2.1 F O R M A T I O N A N D C H A R A C T E R I Z A T I O N O F B I L A D I E N E M E T A L C O M P L E X E S 60a A N D 60b 2.1.1 Syntheses of Complex 60a and 60b To investigate and explore the use of dipyrromethenes in self-assembly processes, we initially used a biladiene-ac, which contains two dipyrromethene moieties and a methylene bridge. This was achieved via condensation of a 2-formylpyrrole and a 5,5'-dicarboxy-2,2'-dipyrromethane (method 1 of biladiene syntheses, section 1.1.2). This method is now routinely used for the preparation of biladienes-ac on a large scale.91 The dipyrromethane used in this method can be synthesized by condensation of the corresponding a-acetoxymethylpyrrole and a-unsubstituted pyrrole. The synthetic route to l,3,17,19-tetramethyl-2,7,8,12,13,18-hexaethyl-biladiene-ac dihydrobromide 57 is shown in Scheme 1-11. Thus, 5-methylpyrrole 62 was oxidized by sulfuryl chloride in methylene chloride to afford a high yield of the trichloromethyl derivative,92 which was hydrolyzed in boiling aqueous acetone and the hydrochloric acid formed was neutralized with sodium acetate. 5-Methylpyrrole 62 was thus converted into 5-carboxypyrrole 63 in >90% yield over two steps. A high yielding process for two phase iodination of 63 then gave iodopyrrole 64. 9 2 ' 9 3 This was reduced with hydrogen under catalysis of P t0 2 to the a-unsubstituted pyrrole 65. Pyrrole 66 was obtained at 60% yield by controlled oxidation of pyrrole 62 with one equivalent of Pb(OAc)4. 9 4 Coupling of pyrrole 65 and 66, catalyzed by TsOH afforded dipyrromethane 67 9 5 which was converted 60a M=Zn 60b M=Co' Scheme 1-11. Synthesis of Zn(II) complex 60a and Co(II) Complex 60b a) 1. S0 2C1 2/CH 2C1 2,1.5 h; 2. (CH 3 )CO/H 2 0, heat; 3. NaOAc/H 2 0; b) I 2/KI, heat, 40 min; c) H 2 /Pt0 2 , 2h; d) 1 eq. Pb(OAc) 4, HOAc/(Ac) 2 0, 60°C; e) TsOH/MeOH, heat; f) H 2 , Pd/C, N(Et) 3, 3h; g) HBr/MeOH, heat, 20min; h) Zn(OAc) 2 or Co(OAc) 2 in MeOH / CHC1 3 , N 2 , 5h. 47 into dicarboxylic acid 68 by 10% Pd/C catalyzed hydrogenation. The resultant dipyrromethane 68 was reacted with 2-formylpyrrole 69 in methanol containing 20% HBr solution (aq., 48%) to afford the desired biladiene-ac 57.9'10 This compound, as a red precipitate, was separated by filtration from the methanol solution. Without any further purification, it was sufficiently pure for the next stage. Biladiene 57 was only sparingly soluble in either MeOH or CHCI3. However, it dissolved very well in a 1:1 mixture of those two solvents. Such a solution was treated with a methanolic solution of zinc acetate hydrate under N 2 at room temperature. After 5h, T L C gave a bright red spot on the solvent front of the plate eluting with CH2CI2. Separation by chromotography afforded analytically pure zinc complex 60a in a yield of 24%. EI mass spectrometry gave the molecular mass for this complex to be 1174, which confirmed its stoichiometry as Zn2(ligand)2. Under the same conditions, cobalt complex 60b was obtained (33% yield) when Co(OAc) 2 was used in place of Zn(OAc) 2. EI mass spectrometry gave the molecular mass at 1162, which corresponds to Co2(ligand)2. 2.1.2 Spectroscopic Analysis of Complexes 60a and 60b Optical Spectra The electronic spectra of Zn complex 60a and Co complex 60b are quite similar (Figure 1-16). Their UV-visible spectra can be divided into two sections. (1) A very intense band is observed in the region of 466-530 nm, assigned to intraligand 71-71* bands. 48 Similar TC-TC absorptions have been observed in the same region of spectra of neutral and protonated ligands and Zn, Co dipyrronato complexes. 9 6 - 9 8 (2) Complexes 60a and 60b both have relatively weak adsorption at the high energy region around 350 nm. These bands are probably due to high-energy n->7t* transition. Since this transition is forbidden, the intensity is lower. Upon the addition of polar solvent DMSO, this signal gave a significant hypsochromic shift (25 nm) compared to the other two (5 nm), thus strongly supporting our assignments of these absorptions. 6.00—, I I I I 350 nm 450 nm 550 nm 650 nm wavelength Figure 1-16. UV-visible spectrum of complexes 60a (solid line) and 60b (broken line) in CH2CI2 NMR spectra: The ' H N M R data of the diamagnetic zinc complex 60a shows it to be largely as expected. As a result of complexation between zinc and pyrrolic nitrogen, the signal at low-field (-13 ppm) disappears. One noticeable exception is the up-field shifts of the bridging methylene hydrogens (3.90 ppm) and the methine hydrogens (6.64 ppm) in 60a 49 compared to 5.06 ppm and 7.02 ppm in the protonated ligand. The meso-methylene hydrogen atoms of biladiene-ac hydrobromide exhibit significant acidity due to the hyperconjugation across the meso-carbon, as has been proven by H-D exchange experiments with 1,19-diethoxy carbonyl biladiene-ac (Scheme 1-12)." However, in the zinc complex, this kind of resonance may be constrained by two factors: (1) electron-rich zinc reduces the acidity of the meso-methylene hydrogens, (2) the tetrahedral geometry of zinc nitrogen coordination restrains the two planar dipyrromethenes from rotating around the methylene freely and thus reduces the possibility of hyperconjugation, since planarity across the whole ligand is no longer maintained. 2.1.3. Crystal Structure Analysis of Complex 60b Slow diffusion of methanol into a methylene chloride solution of 60b provided green plate crystals with a metallic luster. These crystals were suitable for X-ray crystallography. An ORTEP representation of the structure is shown in Figure 1-17. F3CCOOD Scheme 3-12. H-D exchange at the mesoposition of 1,19-dideoxybiladiene- ac 50 Experimental details of the structure are listed in Table 1-1, the final atomic coordinates are listed in Table 1-2 and selected properties in Table 1-3 and Table 1-4 (Appendix). The molecule has a dimeric structure containing two tetrahedral cobalt atoms and two tetradentate biladiene-ac ligands. A molecular structure shows two-fold symmetry with the cobalt atoms positioned on a crystallographic axis. Each of the two ligand strands displays a similar conformation to that observed for a biladiene dihydrobromide100 and bilirubin. 1 0 1 However, the inter-planar angle between the two Figure 1-17. ORTEP representation of complex 60b. (For clarity, hydrogen atoms are omitted.) 51 dipyrromethene moieties is 89° in 60b, significantly smaller than those of 107° and 98°, which have been measured in crystal structures of dihydrobromide100 and bilirubin 1 0 1 respectively. This is the result of contrasting degrees of twist of the individual dipyrromethene units in 60b with respect to the central methylene bridge. The four Co-N bond lengths at one cobalt center are 1.990 A, 1.993 A, 1.988 A and 2.00 A respectively and bond angles of Nl-Co-N2, NI-C0-N6, N5-Co-N6, Nl-Co-N5, N2-Co-N6, N2-Co-N5 are 98.7, 106.7, 97.8, 108.3, 135.8 and 108.0° respectively. This indicates a slightly distorted tetrahedral geometry at the cobalt atoms. The methylene bridge has a bond angle of 121° with its two adjacent carbons. As a result, the methylene carbon bears some o strain. The Co-Co distance is 4.33A in this structure. On the basis of its mass spectrometry, zinc complex 60a is also a dimer which is consistent with an analogous zinc complex previously characterized by X-ray crystallography and shown to be a helical structure by Sheldrick and Engel. 1 0 2 In our study, we also carried out molecular modeling 1 0 3 of 60b. The structure obtained by simulation of this complex is similar to the X-ray structure, as shown by comparison of the structural parameters in Table 1-5. Table 1-5. Selected structural parameters of X-ray and simulated structures of complex 60b Complex 60b (A) X-ray (B) model Bond Angle (°) N l - C o l - N 2 98.7 103.4 1 * 52 Nl-Co l -N5 108.3 110.3 Distance (A) Co-Co 4.331 4.465 Dihedral Angle (°) Two planar di-pyrromethenes in same strand 89 91 It is well known that Ni and Pd-complexes of biladiene-ac are monomers and yield tetradehydrocorrin metal complexes upon cyclization under oxidative conditions and that the analogous Cu-complex is also a monomer and yields a metalloporphyrin upon ring closure.1 0 4 However, the observation of dimeric structures for 60b and the related zinc complex mentioned above is not surprising if we take into consideration that zinc(II) and Co(II) prefer a tetrahedral coordination geometry. Thus, the helical structure will be considerably less strained than would be a cyclic monomer. 53 2.2 FORMATION AND CHARACTERIZATION OF 1,1'-BIS-(DIPYRRINYL) ZINC COMPLEX 59 In 1966, Dolphin et al. observed that 1,19-dideoxy-b-norbilene-a 21 reacted with copper(II) acetate in alcoholic solution to give two copper complexes of the 1,1'-bis-(dipyrrinyls) (Scheme 1-13).21 The proportion of each varied with the nature of the solvent. On the basis of molecular weight determinations, one was regarded to be monomer 22 and the other, the formation of which was favored in methanolic rather than ethanolic solution, was proposed to be the dimer 23. Br' 21 1 C2H5O2C CO2C2H5 22 23 Scheme 1-13. Complexation of 1,19-dideoxy-b-norbilane-a 21 with Cu(II) 54 2.2.1 Synthesis of Complex 59 We re-examined this reaction and tried to determine the exact structure of the dimer. In the previous experiment (Scheme 1-7), 5,5'-unsubstitued-2,2'-bipyrrole 17 was used to condense with 2-formylpyrrole 18 in the presence of HBr. The reaction gave a 5-(pyrrol-2-yl)-dipyrromethene hydrobromide but a second mole of 2-formylpyrrole did not react further to give the corresponding l,l'-bis-(dipyrrinyl) dihydrobromide because of the general deactivation towards further electrophilic attack upon formation of a dipyrromethene unit.21 To construct a l,l'-bis-(dipyrrinyls) directly, we used a 5,5'-diformyl-2,2'-bipyrrole as a key synthetic intermediate. This compound, unlike 5,5'-unsubstituted bipyrroles which require the reduction of l,l'-bis-(dipyrrinyl) dihydrobromides before yielding the tetrapyrrolic compound 21,21 condensed with two equivalents of 2-unsubstituted pyrroles to directly give the desired ligand. This alternative strategy thus avoids the need for additional redox steps as detailed above. The synthetic route for preparation of complex 59 is shown in Scheme 1-14. Pyrrole 70 was oxidized reasonably by two equivalents of SO2CI2 to the corresponding 5-(dichloromethyl)pyrrole92 which was then hydrolyzed smoothly in crude form to the corresponding 2-formylpyrrole 71 upon treatment with saturated aqueous NaOAc solution. To protect the aldehyde, it was converted to 5-(cyanovinyl)pyrrole 72 by reacting with ethyl cyanoacetate ester.105 The catalytic hydrogenolysis of the 5-(cyanovinyl)pyrrole benzyl ester 72 was effected in the presence of 10% Pd-C catalyst in THF at room temperature and atmospheric pressure. This reduction required some care since the cyanovinyl group is also reducible, albeit at a far slower rate than the benzyl 55 59 Scheme 1-14. Synthesis of l,l'-bis-(dipyrrinyl) Zn(II) complex 59 a) 1.2 eq.S0 2 Cl 2 / CH 2 C1 2 , 1.5 h; 2. H 2 0 / NaOAc, heat; b) N C C H 2 C 0 2 C 2 H 5 / toluene, heat, 2 h; c) 1. H 2 /10% Pd/C,3 h; 2.12, Nal / H 2 0 , heat, 45 min, C1CH 2CH 2C1; d) Cu/DMF, 110°C, 3 h; e) K O H / N 2 , reflux, 3h; f) HBr / CH 2 C1 2 ; g) Zn(OAc) 2 ,CHCl 3 / MeOH, reflux, 3h. 56 ester. The hydrogen uptake slows dramatically after the first equivalent has been consumed, and the reaction must be terminated at this point to ensure maximum yield. 1 0 5 The resultant a-pyrrole carboxylic acid was converted into iodopyrrole according to the method described previously. The bipyrrole 74 was prepared via Ullman coupling reaction of iodopyrrole 73 in the presence of copper.1 0 6 Removal of the protecting group gave the corresponding 5,5'-diformyl-2,2'-bipyrrole 75. It was then reacted with an a-unsubstituted pyrrole to give bis-(dipyrrinyl) hydrobromide 54. It was unnecessary to isolate the ligand from solution. After consumption of mineral acid by addition of aqueous ammonia solution, a methanol solution of zinc acetate was added to the mixture. Small metallic green crystals precipitated slowly during stirring at room temperature. Refluxing significantly accelerated the formation of the precipitate but did not increase the yield. The precipitate was separated and recrystallized from CH 2 C1 2 / MeOH to give metallic green crystalline prisms of zinc complex 59. Mass spectrometry of complex showed 59 to have m/z = 1035, which corresponds to C6oH 7 2NgZn 2 , indicating that the compound consists of two zinc atoms and two ligands. In this reaction, no monomeric zinc complex was observed. High resolution mass spectrometry and microanalysis confirmed the molecular constitution. The proton N M R spectrum shows the absence of the N - H signal of the dipyrromethene moiety, which usually appears at low field, 11-13 ppm. A signal at 6.88 ppm corresponds to the hydrogen at the meso-position. 1 3 C N M R spectroscopy is fully consistent with the structure proposed above. Both ' H and 1 3 C N M R spectra attested to the high symmetry within the complex. 2.2.2 Crystal Structure Analysis of Complex 59 Slow diffusion of methanol into CH2CI2 solution of complex 59 provided plate green plates with a metallic luster, suitable for X-ray crystallography. An ORTEP representation of the structure is shown in Figure 1-18. Experimental details of the Figure 1-18. ORTEP representation of complex 59 (For clarity, hydrogen atoms are omitted.) 58 structure are listed in Table 1-6. Selected bond lengths and selected bond angles are listed in Table 1-7 and Table 1-8 (Appendix). The X-ray structure (Figure 1-18) shows 59 to crystallize as a dinuclear helix, satisfying the tetrahedral coordination geometry of each Zn 1 1 center. The four Zn-N bonds at one coordination center are 1.9956 A, 1.9964 A, 1.9812 A, 1.9804 A of length respectively. The bond angles of N-Zn-N are 106.30°, 95.51°, 113.30°, 95.56°. The two planar dipyrromethenes binding to the same zinc ion give a dihedral angles of 112°, which is much bigger than 89° in 60b. These data indicate the Zn-N tetrahedral coordination suffers significant distortion. Meanwhile, as a result of differing spacer length from 60b, two main differences occur. The angle between two flat dipyrromethenes in the same strand changes from 89° in complex 60b to 108° in complex 59. Correspondingly, the distance between two metal centers within each helix is different, being 4.33A in 60b and 4.89A in 59. This demonstrates that the spacers play an important role upon the exact parameters of these helical structures. The helical structure of this assembly seems surprising if we take into account that maximum derealization between two directly-linked dipyrromethene moiety would be achieved through co-planarity, which would not lead to the helical structure observed. However, steric hindrance between two p-methyl groups would force the two dipyrromethene planes away from co-planarity. Computer-modeling gave a structure of ligand 54 which exhibits the two dipyrromethene planes to be almost perpendicular (Figure 1-19). Therefore, it is reasonable that the dihedral angle adopted between the two dipyrromethene planes within the same strand is 108°. 59 H H H H A B Figure 1-19. two possible conformations of bis-(dipyrromethene) (A) represents coplanar conformation (B) represents perpendicular conformation The simulated structure of this compound is consistent with the observed X-ray structure. Some selective comparisons are shown in Table 1-9. Table 1-9. Selected structural parameters of X-ray and simulated structures of complexes 59. Complex 59 " (C) X-ray (D) model Bond Angle (°) Nl-Zn-N2 95.53 102.01 N l - Z n - N l ' 106.32 108.75 Distance (A) Zn-Zn 4.890 4.766 Torsional Angle(°) N2-C9-C9'-N2' -112.3 -110.6 60 2.3 FORMATION OF 1,2-BIS-(DIPYRRIN-1'-YL)ETHANE COBALT COMPLEX 61 Although complexes 60b and 59 both exhibit helical structures, their structural parameters are quite different. With the complex 60a, 60b (n = 1) and complex 59 (n = 0) in hand, our next investigation concerned ligands with larger, more flexible spacer, ethylene (n = 2). Using a-formylpyrrole 78, which was synthesized according to the method used previously, as the starting material, we synthesized l,2-bis-(5'-carbonylpyrrol-2'-yl)ethene 79 through a McMurry coupling reaction.106 The compounds formed consisted of a mixture of the (Z)- and (E)-diastereomers. For our studies, it was unnecessary to separate them. They were smoothly reduced by magnesium in the presence of Pd/C to give 1,2-bis-pyrrolylethane 80. After hydrolysis of the ethyl ester by heating in aqueous NaOH solution, the resulting carboxylic acid 81 was condensed with a-formylpyrrole 82 in the presence of HBr (48% in acetic acid) to form ligand 58 (Scheme 1-15). The compound was crystallized from methanol upon slow evaporation under reduced pressure. Analysis of this compound by mass spectrometry and ' H N M R confirmed the formation of the required ethylene bridged bis(dipyrrinyl) HBr salt. The hydrobromide salt 58 was dissolved in a 1:1 mixture of methanol and chloroform to give a dark red solution. This solution was treated with a methanolic solution of cobalt acetate and the reaction solution was stirred under an atmosphere of nitrogen. Two apolar red compounds were detected by T L C after 2 h. These two compounds have very similar Rf values in most eluting solvents thus rendering 61 61a 61b Scheme 1-15. Synthesis of 1,2-bis-(dipyrrinyl)ethane Co(II) complex 61 a) Pb(OAc) 4, HOAc/(Ac) 2 0, 85°, 1 h; b) 1. Zn/TiCl 4 , Py, THF, reflux, 5 h; 2. H 2 0 /NaHC0 3 ; c) Mg, Pb/C, 24 h; d ) l : l EtOH/NaOH (aq.); e) H B r / M e O H , 5 h; f) Co(OAc) 2 H 2 0 , MeOH/CHCl 3 , 3 h. 62 separation very difficult via chromatography. On the basis of molecular weight determinations (FAB-MS), the first compound (m/z 568), which is major product of the reaction, is regarded as monomer (Co(ligand)), and the second (m/z 1133) as dimer which comprises two ligands and two cobalt atoms (Co2(ligand)2). Attempts to crystallize the complex in various solvents failed to give X-ray quality crystals. We carried out molecular modeling of these two complexes using HyperChem Release 5.01. 1 0 3 It shows that the dimer 61a has a helical structure and the monomer 61b possesses a tetrahedral conformation (Figure 1-20). In both cases, the cobalt ion centers do not have perfect tetrahedral coordination geometry, and instead possess a distorted tetrahedral coordination geometry. Compared to ligands of complexes 60b and 59, the ethylene-bridged ligands have more flexibility through the spacer. The two dipyrromethene moieties of an individual ligand can fold around one cobalt ion to form a distorted tetrahedral monomer. It can also coordinate with two different cobalt ions, together with the other molecule of ligand, to form the helix. This rationalizes the formation of both 61a and 61b from this reaction. Figure 1-20. Simulated Models for 61a and 61b (For clarity, hydrogen atoms and substitutents are omitted.) A represents model for dimeric complex 61a B represents model for monomeric complex 61b 63 2.4. FORMATION AND CHARACTERIZATION OF TRINUCLEAR DUPLEX HELICAL COMPLEX 87 After having successfully prepared dinuclear helicates, we further extended our investigation to larger structures with poly-dipyrromethenes. So far, most studies on linear polypyrrolic compounds have been focused on the derivatives built from two, three or four pyrrole rings linked directly or by methylene, methines or other bridges. Systems of this kind with more than four pyrroles are rare and the linear polypyrrolic chain containing three or more dipyrromethene units are not reported in the literature. We thus need to develop a convenient and efficient method to synthesize these compounds. 5,5'-Dimethyl dipyrromethene 83 was synthesized from two equivalents of ethyl 2-pyrrolecarboxylate 77 as shown in Scheme 1-16. This compound is symmetrically substituted and, when treated with a slightly excess two equivalents of bromine in hot formic acid, was brominated on both a-methyl substituents to give 84. Mass spectrometry confirmed the dibromo substitution on the molecule. The *H N M R spectrum exhibits symmetry, which indicates the symmetrical methyl groups at both sides of dipyrromethene were brominated. Reaction of 5,5'-di(bromomethyl)-2,2'-dipyrromethene 84 with 2 equivalents of 5-unsubstituted 5'-alkylpyrromethene 85 in the presence of tin tetrachloride in methylene chloride afforded the six membered pyrrolic linear compound 86 which precipitated from solution. Filtration, followed by several washes with ether and methanol gave compound 86. Mass spectrometry of this compound was indicative of the formation of hexapyrrole 64 hydrobromide salt 86. ' H and 1 3 C N M R spectroscopy was most informative for the identification of 86. The observed number of signals of this compound indicates the high symmetry of the molecule. Three signals at low-field region of 12-14 ppm with equal integration represent the strongly hydrogen-bonded N H protons and are consistent with the inclusion of three dipyrromethene units in the molecule. Two methine proton signals with integration 1 and 2 respectively and one signal of methylene protons with integration of 4 are fully in agreement with the linear structure 86. Their chemical shift values are similar to those of their counterparts in biladiene-ac.9 UV-visible spectra show the molecule is a highly conjugated system. Hexapyrrole 77 dissolved in a 1:1 mixed solution of C H 2 C l 2 / M e O H was treated with Zn(OAc) 2-2H 20 in methanol. A red metallic compound 86 was obtained after chromotography over silica gel, eluting with 1:9 ethyl acetate:pet. ether (35 - 60°C). It was recystallized from CHCI3 and methanol to afford a metallic red crystalline solid. LS-EI mass spectrometry shows the product to have molecular mass of 1664, which corresponds to the trinuclear bis-ligand complex (Zmligand ratio of 3:2). Structural simulation as before indicates that it also exists as a helical double-stranded structure. High resolution Mass Spectometry (HR-LSIMS) gives a molecular mass of 1662.7823 and, together with microanalysis, suggests the molecular stoichiometry to be C98Hi 2 2 Ni 2 Zn3, which in consistence with the structure 86. The *H N M R spectrum of 87 (Figure 1-21) is reminiscent of dinuclear zinc complex 60a and thus, supports the proposed helical structure, namely tri-nuclear array 65 Scheme 1-16. Synthesis of trinuclear duplex helicate 87 a) HBr/HCOOH, heat, 3h; b) Br 2 /HCOOH; c) SnCl4/CH 2 Cl 2 , 12h; d) Zn(OAc) 2 in M e O H / C H 2 C l 2 of metals held together by two helically arranged methylene-bridged dipyrromethenes. The lack of N H low-field signals is supportive of coordination between zinc ions and pyrrolic nitrogens. The spectrum also shows the up-field shift of meso-methylene protons. However, these protons experience further up-field chemical shift (3.54 ppm in complex 87 and 3.90 ppm in 60a). This shift indicates that the helical structure of 87 is more rigid so that hyperconjugation across the methylene bridge is more unlikely to take place. The other characteristic of this signal is that the signal is split into A B system the coupling constant of which is 18 Hz. This indicates that the two protons of bridging methylene group are in different molecular environments. In the lH N M R spectrum, we 'i.S ' ' Y.O ' 6.5 ' " ' 6.0 5.5. 5.0 ' 4 5 ' ' 4.0 '." 3.5 3!o 2'.5 ' 2'.0 ' \'.5 lit) .5 6 Figure 1-21. ' H N M R spectrum (CDC13) of complex 87 observed two kinds of mesomethine proton signals which corresponded to ones in the central and side dipyrromethenes of the linear ligand strand. Their chemical shifts are quite different. The one due to the side dipyrromethenes is seen at 6.62 ppm and the other which is from the central dipyrromethene is observed at 6.12 ppm. The optical spectrum of complex 87 is shown in Figure 1-22. The three-band pattern is analogous to that of biladiene zinc complex 60a, but with a 15 nm hypsochromic shift of the mid band and slightly higher extinction coefficients. 6.00 H 350 450 550 650 wavelength (nm) Figure 1-22. UV-visible spectrum of complex 87 in methylene chloride Slow diffusion of methanol into a methylene chloride solution of 87 provided plate green crystals with metallic luster, which were used for X-ray crystallography. Due to solvent disorder, the R-value observed was high. The program SQUEEZE/PLATON was used to correct the data for disordered solvent.107 An ORTEP representation of the structure is shown in Figure 1-23. Experimental details of the structure are listed in Table 68 1-10. The final atomic coordinates are listed in Table 1-11 and selected properties in Table 1-12 and Table 1-13. The X-ray structure of complex 87 shows a linear arrangement of the three zinc atoms, and the two ligand strands fold around the axis to form a double-stranded helical geometry. This conforms to the structure obtained from molecular modeling. Resembling complex 59 and 60b, the two ligands of complex 87 are twisted around the linking methylene bridge as they coil around the linear zinc axis. As a result, the ligands are Figure 1-23. ORTEP representation of complex 87 69 divided into 3 folds and each fold is constituted by a flat dipyrromethene subunit which chelates the zinc11 by two Zn-N bonds. The structural properties of this complex are nearer to those of complex 60b than complex 59. The distances of Zni-Zn 2 and Zn 2 -Zn 3 are 4.38 A and 4.17 A respectively. The angle between two adjacent dipyrromethenes in the same strand is 85° in complex 87, which is close to that of 89° in complex 60b, rather than 59 (108°). 70 2.5 FORMATION AND CHARACTERIZATION OF BIS-(1,3,7,9-TETRAMETHYL-8-ETHYL-DIPYRRIN-2-YL) SULFIDE ZINC(II) COMPLEX 92 We have also investigated a (3,(3'-linked dipyrromethene ligand-metal complex. We chose &w-(l,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-yl) sulfide 91 as the candidate of ligands because of the ease of preparation of such compounds. These compounds can be obtained either from condensation of bis(5-carboxy-2,4-dimethyl-pyrrol-3-yl)sulfide and a 2-formylpyrrole or from condensation of bis(5-formyl-2,4-dimethylpyrrol-3-yl)sulfide and a 2-carboxypyrrole. Both synthetic routes work effectively and efficiently.1 0 8 The latter method was used in our study (Scheme 1-17). When the starting material a,a'-substituted, (3-unsubstituted pyrrole 88 was treated with sulfur dichloride at low temperature (-70 °C), the substitution occurred at the (3-position to smoothly form bis-(5-1OR formyl-2,4-dimethylpyrrol-3-yl) sulfide 89. This reaction was performed in high yield (95%) and the formyl group remained unaffected. The precipitate resulting from the addition of methanol was separated from the solution by suction filtration. It was readily condensed with a a-carboxypyrrole, which was obtained by hydrolysis of f-butyl a-pyrrolecarboxylate 90 in TFA in the presence of HBr. Bis-(l,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-yl)sulfide dihydrobromide salt 91 was thus obtained. This compound precipitated from the solution as a dark red powder which was sufficiently pure for analysis and the complexation reaction. 71 Scheme 1-17. Synthetic Route to Bis(l,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-yl) Sulfide Zinc Complex 92 a) SC1 2/CH 2C1 2, -70 °C; b) TFA, HBr/HOAc, 3h; c) Zn(OAc) 2/MeOH, lh 72 Dihydrobromide salt 91 was reacted with zinc(II) acetate hydrate in methanol with a small amount of triethylamine to form Zn-complex 92. After separation from the reaction mixture by chromotography, the Zn complex was recrystallized to give a dark red crystal from methanol. Fast Atom Bombardment mass spectrometry suggests that the complex 92 is a dimer containing two ligands and two metal centers. Slow evaporation of C H C I 3 from C H C I 3 and methanol solution provides dark red crystals with metallic luster. These crystals were suitable for X-ray crystallography analysis. An ORTEP representation of the structure (Figure 1-24) shows that it exists in crystals as a helical structure with C2 symmetry. Experimental details of the structure are listed in Table 1-14. Selected bond lengths and bond angles are listed in Table 1-15 and Table 1-16 respectively. Figure 1-24. ORTEP representation of complex 92 (For Clarity, hydrogen atoms and substituents are omitted) The X-ray structure of complex 92 shows the molecule has duplex helical structure containing two zinc atoms and two bis-(l,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-73 yl) sulfide ligands. The bond lengths for Zn-N are 1.972(5), 1.976(4), 1.976(4) and 1.991(4)A and bond angles for N-Zn-N 121.6(2), 96.4(2), 118.7(2) and 112.4(2)°. These data shows the central zinc atoms have distorted tetrahedral geometry. The ligand conformation, like that of complex 6 0 b , is divided into two-fold at sulfur atom. The bond angle for N(8)-S-N(ll) is 101.7(2). The molecular structure has a symmetry with the zinc atoms positioned on a crystallographic axis. 74 2.6 C U R R E N T D E V E L O P M E N T S Recently, Thompson in our group also reported the self-assembly of metal ion and ligands comprising two dipyrromethene units linked at the /^-position to form circular trimeric zinc ion and cobalt ion complexes. 1 0 9 When a solution of Zn(OAc)2 and NaOAc in MeOH was added to a solution of 93110 (Figure 1-25) in CHCI3 a single trimeric 94n=3 100 h=3, m=2 95 n=4 101 n=4, m=2 96n=6 102 n=6, m=1and2 97n=8 103 n=8, m=1 98n=12 104n=12, m=1 Figure 1-25. (3-Linked multi-dipyrromethene ligands complex 99 [Zn3(ligand)3] was formed. The X-ray structure of trinuclear complex 99 (Figure 1-26) shows that it crystallizes in a triangular fashion, with the ligands linking the three metal centers overlapping, in a progressive manner: in each ligand, one dipyrromethene binding unit lies above the averaged plane of the molecule, whilst the other dipyrromethene unit lies below the averaged plane. The distances between the three 75 metal centers are 9.27, 9.34 and 9.36 A. Each metal center has distorted tetrahedral geometry in order to permit the triangular arrangement. The six planar dipyrromethene sub-units lie at 63.6, 85.6 and 68.7° to their partner dipyrromethene unit within the same ligand. C 2 5 C 2 6 C75 Figure 1-26. X-ray structure of complex 99 Similarly, 93 was reacted with Co(OAc)2 to give the corresponding cobalt trimer, with molecular mass of 1449, again in high yield. Several reaction conditions were applied to these preparations including temperature variations (-10 °C to reflux in methanol) and different lengths of addition period (metal salts added in one portion, to over 24 hours). In each and every case, the same trimers were formed in high yield, suggesting them to be both the kinetic and thermodynamic products. 76 To investigate the role that the spacers play in the assembly of architectures, several hydrobromide salts 100-104 with varying hydrocarbon spacer lengths (n = 3, 4, 6, 8, 12) between the ^-positions were subjected to standard complexation reactions with Zn(OAc)2. It was observed that self-assembly of zinc ion with 94 and 95 (n = 3, 4) gave almost exclusively dimeric products. As the length of the hydrocarbon chain between the pyrrole moiety increases (96, n = 6), monomer begins to form as a fraction of products. When the length of the bridging chain is more than 6 carbons, formation of monomer dominates. Molecular modeling rationalizes this, as the longer spacer length allows the dipyrromethene units to fold back against each other, thus fulfilling the tetrahedral geometry requirements for binding to the metal ion. In the case of the ligand derived from salt 93 (n = 0) this folding is clearly not allowed and the trimer is formed. For 94 and 95 (n = 3 and 4 respectively) the spacer chain length permits sufficient folding for the corresponding dimers to be formed, but when n = 6 or greater, folding is sufficient to allow the monomer to form 2.7 SUMMARY AND FUTURE WORK In our study, several transition metal tetrapyrrole complexes 59-61 were prepared. According to the terminology developed by Lehn, '" these assemblies amount to a "tetrahedral reading by metal ion, Zn 1 1 or Co 1 1 of the molecular information stored in the strands". The ligands 56-58 used in our study differ mainly in the length of spacers. Zn" complex 59, whose ligand is directly linked bis-dipyrromethene (n = 0), has a 2:2 Zn: ligand helical structure. This structure was driven by minimization of steric constrain 77 seen both within and between ligands, and zinc tetrahedral coordination. The tendency of zinc ion to exhibit tetrahedral geometry led the linear ligand to bend a little at the methylene position. As a counteraction, the zinc ion compromised to give a distorted tetrahedral geometry due to the rigidity at the spacer. In the helical structure of cobalt complex 6 0 b , the spacer C H 2 is more flexible than that of complex 5 9 . Thus, the central metal ion Co" possesses an almost perfect tetrahedral geometry. As a result, the distance between metal ions is relatively short (4.33 A) in comparison to that of complex 5 9 (4.87 A). When spacer was changed to ethylene (n = 2), complexation was more complicated, because the two bipyrromethene units of the ligand, on one hand, can wrap about two different metal ions to undergo helication and on the other hand, they can turn around to wrap up the same metal ion to give a monomer. In conclusion, the spacer is one of the key factors to determine and impact the structures of these compounds during their assembling. The methylene spacer in our examined polypyrromethene system is rigid enough to prevent the coordination of two bonding units of one strand to the same metal ion, but flexible enough to wrap about the metal ions to produce helical complexes. Based on this observation, we extended our study to trinuclear complexation of polypyrromethene in which the dipyrromethene units are linked by methylene. We synthesized the linear tris-didentate ligand 8 6 in a convenient method and in moderate yield. The resulting Zn 1 1 complex has a 3:2 metal ion: ligand helical structure in which three metal ions are in the same axis. In the study of p-linked polydipyrromethene system, the same conclusion can be drawn. However, the (3-direct linked complex has a circular structure rather than a helix. 78 As a result, zinc ion tetrahedral coordination can be satisfied without sacrifice of losing linearity of the ligand. In our study, we also observe that the heteroatomic spacer has little effect on the structures of self-assembled complexes. In order to study the structures of transition metal complexes of this kind, we also use molecular modeling when X-ray crystal structures were unavailable. The structures obtained by simulation of complexes 59, 60b and 87 are extremely similar to their X-ray structures, as shown by comparison of the structural parameters in Table 4 and 8. The strong correlation between the actual (X-ray) and simulated parameters indicates the usefulness of this type of modeling for these complexes. In summary, our research demonstrates the usefulness of linked dipyrromethene ligands in the self-assembled preparation of uncharged novel architectures. The current research work is focussed towards the construction of novel supramolecular helicates using dipyrromethene ligands. However, there are many topics that are left untouched. For example, can we synthesize triple helicates by using polydipyrromethenes and octahedral metal ions? Can we use the polydipyrromethene-like compounds to construct the homostranded heterotopic helicates? Meanwhile, the polydipyrromethene metal complexation also promises us a perspective possibility to construct novel grids, ladders, cylinders and other structurally organized supramolecular architectures. We have no doubts that, in the future, further developments toward the self-assembly of these novel intriguing supramolecular structures will be made. 79 Recently numerous other intriguing assemblies have been made and these novel nanostructures present unique approaches to new materials. As our understanding of structural programming increases and our control over the synthesis of materials improves, we optimistically anticipate that the ability to hold metal ions at close proximity to each other by highly colored conjugated ligands will result in exciting new electronic and photonic applications. 80 3. E X P E R I M E N T A L 3.1 PREPARATION The reagents were purchased from Aldrich Chemical Co. and were used without further purification. CH2CI2 and CHCI3 were distilled from calcium hydride. Anhydrous Na 2 S04 was used to dry the organic solutions during workups. Flash column chromatography was performed using 230-400 mesh silica gel (Merck). Analytical thin-layer chromatography was done on pre-coated silica gel aluminum plates containing a fluorescent indicator (GF-254 Merck). Melting points were determined on a Bristol Hot Stage and stand uncorrected. *H N M R spectra were recorded at 200 or 400 MHz; l 3 C N M R spectra at 75 MHz. CDCI3 was used as solvents. UV-Vis spectra were recorded using a HP8452A photo diode array spectrophotometer (instrumental precision ± 2 nm) in the solvents indicated. The starting pyrroles 62, 69, 70, 76, 77, 82, 88 and 90 and dipyrromethene hydrobromide 85, were previously prepared in our lab and were used without any further purification. 2,7,8,12,13,18-Hexaethyl-l,3,17,19-tetramethylbiladiene-5,15 hydrobromide (57) 2,7,8,12,13,18-Hexaethyl-l,3,17,19-tetramethylbiladiene-5,15 hydrobromide 57 was prepared according to the procedure described by Johnson.9 m.p. >250°; ' H N M R (200 MHz, CDC13): 5 13.20 (br s, 2H, 2 X NH), 13.04 (br s, 2H, 2 X NH), 7.02 (s, 2H, 2 X meso-CH), 5.06 (s, 2H, meso-CH2), 2.70 (q, J = 7.3 Hz, 4H, 2 X C//2CH3), 2.60 (s, 6H, 2 X CHj), 2.44 (q, J = 7.3 Hz, 4H, 2 X C H 2 C H 3 ) , 2.38 (q, J = 81 7.3 Hz, 4H, 2 X C/72CH3), 2.22 (s, 6H, 2 X CH 3), 1.10 (t, J = 7.3 Hz, 6H, 2 X CH 3), 1.03 (t, J = 7.3 Hz, 6H, 2 X CH 3), 0.96 (t, J = 7.3 Hz, 6H, 2 X CH 3); LR-MS (EI) m/z = 525 ([M-2Br-H]+); Anal. Calcd for C 3 5H5oBr 2N4: C, 61.22; H, 7.29; N, 8.16; Found: C, 61.34; H, 7.12; N, 8.00; UV-visible ( 5 % MeOH/CHCl3), X m a x. (log e): 288 (3.35), 374 (4.08), 454 (4.35), and 522 (5.31) nm. Cobalt(II) complex (60b) 2,7,8,12,13,18-Hexaethyl-l,3,17,19-tetramethylbiladiene-5,15 hydrobromide 57 (0.1 g, 0.15 mmol) was dissolved in 1:1 CHC13 and methanol (20 mL). The red solution was purged with N 2 for 10 min. Co(OAc)2-4H20 (50 mg, 0.2 mmol) dissolved in methanol (5 mL) was added. The reaction mixture was allowed to stir at room temperature under N 2 for 5 h, whereby TLC (CH2C12) showed a new red spot (Rf = 0.95) and absence of starting material (Rf = 0.32). The solvent was removed in vacuo and the crude product purified by chromatography, using CH 2C1 2 as eluant. The first elute was collected and evaporated to dryness^ It was recrystallized from CH 2Cl 2/MeOH to afford product of green metallic luster (0.021 g, 24 % ) . A crystal for X-ray crystallography was grown by methanol diffusion into a methylene chloride solution of 60b. Rf (CH2C12) = 0.95; UV-visible (CH2C12) X m a x (log e) 366 (4.12), 488 (5.31), 526 (5.33); LR-MS (EI) m/z = 1162(M+); Anal. Calcd for C 7 0H 9 2N 8Co 2: C, 72.29; H, 7.92; N, 9.60; Found: C, 72.53; H, 7.96; N, 9.44. Zn (II) complex (60a) 2,7,8,12,13,18-Hexaethyl-1,3,17,19-tetramethylbiladiene-5,15 hydrobromide 57 was reacted with Zn(OAc) 2.2H 20 under the same condition as for 60b. After chromotography 8 2 and crystallization, the Zn complex was obtained as green metallic shiny prisms (0.039 g, 33%). Rf (CH2C12) = 0.95; 'H NMR (200 MHz, CDC13): 5 6.63 (s, 2H, 2 X meso-CH), 3.90 (s, 2H, meso-CH2), 2.20-2.40 (m, 8H, 4 X C# 2CH 3), 2.15 (s, 6H, 2 X C7f3), 2.08 (q, J = 7.8 Hz, 4H, 2 X CH 2CH 3), 2.06 (s, 6H, 2 X Ctf 3), 1.08 (t, J = 7.8 Hz, 6H, 2 X CH 2CH 3), 0.92 (t, J = 7.8 Hz, 6H, 2 X CH 2Ctf 3), 0.88 (t, J = 7.8 Hz, 6H, 2 X CH 2CH 3); 1 3C NMR (75 MHz, CDC13): 5 157.04, 153.83, 143.62, 136.54, 135.58, 134.88, 129.79, 128.72, 120.60, 30.62, 18.21, 18.13, 17.97, 17.63, 16.19, 15.02, 13.06, 9.82; LR-MS (EI) m/z = 1174 (M +); UV-visible(CH 2Cl 2) X m a x (log e): 388 (3.71), 466 (5.14), 524 (5.08); Anal. Cald. for C 7 oH92N 8Zn 2: C, 71.55; H, 7.83; N, 9.54. found: C, 71.46; H, 7.93; N, 9.66. 4,4'-Diethyl-5,5'-diformyl-3,3'-dimethyl-2,2'-bipyrrole(75) 4,4'-Diethyl-5,5'-dialdehyde-3,3'-dimethyl-2,2'-bipyrrole 75 was prepared from 70 according to the procedure described by Guilard. 1 0 6 For analysis, 0.5g of above crude compound was taken up into hot ethanol and recrystallized to give pure bipyrrole 75. 'H NMR (200 MHz, CDC13): 6 9.86 (s, br, 2H, NH), 9.64 (s, 2H, CM)), 2.76 (t, J = 7.5 Hz, 4H, Cf7 2CH 3), 2.09 (s, 6H, CH3), 1.11 (t, J = 7.5 Hz, 6H, CH 2CH 3); LR-MS (EI) m/z = 272 (M +). Anal. Cald for Ci 6 H2oN 20 2: C, 70.56; H, 7.40; N, 10.29; found: C, 70.23; H, 7.72; N, 10.18. Zinc(II) complex (59) 83 4,4'-Diethyl-5,5'-diformyl-3,3'-dimethyl-2,2'-bipyrrole 75 (0.272 g, 1 mmol) and 3,4,5-trimethylpyrrole (0.218 g, 2 mmol) were dissolved in methylene chloride (25 mL). To this was added 45 % HBr in acetic acid (0.5 mL). The yellow solution turned to green immediately. The mixture was stirred for 30 min and the solvent was then removed in vacuo. The residue was dissolved in 1:1 MeOH:CHCl3 (20 mL) and zinc acetate dihydrate (0.210 g, 1 mmol) in a minimal amount of methanol was added. The reaction mixture was refluxed for 3 h. The product of green metallic luster precipitated upon cooling and was separated by suction filtration (0.11 g, 21%). The crystal for X-ray crystallography was grown by slow diffusion of methanol into methylene chloride solution. ' H N M R (400 MHz, CDC13): 5 6.88 (s, 4H, 4 X meso-CH), 2.58 (q, 7 = 7.5 Hz, 8H, 4 X C H 2 C H 3 ) , 2.04 (s, 12H, 4 X CH 3 ) , 1.90 (s, 12H, 4 X CH 3 ) , 1.75 (s, 12H, 4 X CH 3 ) , 1.68 (s, 12H, 4 X CH 3 ) , 1.12 (t, / = 7.5 Hz, 12H, 4 X CH 2 C7/ 3 ); LR-MS (EI) m/z = 1035 (M + ); Anal. Calcd for C 6oH72N 8Zn 2: C, 69.59; H , 6.96; N , 10.82. Found: C, 69.32; H , 7.06; N , 10.42; UV-visible (CH 2C1 2) ? i m a x (relative intensity) 380 (1.0), 440 (1.84), 596 (1.20). l ,2-Bis-(5-carboxyl-3-ethyl-4-methylpyrrolyl)-ethane (81) l,2-Bis-(5-ethoxycarbonyl-3-ethyl-4-methylpyrrolyl)-ethane 80* was hydrolyzed to give 81 (0.42g, 96%) according to the procedure described by Falk." 2 'H N M R (200 MHz, CDC13): 5 8.67 (br s, 2H, 2 X NH), 2.80 (s, 4H, CH2CH2), 2.39 (q, J Prepared by Jack Chow in our lab according to the procedure described by Falk 1 1 2 84 = 7.2 Hz, 4H, 2 X C//2CH3), 2.01 (s, 6H, 2 X pyrrole-Cr73), 1.03 (t, J = 7.2 Hz, 6H, 2 X CH2C//3); LR-MS (EI) m/z = 332 (M+). l,2-Bis-(2',8'-diethyl-3',7',9'-trimethyldipyrrin-l'-yl)-ethane dihydrobromide (58) l,2-Bis-(5-carboxyl-3-ethyl-4-methylpyrrolyl)-ethane 81 (0.41 g, 1.23mmol) and4-ethyl-2-formyl-3,5-dimethyl pyrrole (0.37g, 2.47 mmol) 82 was dissolved in 10 mL of methanol. Under vigorously stirring, 0.5 mL of 48% HBr aqueous solution was added dropwise. After stirring for 5 hours, the red precipitate was filtered out and washed with 5 mL of methanol containing 2-3 drops of HBr solution and diethyl ether ( 2 X 5 mL) respectively. Analytically pure red crystals were obtained (0.38 g, 46%). 'H NMR (200 MHz, CDC13): 5 13.18 (br s, 2H, NH), 12.90 (br s, 2H, NH), 7.04 (s, 2H, meso-CH), 3.46 (s, 4H, meso-Ctf2), 2.78 (q, J = 7.5 Hz, 4H, C/72CH3), 2.64 (s, 6H, CH 3), 2.38(q, / = 7.5 Hz, 4H, CH2CR3), 2.24 (s, 3H, CH3), 2.20 (s, 3H, C/73),1.10 (t, / = 7.5 Hz, 6H, CH 3), 1-08 (t, J = 7.5 Hz, 6H, CH3); LR-MS (EI) m/z = 512 ([M-2Br]+). UV-visible (CH2C12) X m a x (relative intensity) 370 (0.228), 464 (0.99), 502 (1.00). Anal. Cald for C 3 4 H48N 4Br 2: C, 60.71; H, 7.14; N, 8.33; Found: C, 60.46; H, 7.15; N, 8.02. Co(II) complex (61a and 61b) Compound 58 (0.2 g, 0.3 mmol) was suspended in 1:1 MeOH:CH 2Cl 2 (10 mL). Co(OAc) 2-2H 20 (0.65 g, 0.31 mmol) in 3 mL of methanol was added. With stirring, this mixture was refluxed for 3 hours. TLC showed that a new bright red spot appeared. It was separated from the reaction mixture by chromotography. Further separation by 85 preparative T L C (methylene chloride: methanol 99.5:0.5) gave two compounds whose Rf were very close. The first compound: 61a (0.018g, 10.6%) LSIMS m/z: 1133; UV-visible (CH 2C1 2) X m a x (relative intensity) 382 (0.28), 504 (0.56), 524(1.00). The second compound: 61b (0.012g, 7%) L R - M S (LSIMS) m/z = 568; UV-visible (CH2C12) lmax (relative intensity) 376 (0.28), 520 (1.00). 3,3'-Diethyl-2,2'4,4'- tetramethyl-5,5'-dipyrromethene hydrobromide (83) Ethyl 4-ethyl-3,5-dimethylpyrrole-2-carboxylate 77 was dissolved in hot 90 %.formic acid (5 mL), and 48 % hydrobromic acid (4 mL) was added. The mixture was heated on a steam bath for 3 h, then allowed to cool. The product was separated and washed with water, methanol and ether. The hydrobromide crystallized from chloroform-petroleum as dark red prisms. ' H N M R (200 MHz, CDC13): 8 13.21 (br s, 2H, 2 X NH), 6.98 (s, 1H, meso-CH), 2.87 (q, J = 7.5 Hz, 4H, 2 X C / / 2 CH 3 ) , 2.37 (s, 6H, 2 X C# 3), 2.23 (s, 6H, 2 X CH3), 1.19 (t, J = 7.5 Hz, 6H, 2 X C H 2 C H 3 ) ; LR-MS (+LSIMS) m/z = 256 ([M-Br-H] +). Anal. Cald for C i 7 H 2 5 N 2 B r : C, 60.54; H , 7.47; N , 8.31; Found: C, 60.38; H , 7.35; N , 8.33. UV-visible (CH 2C1 2) ? i m a x ( l o g 8) 286 (3.26), 362 (3.69), 486 (4.79). 86 2,2'-Dibromomethyl-3,3'-diethyI-4,4'-dimethyl-5,5'-dipyrromethene hydrobromide (84) 3,3'-Diethyl-2,2'4,4'- tetramethyl-5,5'-dipyrromethene hydrobromide 83 (1.5 g, 4.4 mmol) was suspended in 96% formic acid (15 mL). To this was added bromine (0.5 mL, 2.2 mmol). The mixture, protected from moisture, was heated on a steam bath for 15 min, then was allowed to cool. The precipitate was filtered off and washed with acetic acid and then diethyl ether. The dipyrromethene hydrobromide was recrystallized from chloroform -light petroleum as red solid (1.47 g, 82%). f H N M R (200 MHz, CDC13): 5 13.33 (br s, 2H, 2 X NH), 7.18 (s, 1H, meso-CH), 4.92 (s, 4H, 2 X CH 2 Br), 2.67 (q, J =7.5 Hz, 4H, 2 X C / / 2 CH 3 ) , 2.22 (s, 6H, 2 X CH 3 ) , 1.28 (t, / = 7.5 Hz, 6H, 2 X C H 2 C H 3 ) , m/z (+LSIMS): 414 ([M-Br-H]+), 334 ([M-2Br-H]+). U V -visible (CH 2C1 2) X r a a x(loge) 386 (3.35), 512 (4.69). Trihydrobromide salt (86) 5,5'-Di-(bromomethyl)-4,4'-diethyl-3,3'-dimethyl-dipyrromethene hydrobromide 84 (0.495 g, 1 mmol) and 3',4'-diethyl-3,4,5'-trimethyl-dipyrromethene hydrobromide 85 (0.65 g, 2 mmol) were dissolved in dry methylene chloride (50 mL). To this was carefully added tin (IV) chloride (2 mL). The reaction mixture was stirred for 12 h at room temperature. The solvent was removed in vacuo. 4:1 Methanol : 48 % HBr in acetic acid (100 mL) was then added and the mixture kept at 0 °C for 2 h. The red precipitate was filtered off and rinsed with diethyl ether to give 86 (0.81 g, 84%). ' H N M R (200 MHz, CDC13): 5 13.36 (br s, 2H, 2 X N/J), 13.20 (br s, 2H, 2 X NH), 13.05 (br s, 2H, 2 X NH), 7.12(s, 1 H, meso-CH), 7.02 (s, 2H, 2 X meso-CH), 5.16 (s, 4H, 2 X meso-C//2), 2.80 (s, 6H, 2 X CH3), 2.72 (q, J = 7.4 Hz, 4H, 2 X Gr7 2CH 3), 2.50 87 (q, J = 7.5 Hz, 4H, 2 X C# 2 CH 3 ) , 2.42 (q, J = 7.5 Hz, 4H, 2 X C/7 2 CH 3 ), 2.30 (s, 6H, 2 X CH3), 2.20 (s, 6H, 2 X CH3), 1.98 (s, 6H, 2 X CH3), 1.20 (t, J = 7.5 Hz, 6H, 2 X CH 2 C# 3 ) , 1.08 (t, / = 7.5 Hz, 6H, 2 X C H 2 C / / 3 ) , 0.78 (t, 7 =7.5 Hz, 6H, 2 X CH 2 C/7 3 ) ; LS-MS (+LSTJVIS) m/z = 819 ([M-2Br-H]+), 737 ([M-3Br-2H]+). UV-visible (CH 2C1 2) m^axOoge) 378 (3.96), 446 (4.49), 490 (4.59), 548 (4.66). Zinc(II) complex (87) The hydrobromide salt 86 (0.2 g, 0.2 mmol) was dissolved in 30 mL of methanol. To this was added a solution of zinc acetate dihydrate (0.09 g, 0.4 mmol) in 1:1 M e O H : C H 2 C l 2 (5 mL). The reaction mixture was refluxed for 8 h. The solvent was removed in vacuo and the residue charged onto a silica gel column eluting with 1:9 ethyl acetate:pet. ether (35 - 60°C). The first orange band was collected and the solvent removed in vacuo. The pure product (0.022 g, 13 %) of red metallic luster was obtained upon recrystallization from CH 2 Cl 2 /MeOH. ' H N M R (400 MHz, CDC13): 5 6.62 (s, 4H, 4 X meso-Cfl), 6.12 (s, 2H, 2 X meso-CH), 1.62 (m, 8H, 4 X CH2CU3), 3.54 (AB system, J = 18 Hz, 8H, 4 X meso-CH2), 2.54 (q, J = 7.5 Hz, 8H, 4 X C/7 2CH 3), 2.22 (m, 8H, 4 X C7/ 2 CH 3 ) , 1.98 (s, 12H, 4 X CH 3 ) , 1.86 (s, 12H, 4 X CH3), 1.48 (s, 12H, 4 X CH3), 1.45 (s, 12H, 4 X CH3), 1.16 (t, J = 7.6 Hz, 12H, 4 X CH 2C/7 3),0.95 (t, J = 7.6 Hz, 12H, 4 X CH 2 C/7 3 ) , 0.78 (t, J = 7.6 Hz, 12H, 4 X CH 2 C/7 3 ) ; l 3 C N M R (75Hz, CDC13) 5 155.61, 152.79, 141.60, 136.29, 135.87, 135.80, 135.70, 134.44, 130.32, 128.71, 123.05, 120.53, 119.81, 29.72, 18.03, 17.80, 17.48, 16.79, 15.67, 14.85, 13.28, 10.10, 9.99, 9.16; UV-visible (CH 2C1 2) Xm a x(loge) 380 (4.63), 454 (5.22), 522 (5.33); LR-MS (+LSTMS) m/z = 1663 (M + ); HR-MS (+LSIMS) m/z 88 Calcd. for C 9 8 H 1 2 2 N 1 2 Z n 3 1662.7823, found 1662.7839. Anal. Calcd for C 9 8 H i 2 2 Ni 2 Zn 3 -6H 2 0: C, 66.40; H, 7.56; N , 9.49; Found: C, 66.31; FL7.35; N.9.39. Zn(II) complex (92) Bis(l,3,7,9-tetramethyl-8-ethyl-dipyrrin-2-yl)sulfide dihydrobromide 91* (2 g, 3.2 mmol) was dissolved in a 1:1 mixture of methylene chloride and methanol (20 mL). Zinc acetate hydrate (1 g, 4.5 mmol) in methanol (5 mL) was added. After stirring the mixture for 1 h at room temperature, the solvent was removed in vacuo. The residue was charged onto a short silica gel column and eluted with methylene chloride. The first red band was collected and solvent removed in vacuo to give dark red crystals (1.11 g, 63 %). ' H N M R (200MHZ, CDC1 3): 6.50 (s, 2H, 2 X meso-CH), 2.00 (s, 6H, 2 X CH3), 1.85 (q, J = 7.32 Hz, 4H, 2 X C/73), 1.80 (s, 6H, 2 X CH 3 ) , 1.55 (s, 6H, 2 X CH3), 1.45 (s, 6H, 2 X CH3), 0.85 (t, J = 7.32 Hz, 6H, 2 X CH3); UV-visible (CH 2C1 2): X (log e) 374 (4.31), 480 (5.11), 515 (5.27) nm. LR-MS (FAB) m/z = 1100 (M++2). Anal. Calcd for C 6 0 H 7 2 N 8 S 2 Z n 2 : C, 65.50; H, 6.60; N , 10.19; Found: C, 65.24; H , 6.69; N , 9.88. Prepared by Dr. Qingqi Chen in our lab. 89 3.2 X - R A Y C R Y S T A L L O G R A P H I C A N A L Y S I S OF 60b, 59, 87 A N D 92 Table 1-1. Crystallographic data for 60b Compound 60b formula C70H92C02N8 fw 1163.42 color, habit green, plate crystal size, mm 0.04x0.18x0.25 crystal system triclinic space group PI (#2) a, A 12.306(2) b, A 14.666(3) c, A 19.939(6) oc° 86.751(7) p° 76.854(4) 7° 66.157(2) V , A 3 3882.7(8) Z 2 D c a l c , g/cm3 1.206 F(000) 1244.00 radiation MoKa(A=0.71069 A) p( MoKa) , cm"1 5.65 Table 1-2. Atomic coordinates and Bea for 60b atom X y z Beq C o l 0.10340(7) 0.23793(7) 0.25941(5) 2.64(3) Co2 0.49054(7) 0.13120(7) 0.24859(5) 2.79(3) NI 0.0287(4) 0.2547(4) 0.1784(3) 2.14(14) N2 0.1730(4) 0.3400(4) 0.2388(3) 2.3(2) N3 0.3658(4) 0.1567(4) 0.3361(3) 2.5(2) N4 0.5922(4) -0.0081(4) 0.2666(3) 2.8(2) N5 -0.0304(4) 0.2849(4) 0.3438(3) 2.03(14) N6 0.1680(4) 0.0923(4) 0.2779(3) 2.4(2) N7 0.4234(4) 0.1432(4) 0.1656(3) 2.07(14) N8 0.5571(4) 0.2342(4) 0.2235(3) 2.10(14) C l -0.0297(5) 0.2072(5) 0.1553(4) 2.4(2) C2 -0.0727(5) 0.2514(5) 0.0950(4) 2.3(2) C3 -0.0343(5) 0.3284(5) 0.0808(4) 2.6(2) C4 0.0270(5) 0.3324(5) 0.1320(4) 2.1(2) C5 0.0801(5) 0.3977(5) 0.1377(4) 2.3(2) C6 0.1413(5) 0.4052(5) 0.1853(4) 2.3(2) C7 0.1831(5) 0.4822(5) 0.1893(4) 2.4(2) C8 0.2382(5) 0.4641(5) 0.2443(4) 2.6(2) C9 0.2256(5) 0.3779(5) 0.2752(4) 2.0(2) CIO 0.2639(5) 0.3417(5) 0.3404(4) 2.6(2) C l l 0.2718(6) 0.2422(5) 0.3664(4) 2.5(2) Table 1-2. .Atomic coordinates and Becj for 60b (continued) atom X y z Beq C12 0.1994(6) 0.2201(6) 0.4260(4) 2.5(2) C13 0.2525(6) 0.1187(6) 0.4324(4) 3.0(2) C14 0.3559(6) 0.0789(6) 0.3760(4) 2.7(2) C15 0.4379(6) -0.0196(5) 0.3647(4) 2.7(2) C16 0.5477(6) -0.0629(5) 0.3166(4) 3.0(2) C17 0.6309(6) -0.1654(5) 0.3100(4) 3.2(2) C18 0.7268(6) -0.1713(6) 0.2556(4) 3.9(2) C19 0.6997(6) -0.0749(6) 0.2292(4) 3.4(2) C20 -0.0480(5) 0.1216(5) 0.1910(4) 3.4(2) C21 -0.1357(6) 0.2124(6) 0.0540(4) 3.1(2) C22 -0.0517(7) 0.1458(7) -0.0070(5) 7.9(3) C23 -0.0552(5) 0.3974(5) 0.0219(4) 3.6(2) C24 0.1668(5) 0.5720(5) 0.1436(4) 3.1(2) C25 0.2792(5) 0.5593(5) 0.0889(4) 4.0(2) C26 0.2866(6) 0.5295(5) 0.2724(4) 2.9(2) C27 0.1893(7) 0.6097(6) 0.3234(5) 4.8(3) C28 0.0948(6) 0.2968(6) 0.4743(4) 3.4(2) C29 0.1296(6) 0.3260(6) 0.5351(4) 4.7(2) C30 0.2142(6) 0.0577(5) 0.4889(4) 3.7(2) C31 0.2955(6) 0.0239(6) 0.5417(4) 4.7(2) C32 0.6115(6) -0.2467(5) 0.3546(5) 4.8(2) Table 1-2. Atomic coordinates and Beq for 60b (continued) atom X y z Beq C33 0.8364(6) -0.2665(6) 0.2272(5) 5.2(3) C34 0.8084(7) -0.3237(7) 0.1752(5) 7.2(3) C35 0.7780(6) -0.0443(6) 0.1726(5) 4.8(2) C36 -0.1122(6) 0.3787(6) 0.3661(4) 2.7(2) C37 -0.2067(6) 0.3757(6) 0.4210(4) 2.9(2) C38 -0.1818(5) 0.2781(6) 0.4329(4) 3.0(2) C39 -0.0697(5) 0.2197(5) 0.3846(4) 2.4(2) C40 -0.0093(6) 0.1180(6) 0.3764(4) 2.6(2) C41 0.0925(6) 0.0605(5) 0.3288(4) 2.3(2) C42 0.1392(6) -0.0459(5) 0.3238(4) 3.2(2) C43 0.2402(6) -0.0770(5) 0.2714(4) 2.9(2) C44 0.2533(6) 0.0105(5) 0.2434(4) 2.5(2) C45 0.3588(5) 0.0055(5) 0.1841(4) 2.9(2) C46 0.3625(5) 0.0960(5) 0.1471(4) 2.3(2) C47 0.3184(5) 0.1339(5) 0.0883(4) 2.2(2) C48 0.3545(5) 0.2117(5) 0.0682(4) 2.4(2) C49 0.4210(5) 0.2164(5) 0.1180(4) 2.3(2) C50 0.4776(5) 0.2831(5) 0.1191(3) 1.9(2) C51 0.5390(5) 0.2931(5) 0.1666(4) 2.2(2) C52 0.5928(5) 0.3632(5) 0.1661(4) 2.5(2) C53 0.6386(6) 0.3474(5) 0.2247(4) 2.8(2) Table 1-2. Atomic coordinates and Beq for 60b (continued) atom X y z Beq C54 0.6138(5) 0.2698(5) 0.2590(4) 2.9(2) C55 -0.0999(5) 0.4699(5) 0.3331(4) 3.5(2) C56 -0.3162(6) 0.4652(6) 0.4567(4) 4.1(2) C57 -0.3078(7) 0.4919(7) 0.5274(5) 6.1(3) C58 -0.2545(5) 0.2350(5) 0.4861(4) 3.8(2) C59 0.0796(6) -0.1119(5) 0.3630(4) 4.2(2) C60 -0.0165(6) -0.1184(6) 0.3286(5) 5.5(3) C61 0.3251(7) -0.1828(6) 0.2439(5) 5.5(3) C62 0.2915(7) -0.2155(7) 0.1859(6) 8.3(4) C63 0.2527(6) 0.0923(5) 0.0518(4) 3.4(2) C64 0.3338(6) 0.0158(6) -0.0051(4) 5.1(3) C65 0.3287(6) 0.2771(6) 0.0090(4) 3.6(2) C66 0.4353(6) 0.2525(6) -0.0531(4) 5.3(2) C67 0.5971(5) 0.4380(5) 0.1126(4) 3.4(2) C68 0.6997(6) 0.4041(6) 0.2489(4) 4.3(2) C69 0.6091(6) 0.5042(6) 0.2844(5) 5.4(3) C70 0.6433(6) 0.2271(6) 0.3254(4) 4.3(2) Beq = %l3T?(U\\{act)2+U22(bb*)2+\Jw(cc)2+2U\2adbb*cos^ 94 Table 1-3. Selected bond length (A) for the compound 60b Co(l)-N(l) 1.990(5) Co(l)-N(2) 1.988(5) Co(l)-N(5) 1.993(5) Co(l)-N(6) 2.000(5) N(l)-C(l) 1.344(7) N(l)-C(4) 1.424(7) N(2)-C(6) 1.397(3) N(2)-C(9) 1.335(3) C(l)-C(2) 1.402(4) C(2)-C(3) 1.367(4) C(3)-C(4) 1.421(4) C(4)-C(5) 1.392(4) C(5)-C(6) 1.394(4) C(5)-C(10) 1.490(4) C(6)-C(7) 1.421(3) C(7)-C(8) 1.357(4) C(8)-C(9) 1.407(4) 95 Table 1-4. Selected Bond Angles (°) for 60b atom atom atom angle atom atom atom angle NI C o l N2 98.7(2) NI C o l N5 108.3(2) NI • C o l N6 106.7(2) N2 C o l N5 108.0(2) N2 C o l N6 135.8(2) N5 C o l C6 97.8(2) C o l NI C l 133.5(4) C o l NI C4 120.8(4) C l NI C4 105.7(5) C o l N2 C6 118.9(4) C o l N2 C9 132.4(5) C6 N2 C9 106.8(5) C o l N5 C36 130.3(5) C o l N5 C39 121.9(5) C36 N5 C39 106.8(5) C o l N6 C41 116.6(4) C o l N6 C44 134.9(5) C41 N6 C44 106.3(5) NI C l C2 112.1(6) NI C l C20 121.8(6) C2 C l C20 126.1(6) C l C2 C3 105.2(6) C l C2 C21 125.4(6) C3 C2 C21 129.2(6) C2 C3 C4 108.3(6) C2 C3 C23 126.2(6) C4 C3 C23 125.5(6) NI C4 C3 108.7(5) NI C4 C5 123.5(6) C3 C4 C5 127.8(6) C4 C5 C6 130.9(6) N2 C6 C5 126.5(6) N2 C6 C7 108.0(5) C5 C6 C7 125.6(6) C6 C7 C8 107.8(6) C6 C7 C24 127.5(6) C8 C7 C24 124.6(6) C7 C8 C9 106.1(5) C7 C8 C26 127.2(6) C9 C8 C26 126.3(6) N2 C9 C8 111.1(6) N2 C9 C10 127.5(6) C8 C9 CIO 121.4(6) C9 C10 C l l 121.5(6) 96 Table 1-6. Crystallographic data for 59 Compound 59 formula C6oH-72Zn2Ng fw 1036.04 color, habit metallic green, irregular crystal size, mm 0.35 x 0.30 x 0.20 crystal system orthorhombic space group Fddd(#70) a, A 14.3337(12) b, A 26.1271(4) c, A 28.7758(7) V, A 3 10776.5(7) Z 8 D c a l c , g/cm3 1.277 F(000) 4384.00 radiation MoKoc(?i=0.71069 A) (i( MoKa) , cm"1 9.36 Table 1-7. Bond Length (A) for 59 atom atom distance atom atom distance Zn(l) N(l) 1.996(2) Zn(l) N(2) 1.982(2) N(l) C(l) 1.328(3) N(l) C(4) 1.407(2) N(2) C(6) 1.394(3) N(2) C(9) • 1.345(3) C(l) C(2) 1.433(3) C(l) C(10) 1.480(3) C(2) C(3) 1.362(3) C(2) C ( l l ) 1.513(3) C(3) C(4) 1.430(3) C(3) C(12) 1.495(3) C(4) C(5) 1.379(3) C(5) C(6) 1.390(3) C(6) C(7) 1.419(3) C(7) C(8) 1.391(3) C(7) C(13) 1.508(3) C(8) C(9) 1.418(3) C(8) C(15) 1.492(3) C(9) C(9)" 1.465(4) C(13) C(14) 1.524(4) 98 Table 1-8. Bond Angles (°) for 59 atom atom atom angle N(l) Zn(l) N( l ) ' 106.32(11) N(l) Zn(l) N(2)' 124.26(7) Zn(l) N(l) C(l) 129.6(2) C(l) N(l) C(4) 107.2(2) Zn(l) N(2) C(9) 128.86(14) N(l) C(l) C(2) 110.5(2) C(2) C(l) C(10) 127.3(2) C(l) C(2) C ( l l ) 124.7(3) C(2) C(3) C(4) 107.1(2) C(4) C(3) C(12) 125.7(2) N(l) C(4) C(5) 123.9(2) C(4) C(5) C(6) 130.0(2) N(2) C(6) C(7) 109.0(2) C(6) C(7) C(8) 107.1(2) C(8) C(7) C(13) 125.6(2) C(7) C(8) C(15) 128.2(2) N(2) C(9) C(8) 111.3(2) C(8) C(9) C(9)" 127.7(2) atom atom atom angle N(l) Zn(l) N(2) 95.53(7) N(2) Zn(l) N(2)' 113.29(10) Zn(l) N(l) C(4) 122.82(15) Zn(l) N(2) C(6) 123.45(15) C(6) N(2) C(9) 106.6(2) N(l) C(l) C(10) 122.2(2) C(l) C(2) C(3) 107.1(2) C(3) C(2) C ( l l ) 128.2(3) C(2) C(3) CQ2) 127.2(2) N(l) C(4) C(3) 108.2(2) C(3) C(4) C(5) 127.9(2) N(2) C(6) C(5) 123.8(2) C(5) C(6) C(7) 127.2(2) C(6) C(7) C(13) 127.3(2) C(7) C(8) C(9) 105.9(2) C(9) C(8) C(15) 125.9(2) N(2)) C(9) C(9)" 121.0(2) G(7) C(13) C(14) 112.8(2) 99 Table 1-10. Crystallographic data for 87 Compound 87 formula C98H122 Ni2Zn3 fw 1664.19 color, habit block, red crystal size, mm 0.32x0.18x0.12 crystal system monoclinic space group C2/c o a, A 21.223(3) b, A 34.635(3) c, A 13.1746(13) v, A 3 9499(2) z 4 D c a l c , g/cm3 1.164 F(000) 3536 Absorption coefficient 0.801 mm"1 100 Table 1-11 Atomic coordinates and Bea for 87 atom X y z U(eq) SOF Zn(l) 0 6076(1) 2500 29(1) Zn(2) 0 7339(1) 2500 33(1) 1 Zn(3) 0 8542(1) 2500 41(1) 1 N(l) 484(2) 5735(1) 3625(3) 29(1) 1 N(2) -642(2) 6296(1) 3269(3) 27(1) 1 N(3) -449(2) 7025(1) 1317(3) 32(1) 1 N(4) 636(2) 7588(1) 1798(3) 30(1) 1 N(5) 448(2) 8278(1) 3776(3) 34(1) 1 N(6) -616(2) 8845(1) 3156(3) 43(1) 1 C(l) 992(2) 5505(1) 3672(4) 31(1) 1 C(2) 1103(3) 5277(2) 4600(4) 37(1) 1 C(3) 634(3) 5387(2) 5143(4) 37(1) 1 C(4) 250(2) 5669(1) 4545(4) 31(1) 1 C(5) -273(2) 5862(1) 4778(4) 32(1) 1 C(6) -672(2) 6143(1) 4252(4) 32(1) 1 C(7) -1210(3) 6308(2) 4583(4) 39(1) 1 C(8) -1496(3) 6559(2) 3842(4) 38(1) C(9) -1142(2) 6539(2) 3036(4) 33(1) C(10) -1345(2) 6752(2) 2053(4) 39(1) 1 C ( l l ) -937(2) 6773(2) 1241(4) 33(1) 1 101 Table 1-11. Atomic coordinates and Be(j for 87 (continued) C(12) -1058(3) 6584(2) 258(4) 34(1) 1 C(13) -612(3) 6734(2) -279(4) 35(1) 1 C(14) -233(2) 7006(1) 385(4) 32(1) 1 C(15) 249(2) 7242(1) 145(4) 32(1) 1 C(16) 636(3) 7512(2) 734(3) 34(1) 1 C(17) 1136(3) 7731(2) 443(4) 41(1) 1 C(18) 1439(3) 7935(2) 1287(4) 39(1) 1 C(19) 1123(3) 7836(2) 2112(4) 34(1) 1 C(20) 1368(3) 7955(2) 3197(4) 54(2) 1 C(21) 941(3) 8036(2) 3953(4) 35(1) 1 C(22) 1078(3) 7906(2) 5000(4) 40(1) 1 C(23) 635(3) 8086(2) 5465(4) 49(2) 1 C(24) 239(3) 8320(2) 4710(4) 45(2) 1 C(25) -252(3) 8564(2) 4859(4) 52(2) 1 C(26) -637(3) 8809(2) 4194(4) 48(2) 1 C(27) -1150(4) 9053(2) 4399(5) 80(2) 1 C(28) -1433(4) 9224(2) 3492(5) 70(2) 1 C(29) -1070(3) 9093(2) 2732(5) 57(2) 1 C(30) 1399(3) 5513(2) 2868(4) 39(1) 1 C(31) 1642(3) 4991(2) 4916(5) 51(2) 1 C(32) 2247(3) 5167(2) 5427(6) 83(2) 1 C(33) 532(3) 5238(2) 6180(4) 42(1) 1 102 Table 1-11. Atomic coordinates and Beq for 87 (continued) C(34) 848(3) 5490(2) 7080(4) 62(2) 1 C(35) -1409(3) 6229(2) 5623(4) 61(2) 1 C(36) -2080(3) 6809(2) 3815(4) 53(2) 1 C(37) -1574(2) 6296(2) -112(4) 42(1) 1 C(38) -2179(3) 6478(2) -724(5) 69(2) 1 C(39) -541(3) 6637(2) -1360(4) 49(2) 1 C(40) 1287(4) 7741(2) -631(5) 72(2) 1 C(41) 2030(3) 8182(2) 1370(4) 52(2) 1 C(42) 2633(3) 7950(2) 1742(6) 72(2) 1 C(43) 1604(3) 7636(2) 5485(5) 58(2) 1 C(44) 575(4) 8040(3) 6592(5) 84(3) 1 C(45) -1404(6) 9050(4) 5451(7) 162(6) 1 C(46) -1077(9) 9386(5) 5965(12) 286(12) 1 C(47) -2007(5) 9488(3) 3278(7) 105(3) 1 C(48) -1808(6) 9893(4) 3246(8) 158(5) 1 C(49) -1173(4) 9197(3) 1620(5) 86(3) 1 103 Table 1-12. Bond lengths (A; for 87 Zn(l)-N(2) 2.001(4; Zn(l)-N(2) 2.001(4; Zn(l)-N(l) 2.013(4; Zn(l)-N(l) 2.013(4; Zn(2)-N(4) 1.977(4; Zn(2)-N(4) 1.978(4; Zn(2)-N(3) 1.986(4; Zn(2)-N(3) 1.986(4; Zn(3)-N(5) 1.986(4; Zn(3)-N(5) 1.986(4; Zn(3)-N(6) 1.999(4" Zn(3)-N(6) 1.999(4 N(l)-C(l) 1.333(6 N(l)-C(4) 1.415(6 N(2)-C(9) 1.343(6 N(2)-C(6) 1.413(6 N(3)-C(ll) 1.343(6 N(3)-C(14) 1.393(6 N(4)-C(19) 1.345(7 N(4)-C(16) 1.427(6 104 Table 1-13. Bond angles (°) for 87 atom-atom-atom angle [°] atom-atom-atom angle [°] N(2)'-Zn(l)-N(2) 135.2(2) N(4)-Zn(2)-N(4)' 128.2(2) N(2)'-Zn(l)-N(l) 107.2(2) N(4)-Zn(2)-N(3) 98.0(2) N(2)-Zn(l)-N(l) 98.7(2) N(4)'-Zn(2)-N(3) 109.8(2) N(2)'-Zn(l)-N(l)' 98.7(2) N(4)-Zn(2)-N(3)' 109.8(2) N(2)-Zn(l)-N(l)' 107.2(2) N(4)'-Zn(2)-N(3)' 98.0(2) N(l)-Zn(l)-N(l) ' 108.2(2) N(3)-Zn(2)-N(3)' 113.5(2) C(l)-N(l)-C(4) 106.2(4) N(3)-C(14)-C(15) 123.7(5) C(4)-C(5)-C(6) 132.0(4) C(15)-C(14)-C(13) 127.0(5) C(5)-C(6)-N(2) 126.2(4) C(16)-C(15)-C(14) 130.8(5) C(5)-C(6)-C(7) 125.5(4) C(15)-C(16)-C(17) 127.3(4) N(2)-C(9)-C(10) 126.5(4) C(15)-C(16)-N(4) 124.4(5) C(8)-C(9)-C(10) 121.8(5) C(18)-C(19)-C(20) 122.8(5) C(9)-C(10)-C(ll) 122.5(4) C(19)-C(20)-C(21) 123.4(5) C(12)-C(ll)-C(10) 126.6(5) 105 Table 1-14. Crystallographic data for 92 Compound 92 formula C 6 2H 7 6 N 8 S 2 Zn 2 Cl4 fw 1027.03 color, habit red, needle crystal size, mm 0.40x0.20x0.15 crystal system monoclinic space group P21/c(#14) o a, A 14.4307(6) b, A 12.4665(6) c, A 36.281(2) (3° 101.220(2) O T V, A 6402.3(5) Z 4 D c a i c , g/cm3 1.318 F(000) 2656.00 radiation MoKa(X=0.71069 A) [i( MoKa) , cm"1 10.25 106 Table 1-15. Selected Bond Lengths (A) for 92 atom atom distance atom atom distance C1(4A) C(32) 1.63(1) C1(4B) C(32) 1.33(1) Zn(l) N(l ') 1.972(5) Zn(l) N(l) 1.991(5) Zn(l) N(2') 1.976(4) Zn(l) N(2) 1.990(4) Zn(l) N(2') 1.976(4) Zn(l) N(2) 1.990(4) Zn(l') N(3) 1.991(4) Zn(l') N(4) 1.976(5) Cl(l) C(31) 1.738(9) Cl(2) C(31) 1.734(9) Cl(3) C(32) 1.91(1) S(l') C(8') 1.760(6) S(l) C(8) 1.756(5) S(l) C(ll) 1.758(6) N(l) C(l) 1.352(7) C(4) C(5) 1.405(7) N(l) C(4) 1.399(7) N(2) C(6) 1.419(7) N(2) C(9) 1.338(7) N(3) C(10) 1.333(7) N(3) C(13) 1.398(7) N(4) C(15) 1.410(7) N( 4) C(18) 1.335(7) C(l') C(19') 1.484(8) C(l) C(2) 1.413(9) C(10) C(25) 1.493(8) C(l) C(19) 1.494(8) C(2) C(3) 1.377(8) C(2) C(20) 1.497(8) C(3) C(4) 1.423(8) C(3) C(22) 1.492(9) C(5) C(6) 1.374(7) C(6) C(7) 1.425(7) CC(12) C(13) 1.409(7 C(7') C(23') 1.489(8) C(7) C(8) 1.390(8) C(7) C(23) 1.498(8) C(8) C(9) 1.429(7) C(9) C(24) 1.462(8) C(10) C(ll) 1.431(8) 107 Table 1-15. Selected Bond Lengths (A) for 92 (continued) atom atom distance atom atom distance C(ll) C(12) 1.371(8) C(12) C(26) 1.497(9) C(13) C(14) 1.400(8) C(14') C(15') 1.377(7) C(14) C(15) 1.361(7) C(15) C(16) 1.452(8) C(16) C(17) 1.389(8) C(28) C(29) 1.537(9) C(16) C(27) 1.490(8) C(17) C(18) 1.415(8) C(17) C(28) 1.491(8) C(18) C(30) 1.491(8) 108 Table 1-16. Selected Bond Angles (°)for 92 atom atom atom angle atom atom atom angle N( l ' ) Zn(l) N(l) 121.6(2) N( l ' ) Zn(l) N(2) 118.7(2) N(l) Zn(l) N(2) 96.4(2) N(2') Zn(l) N(2) 112.4(2) C(8) S(l) C ( l l ) 101.7(2) Zn(l) N(l) C(l ) 130.4(4) Zn(l) N(l) C(4) 123.2(4) C(l) N(l) C(4) 105.8(5) Zn(l) N(2) C(9) 129.7(4) C(6) N(2) C(9) 108.0(4) Zn(l) N(2) C(6) 122.1(4) C(10) N(3) C(13) 106.1(5) N(l) C(l ) C(2) 111.4(5) N(l) C(4) C(5) 122.8(5) C(3) C(4) C(5) 127.9(5) C(l) C(2) C(3) 106.7(5) N(l) C(4) C(5) 122.8(5) C(4) C(5) C(6) 131.1(5) N(2) C(6) C(5) 123.9(5) N(2) C(6) C(7) 108.0(5) C(5) C(6) C(7) 128,0(5) S(l) C(8) C(7) 125.5(4) S(l) C(8) C(9) 127.0(5) C(7) C(8) C(9) 107.9(5) N(2) C(9) C(8) 109.8(5) 109 References: 1 Longo, F. R.; Thorne, E. J.; Adler, A . D.; Dym, S. J. Heterocycl. Chem. 1975, 12, 1305. 2 Treibs, A. ; Haberle, N . ; Justus Liebigs Ann. Chem. 1968, 718, 183. 3 a) Fischer, H. ; Orth, H. Die Chemie des Pyrrols, Vol . 2, erste Halfte 1-151 and references therein., Akademische Verlagsgesellschaft m.b.H., Leipzig 1940. b) Falk, H. The Chemistry of Linear Oligopyrroles and Bile Pigments, Springer Verlag, Wien, New York 1989. 4 March, F. C ; Couch, D. A. ; Emerson, K.; Fergusson, J. E.; Robinson, W. T. J. Chem. Soc. (A) 1971,440. 5 Harris, R. L . N . ; Johnson, A. W.; Kay, I. T. J. Chem. Soc. 1966, 22. 6 Fleiderman, L . I.; Mironov, A . F.; Evstigneeva, R. P. Zh. Obshch. Khim. 1975, 45, 197; Fleiderman, L. I.; Mironov, A. F.; Evstigneeva, R. P. Zh. Obshch. Khim. 1973, 43, 886. 7 Morosini, P.; Scherer, M . ; Meyer, S.; Lynch, V. ; Sessler, J. L. J. Org. Chem. 1997, 62, 8848. 8 Fleiderman, L . I.; Mironov, A . F.; Evstigneeva, R. P Khim. Geterotsikl. Soedin, 1973, 1146. 9 Johnson, A. W.; Kay, I. T. J. Chem. Soc. 1961, 2418. 10 a) Engel, J. Gossauer, A . J. Chem. Soc, Chem. Comm. 1975, 570. b) Engel, J.; Gossauer, A . / . Chem. Soc, Chem. Comm. 1975, 713. c) Engel, J.; Gossauer, A. Justus Liebigs Ann. Chem. 1976, 1637. 11 Clezy, P. S.; Lim, C. L. ; Shannon, J. S. Aust. J. Chem. 1974, 27, 1130. 12 Harris, R. L . N . ; Johnson, A. W.; Kay, I. T. Q. Rev. Chem. Soc. 1966, 20, 211. 13 Grigg, R.; Johnson, A . W.; Kenyon, R.; Math, V. B.; Richardson, K. J. Chem. Soc. (C) 1969, 176. 14 Mironov, A . F.; Kulish, M . A. ; Kobak, V . V. ; Rozynov, B. K ; Evstigeeva, R. P. 110 Zh. Obshch. Khim, 1974, 44, 1407. 15 Baptista de Almeida, J. A . P.; Kenner, G. W.; Smith, K. M . ; Sutton, L. J. Chem. Soc. Chem. Comm. 1975, 111. 16 Johnson, A . W. Pure Appl. Chem. 1963, 23, 511. 17 Dolphin, D.; Johnson, A. W.; Leng, J.; van den Broek, P. J. Chem. Soc. (C) 1966, 880. 18 Clarke, D. A. ; Grigg, R.; Harris, R. L. N . ; Johnson, A . W.; Kay, I. T.; Shelton, K. W. J. Chem. Soc. (C) 1966, 1436. 19 Dolphin, D.; Harris, R. L . N . ; Johnson, A. W.; Huppatz, J. L.; Kay, I. T. J. Chem. Soc. (C) 1966, 30. 20 Bamfield, P.; Harris, R. L. N . ; Johnson, A.W.; Kay, I. T.; Shelton, K. W. J. Chem. Soc. (C) 1967, 1436. 21 Dolphin, D.; Harris, R. L . N . ; Huppatz, J. L.; Johnson, A . W.; Kay, I. T.; Leng, J. J. Chem. Soc. (C) 1966, 98. 22 Johnson, A . W.; Price, R. J. Chem. Soc. 1960, 1649 23 Johnson. A . W.; Kay, I. T.; Podrigo, R. J. Chem. Soc. 1963, 2336 24 Lehn, J .-M. Supramolecular Chemistry: Concepts and Perspectives V C H , Weinheim, 1995. 25 Fox, S.W.; Dos, K. Molecular Evolution And The Origin Of Life. Marcel Decker, Inc.: New York, 1977, Chapterl and Chapter 6. 26 Anderson, S.; Anderson H. K.; Sanders, J. K. Accounts Chem. Res. 1993, 26, 469. 27 Lindsey, J. S. New J. Chem. 1991, 15, 153. 28 Kroto, H. W. Angew. Chem. 1992, 104, 113; Angew. Chem. Int. Ed. Engl. 1992, 31, 111. 29 Ebbesen, T. W. Annu. Rev. Mater. Sci. 1991, 113, 4696. 30 Lawrence, D.; Jiang, T.; Levett M . Chem. Rev. 1995, 95, 2229. 31 Wald, G. Sci. Am. 1954, 191, 44. i l l 32 Whitesides, G. M . ; Mathias, J. P.; Seto, C. T. Science 1991, 254, 1312. 33 Bovey F.A.; Winslow, F. H. Macromolecules Acedamic Press, New York, 1979. 34 Wright, J. D. Molecular Crystals Cambridge Univ. Press, 1987. 35 Pershan, P. S. Structure of Liquid Crystal phases Wold Scientific, 1988. 36 Kortan, A . R; Hull, R.; Opila, R. L. ; Bawendi, M . G.; Steigerwlad, M . L . ; Brus, L . E. J. Am. Chem. Soc. 1990, 112, 1327. 37 Vinson, P. K.; Bellare, J. R.; Davis, H. T.; Miller W. G.; Scriven, L. E. J. Collid Interface Sci. 1991, 142, 74. 38 Shinoda, K. Friberg, S. Emulsion and Solubilization Wiley, 1986. 39 Frankel, D. A. ; Lamprarski, H. ; Liman, U . ; O'Brien, D. F. / . Am. Chem. Soc. 1989, 111,9262. 40 Ringschorf, H. ; Schlarb, B.; Venzmer, J. Angew. Chem. Int. Ed. Engl. 1988, 27, 114. 41 Whitesides, G. M . ; Plainbins, P. E. Langmir 1990, 6, 87. 42 Funeriu, D. P.; Lehn, J.-M.; Baum, G.; Fenske, D. Chem. Eur.1991, 3, 99. 43 Sharp, K. A. ; Nichollis, A. ; Fine, R. F.; Honing, B. Science 1991, 252, 106. 44 Kramer, R.; Lehn, J.-M. Marquis-Rigault, A. Proc. Natl. Acad. Sci. USA, 1993, 90, 5394. 45 Baxter, P. N . W.; Lehn, J.-M.; De Cian, A. ; Fischer, J. Angew. Chem.1993, 105, 92; Baxter, P.; Lehn, J .-M.; DeCian, A. ; Fischer, J. Angew. Chem. Int. Ed. Engl. 1993, 32, 69. 46 a) Desiraju, G. R.Crystal Engineering, in The Design of Organic Solids Elsevier, Amsterdam, 1989. b) Jeffrey, G. A. ; Saenger, W. Hydrogen Bonding in Biological Structures, Springer, Berlin, 1991. c) Bernstein, J. Etter, M . C ; Leiserowitz, L. in Structure Correlation, Eds, Duntitz, J. D. Burgi, H. B. V H C , Weinheim, 1993. 47 Piguet, C ; Bernardinelli, G.; Hopfgartner, G. Chem. Rev.1991, 97, 2005. 48 Funeriu, D.-P.; He, Y . -B . ; Bister, H.-J.; Lehn, J .-M. Bull. Soc. Chim. Fr. 1996, 133,673. 112 49 Baxter, P. N . W.; Hanan, G. S.; Lehn, J.-M. J. Chem. Soc, Chem. Commun. 1996, 2019. 50 Weissbuch, I.; Baxter, P. N . W.; Cohen, S.; Cohen, H. ; Kjaer, K.; Howes, P. B. ; Als-Nielsen, J.; Hanan, G. S.; Schubert, U . S.; Lehn, J.-M.; Leiserowitz, L. ; Lahav, M . J. Am. Chem. Soc. 1998, 120, 4850. 51 Baxter, P. N . W.; Lehn, J.-M.; Kneisel, B. O.; Fenske, D. J. Chem. Soc, Chem. Commun. 1997, 2231. 52 Baxter, P. N . W.; Lehn, J.-M.; Fischer, J.; Youinou, M.-T. Angew. Chem. Int. Ed. Engl. 1994, 33, 2284. 53 Fujita, M ; Yu, S.-Y; Kusukawa, T; Funaki, H ; Ogura, K; Yamaguchi, K. Angew. Chem. Int. Ed. Engl. 1998, 37, 2082 54 Hasenknopf, B.; Lehn, J .-M.; Baum, G.; Kneisel, B. O.; Fenske, D. Angew. Chem. Int. Ed. Engl.1996, 35, 1838. 55 Lehn, J. M . ; Sauvage, J. P.; Ziessel, R.; Piccinni, C ; Germain, G.; Declerq; Van Meerssche, M . Nouv. J. Chim. 1983, 7, 413. 56 Seto, C. T.; Whitesides, G. M . J. Am. Chem. Soc. 1991, 113, 2440. 57 Schumacher, R. Angw. Chem. Int. Ed. Engl. 1990, 29, 329. 58 Lehn, J .-M. Angew. Chem. Int. Ed. Engl. 1990, 29, 1304. 59 Berger, R. D.; Gibson, J. M . Appl. Phys. Lett. 1990, 57, 153. 60 Shull, K. R.; Winey, K. I.; Thomas, E. L. ; Kramer, E. J. Macromolecules 1991, 24, 2748. 61 Penner, R. M . ; Heben, M . J.; Longin, T. L. ; Lewis, N . S. Science 1990, 25, 1118; Rusling, J. F. Acc. Chem. Res. 1991, 24, 75. 62 Ozin, G. A. ; Kuperman, A. ; Stein, A. Angew. Chem. Int. Ed. Engl. 1989, 28, 359. 63 Kern, K. Phys. Rev. Lett. 1991, 67, 855. 64 Crans, D. C. Comments Inorg. Chem. 1994, 16, 35. 65 Schoentjes, B.; Lehn, J.-M.; Helv. Chim. Acta, 1995, 78, 1. 113 66 Tabushi, I.; Kuroda, Y . Tetrahedron Lett. 1982, 23, 4601. 67 Kobuke, Y. ; Ueda, K.; Sokabe, M . ; / . Am. Chem. Soc. 1992, 114, 7618. 68 Barr, R. G.; Pinnavaia T. J. J. Phys. Chem. 1986, 90, 328. 69 Zerkowski, J. A. ; Seto, C. T.; Whitesides, G. M . J. Am. Chem. Soc. 1992, 114, 5473. 70 a) Ghadiri, M . R.; Granja, J. R.; Milligan, R. A. ; McRee, D. E. Khazanovich, N . Nature, 1993, 366, 324. b) Kim, H. S.; Hartgerink; Ghadiri, M . R. J. Am. Chem. Soc. 1998, 120, 4417. 71 Ghadiri, M . R.; Granja, J. R.; Buchler, L. K. Nature 1994, 369, 301 72 Baxter, P. N . W. In Comp. Supramol. Chem. Volum 9, Ed: Sauvage, J.-P.; Hosseini, M . W. Ed., Elsevier: Oxford, UK. 1996, 165-211. 73 Lehn, J. M . ; Sauvage, J. P.; Ziessel, R.; Piccinni, C.; Germain, G.; Declerq; Van Meerssche, M . Nouv. J. Chim. 1983, 7, 413. 74 Lehn, J .-M.; Rigault, A. ; Siegel, J.; Harrowfield, J.; Chevrier, B.; Moras, D. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 2565. 75 Lehn. J .-M.; Rigault, A. Angew. Chem. Int. Ed. Engl. 1988, 27, 1095. 76 Kramer, R.; Lehn, J.-M.; DeCan, A. ; Fischer, J. Angew. Chem. Int. Ed. Engl. 1993, 32, 705. 77 Baxter, P. N . W.; Hanan, G. S.; Lehn, J.-M. unpublished source. 78 Kramer, R.; Lehn, J .-M.; Marquis-Rigault, A. Proc. Natl. Acad. Sci. USA. 1993, 990, 5394. 79 Fynman, R. Science 1991, 254, 1300. 80 Baxter, P. N . W.; Lehn, J.-M.; Fisher, J.; Youinou, M . Angew. Chem. 1994, 106, 2434. 81 Tomalia, D. A. ; Naylor, A. M . ; Goddard, W. A. Angew. Chem. 1990, 102, 119; Angew. Chem. Int. Ed. Engl. 1990, 29, 138. 82 Huck, W. T. S.; van Veggel, J. M . ; Kropman, B. L. ; Blank, D. H. A. ; Keim, E. G.; Smithers, M . M . A. ; Reinhould, D. N . J. Am. Chem. Soc. 1995, 117, 8293. 114 83 Harding, M . M . ; Koert, U . ; Lehn, J. M . ; Marquis-Rigault, A. ; Pigurt, C ; Siegel, J. Helvetica Chimica Acta 1991, 74, 594. 84 Zhang, Y. ; Thomposon, A. ; Dolphin, D. Ann. Report of Venture Business, Laboratory, Kobe University, 1999, 4, 166. 85 Fisher, H.; Orth, H. , Die Chemie des pyrroles. Akademische Verlasgesellschaft; Leipzig, 1940; reprint by Johnson Reprint Corporation, New York, 1968, Vol.2 erste Halft pp 1-151 and references therein. 86 Struckmeier, G.; Thewalt, U . ; Fuhrhop. J.-H. J. Am. Chem. Soc. 1976, 98, 278. 87 Dolphin, D. Ph.D. Thesis, The University of Nottingham, 1965. 88 Dolphin, D.; Harris, R. L. N . ; Huppatz, J. L. ; Johnson, A.W.; Kay, I. T. / . Chem. Soc. (C) 1966, 30. 89 Sheldrick, W. S.; Engel, J. 7. Chem. Soc. Chem. Commun. 1980, 5. 90 Zhang, Y. ; Thompson, A. ; Rettig , S. J.; Dolphin, D. / . Am. Chem. Soc. 1998, 120, 13537. 91 Battersby, A. ; Hunt, R.; McDonald, E.; Paine, J. B.; Saunders, J. J. Chem. Soc. Perkin 11976, 1010. 92 Wijesekera, T. P.; Paine III, J. B.; Dolphin, D. J. Org. Chem. 1985, 50, 3832 93 Smith, K. M . ; Pandey, R. K. J. Chem. Soc. C1971, 502. 94 Smith, K. M . ; Pandey, R. K. J. Heterocyclic Chem. 1983, 20, 1383. 95 Mironov, A. F.; Ovsepyan, T. R.; Evstigneeva, R. P.; Preobrazhenski, N . A . Zh. Obshch. Khim. 1965, 35, 324. 96 March, F. C ; Couch, D. A. ; Emerson, K.; Fergusson, J. E.; Robinson W. T. J. Chem. Soc. (A) 1971, 440. 97 Sunramanian, J.; Fuhrhop, J.-H.; The Porphyrins, Dolphin, D. Academic Press, New York, Volume II. Chapter 8. 256. 98 Bruckner, C. Zhang, Y. ; Rettig, S.; Dolphin, D. Inorg. Chim. Acta. 1997, 263, 279 99 Dolphin, D.; Johnson, A. W.; Leng, J. van der Broek P. J. Chem. Soc. (C) 1966, 880. 115 100 Struckmerier, G.; Thewalt, U . ; Engel, J. / . Chem. Soc. Chem. Comm. 1976, 963. 101 Bonnett, R.; Davis, J. E.; Hursthouse, M . B. Nature 1976, 262, 1298. 102 Sheldrick, W.; Engel, J. J. Chem. Soc. Chem. Comm. 1980, 5. 103 HyperChem Release 5.01 MM+ force field optimized by PolakRibiere 104 Johnson, A. W. Quart. Rev. 1975, 29, 1. 105 Battersby, A. ; Hunt, R.; McDonald, E. / . Chem. Soc. Perkin 11976, 1008. 106 Guilard, R.; Aukauloo, M . A. ; Tardieux, C ; Vogel, E. Synthesis 1995, 1481. 107 Data collection and structure solution were conducted by Maren Pink, Victor G. Young, Jr. at the X-Ray Crystallographic Laboratory, 160 Kolthoff Hall, Chemistry Department, The University of Minnesota. A l l calculations were performed using SGI INDY R4400-SC or Pentium computers using the S H E L X T L V5.0 suite of programs. The poor data are due to solvent loss and/or heavily disordered solvent and/or a modulation of the structure. The program SQUEEZE/PLATON was used to correct the data for disordered solvent. The overall model ameliorated and R-value improved by approximately 3%. The final full matrix least squares refinement converged (only after applying a damping factor which could be taken out in the last refinement)to R= 0.0784. 108 Chen, Q.-Q.; Dolphin, D. unpublished source. 109 Thompson, A. ; Rettig, S. J.; Dolphin, D. J. Chem. Soc. Chem. Comm. 1999, 631. 110 Paine J. B.; Dolphin, D. Can. J. Chem., 1978, 56, 1710. 111 Lehn, J . -M.; Supramolecular Chemistry V C H : Weinheim, 1995 112 Chen. Q.-Q.; Falk, H ; Micura, R. Monatshefte fur Chemie 1995, 126, 473. Part 2 SYNTHESIS, DERIVATIZATION AND STRUCTURAL CHARACTERIZATION OF OCTAHEDRAL TRIS(5-PHENYL-4,6-DIPYRRINATO) COMPLEXES OF COBALT(III) AND IRON(III) 117 1. I N T R O D U C T I O N 1.1 DIPYRROMETHENES AND THEIR PROPERTIES a,a-Dipyrromethenes 1, also known as 4,6-dipyrrins, are basic, brightly colored, fully conjugated flat bipyrrolic molecules. Their skeleton structure, atom numbering scheme and formal nomenclature is shown in Figure 2-1. Carbons 1 and 9 are also 3 5 7 1 10 11 9 1 4,6-Dipyrrin 2-(2-H-Pyrrol-2-ylidenemethyl)pyrrole Figure 2-1. Structure, atom numbering scheme and formal nomenclature of oc,a'-dipyrromethene referred to as oc-carbons, positions 2, 3, 7 and 8 as 6-positions and position 5 as the meso-position. Since they are requisite intermediates in Fischer's porphyrin synthesis,1 which is historically the first discovered method for the preparation of tetrapyrrolic porphyrins, their syntheses and properties have been intensively investigated. ' ~ Dipyrromethenes are usually obtained as the corresponding salt of the mineral acid present in the reaction mixture.4 These salts are usually stable red-colored crystalline compounds showing a characteristic green iridescent luster. The corresponding dipyrromethene free bases, which are usually less stable than the protonated species, 118 particularly when they carry only alkyl substituents may be liberated from the latter by treatment with aqueous ammonia.4 The characteristic light absorption of dipyrromethenes lies in the range 430-470 nm.5'6 Protonated species absorb at somewhat (about 20-40 nm) longer wavelengths. This bathochromic shift is particularly pronounced in the case of mesc-substituted dipyrromethenes due to the steric interaction.7 Free base dipyrromethenes are often difficult to purify, probably because of their facile protonation and deprotonation. However, some dipyrromethene-free bases have been conveniently purified by sublimation under high vacuum. Dipyrromethenes behave as valence tautomers whose structures 105 and 106, Scheme 2-1) are interchanged readily by proton-shift from one nitrogen atom to the other. 105 106 Schem 2-1. Valence-tautomeric structures of dipyrrins. (For clarity, substituents are omitted.) The predominating tautomer at equilibrium depends mainly on the substituents present at the ring positions. Obviously, both tautomers are equally represented in the case of symmetrically substituted derivatives. In contrast to pyrroles, dipyrromethenes react only rarely with electrophiles, but they can react reversibly with many nucleophiles such as water,9 alcohols,6'10 bromine,7 hydrogen cyanide," sodium bisulfite,12 methyl magnesium bromide,13 triethyl phosphite,8 etc. This phenomenon is rationalized by that, owing to the ability of the "pyridine-like" nitrogen atom on the pyrrolenine ring to withdraw electrons in contrast to the pyrrolic nitrogen atom which is a weak electron donor, the reactivity of 119 the dipyrromethene molecule as a whole resembles that of a 7t-electron-deficient heteroaromatic systems which are relatively unreactive toward electrophiles. As an important class of chemicals with unique properties, syntheses of dipyrromethenes have been extensively explored as detailed below. 1 4 1. The "classic" and most versatile reaction is the acid catalyzed condensation of a pyrrole aldehyde with an a-unsubstituted pyrrole derivative (Scheme 2-2). 3 ' 1 5 In certain cases, pyrrole derivatives bearing a halogen atom at the reactive a-position have been employed.16 This method enables the synthesis of dipyrromethenes carrying different substituents on each ring. 107 108 109 Scheme 2-2. Preparation of dipyrromethene (Method 1) 2. Dipyrromethanes can be oxidized into the corresponding dipyrromethenes by a wide variety of oxidants, such as bromine17, tert-butyl hypochlorite18, N -bromosuccinimide, ferric chloride3, lead dioxide, lead tetraacetate19 and DDQ (2,3-dichloro-5,6-dicyano-l,4-quinone)20 (Scheme 2-3). 110 111 Scheme 2-3. Preparation of dipyrromethene (Method 2) 120 3. Halogenation of a 5-unsubstituted 2-methylpyrroles21 or the corresponding 5-pyrrolecarboxylic acid 2 2 leads to the formation of 5-methyl-5'-bromo-dipyrromethenes by head-to-tail condensation of the two molecules of the same pyrrole derivative (Scheme 2-4). H Br 2 /AcOH 112 113 Scheme 2-4. Preparation of dipyrromethene (Method 3) 4. Condensation of two molecules of a-unsubstituted pyrroles or the corresponding a-pyrrolecarboxylic acids with formic acid or orthoformate affords symmetrically substituted dipyrromethenes (Scheme 2-5).4 Id H c o o H / H c i O d N H 114 Scheme 2-5. Preparation of dipyrromethene (Method 4) 5. Symmetric substituted dipyrromethenes can also be obtained by acid-catalyzed self-condensation of a-pyrrole aldehydes with elimination of one formyl group (Scheme 2-6) 23 HBr/EtOH N " C H O H 116 Scheme 2-6. Preparation of dipyrromethene (Method 5) 121 1.2 PREPARATION AND CHARACTERIZATION of meso-PHENYLDIPYRROMETHENES Recently, both Bruckner in our group 2 4 and in an independent study, Lindsey 2 5 have reported the preparation of 5-phenyl substituted oc,6-unsubstituted dipyrromethanes such as 121, 122 and 123 by the acid catalyzed condensation of benzaldehyde 119 or p-substituted benzaldehyde 118 and 120, with pyrrole (Scheme 2-7). For an optimized CHO Y 1 1 8 R = N 0 2 121 R=NOo 124 R = N 0 2 119 R = H 122 R=H 125 R = H 1 2 0 R = C O 2 C H 3 1 2 3 R = C 0 2 C H 3 126 R = C 0 2 C H 3 Scheme 2-7. Synthesis of raeso-phenyldipyrranes and mesophenyldipyrrines Reaction conditions: i) TFA, at room temperature; ii) benzene, DDQ; iii) NaBH 4 /MeOH. yield, Lindsey 2 6 and Carrell 2 7 suggested use of a large amount of extra pyrrole as solvent (benzaldehyde:pyrrole 1:50-100). Thus, the yields for 121, 122 and 123, after separation and purification by chromatography, were 82%, 49% and 49% respectively.28 The relatively high yield of 121 is presumably due to the electron-withdrawing nitro group favoring the electrophilic condensation and also stabilizing the resulting dipyrromethane toward acid catalyzed polymerization. 122 The resulting dipyrromethanes were oxidized with DDQ to afford the corresponding dipyrromethenes 124, 125 and 126.24 DDQ is a useful oxidant in the synthesis of pyrrolic pigments since dehydrogenation only occurs at meso position, without further oxidization at 1,9 position, p- and ochloranil are equally well suited to perform the conversion. Dipyrromethenes 124, 125 and 126 can be reduced with NaBFLt to regenerate the corresponding dipyrromethanes. The optical spectra of dipyrromethenes 124 and 125 are shown in Figure 2-2. The two-band pattern of the protonated meso-phenyldipyrromethene resembles that of 3,3',4,4',5,5'-tetramethyldipyrromethene, but -14 nm hypsochromically shifted with slightly lower extinction coefficients. The bands have been assigned to 71-71* transitions and are indicative of marked planarity of these fully conjugated aromatic systems.24 Comparison of free base and corresponding protonated dipyrromethenes shows a 6E+04 2E+04H OE+00 550 —I 650 Figure 2-2. UV-visible spectra of complexes 124 (solid line) and 125 (broken line). 123 bathochromic shift of 42 nm and a doubling of extinction coefficient of the latter compounds.30 Due to steric interactions, the phenyl moiety is inferred to be approximately perpendicular to the plane of dipyrromethene and, thus, not in the full conjugation with the pyrrolic systems. Consequently, the substituent on the phenyl moiety has minimal influence on the n electron density of derivatives 1 2 5 . Thus, the optical spectra of 1 2 4 , 1 2 5 and 1 2 6 are observed to be markedly similar.24 The observed number of signals in 'H and 1 3 C-NMR of raeso-phenyl-dipyrromethenes indicates symmetric structures.25 This is consistent with formulating the dipyrromethenes as adopting a planar conformation and a rapid tautomeric exchange of the NH-proton between the two nitrogens. As a result of this fast exchange, the proton is effectively held in plane by strong hydrogen bonding causing a low field resonance of 12.5 ppm. As a conclusion of this, meso-phenyldipyrromethenes can be assigned Z configuration (based on the exo-cyclic double bond) and syn conformation (based on exo-cyclic single bond).31 The chemical shifts for the (3-protons of 1 2 4 of 6.39 and 6.47 ppm and for the oc-protons of 7.78 ppm confirm the aromatic character of these compounds. 1.3 F O R M A T I O N A N D C H A R A C T E R I Z A T I O N O F D I V A L E N T T R A N S I T I O N M E T A L C H E L A T E S O F m e s o - P H E N Y L D I P Y R R O M E T H E N E S A characteristic property of dipyrromethenes is their ability to form quite stable metal chelates.32 Some meso-phenyl and at least one oc-substituted 3 3 ' 3 4 as well as some meso- and a- or 8-unsubstituted dipyrromethenes have long been known to form 2:1 124 (ligand:metal) complexes with most bivalent transition metal ions (e.g. Mn , Co , N i , C u 2 + , Z n 2 + ) . 3 5 - 3 7 These complexes have some unique properties which have made them objects of several recent studies.21'38 Derivatisation of the meso-phenyl groups gives rise to a range of substituents, making these complexes amenable to incorporation into larger structures such as photonic wires,3 9 optoelectronic gates40 and light-harvesting arrays 2 1 or other partial models of the photosynthetic apparatus.29'41 The lack of any a-substituent in dipyrromethenes and thus the lack of unfavorable inter-ligand steric interactions when several ligands chelate one metal, has a pronounced effect on the stereochemistry of the ligands around the metal centre.20 Recently, Bruckner in our group reported a series of divalent transition metal a-unsubstituted meso-phenyldipyrromethene complexes 4 2 They were prepared by treatment of a concentrated MeOH solution of the meso-phenyldipyrromethenes 124 or 125 with the divalent metal ions C o 2 + , N i 2 + , C u 2 + , and Z n 2 + as their acetates (Scheme 2-8). Mass-spectrometry I I I I >i (£_ j ' V I — I " i I 130 R=H, M=Ni Scheme 2-8. Formation of meso-phenyldipyrrinato complexes from dipyrromethenes. Reaction conditions: M(II)(OAc)2/MeOH/N(Et)3 125 analyses confirmed the stoichiometry of the complexes as M(Ligand)2. The nickel, copper, and zinc complexes are stable and form X-ray quality dichroic (metallic dark green/red) crystals upon slow evaporation of a CHCl3 / l% MeOH solution. Figure 2-3 shows the X-ray crystal structure of bisfmeso-phenyl-dipyrrinato] Ni(U) 130. The molecule has D2-symmetry, which makes the two ligands equivalent and endows a C2-axis passing through the p-hydrogens of the meso-phenyl, the methine carbons and the central metal. The planes of the two essentially planar dipyrromethene ligands enclose a dihedral angle of 38.5°. The corresponding angle in [3,3',5,5'-tetramethylpyrrinato]nickel(U) is 76.3°. Bruckner concluded that complex 130 bears a Figure 2-3. X-ray structure of complex 130 126 distorted squared planar geometry with relatively small distortion angle resulting directly from the smaller size of the a-H as compared to the oc-methyl group.42 The bite angle N-Ni-N a of the ligand is 94.3° and N-Ni-N b angle 152.5°. The four Ni-N distances are equal (1.879 A) and unusually short for complexes of this kind. This effect is regarded to be partially due to the reduced ionic radius of the d 8 low spin ion and partially due to the reduced inter-ligand steric interactions. Inspection of the intra-ligand bond lengths reveals that two types of C-N bonds exist, a short C a - N bond and a long C a -N bond. The differences can be accounted for in terms of a resonance description of the Tt-electrons in the ligand molecule that the C a - N bond would receive partial %-contribution, the Ca-N bond would not. Figure 2-4 shows the two limiting resonances and the associated bond lengths.42 The mean plane of the meso-phenyl group is tilted 58.1° with respect to the mean plane of the dipyrromethene unit. This deviation from the expected orthogonal finds its parallels in the structure of TPPs. Figure 2-4. Limiting resonance forms of dipyrrinato ligands. Bond distances are given in A. 127 The optical properties of dipyrromethene metal complexes bear great similarity. Their optical spectra are nearly indistinguishable in non- or weakly coordinating solvents such as benzene, methanol, methylene chloride or chloroform but show variations in pyridine. Figure 2-5 shows the UV-visible spectra of complexes zinc chelate 127 and copper chelate 128. Their optical spectra can divided into two patterns: a very intensive absorption in the region of 460-500 nm, which is assigned to %-n* transitions and relatively weak bonds in the 260 -350 nm region which is attributed by charge transfer transition. From the intensity of the longest wavelength transition, the geometry of the ligand field has been evaluated using Martell's model (Equation 2-1), which is based on the assumption that the intensity of such transition changes with the tetrahedral angle between the ligands.43 = Sin 0 Equation 2-1 0 is the tetrahedral angle, e is the extinction coefficient of a reference compound known to be tetrahedral and e e is the extinction coefficient of a similar compound whose geometry is to be determined. According to the above equation, the calculated tetrahedral angle in zinc complex 127 is 90°, in copper complex 128 48° and in nickel complex 130 is 42° , 4 2 which is close to 38.5° observed in X-ray structure. The ' H and I 3 C N M R spectra of the diamagnetic metal complexes 127, 129, 130 have been obtained. Analyses of these data show that the spectrum of the zinc chelate is, as expected, very similar to that of protonated ligand. However, a large low field shift of the oc-protons is observed in the nickel chelates 129 and 130, i.e. a shift of +2.48 ppm for 129 as compared to the zinc analog 127. This is considered to be supporting evidence for the small dihedral angle of the ligand mean planes in the nickel complex. The oc-protons experience shielding effects of both the aromatic dipyrrinato systems and thus are more shielded when compared to the tetrahedral zinc complex, where such 'double' shielding cannot occur. Analysis of other aspects of the N M R spectra also supports the conclusion regarding existence of the square planar coordination geometry of nickel(U) chelates. Square planar nickel(U) complexes are typically diamagnetic and tetrahedral Ni(U) complexes paramagnetic. Nickel(H) chelates of a-substituted dipyrromethenes have been 129 described as paramagnetic and, therefore, their description as distorted tetrahedral rather than distorted square planar is plausible regardless of the actual dihedral angle between the ligands. It was, therefore, surprising, that the nickel chelates 129 and 130, as judged by their sharp *H and 1 3 C - N M R spectra, proved to be diamagnetic. This suggests that they are (distorted) square planar. 1.4 G O A L O F P R O J E C T Dipyrrinato complexes in which the coordination number of the central metal is higher than four are either mixed ligand systems44 or those where carbonyl oxygens of a-substituents in bisdipyrrinato complexes act as additional donors, forming N 4 O 2 - and N4.04-donor sets. 4 5 ' 4 6 Dipyrrinato tris-chelates (Ng-donor set) are for steric reasons only attainable using a-unsubstituted dipyrromethenes. We are aware of only two earlier reports of such complexes; two 1974 reports by Murakami et al. of the tris(3,3',4-trimethyldipyrrinato) complexes of manganese(in)47 and iron(IH)48. However, their molecular structures were not determined and, in fact, some spectroscopic properties indicated less than octahedral symmetries. We present in the next chapter another example in which the lack of inter-ligand steric interactions in meso-phenyl dipyrromethene ligands markedly distinguishes their coordination properties from those of a, 6-alkylated systems, namely the ability to form neutral octahedral dipyrrinato tris-complexes with trivalent metals. We chose cobalt(III) and iron(in) as typical examples. An X-ray single crystal structure analysis of the 130 cobalt(IU) complex illustrates their high symmetry and represents the first structurally characterized tris-dipyrrinato chelate, and, in fact, the first report on cobalt(IJJ) dipyrrinato complexes in general. Furthermore, we have been able to demonstrate the functional group interconversion of selected p-substituents of cobalt(III) trispyrrinato complexes, thereby underlining their stability and potential for incorporation into larger structures. 131 2. RESULTS AND DISCUSSION 2.1 FORMATION OF meso-PHENYLDIPYRROMETHENES The dipyrromethenes 124, 125, and 126 used in this study were formed by oxidation of the corresponding dipyrromethanes 120, 121, and 122 using DDQ 2 0 ' 2 1 (Scheme 2-2). The dipyrromethanes were synthesized using the general procedures 91 99 provided by Lee and Lindsey or Carrell . However, isolation and purification of the dipyrromethenes is not mandatory for their use as ligands and, thus, a simplified procedure involves the one-pot oxidation of the dipyrromethane, basification of the reaction mixture and subsequent metallation of the resulting dipyrromethene. 2.2 FORMATION OF TRISPYRRINATO COMPLEXES OF Co(III) AND Fe(III) The (distorted) tetrahedral cobalt(II) complexes of both a- and 8-alkyl substituted dipyrromethenes have been frequently reported, and in every case they have been described as air-stable. 1 3 ' 2 9 ' 4 9 ' 5 0 Following a general protocol,2 0 we found that, as expected, a solution of cobalt(U) acetate in methanol yielded an orange precipitate when reacted with meso-phenyldipyrromethene 122 and base. Analysis using mass spectrometry indicated the formation of the anticipated product bis(meso-phenyldipyrrinato)cobalt(n) (HR-MS (+EI, 200 °C) m/z calcd for C 3 0 H 2 2 N 4 C 0 : 497.1176, found: 497.1171). The optical spectrum of a freshly prepared sample showed a 132 clear resemblance in shapes and values to the previously prepared nickel(U), copper(U), and zinc(II) complexes. However, we were unable to purify the product due to its decomposition into orange polar products. Upon standing open to air, a mixture containing an excess of ligand formed a non-polar orange pigment with a mass corresponding to the oxidized cobalt(UI) trispyrrinato complex 131 (HR-MS (+EI, 220°C) m/z calcd. for C 4 5 H 33N 6 C o : 716.2098; found: 716.2095). We thus decided to directly use a relatively labile cobalt(UI) salt ([Co m (py)4Cl2]Cl) 4 1 (py = pyridine) and, the use of this complex rather than Co(OAc)2 produced the desired trisdipyrrinato complexes 131-135, as precipitates, with dipyrromethenes 124, 125, and 126. The precipitates were purified by chromatography to produce bright orange crystalline materials. The combined yield for the oxidation of the dipyrromethanes to the dipyrromethene and the metal complexation performed as an one-pot synthesis is -30 % and the reaction can be scaled up to produce 1 gram of metal complex per run. Column chromatography is a particularly convenient method to isolate the resulting complexes since they are neutral and generally 124 R=N0 2 125 R=H 126 R = C 0 2 C H 3 131 R=H, M=Co 132 R=N0 2 , M=Co 133 R = C 0 2 C h 3 , M=Co 134 R=H, M=Fe 135 R=N0 2 , M=Fe Scheme 2-9. Formation of trisfmeso-phenyldipyrronato] complexes from dipyrrins. Reaction conditions: M(III)(OAc)2/MeOH/N(Et)3 133 the least polar component in the reaction mixtures. An excess of metal lowered the yields and caused the formation of polar orange products which we assume to be mixed ligand systems, containing dipyrromethene units as well as acetate, hydroxo- chloro groups or other potential ligands present in the reaction mixture. Cobalt(JH) shows a particular affinity for nitrogen donor atoms. In fact, the preferred oxidation state of cobalt in the presence of nitrogen donor ligands is III. Examples of this type of complex are profuse. ' It has been noted by Murakami et al44' 4 5 that the bulkiness of cc-substituents in dipyrrinato complexes of iron and manganese determines the preferred oxidation states of the metal ion. Bulky methyl substitution allowed only the formation of bisdipyrrinato complexes with a tetrahedral conformation, resulting in the formation of divalent metal complexes with a 1:2 metal to ligand molar ratio, while the lack of any substituents at these positions allowed the formation of octahedral 1:3 complexes with a trivalent metal. In other words, the ligand, by allowing or disallowing certain coordination geometries, stabilizes the metal in the oxidation state which best suites the enforced coordination geometry. This effect of control of the oxidation state brought about by the a-substituents (or the lack thereof) parallels the effect of control of spin state in the series of nickel(U) bisdipyrrinato complexes 4 2 This rationalization suggests that mesophenyldipyrromethenes should form stable iron(ni) triscomplexes and this was, indeed, observed in my study. Conversely, iron(U) formed only oxidatively unstable bis(dipyrrinato) complexes and, in the presence of 134 excess ligand, spontaneously formed the iron(in) tris(dipyrrinato) complexes. Thus, iron(IIT), as its chloride, readily reacted in methanol and in the presence of base, with the dipyrromethenes 124 and 125 to provide precipitates which, after purification by recrystallization or chromatography, was analyzed and characterized (elemental analysis, mass spectrum) to be of the trisdipyrrinato structures 134 and 135 (Scheme 2-4). 2.3 SPECTROSCOPIC PROPERTIES OF TRIS-DIPYRRINATO COMPLEXES OF Fe(III) AND Co(III) The UV-visible spectra of cobalt and iron complexes 132 and 135 are shown in Figure 2-6. They resemble those of structurally characterized bivalent transition metal complexes of raeso-phenyldipyrromethenes42 and, therefore, it can be assumed that in -r 1 1 1 1 1 200 300 400 500 600 700 wavelength [nm] Figure 2-6. UV-visible spectra of complexes 132 (solid line) and 135 (broken line) 135 these trivalent metal complexes the dipyrromethene moieties are not distorted but, as expected, flat. Two sets of peaks can be distinguished. The higher energy transitions, which are identical in both p-nitro substituted complexes but which differ with the type of phenyl substituents present, can be attributed to n-n* transitions of the p-nitrophenyl group and those between 400 and 540 nm to ligand-to -metal charge transfer and n-n* transitions of the dipyrrinato moiety. The spectrum of 135 is slightly hypsochromically shifted compared to that of the B-alkylated analog.39 This is a general trend in the more electron deficient meso-phenyldipyrromethenes.20 The simple L H and 1 3 C N M R spectra of the diamagnetic cobalt(UI) species resemble those of the non-metallated ligands and lay testimony to the assumed high symmetry of the complex. One feature of the N M R spectra actually suggests an octahedral structure, namely the particularly large high-field shift of the oc-protons (8= 6.43 ppm) of the ligand in, for instance, 132 compared to that of the analogous distorted square planar nickel(U) (8 = 10.83 ppm) or the tetrahedral zinc(U) bis(dipyrrinato) complexes (p8 = 7.59 ppm).2 0 Models demonstrate that in a tetrahedral arrangement the oc-protons of the dipyrrinato units are the furthest away from the opposing pyrrolic ring. A planar arrangement 'overlaps' the protons, which are then subject to a 'double shielding' effect by two aromatic dipyrrinato systems. An octahedral arrangement, however, would 'push' the oc-protons towards a face of a neighboring aromatic unit, thereby affecting the observed deshielding. 136 2.4 CRYSTAL STRUCTURE OF TRIS[(5-PHENYL) DIPYRRINATO] Co(III) • ACETONE A crystal of 131 (as its acetone solvate) suitable for X-ray crystallography was grown and its analysis fully confirmed the proposed octahedral structure. An ORTEP representation of the molecule is shown in Figure 2-7 and some relevant experimental data are listed in Table 2-1. Positional parameters and equivalent isotropic thermal parameters, selected bond length and angles are listed in Tables 2-2 to 2-4 (Appendixes). Figure 2-7. X-ray structure of 131 137 The six coordinating nitrogens in complex 131 form an almost perfect octahedral coordination sphere around the central metal. The complex 131 has exact C2-symmetry. The distortion from the, perhaps, expected Cj-symmetry is, however, very small. The acetone is 1:1 disordered about the two-fold axis with the terminal atoms C(26) and C(27) located on the twofold axis. The weso-phenyldipyrrinato ligand molecules are flat and they enclose dihedral angles of only 1.1 and 2.2° off the ideal 90° and the cobalt-nitrogen bond lengths (1.945(2) A) are, within the experimental uncertainty, equal and within the expected range. The bite angles of the two non-equivalent ligands are 87.25(9) and 92.04(9)°. The trend in the bond length differences between the two pyrrolic Coc-Cft bond lengths is equivalent to those observed before, and they find the same explanation in 90 the resonance structures involved. o The short distance (2.42 A) from the a-hydrogens to the nitrogens of the opposing ligands is remarkable, e.g. HI (attached to C l ) to N3. As explained above, this short distance is a result of the octahedral arrangement of the ligands and the given length of the metal-nitrogen bond. However, the lack of any appreciable distortion within the ligands, or within the arrangement of the ligands around the central metal to prevent such a close contact, allows speculations about the existence of a stabilizing hydrogen-bond interaction between these atoms. Such weak C - H - - N interactions have been described before and are also known to influence solid state structures.53 138 2.5 CHEMICAL TRANSFORMATION OF ^-PHENYL SUBSTITUENTS OF TRIS[(5-PHENYL)DIPYRRINATO] Co(III) COMPLEXES The use of trisdipyrrinato complexes in larger structures such as model systems for certain aspects of the photosynthetic apparatus2 1'3 3'5 4'5 5 or for the construction of large supramolecular structures56 requires simple functionalization or functional group interconversion of existing functionality. The potential for such reactions is shown with some example reactions performed on cobalt(in) complexes (Scheme 2-10). 132 ^ 2 ' p d ^ C » 136, n=3, M = Co(lll), R = NH 2 1 2 7 H 2 • p t o 2 » . 137 n=2, M = Zn(ll), R=NH 2 H O O H 2 0/OH" 133—s ^ 138, n=3, M = Co(lll), R = C Q 2 H SOCI, 132 n=3, M = Co(lll), R = N 0 2 M e O H 133 n=3, M = Co(lll), R = C O 2 C H 3 - * 139 n=3, M = Co(lll), R = COCI 127 n=2, M = Zn(ll), R = N 0 2 Scheme 2-10. Chemical reactions of p-phenyl substituents of dipyrrinato metal complexes Hydrogenation of the p-nitro groups of 132 in presence of Pd/C produced the p-amino-substituted complex 136, accompanied by a small amount of other metal-reduced complexes. Under the same conditions, attempted hydrogenation of bis(/?-nitro-phenyldipyrrinato) Zn(U) complex 127 gave no desired p-amino-substituted complex 137, only the product of decomposition. The use of more selective catalyst, Pt02, smoothly gave product 137 without any reduction of the metal center or loss of chromophore. 139 Interestingly, it has been reported that free base dipyrromethenes are susceptible to reduction to give the corresponding leuko-compounds.19 Mass spectrometry and ' H N M R confirmed the successful reduction of the nitro- group to an amino- group. UV-visible spectrum of compound 137 was quite similar to that of the p-nitrophenyldipyrrinato zinc(U) complex 127. The saponification of methyl ester 133 produced the free acid 138 which was converted to the corresponding acid chloride 139. This was directly, for mere demonstration purposes, smoothly converted back to the methyl ester 133. The acid chloride 139 was later reacted with amines to form amides. One promising application of dipyrronato metal complexes is to construct large structures such as dendrimers. A key step in the synthesis of dendritic fragments is Zn 141 Scheme 2-11. Reaction of 137 with benzoic acid under catalytic conditions using D M A P / D C C 140 generation growth, which requires that the coupling step, such as an amidation reaction or esterification, be optimized. In order to utilize this type of complex to construct dendrimers, we first carried out some standard investigations on the amidation or esterification. Reaction of benzoic acid with compound 137 in the presence of 4-(dimethylamino)-pyridine (DMAP) and DCC under a variety of conditions gave compound 141 (Scheme 2-11), rather than the expected amide 140. In order to overcome this and avoid the additional step of acid chloride formation, direct coupling of acid to the amide was investigated. The condensation of benzoic acid with 137 in the presence of dicyclohexylcarbodiimide (DCC) and D M A P under a variety of reaction conditions still gave no compound 140, but 42% yield of acylurea compound 141. As the formation of the urea is known to be favored at high pH, D M A P was replaced by its /7-toluenesulfonic acid salt, as described by Moore and Stupp.57 Reaction of benzoic acid and 137 with DCC and 4-(dimethylamino)-pyridinium p-toluenesulfonate (DPTS) in THF gave 67 % desired compound 140 as well as trace amounts of side products. The mechanism of the last two reactions is illustrated in Scheme 2-12. It is believed that carbodiimide condensation reaction involves the O-acylisourea intermediate 142. This intermediate can follow several possible reaction pathways. Intramolecular oxygen to nitrogen acyl group transfer accounts for the formation of N-acylurea 141 (path A). Bimolecular reaction between the O-acylisourea and a second carboxylic acid led to the formation of the acid anhydride 143 and urea 144 (path B). The extent of N-acylurea relative to anhydride in the absence of other nucleophilic species is known to depend strongly on the solvent, pK a of the carboxylic acid and the pH of the reaction medium. At lower pH values N-acylurea 141 formation is known to be suppressed. This may account, in part, for the high conversions to ester obtained in the presence of TsOH. o H9 A R - N - C - N , R' 141 R Path A O A R'COOH R-N=C=N-R • H ? R R'COOH R-N-C=N-R »• Path B • j O j , o o H II H II II R - N - C - N - R + R'—COC-R' 142 143 144 Me. Me H 9 H R - N - C - N - R * V - C O R ' R"NH 2 v 137 R'COOH 145 R"NH 2 137 * ~ R'CONR" DMAP/H+ 140 Scheme 2-12. Reaction pathways in carbodiimide condensation R=cyclohexyl, R'=benzyl, R"=bis(/?-aminophenyl diprrinato) Zn(II) Optimization of the reaction conditions showed that the solvent of choice was dichloromethene. However, since amino complex 137 was only very sparingly soluble in dichloromethene, the reaction was conducted in THF. In conclusion, neither the strongly basic reactions required for the ester cleavage nor the harshly acidic conditions of the chlorination affected the complex. This is not entirely unexpected as cobalt(IU) complexes have generally high stability and are kinetically inert. 142 2.6 F U T U R E P E R S P E C T I V E The unique structure of dendrimers gives rise to properties that have, in turn, led to applications ranging from bulk polymer additives to biomedical agents. Synthesis of this type of macromolecule has attracted great attention and made rapid progress. Encouraged by successful interconversion of peripheral functionality without any reaction of metal center or lose of chromophore, we expect the construction of dendritic macromolecules by using meso-phenyldipyrrinato metal complexes through the simple reaction, e.g. direct condensation of acids and amines. These dendritic compounds are of great interest since the building block dipyrrinato metal complex has interesting physical properties such as electro- and photo- properties and corresponding dendrimers are expected to have similarly interesting electro- and chemical properties. Based on the DCC-DPTS catalyzed condensation of carboxylic acid and alcohol shown herein, second generation dendrimer 146 (Figure 2-8) may be produced by the combination of tri-alcohol and three equivalent carboxylic acid. Moreover, recursively conducting the coupling reaction on this compound, it is expected that the central core can "grow" convergently to produce di-nucleic (Co(UI) and Zn(U)) dendritic macromolecules. This convergent growth approach requires a high control over the number and placement of functional groups at the periphery of individual complexes. This can be overcome by using an excess amount of one of the reactants. It is envisaged that the proposed dendritic molecules may have new and unusual characteristics, owing to their unique structure and large size. 143 146 Figure 2-8. Model structure for 2nd generation dendrimer We have developed efficient strategies with which to synthesize dendritic macromolecules by assembling the building blocks through a convergent growth approach and we anticipate further developments in this area in the near future. 144 3. E X P E R I M E N T A L S E C T I O N [Co(pyridine)4Cl2]Cl was prepared as previously reported. A l l other reagents and solvents were purchased and used as received. The silica gel used in flash chromatography was Merck Silica Gel 60, 230-400 mesh whilst Rf-values were measured on Merck silica TLC aluminum sheets (silica gel 60 F 2 5 4 ) . Melting points were determined on a Thomas Hot Stage and stand uncorrected. *H and 1 3 C N M R spectra were recorded on Bruker AC-200 or AMX-300 instruments. UV-visible spectra were recorded using a HP8452A photo diode array spectrophotometer (instrumental precision ± 2 nm) in the solvents indicated. Elemental analyses were performed by Mr. P. Borda of the departmental microanalytical laboratory. The high and low resolution mass spectra were obtained by the departmental mass spectrometry service laboratories (G. Eigendorf, Director). 3.1 P R E P A R A T I O N 5-Phenyldipyrromethane (122) The compound 122 (7.20 g, 55 %) was prepared according to the method described by Lindsey. 2 5 5-Phenyldipyrromethene (125) 145 meso-Phenyl dipyrromethene 125 (240 mg, 55.0% based on crude material) was prepared according to the procedure described by Briickner.4 2 M W = 220.26; mp = 184 °C (Lit. 184 °C); R f = 0.39 (CH 2Cl 2-silica gel); 'H-NMR (200 MHz, CDC1 3): 5 12.9 (br s, 1H), 7.95 (d, J =8 Hz, 2H) 7.55 (d, J = 8 Hz, 2H), 7.36 (d, J = 8 Hz, 2H), 6.40 (m, 2H), 6.21 (m, 2H); 1 3 C - N M R (50 MHz, CDC1 3): 5 144.6, 144.6, 140.9, 139.5, 132.4, 128.9, 128.8, 123.7, 119.0; UV-visible (MeOH) Xmax (relative intensity): 222 (1.0), 266 (0.77) nm; LR-MS (EI, 180 °C) m/z = 220 (77, M + ) ; HR-MS (EI, 180 °C) calcd. for C i 5 H 1 3 N 3 0 2 : 267.1007; found: 267.1008; Analysis calcd. for C 1 5 H 1 3 N 3 0 2 : C 67.41, H 4.9, N 15.72; found: C 67.23, H 4.98, N 15.62. 5-(4-Nitro-phenyl)-4,6-dipyrromethane (121) 5-(4-Nitro-phenyl)-4,6-dipyrromethane (3.94 g, 81.8 %) was prepared was prepared according to the procedure described by Briickner.4 2 mp = 158 °C (Lit. 160 °C); R f = 0.36 (CHCl 3-silica gel); 'H-NMR (200 MHz, acetone-d6): 5 8.14 (d, J = 11.8 Hz, 2H), 7.95 (br s, 1H), 7.37 (d, J = 11.8 Hz, 2H), 6.18 (dd, J = 11.8, 2.5 Hz, 2H), 5.87 (m, 2H), 5.58 (s, 1H); l 3 C - N M R (50 MHz, CDC1 3): 5 149.7, 146.9, 130.8, 129.2, 123.8, 118.0, 108.8, 107.8, 43.8; UV-visible (MeOH) X m a x (rel. intensity): 222 (1.0), 266 (0.77) nm; LR-MS (EI, 180 °C) m/z= 267 (100, M + ) , 220 (9.7, M H + - N 0 2 ) ; HR-MS (EI, 180 °C) calcd. for C 1 5 H 1 3 N 3 0 2 : 267.1008; found: 267.1010. 5-(4-Nitrophenyl)-4,6-dipyrromethene (124) 146 This compound was prepared by a method analogous to that used for compound (125). Yield after chromatography: 59 %. mp = 189-191 °C; R f = 0.83 (CH 2Cl 2-silica gel); 'H-NMR (200 MHz, acetone-d6): 8 -12.0 (br, s, IH), 7.96 (d, 2H, J = 8 Hz, 2H), 7.55 (s, 2H), 7.36 (d, J = 8 Hz, 2H), 6.40 (m, 2H), 6.21 (m, 2H); l 3 C - N M R (50 MHz, CDC13): 5 145.6, 144.6, 140.9, 139.5, 132.4, 128.9, 128.8, 123.7, 119.0; IR (neat): 1555, 1520, 1515, 1510, 1450, 1340, 1050 cm"1; UV-visible (MeOH/trace NH 3 ) X m a x (log £): 264 (3.99), 300 (4.08) 434 (4.38) nm; U V -visible (MeOH/trace HC1) X m a x (log e): 258 (4.16), 336 (4.17), 470 (4.74) nm; LR-MS (EI, 150 °C) m/z = 265 (100, M + ) , 234 (19, MH + -20) , 218 (16, M+-HN0 2 ) ; HR-MS (EI, 150 °C) calcd. for C i s H n N ^ : 265.0851; found: 265.0850. 5-(4-Methoxycarbonylphenyl)dipyrromethane (123) A solution of methyl 4-formylbenzoate (5.55 g, 33.8 mmol) and pyrrole (22.6 mL, 340 mmol, 10 equiv.) in toluene (200 mL) was purged with N2 for 30 min before a saturated solution of p-toluenesulfonic acid in warm toluene (1 mL) was added and the mixture heated to reflux for 1.5 h. After cooling, the pale yellow solution was washed with dilute aqueous Na2CC>3 and then H2O. The organic phase was dried over Na2C03 and the solvents evaporated in vacuo to give a brown oil which was charged onto a flash column (6 X 25 cm, CHCI3 ) . Following residual pyrrole, the main colorless fractions were collected. The solvent was removed in vacuo to give a tan solid (4.65 g, 49 %). An analytical sample was recrystallized from CHC^/hexane to provide a white powder. Rf (CHCl3-silica) = 0.15 (spot stains bright red when fumigated with Br2) ; mp = 158 °C; 1H N M R (300 MHz, DMSO-d 6 ) 10.6 (br s, 2H), 7.87 (d, J = 9 Hz, 2H), 7.31 (d, J = 9 Hz, 2H), 6.62 (s, 2H), 5.91 (m, 2H), 5.66 (s, 2H), 5.45 (s, 1H), 3.82 (s, 3H, C H 3 ) ; 13c N M R (50 MHz, CDC1 3) 167.4, 147.9, 132.1, 130.3, 129.1, 128.9, 118.0, 108.9, 107.9, 52.6, 44.4; Analysis calcd. for C i 7 H i 6 N 2 0 2 : C, 72.84, H 5.75, N 9.99; found: C 72.45, H 5.88, N 9.68. 5-(4-Methoxycarbonylphenyl)-4,6-dipyrromethene (126) Dipyrromethane 123 (1.0 g, 3.57 mmol) was dissolved in C H C I 3 (20 mL). DDQ (891 mg, 1.1 equiv.) dissolved in warm benzene (2 mL) was added and the instantly darkening mixture was gently warmed for 30 min. Et3N (1 mL) was then added to the black solution containing precipitated material, resulting in a dark yellow homogenous solution. The solution was passed through a short column (CHCl 3-silica) and the first bright yellow fraction was collected, reduced to a yellow residue in vacuo and immediately used in subsequent metal complexation experiments. Rf (CH2CI2 / 2%MeOH-silica) = 0.2-0.3 (bright yellow spot stains slowly pink when fumigated with Br 2 ) , UV-visible (MeOH-trace NH4OH) X m a x 310, 430 nm; UV-visible (MeOH-trace HCl) X m a x 360, 472 nm; LR-MS (+EI) m/z = 278 (100, M+), 263 (15, M+-C H 3 ) , 219 (60, M + - C 0 2 C H 3 ) . 148 Tris(5-phenyl-4,6-dipyrrinato)Co(III) (131) To a stirred solution of 5-phenyldipyrromethane (111 mg, 0.5 mmol) in CHCl3:MeOH (1:1, 10 mL) were added DDQ (113 mg, Lo equiv.), dissolved in warm benzene (2 mL). After TLC examination revealed full consumption of the starting material (Rf (CH2C12:CC14, 1:1) = 0.78) (~1 hr), Et 3N (1 mL) and a solution of [Co(py) 4Cl 2]Cl (80 mg, 0.16 mmol) in MeOH (3 mL) were added. The dark mixture was gently warmed for 30 min. and then evaporated to dryness in vacuo to produce a black solid. This crude product was charged onto a flash column ( 1 5 X 3 cm, CH2CI2) . The first bright orange band was collected and evaporated to dryness to produce 131 (35 mg, 29 % based on 122). mp = 243° C; Rf (CH 2C1 2) = 0.86 ; !H-NMR (200 MHz, CDC13) 7.46 (m, 5H), 6.72 (dd, J = 1.2, 4.4 Hz, 2H), 6.40 (s, 2H), 6.3 (dd, J = 1.2, 4.4 Hz, 2H); l 3C NMR (50 MHz, C D C I 3 ) 151.7, 145.3, 138.1, 135.7, 133.0, 130.4, 128.1, 127.2, 118.5; IR (film on KBr): 1553, 1376, 1338, 1246, 1204, 1032, 994 cm"1; UV-visible (CH 2C1 2) Xmax (log e) 318 (4.34), 398 (4.26), 466 (4.83), 504 (4.77) nm; LR-MS (+FAB, thioglycerol) m/z = 111 (16, MH+), 603 (5, M+- 2 x C 6H 5), 497 (68, M+- C i 5 H u N 2 ) , 278 (43), 220 (35, ligandH+); HR-MS (+EI, 220 °C) m/z calcd. for C 4 5 H33N 6Co: 716.2098; found: 716.2095; Analysis calcd. for C 4 5 H 3 3 N 6Co: C, 75.40, H 4.64, N 11.72; found: C 74.91, H4.71, N 11.55. Slow evaporation of an acetone solution produced 131 as its mono acetone solvate in X-ray quality ruby-red dichroic (red-green) crystals. Analysis calcd for C 4 5H33N 6Co-C 3H 60: C 74.41, H 5.07, N 10.85; found: C 74.73, H 4.90, N 11.31. 149 Tris[5-(4-methoxycarbonylphenyl) -4 ,6-dipyrrinato]Co(III) (133) The 5-(4-methoxycarbonylphenyl)dipyrromethene 126 prepared as described above was dissolved in MeOH (10 mL) and E t 3 N (1 mL). To the heated solution, [Co(py)4Cl2]CI (230 mg, 0.54 mmol, 0.15 equiv. based on dipyrromethane 123) dissolved in methanol (2 mL) was added dropwise over a period of 30 min, until TLC evaluation indicated that all the dipyrromethene was consumed. Work-up of the bright orange solution was analogous to that described for 131, albeit with a more polar eluent (CH2CI2, then 2% MeOH/CH2Cl2), to produce 133 as an orange microcrystalline powder (350 mg, 33 %). mp > 280°C; Rf (CH 2C1 2) = 0.7 (bright orange spot); ' H N M R ( C D C I 3 , 300 MHz) 8.10 (d, J= 9 Hz, 2H), 7.53 (d, 7= 9 Hz, 2H), 3.97 (s, 3H), 6.66 (m, 2H), 6.42 (s, 2H), 6.35 (m, 2H); 13'c N M R (50 MHz, C D C I 3 ) 167.2, 152.6, 145.3, 143.0, 135.6, 133.4, 130.9, 130.7, 129.0, 119.6, 52.8; IR (KBr): 1721, 1548, 1557, 1378, 1346, 1280, 1249, 1110, 1048, 997, 890, 829, 771, 759, 721 cm"1; UV-visible (CH 2C1 2) Xmax (rel. intensities) 270 (sh), 310 (0.45), 400 (0.33), 470 (1.0), 506 (0.85) nm; MS (+FAB, NBA) m/z = 891 (8, MH+), 613 (10, MH+-ligand); Analysis calcd. for C 5 i H 3 9 N 6 0 6 C o : C, 68.76, H 4.41, N 9.43; found: C 68.47, H 4.45, N 9.30. Tris[5-(4-carboxylphenyl)-4 ,6-dipyrrinato]Co(III) (138) Ester 133 (200 mg, 0.225 mmol) was suspended in a 1:1 mixture of THF and 4% K O H (aq.) (10 mL) and the suspension heated to reflux for 3 h during which time all material dissolved. After cooling, 6N aqueous H C l was added dropwise until the free acid 150 precipitated as a red solid. This was filtered and dried in vacuo at 60°C to produce the title compound as a red solid (170 mg, 90%). Elemental analysis indicated the presence of varying amounts of solvent in this hygroscopic solid. Rf ( C H 2 C l 2 / 5 % MeOH-silica) = 0 (bright orange-red spot); ' H - N M R (200 MHz, DMSO-d6) 6.38 (m, 1H), 6.44 (s, 1H), 6.66 (m, 1H), 7.62 (d, J= 9 Hz, 1H), 8.20 (d, J= 9 Hz, 1H); IR (KBr): 3440 (br), 1687, 1547, 1390, 1350, 1257, 1035, 994, 883, 766, 720 cm" 1; MS (EI) m/z = 848 (M+); UV-visible (MeOH/H 2 0 1:1) ? i m a x (rel. intensities) 308 (0.45), 400 (0.33), 468 (1.0), 506 (0.85) nm. Tris[5-(4-chlorocarbonylphenyl)-4,6-dipyrrinato]Co(III) (139) Acid 138 (120 mg, 0.135 mmol) was dissolved in a little dry THF, oxalyl chloride (17 mg, 0.135 mmol) and a trace of D M F were added and the orange solution was warmed for 30 min. A TLC of this solution revealed one low polarity streaking orange spot and consumption of all starting material (89 mg, 73 %). The mixture was then evaporated to dryness in vacuo to produce an orange film. Alternatively, neat S O C l 2 (followed by evaporation to dryness while small amounts of CCI4 were added) was used to accomplish the conversion of 138 to 139. The crude acid chloride was used immediately. LR-MS (+EI) m/z = 904 (M+). The identity of 139 was also shown by dissolution in MeOH containing Et3N, to quantitatively generate the ester 133. Tris[5-(4-nitrophenyl)-4,6-dipyrrinato]Co(III) (132) 151 Following the procedure as for the preparation of 131, the title compound was obtained from 5-(4-nitrophenyl)dipyrromethane 121 as a brick-red material (0.24 g, 38 %). mp >320° C; Rf (CH 2C1 2) = 0.82; iH-NMR (200 MHz, CDC13) 8.32 (d, J = 8.0 Hz, IH), 6.38 (dd, / = 1.6, 4.6 Hz, IH), 7.62 (d, 8.0 Hz, IH), 6.61 (dd, 7 = 1.6, 4.6 Hz, IH), 6.43 (s, IH); IR (film on KBr): 1534, 1341, 1243, 1033, 994, 889 cm"1; UV-visible (CH 2C1 2) ^max Gog e) 306 (4.57), 396 (4.33), 472 (4.73), 508 (4.67) nm; LR-MS (+FAB, NBA) m/z = 851 (M+); Analysis calcd. for C 4 5H3oN 9 0 6 Co .C3H 6 0: C, 63.46, H 3.99, N 14.80; Found: C 63.09, H 3.77, N 13.94. Tris[5-(4-aminophenyl)-4,6-dipyrrinato]Co(III) (136) The trinitro complex 132 (160 mg, 0.19 mmol) dissolved in THF (10 mL) containing Et3N (1 drop) was stirred with 10%Pd on carbon (30 mg) while H 2 was bubbled through the solution for 3 h. No starting material could be detected after this time by TLC. The solution was filtered over Celite® and evaporated to dryness in vacuo. The residue was taken up in C H C I 3 and the product precipitated by addition of E t 2 0 . The solids were separated by filtration, washed with a little cold EtOH followed by E t 2 0 and then dried under vacuum at 60°C to provide title compound 136 as a red powder (112 mg, 78 %). Rf (CH 2C1 2) = 0.1 (orange spot); mp > 200 °C (dec); !H-NMR (300 MHz, DMSO-d 6 ) 7.77 (dd,7= 1.5, 4.5 Hz, 1H),7.11 (d ,7=8Hz, IH), 6.64 (d, J = 8 Hz, IH), 6.33 (dd,7 = 1.5, 4.5 Hz, IH) 6.14 (s, IH), 5.51 (s, IH); 1 3 C - N M R (50 MHz, DMSO-d 6 ) 149.9, 149.8, 147.7, 135.4, 132.5, 131.8, 124.5, 118.3, 112.5,; IR (KBr): 3456, 3363, 1707, 1618, 1543, 1380, 1344, 1247, 1205, 1043, 1002, 885, 809, 771, 735, 714 cm" 1; UV-visible 152 (CH2CI2) ? t m a x (e) 394 (4.41), 466 (4.75), 502 (4.66) nm; LR-MS (+FAJ3, thioglycerol) m/z = 762 (30, MH+), 528 (M+-ligand); Analysis calcd. for C 4 5 H 3 6 N 9 C o : C, 70.95, H 4.76, N 16.55; Found: C 70.58, H 4.66, N 16.32. Tris(5-phenyl-4,6-dipyrrinato)Fe(III) (134) Prepared in a yield of 51 % yield as dark green crystalline material with a metallic luster from anhydrous FeCl3 and 123 by a method similar to the procedure described for 131. mp 286 0 C; Rf (CH 2C1 2) = 0.79 (brown spot); IR (film on KBr): 1553, 1378, 1330, 1039, 994, 835 cm" 1; UV-visible (CH2CI2) A , m a x Gog e) 316 (4.34), 442 (4.72), 490 (4.59) nm; LR-MS (+EI, thioglycerol) m/z =714 (3.5, M+), 600 (3), 494 (35, M+-ligand), 275 (20); Analysis calcd. for C 4 5 H 3 3 N 6 F e : C, 75.73, H 4.66, N 11.77; found: C 75.92, H 4.75, N 11.87. Tris[5-(4-nitrophenyl)-4,6-dipyrrinato]Fe(III) (135) Prepared in 53% as dark green microcrystalline material from 124 as described for 134. mp>320 0 C; Rf (CH 2C1 2) = 0.76 (brown spot); IR (film on KBr): 1552, 1340, 1238, 1040, 994, 822 cm- 1; UV-visible (CH 2C1 2) ? i m a x (log e): 272 (4.82), 304 (4.70), 448 (4.81), 498 (4.69) nm; LR-MS (+FAB, thioglycerol) m/z = 848 (4.2, M+), 584 (2, M+-ligand); Analysis calcd. for C 4 5 H 3 o N 6 0 6 F e C 3 H 6 0 : C 66.67, H 4.2, N 9.72; found: C 63.48, H 3.93, N 13.86. 153 4-(DimethyIamino)-pyridinium /7-toluenesuIfonate (DPTS) 4-(Dimethylamino)-pyridinium p-toluenesulfonate (DPTS) was prepared (6.9 g, 88%) according to the procedure described by Moore.57 The reagent is not hygroscopic and is stable indefinitely when stored at room temperature. mp. 165 °C (Lit. 165 °C); 1H-NMR (300 MHz, DMSO-d6): 14.42 (s, br, 1H) , 7.8-8.3 (m, 4 H ) , 7.16 (m, 2 H ) , 6.72 (m, 2 H ) , 3 .22 (s, 6 H ) , 2 .33 (s, 3 H ) ; LR-MS (+FAB, thioglycerol) m/z = 294 (3.6, M+), 111 (24, M+-TsO); Anal, calcd. for C 1 4 H 1 8 N 2 S O 3 : C 57.14, H 6.12, N 9.52; found: C 56.95, H 6.16, N 9.37. Bis[5-(4-nitro-phenyl)-4,6-dipyrrinato]Zn(II) (127) This compound was prepared according to the method described in Christian Briichner's PhD. Thesis.2 8 Bis[5-(4-amino-phenyl)-4,6-dipyrrinato]Zn(II) (137) Complex 127 (160 mg, 0.19 mmol) was dissolved in THF (10 mL) containing Et^N (1 drop) and the solution stirred with 10 % Pt0 2 on carbon (30 mg) under 1 atm H 2 . After 3 h, no starting material was detected by TLC. The solution was filtered through Celite® and the solvents evaporated to dryness in vacuo. The residue was dissolved in CHCI3 and the product precipitated by addition of E t 2 0 . The solids were filtered off, washed with a little cold EtOH followed by E t 2 0 and then dried under vacuum at 60°C to provide (23) 154 as a yellow powder (112 mg, 78 %). Rf ( C H 2 C 1 2 ) = 0.1 (orange spot); mp > 2 0 0 °C (dec.); 1 H-NMR (300 MHz, acetone-d6): 6.72-7.44 (m, 4 H ) , 6.80 (m, 4 H ) , 6.42 (m, 2 H ) , 5.16 (s, 2 H ) ; UV-visible ( C H 2 C I 2 ) A . m a x (log e): 458 (4.74), 478 (5.10) nm; LR-MS (+FAB, thioglycerol) m/z = 533 (M+-1); Analysis calcd. for C 3 o H 2 4 N 6 Z n : C, 67.41, H 4.49, N 15.73; found: C 67.83, H 4.23, N 16.02. Bis[5-(4-benzylamide phenyl)-4,6-dipyrrinato]Zn(II) (140) A solution of Bis[5-(4-amino-phenyl)-4,6-dipyrrinato] Zn(IJ) 137 (53 mg, 0.1 mmol) in dry THF (5 mL) was slowly added to a solution of benzoic acid (122 mg, 1 mmol), D C C (50 mg) and a small amount of D M A P in THF (10 mL). The mixture was stirred for 5 h and T L C showed no more red acid 138 was left. The solvent was removed in vacuo and the resulting pink residue charged onto a short silica gel column eluting with 5% methanol in dichloromethane gave a yellow band which was collected and the solvents removed in vacuo. Recrystallization of resulting solid from an acetone solution gave yellow crystals. ' H - N M R (300 MHz, C D C 1 3 ) : 7.90 (m, 3 H ) , 7.76 (m, 2 H ) , 7.50-7.65 (m, 7 H ) , 6.78 (d, 2 H ) , 6.42 (d, 2 H ) ; ' H - N M R ( C D C 1 3 , add 1 drop of DMSO-d 6): 9 .34 (s, 1H) , 7.88 (m, 2 H ) , 7.78 (m, 2 H ) , 7.45-7.60 (m, 7 H ) , 6.74 (d, 2 H ) , 6.38 (d, 2 H ) , LR-MS (EI) m/z = 741 (M+); UV-visible ( C H 2 C I 2 ) ? w ( l o g e ) : 4 5 6 (4-64)> 4 7 8 ( 5 - 0 6 ) n m -Tris[5-(4-(2'-hydroxyethoxycarbonyl)phenyl)-4,6-dipyrrinato]Co(III) (144) 155 Cobalt complex tricarboxylic acid 138 (30 mg, 0.035 mmol), DPTS (0.031 g, 0.106 mmol) and ethylene glycol (0.22 g, 3.5 mmol) in dry THF (10 mL) was stirred under N 2 atmosphere for 15 min. DCC (28 mg, 0.135 mmol) in dry THF (5 mL) was added slowly by syringe. The reaction mixture was stirred under N 2 for 24 h. At the end of this period, TLC showed a main red spot and two minor red spots. The reaction mixture was filtered, and solvent was removed in vacuo to obtain an orange residue. It was purified by chromatography (silica gel) eluting with 1:10 methanol : methylene chloride. The main red band was collected and the solvents were removed in vacuo. The residue was crystallized from chloroform to afford red crystalline prisms (15 mg, 43.7 %). 'H N M R (200 MHz, C D C 1 3 ) : 8.06 (m, 2 H ) , 7.50 (m, 2 H ) , 6.62 (m, 2 H ) , 6.34 (m, 2 H ) , 6.22 (m, 2 H ) , 4.48 (t, 2 H ) , 3.96 (t, 2 H ) , 1.92 (s, br, IH); 1 3C N M R (75 MHz, C D C 1 3 ) : 203.23, 166.57, 144.79, 142.81, 135.13, 132.90, 130.46, 130.00, 128.72, 119.17, 102.72, 102.49. LR-MS (EI) m/z = 981 ([M+l]+); UV-visible ( C H 2 C I 2 ) ? W (log e): 456 (4.64), 478 (5.06) nm. 156 3.2 X - R A Y C R Y S T A L L O G R A P H I C A N A L Y S I S O F 1 3 1 - A C E T O N E Crystallographic data are summarized in Table 2-1. The final unit-cell parameters were obtained by least-squares on the setting angles for 25 reflections with 20 = 28.5-41.8°. The intensities of three standard reflections, measured every 200 reflections throughout the data collections, decayed uniformly by 5.2%. The data were processed, corrected for Lorentz and polarization effects, decay, and absorption (empirical: v|/-scans) 6 0. The structure was solved by conventional heavy atom methods. The structure analysis was initiated in the centrosymmetric space group C2/c on the basis of the E -statistics, this choice being confirmed by subsequent calculations. The metal complex has exact C 2 symmetry. The acetone molecule was modeled as 1:1 disordered about the twofold axis with the terminal carbon atoms C(26) and C(27) located on the twofold axis. The non-hydrogen atoms were refined with anisotropic thermal parameters. Hydrogen atoms were fixed in idealized positions ( C - H = 0.98 A , 5 H = 1.2 ^bonded a t o m ) . A correction for secondary extinction (Zachariasen type 2 isotropic) was applied, the final value of the extinction coefficient being 1.67(10) x 10" 7. 5 1 Neutral atom scattering factors and anomalous dispersion corrections for all atoms were taken from the International Tables for X-Ray Crystallography61. Final atomic coordinates and equivalent isotropic thermal parameters are given in Table 2-2 and selected bond lengths and angles appear in Tables 2-3 and 2-4 . Complete tables of bond lengths and angles, calculated hydrogen 157 atom parameters, anisotropic thermal parameters, torsion angles, intermolecular contacts, and least-squares planes are included as supplementary material. 158 Table 2-1. Crystallographic data for 131-Acetone.a compound formula fw color, habit crystal size, mm crystal system space group a, A b, A c, A o a V, A Z T, °C p g/cm3 F(000) radiation transmission factors scan type 1 3 1 C 3 H 6 0 C48H 3 9 CoN 6 0 774.81 red, prism 0.35 x 0.40 x 0. monoclinic C2/c 13.539(2) 22.348(2) 13.467(1) 107.654(9) 3882.7(8) 4 21 1.325 1616 Mo 4.88 0.91-1.00 (0-29 scan range, deg in co 1.25 + 0.35 tan 6 scan speed, deg/min 32 (up to 8 rescans) data collected +h, +k, ±1 29 m a x , deg 60 crystal decay, % 5.2 total no. of reflections 6017 unique reflections 5786 Rmerge 0.053 no. with / > 3D(7) 2309 no. of variables 266 R 0.039 Rw 0.033 gof 1.59 max A/a (final cycle) 0.0008 residual density e/A 3 -0.28, 0.35 a Rigaku AFC6S diffractometer, takeoff angle 6.0°, aperture 6.0 x 6.0 mm at a distance of 285 mm from the crystal, stationary background counts at each end of the scan (scan/background time ratio 2:1), Mo Ka (k = 0.71069 A) radiation, graphite monochromator, (^{F2) = [S2 (C + 45)]/Lp2 (S = scan speed, C = scan count, B = normalized background count), function minimized Eu>(IF0l-LFcl)2 where w = 4F 0 2 /o 2 (F 0 2 ) , R = EIIFol-IFJI/ZIFol, Rw = (Zw(\F0\-\Fc\)2/Z,w\F0\2)m, and gof = [Zw(\F0\-\Fc\)2/(m-n)] Values given for R, Rw, and gof are based on those reflections with / > 3a(7). 161 Table 2-2. Atom Coordinates and Beqm [A 2 ] for compound 131-acetone atom X y z *eq.a Co(l) 0.5000 0.24994(3) 0.7500 2.73(1) O(l) 0.4289(5) 0.0350(3) 0.3447(6) 9.6(2) N(l) 0.5052(2) 0.2528(1) 0.6073(1) 2.86(5) N(2) 0.6039(2) 0.18703(9) 0.7868(2) 2.86(6) N(3) 0.3911(2) 0.31020(10) 0.7169(2) 2.72(6) C(l) 0.4484(2) 0.2881(1) 0.5311(2) 3.36(7) C(2) 0.4646(2) 0.2751(1) 0.4354(2) 3.57(7) C(3) 0.5343(2) 0.2291(1) 0.4537(2) 3.28(7) C(4) 0.5613(2) 0.2152(1) 0.5616(2) 2.88(7) C(5) 0.6291(2) 0.1709(1) 0.6149(2) 2.80(6) C(6) 0.6510(2) 0.1583(1) 0.7208(2) 2.88(7) C(7) 0.7267(2) 0.1181(1) 0.7816(2) 3.45(7) C(8) 0.7248(2) 0.1222(1) 0.8815(2) 3.49(7) C(9) 0.6483(2) 0.1647(1) 0.8819(2) 3.41(7) C(10) 0.6808(2) 0.1331(1) 0.5542(2) 3.31(7) C ( l l ) 0.7495(3) 0.1568(1) 0.5063(3) 4.44(8) C(12) 0.7933(3) 0.1211(2) 0.4467(3) 5.6(1) C(13) 0.7697(3) 0.0610(2) 0.4354(3) 6.1(1) C(14) 0.7040(3) 0.0368(1) 0.4838(3) 5.7(1) 162 C(15) 0.6593(3) 0.0725(1) 0.5427(3) 4.54(9) C(16) 0.2888(2) 0.3004(1) 0.6841(2) 3.39(7) C(17) 0.2333(2) 0.3542(1) 0.6624(2) 3.55(7) C(18) 0.3051(2) 0.3993(1) 0.6848(2) 3.29(7) C(19) 0.4040(2) 0.3722(1) 0.7192(2) 2.83(7) C(20) 0.5000 0.4008(2) 0.7500 2.82(9) C(21) 0.5000 0.4677(2) 0.7500 3.1(1) C(22) 0.4963(3) 0.4991(1) 0.6610(2) 4.71(9) C(23) 0.4964(3) 0.5613(2) 0.6612(3) 5.5(1) C(24) 0.5000 0.5914(2) 0.7500 5.5(2) C(25) 0.4734(6) 0.0372(4) 0.2841(7) 6.1(3) C(26) 0.5000 0.0962(3) 0.2500 6.1(2) C(27) 0.5000 -0.0198(3) 0.2500 8.6(2) a 2 Beq = (8/3)7C UlUijaiaj(ai-aj) 163 Table 2-3. Selected bond length (A) for the compound 131-acetone Co(l)-N(l) 1.946(2) Co(l)-N(2) 1.944(2) N(l)-C(l) 1.336(3) N(l)-C(4) 1.395(3) N(2)-C(6) 1.397(3) N(2)-C(9) 1.335(3) C(l)-C(2) 1.402(4) C(2)-C(3) 1.367(4) C(3)-C(4) 1.421(4) C(4)-C(5) 1.392(4) C(5)-C(6) 1.394(4) C(5)-C(10) 1.490(4) C(6)-C(7) 1.421(3) C(7)-C(8) 1.357(4) C(8)-C(9) 1.407(4) Table 2-4 Selected bond angles (°) for the compound 131-acetone N(l)-Co(l)-N(l) ' 176.2(2) N(l)-Co(l)-N(2) 92.04(9) N(l)-Co(l)-N(3) 90.15(9) N(2)-Co(l)-N(3) 176.7(1) C(l)'-N(l)-C(4) 106.3(2) N(l)-C(4)-C(3) 108.2(2) C(2)-C(3)-C(4) 107.4(2) C(l)-C(2)-C(3) 106.4(3) N(l)-C(l)-C(2) 111.7(2) N(l)-C(4)-C(5) 124.3(2) C(3)-C(4)-C(5) 127.5(3) C(4)-C(5)-C(6) 125.0(3) C(5)-C(6)-C(7) 127.5(3) C(6)-C(7)-C(8) 107.5(3) C(7)-C(8)-C(9) 106.5(2) N(2)-C(9)-C(8) 111.5(3) C(6)-N(2)-C(9) 106.2(2) N(2)-C(6)-C(7) 108.2(2) N(2)-C(6)-C(5) 124.2(2) Co(l)-N(2)-C(9) 126.5(2) Co(l)-N(2)-C(6) 127.2(2) References: 1 Fischer, H. ; Orth, H. Die Chemie der Pyrrole, Akad. Verlagsges., Leipzig, 1937, VollIIPart 1. 2 Fischer H. ; Schubert, M . Chem. Ber. 1923, 56, 1202. 3 Gossauer, A. Die Chemie der Pyrrole, Springer-Verlag, Berlin and New York, 1974,180. 4 Fischer, H. ; Orth, H. Die Chemie des Pyrroles, Akad. Verlagsges., Leipzif, 1934, Vol . I. (reprint by Johnson Reprint Corp., New York, 1968). 5 Falk, H. ; Hofer, O.; Gergely, S. Monatsh. Chem. 1974, 105, 1019. 6 Pruchner F.; Stern, A . Z. Phys. Chem., Abt. A 1937, 180, 25. 7 Jeffreys R. A. ; Knott, E. B. J. Chem. Soc. 1951, 1028. 8 Falk, H. ; Gergely S.; Hofer, O. Monatsh. Chem. 1974, 105, 853. 9 Brunings K. J.; Corwin, A . H. J. Am. Chem. Soc. 1942, 64, 2106. 10 Brunings K. J.; Corwin, A . H. J. Am. Chem. Soc. 1944, 66, 337. 11 Treibs, A. ; Kreuzer, F. H. Jusus Liebigs Ann. Chem. 1969, 721, 116. 12 Treibs, A. ; Zimmer-Galler, R. Jusus Liebigs Ann. Chem. 1963, 664, 140. 13 Booth, H. ; Johnson, A . W.; Langdale-Smith, J. J. Chem. Soc. 1963, 650. 14 Gossauser, A. ; Engel, J. Review of the synthesis of dipyrromethene, The Porphyrins, Vol . I l l , Eds. D. Dolphin, Academic Press Inc. 1978, Chapter 7. 15 Rocha Gonsalves, A. M . ; Kenner, G. W.; Smith, K. M . Chem. Commun. 1971, 1034. 16 Treibs, A. ; Kolm, H. G. Justus Liebigs Ann. Chem. 1958, 614, 176. 17 Tarlton, E. J.; MacDonald S. F.; Baltazzi, E. J. Am. Chem. Soc. 1960, 82, 4389. 167 18 Ballantine, J. A. ; Jackson, A. H. ; Kenner G. W.; MacGillivary, G. Tetrahedron, Suppl. 1996,7,241. 19 Shinohara, H. ; Honda, K.; Misaki S.; Imoto, E. Nippon Kanaku Zasshi 1960, 81, 1740; Chem. Abstr. 1962, 56 3441. 20 Jacobi, P. A. ; Strell, M . ; Strell, I.; Grimm, D.; Gieren, A. ; Schanda, F. Liebigs Ann. Chem. 1978, 289. 21 Corwin, A . H. ; Viohl, P. J. Am. Soc. Chem. 1944, 66, 1137. 22 MacDonald, S. F.; Stedman, R. J. Can. J. Chem. 1954, 32, 896. 23 Castro, A . J.; Tertzakian, J. G.; Nakata, B. T.; Brose, D. A . Tetrahedron 1967, 23, 4499. 24 Bruckner, C ; Karunaratne, V. ; Rettig, S. J.; Dolphin, D. Can. J. Chem. 1996, 74. 2182. 25 Wagner, R. W.; Lindsey, J. S. Pure Appl. Chem. 1996, 68.1373. 26 Lee, C.-H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427. 27 Carell, T. Ph.D. thesis, Ruprechts-Karl-Universitat, 1994. This method was mentioned in the footnotes of a paper by Rebek and co-workers. Shipps Jr., G.; Rebek Jr., J. Tetrahedron Lett. 1994, 35, 6823-6826. 28 Bruckner, C ; Sternberg, E. D.; Boyle, R. W.; Dolphin, D. manuscript submitted for publication. 29 Johnson, A . W.; Kay, I. T.; Markham, E.; Price, R.; Shaw, K. B.; J. Chem. Soc. 1959, 3416 30 Brooker, L . G. S.; Sprague, R. H. J. Am. Chem. Soc. 1941, 63, 3203 31 Falk, H. The Chemistry of Linear Oligopyrroles and Bile Pigments; Springer: Wien, New York, 1989 and reference herein. 32 (a) Fischer, H.; Orth, H. Die Chemie des Pyrrols, Vol. 2, erste Hiilfte p. 1-151 and references therein., Akademische Verlagsgesellschaft m.b.H., Leipzig 1940. (b) Falk, H. The Chemistry of Linear Oligopyrroles and Bile Pigments, Springer Verlag, Wien, New York 1989. 168 33 (a) Rogers, M . A. T. J. Chem. Soc. 1943, 596-597. (b) Hi l l , C. L. ; Williamson, M . M . J. Chem. Soc, Chem. Commun. 1985, 1228. (c) Cavaleiro, J. A . S.; Condesso, M . d. F. P. N . ; Olmstead, M . M . ; Oram, D. E.; Snow, K. M . ; Smith, K. M . J. Org. Chem. 1988, 53, 5847. 34 Falk, H. ; Schoppel, G. Monatsh. Chem. 1990, 121, 67. 35 Murakami, Y . ;Sakata, K. Inorg. Chim. Acta 1968, 2, 273. 36 Murakami, Y . ; Sakata, K. Bull. Chem. Soc. Jpn. 1974, 47, 3025. 37 Vos de Wael, E.; Pardoen, J. A. ; van Koevering, J. A. ; Lugtenburg, J. Reel. Trav. Chim. Pays-Bas 1977, 96, 306. 38 Treibs, A. ; Kreuzer, F. H. Liebigs Ann. Chem. 1968, 718, 208. 39 Wagner, R. W. Lindsey, J. S. J. Am. Chem. Soc. 1994,116, 9759. 40 Wagner, R. W.; Lindsey, J. S.; Seth, J.; Palaniappan, V. ; Bocian, D. F. J. Am. Chem. Soc. 1996, 118, 3996. 41 Debreczeny, M . P.; Svec, W. A. ; Wasielewsky, M . R. New J. Chem. 1996, 20, 815. 42 Bruckner C. PhD Thesis 1995. 43 Biswas, K. M . ; Houghton, L . E.; Jackson, A . H. Tetrahedron, Supplement 7, 1966, 22, 261. 44 Murakami, Y . ; Matsuda, Y . ; Iiyama, K. Chem. Lett. 1972, 1069. 45 Clarke, E. T.; Squattrito, P. J.; Rudolf, P. R.; Motekaitis, R. J.; Martell, A . E.; Clearfield, A. Inorg. Chim. Acta 1989, 166, 221. 46 Murakami, Y . ; Matsuda, Y . ; Sakata, K.; Martell, A . E. J. Chem. Soc, Dalton Trans. 1 1973, 1729. 47 Murakami, Y . ; Sakata, K.; Harada, K.; Matsuda, Y . Bull. Chem. Soc. Jpn. 1974, 47,3021. Murakami, Y . ; Matsuda, Y. ; Sakata, K.; Harada, K. Bull. Chem. Soc. Jpn. 1974, 47, 458. 169 49 March, F. C ; Couch, D. A. ; Emerson, K.; Fergusson, J. E.; Robinson, W. T. J. Chem. Soc. (A) 1971,440. 50 (a) Johnson, A . W.; Kay, I. T.; Markham, E.; Price, R.; Shaw, K. B. J. Chem. Soc. 1959, 3416; (b) Fergusson, J. E.; Ramsay, C. A . J. Chem. Soc. (A) 1965 5222; (c) Murakami, Y . ; Matsuda, Y. ; Sakata, K. Inorg. Chem. 1971, 10, 1728. 51 Levason, W.; McAuliffe, C. A. Coord. Chem. Rev. 1974, 12, 151. 52 Cotton, F. A. ; Wilkinson, G. Advanced Inorganic Chemistry, John Wiley & Sons, New York 1988. 53 (a) Taylor, R. Kennard O. J. Am. Chem. Soc. 1982, 104 5063; (b) Sarma, J. A . R. P.; Desiraju G. R. Acc. Chem. Res. 1986, 19, 222. 54 Wagner, R. W.; Lindsey, J. S.; Seth, J.; Palaniappan, V. ; Bocian, D. F. J. Am. Chem. Soc. 1996, 118,3996. 55 Debreczeny, M . P.; Svec, W. A. ; Wasielewsky, M . R. New J. Chem. 1996, 20, 815. 56 for illustrative examples see e.g.: (a) Tour, J. M . Chem. Rev. 1996, 96, 537; (b) Whittle, B. Everest, N . S. Howard, C. Ward, M . D. Inorg. Chem. 1995, 34, 2025; (c) Lehn, J.-M. Supramolecular Chemistry, V C H , Weinheim 1995. 57 Moore, J. S.; Stupp, S. I. Macrocolecules 1990, 23, 65. 58 Elgy, C. N . ; Wells, C. F. J. Chem. Soc, Dalton Trans. 1980, 2405. 59 Paine J. B. I l l ; in The Porphyrins, Vol. 1, Dolphin D. Ed.: Academic Press, New York, 1978, p. 101-234. 60 teXsan: Crystal structure analysis package. Unix version 1.7. Molecular Structure Corporation. The Woodlands, TX, U.S.A. 1995. 61 (a) International Tables for X-Ray Crystallography. Kynoch Press. Birmingham, U K . (present distributor Kluwer Academic Publishers: Boston, M A , USA). 1974. Vol . IV. pp. 99-102; (b) International Tables for Crystallography. Kluwer Academic Publishers. Boston, M A , USA. 1992. Vol . C. p. 200-206. 170 PART 3 OXIDATION OF PORPHYRINS WITH N-SULFONYL OXAZIRIDINES 171 1. I N T R O D U C T I O N 1.1 I N T R O D U C T I O N OF P O R P H Y R I N - R E L A T E D M A C R O C Y C L E S Porphyrins and related tetrapyrrolic systems are among the most widely studied of all macrocyclic compounds due to their biological importance and chemical versatility.1 The term "porphyrin-related compounds" refers to macrocycles consisting of four pyrrole or reduced pyrrole rings linked directly or, more commonly, through methine bridges. Based on the placement and level of oxidation, they can be subdivided into porphyrins 147, chlorins 148, bacteriochlorins 149, isobacteriochlorins 150, corphins 151, pyrrocorphins 152, corrins 153, etc. Their skeletal structures, atom numbering scheme and names are shown in Figure 3-1. 147 148 149 150 porphyrin chlorin bacteriochlorin isobacteriochlorin 17 1 6 1 5 14 13 154 12 8 10 151 152 153 corphin pyrrocorphin cornn Figure 3-1. Chromophores of macrocyclic tetrapyrroles 172 Porphyrin-like compounds widely occur in nature and perform fundamental functions of life on earth. They often incorporate various metal ions and possess substitution patterns around their periphery. These features, along with the specific oxidation level of the tetrapyrrolic macrocycle, determine the fundamental biological functions they perform in the living system. Among these compounds, the most biologically important and well known include heme, chlorophyll a, bacteriochlorophyll, and vitamin B12 (Figure 3-2). Heme 155, the iron-containing red blood pigment, takes the form of a completely unsaturated 18-71 aromatic porphyrin chromophore. It transports oxygen and electrons. Chlorophyll a 156 is a magnesium complex of dihydroporphyrin (a C 0 2 H Heme 155 C 0 2 H C 0 2 R C 0 2 C H 3 156 Chlorophyll a 157 Bacteriochlorophyll 158 Vitamin B 12 Figure 3-2. The most important naturally occurring porphinoids 173 chlorin) and is present in all organisms capable of oxygenic photosynthesis, such as higher plants and many algae, where it occurs in both reaction centers (photosynthetic systems I and II) and in all light-harvesting complexes.2 It is the most important and most abundant chlorophyll in nature. Although one of its peripheral double bonds is reduced, it still maintains aromatic conjugation. Bacteriochlorphyll a 157 is a magnesium-containing tetrahydroporphyrin. It is the most widely distributed bacteriochlorin pigment and the main component of the photosynthetic apparatus of purple and green bacteria. Vitamin B12 158, a cobalt complex of a corrin, is the " anti-pernicious" red pigment essential for numerous biochemically important rearrangement reactions. A striking feature of all these substances is that they share a common origin: they are constructed by living systems via the common starting material uroporphyrinogen III 162, which is biosynthesized from 5-aminolevulinic acid (ALA) 159 by way of porphobilinogen (PBG) and preuroporphynnogen. uroporphyrinogen Hi 162 Scheme 3-1. Biosynthesis of Uroporphyrinogen III 174 1.2 OPTICAL ABSORPTION SPECTRA The free-base and metallated porphinoids with 18-TE conjugation systems exhibit two types of characteristic absorption in the visible region: Soret and Q bands. A l l of these bands are interpreted to be due to TX-TX* transitions.4 The Soret band appears around 400 nm with a very high intensity (log e ~ 5). This band is the most intense absorption found in all fully conjugated tetrapyrrolic systems. The intensity of the Soret band is weaker in chlorins and it is totally absent in non-conjugated tetrapyrroles such as porphyrinogens. However, this kind of band is present in vitamin B12 and in the metal complexes of bile pigments, all of which have interrupted conjugation in the ligand, but the conjugation pathway is maintained through the metal atom. In the spectra of free-base porphyrins, the Q bands are four less intense absorptions, i.e. band I, II, III, IV, which appear in the region 450-650 nm. The relative intensities of these four bands are highly related to the number, position and nature of substitution around the periphery of porphyrins and have been identified as four basic types: etio-type, rhodo-type, oxorhodo-type and phyllo-type (Figure 3-3).5 For example, porphyrins with one strongly electron-withdrawing group displays rhodo-type spectra, while the presence of two electron-withdrawing groups on diagonally opposite rings result in oxorhodo-type spectra. In contrast to free base porphyrins, metalloporphyrins exhibit two-band Q absorptions, designated a and (3, between 500 and 600 nm. The change is attributed to the increasing symmetry of the conjugated ring (from D 2 h symmetry in free base porphyrin to D 4 h symmetry in the metal complex).6 500 550 600 650 500 550 600 650 Figure 3-3. Four types of porphyrin Q-band absorptions. In chlorins, both the Soret and Q bands have significant bathchromic shift from their parent porphyrins. Band I of the Q bands in the visible region is very prominent (Figure 3-4) and about 25 nm longer wavelength than that of porphyrins. The ratio of the Soret band to band I is approximately 5:1 (versus about 50:1 in porphyrins). The extinction coefficients of band IV and the Soret band, in neutral solvents, are comparable with those of the related bands in analogous porphyrins. Bacteriochlorin optical spectra exhibit a further bathchromic shift in band I (around 720 nm). The ratio of the Soret to band I decreases to about 2:1. The strong absorption and bathchromic shift of band I in m e t a l l o c h l o r i n c h l o r i n I 400 500 600 700 800 nm Wavelength Figure 3-4. Typical optical spectra of chlorins and metallochlorins chlorins and bacteriochlorins make them of great interest for studies including photodynamic therapy. 1.3 THE USE OF PORPHYRINS AND THEIR DERIVATIVES IN PHOTODYNAMIC THERAPY (PDT) 1.3.1 General Introduction to PDT Photodynamic therapy (PDT) is a medical treatment which uses a combination of photosensitizer, visible light and oxygen to bring about destruction of diseased tissue and cells. This photomedicinal approach has been practiced since the time of the ancient Egyptians.8 They used the combination of orally ingested plants (including light-activated psoralens) and sunlight to treat vitilago over 4000 years ago. Contemporary PDT application began when Raab showed in 1900 the toxic effect of combination of acridine 177 dyes and dyes with which the unicellular organisms could be effectively killed. 9 In 1925 Policard examined the ability of porphyrins to produce a phototoxic effect.10 In 1946 Auler and Banzer showed the hematoporphyrin accumulated in cancerous tissue.11 Since then an enormous body of research activity has been devoted to the detection and treatment of cancer with porphyrin photodynamic therapy. Today, PDT has become the fourth established modality, next to surgery, chemo- and radio-therapy, for the treatment of cancer.35 The basic idea of photodynamic therapy for the treatment of cancer is this: i) A photosensitizer (PDT drug) with some preferential accumulation or selective localization in tumor cells is introduced into a patient's body in the dark. This drug should have negligible dark toxicity and this will not cause any i l l effect in the dark, ii) After a certain time (which will depend on the photosensitizer and the tumor type), the equilibration of drug delivery is reached between biological compartments. At this stage, photosensitizers localize selectively and/or accumulate preferentially in rapidly growing tissues such as sarcomas and carcinomas while relatively sparingly in the healthy tissue, iii) Irradiate with visible light to generate the cytotoxic effect to kill the diseased tissue. Since the concentration of photosensitizer in the area of healthy tissue is much lower relative to that in the tumor tissue, the irradiation causes only the destruction of the diseased tissue. In this step, a laser source and fiber optics are utilized to target the tumor with some precision. The use of these two technologies also allows irradiation of the internal tumors, so that this method is not restricted to the tumors at or near the surface. An efficient photosensitizer plays the key role in PDT. Modern PDT uses porphyrin related compounds as photosensitizers for the following reasons: (1) 178 OH Porphyrins have absorption bands in the region of visible light or the near infrared. (2) When they are irradiated, porphyrins often generate the cytotoxic singlet molecular oxygen from triplet oxygen. (3) They have no or low (negligible) i l l effect to human body in the dark. (4) Porphyrins selectively localize and/or preferentially accumulate in the tumorous tissue. Since the first investigation of haematoporphyrin derivative (HpD) 163, which is a so called first generation PDT drug, as an anti-cancer drug in I960, 1 2 a number of potential porphyrin related PDT drugs have been developed. A selection of the second-generation PDT drugs are shown in Figure 3-5. 13-15 O H 165 C02H(Me) C02Me(H) 166 1 6 7 Figure 3-5. Structures of selected second-generation photosensitizers 164 mono-L-aspartyl chlorin ; 165 meso-tetra(m-hydroxy)phenylchlorin; 166 tin etiopurpurin; 167 benzoporphyrin derivative monoacid (BPDMA) 179 1.3.2 Mechanisms of PDT PDT is highly dependent on the presence of dioxygen.1 6 It is believed that the singlet oxygen generated from photosensitization of molecular triplet oxygen is the key toxic species formed during the therapy. The photophysical processes for the generation of singlet oxygen are illustrated in a modified Jablonski diagram shown in Figure 3-6.17 They involve: i) the PDT drug (porphyrin-related photosensitizer) is excited by radiation at a specified wavelength and electrons are promoted from the ground state to an excited singlet state (So—>Sn). This transition corresponds to the K-K* transition in a porphyrin, ii) the excited photosensitizer undergoes non-radiative inter-system crossing to convert to the triplet state (Si—>Ti + heat). This transition is spin-forbidden as it requires spin inversion, ('forbidden' means the process is less likely to take place than an 'allowed' UJ Inter-system Crossing ^ — - n o 2 Singlet Oxygen Production Figure 3-6. Modified Jablonski diagram for generation of excited porphyrin states and reactive singlet molecular oxygen 180 transition). A good photosensitizer undergoes ISC with high efficiency, iii) the triplet photosensitizers interact with triplet oxygen in the vicinity through spin exchange, thus generating highly reactive singlet molecular oxygen (Ti->S 0 + hv). This process is called a Type II photoprocess, which competes with Type I photoprocess (electron or hydrogen transfer with other molecules). In PDT, the Type II process (singlet oxygen) 1 o predominates over the Type I. Singlet oxygen is a very cytotoxic species and undergoes several reactions such as oxidation and cycloaddition with biological substrates, including unsaturated lipids and Scheme 3-2. Reactions between singlet oxygen with some biomolecules. 181 a-amino acid, thus disrupting biological processes. The reactions between biomolecules and singlet oxygen are illustrated in Scheme 3-2. Since porphyrin-related photosensitizers localize selectively and/or accumulate preferentially in rapidly growing tissues such as sarcomas and carcinomas, only these unwanted tissues are killed by the produced singlet oxygen.2 0 Macroscopic damage to tumor cells appears to occur by three pathways: (i) direct damage to tumor cells, (ii) damage to endothelial cells of vascular system of the 9 1 tumor, and (iii) macrophage-mediated infiltration of the tumor. In summary, the principal physical and chemical processes for photodamage in PDT can be illustrated by the following expression (P = photosensitizer): P(SQ) \2L • P ^ ) !§£ PfT,) P(T 1) + 3 0 2 P(S 0 ) + 1 O 2 Biomeolecules + 1 0 2 products 1.3.3 Desirable Properties For PDT Drug The drug (photosensitizer) is the essential part in PDT. An ideal drug should have the following properties: 1. proper absorption wavelength: Due to light absorption by endogenous chromophores, mainly hemoglobin and light scattering, the effective light penetration through tissue is very poor in the low wavelength region of the 22 visible spectrum. As the wavelength increases, the effective light penetration increases as well. Experiments indicate the light penetrates effectively through 182 tissue in the red to the near infrared region (> 650 nm). As a result, the ideal drug is one that exhibits a strong absorption in such a region (> 650 nm). 2. high preference for accumulation in the tumor: The drug must have a selectivity for enrichment in the tumorous tissue vs the normal tissue. Since singlet oxygen is also detrimental to the healthy tissues, a differentiation of drug concentration between biological compartments must be achieved before the irradiation. This ensures that the efficient destruction of the diseased tissue takes place while the healthy tissue remains intact or experiences less i l l effect. 3. low dark toxicity and quick metabolization: The PDT drug itself should be non-toxic in the absence of light. The drug should be excreted or metabolized quickly in a way that does not generate toxic metabolites of any kind after the treatment is complete. 4. From the standpoint of chemical synthesis, the drug should be made from readily available materials and the protocol of synthesis should be simple and able to be scaled up to an industrial scale. It should contain groups, such as phenyl group which allows easy derivatization or variation in order to optimize various properties of the drug. 5. It should exhibit some preferred physical or photophysical properties for drug administration, such as good solubility in water and in the body's tissue fluid, 183 easy formulation, high quantum yield of triple formulation, with a triplet energy greater than 94 kJ/mol, and high singlet oxygen quantum yield. 1.4 C H L O R I N S U S E D A S P D T D R U G S A N D C O N V E R S I O N O F P O R P H Y R I N S T O C H L O R I N S Because both light absorption and light scattering by tissue increase as the wavelength decreases, the most efficiently excited photosensitizers are those which have strong absorption bands at the red end of the visible spectrum (> 650 nm). The search for red absorbers has been a major activity in recent years. Fortunately, chlorins and bacteriochlorins show that their Band I (Q bands) have a significant bathochromic shift (porphyrin, -620 nm, chlorin, -640 nm and bacteriochlorins, =720 nm). They meet these requirements perfectly and thus quickly take the place of porphyrins as photosensitizers. Indeed, most of the second generation PDT drugs are chlorins or bacteriochlorins (see Figure 3-5). The importance of chlorins and bacteriochlorins as potential PDT drugs, coupled with the lack of simple and general methods for their synthesis, resulted in profuse efforts to convert porphyrins into their reduced forms. Theoretically, this corresponds to the mere addition of one (or two for the synthesis of bacteriochlorins) pair(s) of hydrogen to the p-positions of a porphyrin. However, an intrinsic problem for such synthesis of a chlorin is the reversible nature and the general non-regioselectivity of the reaction. So far, only in exceptional cases has a selective reduction been observed.25'26 On the other hand, oxidation of one or two (3-(3' double bond(s) of porphyrins also results in a chlorin with a 184 long-wavelength absorption. This methodology not only involves an irreversible reaction, but facilitates the introduction of various functional groups on the periphery of the porphyrins, which helps to introduce amphiphilicity into the molecule and improve photophysical properties. There are several oxidants that are able to perform such oxidation, however their reactions come out with poor yield and low selectivity.2 7'2 8 To date, osmium tetraoxide-mediated dihydroxylation is the most optimal route to |3-(3' dihydroxylation of porphyrins. 1.4.1 Osmium Tetraoxide Oxidation of Porphyrins Osmium tetraoxide selectively attacks (3-p'pyrrolic double bonds of free base porphyrins to give the corresponding dihydroxylchlorin. This reaction dates back to the work of Fischer 2 9 and has been intensively studied by Chang and coworkers.30 Scheme 3-3 shows the osmylation reaction of OEP 168 in the presence of pyridine and likely conformation of the resulting osmate ester 171 (Scheme 3-3). The osmate ester is postulated to result from [2+2] cycloaddition of a Os=0 bond and a (3-|3' C=C bond followed by rearrangement of the resulting metallacycle. 3 1 ' 3 2 Two pyridines coordinate to osmium and stabilize the ester. The osmate ester can be reduced by a wide variety of reductants, such as hydrogen sulfide, lithium aluminum hydride, or sodium bisulfide, thus giving the corresponding diol-octaethylchlorin. Hydrogen sulfide is proven to be the most efficient method of reduction and facilitates a one-pot procedure.34 Recently, Bruckner in our group has reported the reaction of tetraphenylporphyrin (TPP) 172 with osmium tetraoxide in the presence of an amine base.34 An osmate ester 173, analogous to former osmate ester 171 was confirmed by ' H N M R spectroscopy. This dihydroxylation reaction has been applied to TPPs with a wide variety of phenyl substituents, providing many promising candidates for third generation PDT drugs. Ph Ph Ph Ph H Ph 172 173 174 Scheme 3-4. Oxidation of TPP with Osmium tetraoxide Reaction conditions: i) OSO4 (1.1 eq.), CHCi3/10% pyridine. ii)H 2S 186 In the above reactions of OEP and TPP with osmium tetraoxide, the resulting diol chlorins are accompanied by a minor amount of tetraol bacteriochlorins. In a separate experiment, treatment of one equivalent of osmium tetraoxide with diol chlorin also results in the same tetraol bacteriochlorin as a mixture of stereoisomers.36 1.4.2 OVERVIEW OF OXAZIRIDINES Oxaziridines are heterocyclic compounds containing a three-membered carbon-nitrogen-oxygen ring. Since the announcement of their discovery in the mid 1950's,37 they have been widely investigated, principally for two reasons. The presence of an inherently weak N-O bond in a strained ring promised a group of compounds of unusually high reactivity.38 In addition, this ring system possesses the structural elements that seem to be required to observe stereochemical isomerism at nitrogen. The barrier to nitrogen inversion for N-alkyloxaziridines has been estimated to be in the range of 30-35 kcal/mol. 4 5 This establishes the non-inverting nature of the nitrogen in these compounds. Optically active oxaziridines, the asymmetry of which is due solely to nitrogen, have been reported.50 The characteristic reactions of this ring system are associated with the presence of a weak N - 0 bond in a strained ring. Thus, in addition to thermal and acid base- catalyzed ring-opening reactions,39 there are a variety of electron-transfer processes leading to reductive ring opening 4 0 Despite of their unusual reactivity, the practical application of oxaziridines has been highly limited since they are not usually sufficiently reactive. 187 N-Sulfonyl oxaziridines are a new class of oxaziridines with a sulfonyl group attached to the nitrogen atom. They are stable, aprotic and neutral oxidizing reagents of considerable synthetic and mechanistic versatility.41 They are characterized by a highly electrophilic oxaziridine oxygen atom. Their reactivity is similar to that of peroxy acids but with higher selective.42' 4 3 The three classes of the most important commercially available N-sulfonyl-oxaziridines are shown in Figure 3-7. A . A T rr s ° N 2 R 0 2 S H r u R 175 176 177 a. R = Me a. R=Ph Ar=Ph b. R = Ph a. X = H Y= H b. R = p-MePh Ar = Ph b. X = CI Y=H c. R = p-MePh Ar=o-MePh c. R = T ' c. X = MeO Y=H O d. X = H Y= Bn e. X = H Y = p-MeOBn Q . M = - ° T Figure 3-7. N-sulfonyl oxaziridines 2-Benzylsulfonyl-3-phenyloxaziridines are the simplest compounds among N -sulfonyl oxaziridines. The general synthesis of these compounds is outlined in Scheme 3-5. ArS02NH.2 is condensed with benzaldehyde diethyl acetal at 150-180°C, affording sulfonimine 178.44 m-CPBA oxidation of sulfonimine 178 gave the corresponding diastereomeric oxaziridines in overall yield of 65%. 4 6 Since the barriers for syn-anti isomerization in sulfonimines are in the order of 12-13 kcal/mol, the oxidation of 178 would be anticipated to afford E and Z oxaziridine 175. However, the N M R spectroscopic studies showed the N(0)C-H proton remained a sharp singlet on cooling to -80°C. This indicates that oxidation of 178 affords a single oxaziridine. The X-ray crystal 188 R S 0 2 N H 2 + ArCH(OEt)2 130-180°C > N = c , A r m-CPBA ^ N ' _ ^ A r R 0 2 S ' XH NaHC0 3 -H 2 0 R 0 2 S ' *H 178 C H C ' 3 175 S c h e m e 3-5. Synthesis of 2-Benzylsulfonyl-3-phenyloxaziridine structure and N M R spectroscopic analysis revealed that this oxaziridine has the E configuration.45 As a result, just two oxaziridine enantiomers having the S,S and R,R configurations at the three-membered ring were prepared. 2-Benzylsulfonyl-3-phenyloxaziridine and its derivatives have been found to be useful in many applications (Scheme 3-6). For example, N-sulfonyloxaziridines selectively oxidize sulfides to sulfoxides without over-oxidation to sulfones 4 6 Primary, secondary and tertiary amines are oxidized at nitrogen by N-sulfonyloxaziridines to give hydroxyl amines and amine oxides 4 7 N-Sulfonyloxaziridines are among the few reagents S c h e m e 3-6. Oxygen-Transfer Reactions of N-Sulfonyloxazidines 189 available for the oxidation of carbanions to alcohols and phenols, and enolates into oc-hydroxy carbonyl compounds.49 Organometallic reagents (RM) are hydroxylated (ROH) in good yield on treatment of N-sulfonyl oxaziridines.48 Kinetic and mechanistic investigations of the oxidation of sulfoxides to sulfones and the expoxidation of alkenes with N-sulfonyloxazirdines by Davis and Bach 5 0 have revealed that the oxygen transfer of oxaziridine undergoes a nucleophilic SN2 type mechanism. The oxygen atom transfer is facilitated by the relatively weak oxygen-nitrogen bond and by the enthalpy of carbon-nitrogen Tt-bond formation in the transition state. In addition, the transition state conformation is steric in origin, dictated by the substitutents attached to the oxaziridine carbon and nitrogen. This feature of oxaziridines determines their important synthetic use for asymmetry-induced oxygen transfer to a wide variety of nucleophilic substrates including alkenes,41 sulfides,42 sulfoxides,46 and enolates.43 1.6 G O A L O F P R O J E C T Osmylation of porphyrins is one of few available methods to oxidize these highly conjugated macrocyclic compounds. It offers an efficient way to prepare the important PDT photosensitizers, chlorins and bacteriochlorins. However OSO4 and H 2S, which is generally used to reduce the osmate ester, are notorious for their extreme toxicity and operational difficulties. In addition, OSO4 is very expensive. These disadvantages make this synthetic approach to chlorins unrealizable in the industry. To this end, we are seeking an alternative oxidant to overcome these disadvantages. N-Sulfonyl oxaziridines 190 have been examined in our study, and have been shown to be very promising reagents to oxidize the p-(3' double of porphyrins to form chlorins and bacteriochlorins. The reaction is operationally simple. These compounds produced therein exhibit promising spectral characteristic and fairly good solubility in a large number of solvents. 191 2. RESULTS AND DISCUSSION The oxaziridine used in our study is 2-benzylsulfonyl-3-phenyloxaziridine 175a which is the most economic and easily-prepared N-sulfonyl oxaziridine.45 The general synthesis of this compound is shown in Scheme 3-5. The final products obtained as a racemic mixture, consisted of two enantiomers (S,S' and R,R'). The oxaziridine is stable at 5°C in the dark for 3 months. 2.1 R E A C T I O N , I S O L A T I O N A N D C H A R A C T E R I Z A T I O N In our study of the oxidation reaction of porphyrins, 1:1 2-benzylsulfonyl-3-phenyloxaziridine and OEP 168 were dissolved in anhydrous methylene chloride and allowed to stir in the dark. After 24 hours, a peak at 640 nm in the UV-visible spectrum had appeared. As the reaction time increased to 3 days, this peak grew to a maximum and a new peak at 700 nm appeared. Increasing the amount of oxidant 2-benzylsulfonyl-3-phenyloxaziridine and lengthening the reaction time invariably led to the formation of a new product with absorption at 700 nm. Separation of the reaction mixture by preparative thin layer chromotography gave three green pigments 179, 180 and 181 with low 179 + 180 + 181 168 Scheme 3-7. Oxidation of OEP with oxaziridine Reaction conditions: 1.2-1.5 eq. C 6 H 5 C H ( 0 ) N S 0 2 C 6 H 5 , CH2CI2, 3 days, room temperature 192 polarity. The yields of these three products were 16%, 15% and 3% respectively. OEP was recovered in 25% yield. The proposed structures of 179 and 180 are shown in Figure 3-8. These two major products have identical UV-visible spectra and the same molecular masses. Mass spectrometry showed these products 179 and 180 to have molecular mass of 795, which is the sum of the two reactants OEP 170 and 2-benzylsulfonyl-3-phenyloxaziridine 175a. Microanalysis and high resolution mass spectrometry confirmed this stoichiometry. The UV-visible spectra of 179 and 180 (Figure 3-9) exhibited characteristic Soret and Q bands, thus confirming the existence of 18 7t-electron conjugation within the molecule. A bathochromic shift of band I of the Q bands and a decrease of the Soret to band I ratio (5:1 versus OEP 50:1) compared to OEP was observed. This indicates that the reactant porphyrin had been converted into a chlorin, and one double bond had been oxidized. Similar analysis was performed on the third product 181. The UV-visible spectrum of 181 shows a larger bathochromic shift of band I (700 nm) and a significant increment of absorption intensity of Q band I compared to the former two products 179 and 180. Mass spectrometry showed the molecular mass as the 179 M 180 (t) Figure 3-8. Suggested structures of compounds 179 and 180 193 sum of two molecular oxaziridines and one molecular OEP. It was deduced to be a bacteriochlorin, which has two of the periphery double bonds oxidized. i 350 450 550 650 750 W a v e l e n g t h (nm) Figure 3-9. UV-visible spectrum of 179,180 (overlayed, solid line) and 181 (broken line). N M R spectroscopy was most informative for these compounds and suggested interesting structures. The ' H N M R spectrum (Figure 3-10) of compound 179 exhibited signals at 8 -2.58 and -2.80 ppm, which are due to N H protons located inside the macrocycle. Signals at 8 8.88, 9.76, 9.86, 9.98 ppm are indicative of meso-methine protons which are outside the ring. The distinctive chemical shift of these signals is caused by deshielding and shielding effects of the aromatic system and indicates retainment of the 18-71 conjugation system. The fact that the N H protons appear as two peaks and the raeso-methine protons as four indicates the loss of symmetry from the 194 parent porphyrin. The l 3 C N M R (Figure 3-15 of supplemental data) showed there are three groups.of signals. The first group lies between 125.00-160.00 ppm. They are assigned to the carbons within aromatic systems of the porphyrin and phenyl group. The second group is between 80.00 and 102.00 ppm. The third group comprises those below 30.00 ppm. They are assigned to the saturated carbons which are on the side chains of the porphyrin. Analysis of H M Q C N M R spectrum (Figure 3-11. For detailed spectra, see Figure 3-16-Figure 3-22 of supplemental data) gave more detailed structural information. It showed that four methine carbons have chemical shifts at 100.12, 98.02, 95.72, 93.30 ppm, because of their interaction with the four methine protons assigned earlier (Figure 3-21). The two carbon signals at 101.88 and 92.30 ppm are assigned to the (3, (3 carbons which are involved in the reaction with oxaziridine, because they are not correlated with any protons and they are distinguished from carbons within the conjugated system. The carbon signal at 95.12 ppm, which is correlated with the proton signal at 6.82 ppm, is assigned to the carbon which was originally from the oxaziridine ring system. Most surprisingly, the protons of the two phenyl groups have chemical shifts at the region 5.50-7.00 ppm, as determined by their correlation with carbon signals between 125.00 -140.00 ppm. These protons experience a significant up-field shifting compared to the regular phenyl proton chemical shift (8 >7.00 ppm). This indicates that the two phenyl groups are enormously shielded by the aromatic macrocycle. Our proposed possible positions of these two phenyl groups are on the top of planar tetrapyrrolic ring. ' H COSY N M R spectrum (Figure 3-12) is in full agreement with the structure suggested in Figure 3-8. ' H N M R and l 3 C N M R spectroscopic study of 1 8 0 also suggests a chlorin structure (Figure B 10.0 9.0 8.0 7.0 6.0 5.0 4 . 0 ^ 3.0 2I0 1.0 / 0 -1.0 -2.0 -3.0 -4.0 Figure 3-10. ' H N M R spectra of compounds 179 and 180 A is ! H N M R spectrum of compound 179 B is N M R spectrum of compound 180 196 Figure 3-11. H M Q C N M R spectrum of compound 179 Figure 3-12. lH COSY spectrum of compound 179 198 3-8). In the {H N M R spectrum, the protons of one phenyl group in the molecule has chemical shifts at the region of 7.00-8.00 ppm. The protons of the other phenyl still appear in the abnormal up-field region (5.50-6.50 ppm). This indicates one of phenyl groups in 34 is on the top of tetrapyrrolic ring and the other is outside. 2.2 P R O P O S E D M E C H A N I S M The mechanism for this reaction is proposed on the basis of the nature of OEP and the oxaziridine and the structure of the products. As discussed earlier (Section 1.4.2 in Part 3), the dominant factor determining the ring-opening reactions appears to be steric and not electronic. In the molecule of 2-benzylsulfonyl-3-phenyloxaziridine, nucleophilic attack at the carbon or nitrogen atoms of the oxaziridine ring seems unlikely due to the steric hindrance from phenyl and sulfonyl phenyl groups. The oxygen atom in the ring is less hindered than nitrogen and is made more electrophilic by the strongly electron-withdrawing sulfonyl group. It therefore becomes the active site toward nucleophiles. It has been shown by Hata and Watanabe51 that small nitrogen substituents within oxaziridines lead to favored attack at nitrogen; as the steric bulk of the substituent increases the site of attack shifts to oxygen. We propose that the exocyclic double bond of OEP attacks the oxygen atom of oxaziridine and causes the cleavage of the N - 0 bond and ring opening, thus introducing a positive charge on the pyrrolic (3-carbon and a negative charge on the nitrogen atom. The intermediate cation is stabilized via derealization over the macrocyclic 7t-system and the 199 ionic nitrogen is stabilized by the strongly withdrawing sulfonyl group, cyclization gives the five-membered ring (Figure 3-13). Due to the steric interaction, the ionic nitrogen may invert during the cyclization, to keep the bulky sulfonyl phenyl group away from the ethyl side chain, and thus, the sulfonyl phenyl group stays on the top of the porphyrin ring. This rationalizes that the protons of the sulfonyl phenyl group are significantly shielded by the porphyrin ic-system in both compounds 179 and 180. Since diastereomeric oxaziridine was used, the final products were also diastereomers and 179 and 180 were obtained 1:1. Figure 3-13. Proposed Mechanism for Reaction of OEP with 2-Benzylsulfonyl-3 -Phenyloxaziridine Under the same conditions, meso-tetraphenyl porphyrin (TPP) does not react with N-sulfonyl oxaziridines. This phenomenon is related to the distinct conformation of TPP. Although maximum derealization between the porphyrin ring and the meso-phenyl 200 substitute in TPP would be achieved through co-planarity, steric interactions between the (3-hydrogens and the ortho-hydrogens on the phenyl rings force the phenyl plane to be placed approximately perpendicular to the porphyrin mean plane.52 Unlike OEP, whose substituents are all in the porphyrin plane thus not causing steric hindrance to nucleophilic attack at nitrogen atom of oxaziridine, TPP does not undergo oxidation by oxaziridine, because the phenyl groups block the top and bottom faces of the |3-(3' double bond and prevent the approach of oxaziridine (Figure 3-14). 2.3. C H E M I C A L P R O P E R T I E S OF C O M P O U N D S 179 A N D 180 Chlorins 179 and 180 are stable compounds and can kept in the dark for extended periods of time at ambient temperature. They decompose slowly if exposed to light. They have very good solubility in common organic solvents, such as MeOH, methylene chloride, diethyl ether, chloroform and hexanes. The compounds 179 and 180 are also stable under strongly basic conditions. They were each dissolved in 95% ethanol and X Figure 3-14. The Steric Interactions within TPP 201 mixed with one equivalent volume of 6 M NaOH aqueous solution. The mixtures were heated at 60 °C for 1 hour and no reaction was observed. When compounds 179 and 180 were treated with one equivalent of N-sulfonyl oxaziridine, further oxidation to bacteriochlorin 181 with yield of 75 % (the total yield of the diastereomer mixture) occurred. MS spectrometry and microanalysis illustrated that the final product is the combination of OEP and two molecules of N-sulfonyl oxaziridine. The UV-visible spectrum of the final product exhibited characteristic Soret bands at 390 nm and a sharp Q band at 700 nm. The ratio of these bands was observed to be 1.5:1. These features are in support of our proposed structure of bacteriochlorin. The intensity of the red-region absorption is extremely high (log e = 5.60). These light absorption properties have thus made them the potential candidates for PDT drugs, because they allow the effective light penetration through skin. Unfortunately, however, this reaction gives several diastereomers and it is extremely difficult to obtain an optically pure compound. This restricts them from the use of photosensitizer as a PDT drug. 2.4. Summary In my study, OEP has been readily oxidized by 2-benzylsulfonyl-3-phenyloxaziridine to chlorin in a reasonable yield. This new approach has afforded a novel and easy route to prepare chlorins. Compared to the other oxidizing agents used in oxidation of porphyrin, 2-benzylsulfonyl-3-phenyloxaziridine is less expensive and less toxic and its operation is relatively simple. However, it introduces the complexity to the 202 chlorins. Our future work will focus on using simple oxaziridines with less chiral centers and less steric hindrance to give simple chlorin in a higher yield. 203 3. EXPERIMENTAL 3.1 P R E P A R A T I O N S The reagents were purchased from Aldrich Chemical Co. and were used without further purification. CH2CI2 and CHCI3 from calcium hydride. Anhydrous Na2S04 was used to dry the organic solutions during workups. Flash column chromatography was performed using 230-400 mesh silica gel (Merck). Analytical thin-layer chromatography was done on pre-coated silica gel aluminum plates containing a fluorescent indicator (GF-254 Merck). Preparative T L C was performed on pre-coated 20 X 20 cm, 0.5 nm thickness Merck silica gel plates (without fluorescent indicator). ' H N M R spectra were recorded at 200 or 400 MHz; l 3 C N M R spectra at 75 MHz. CDC1 3 was used as solvent for N M R spectroscopy. Benzaldehyde-dimethyl-acetal46 and N-benzylidenebenzenesulfonamide45 were synthesized according to literature methods. OEP was prepared previously in our lab. 2-Benzylsulfonyl-3-phenyloxaziridine (175a) 2-Benzylsulfonyl-3-phenyloxaziridine 175a was prepared according to the method described by Davis. 4 5 ' H N M R (200 MHz, CDCI3): 8.07 (m, 2H, Ar-H), 7.78 (m, 1H, Ai-H), 7.62 (m, 2H, Ar-H), 7 A3 (m, 5H, Ar-fl), 5.49 (s, 1H, CH); 1 3 C N M R (75 MHz, CDCI3): 135.00, 134.76, 130.47, 131.44, 129.37, 129.35, 128.75, 128.24, 76.30; LR-MS (EI) m/z = 261 ( M + , 204 33%), 245 ([M-0] + , 100%) Anal. Calcd for C,3HuN0 3 S: C, 59.77; H, 4.21, N , 5.36; found: C, 60.03; H: 4.16; N , 5.39 . Chlorins 179 and 180 To 150 mL of methylene chloride solution of OEP (200 mg, 0.37 mmol) was added a solution of 2-benzylsulfonyl-3-phenyloxaziridine (145 mg, 0.55 mmol, 1.5 eq.) in methylene chloride (10 mL). The reaction mixture was stirred magnetically. After 24 h. UV-visible spectrum showed a strong peak at 640 nm. The peak increased in intensity as the reaction time increased and grew to a maximum after 3 days. Upon the addition of 2-benzylsulfonyl-3-phenyloxaziridine (55 mg, 0.18 mmol), a second peak at 700 nm appeared. The solvent was removed in vacuo, and the solid residue was washed with hexanes (3 X 50 mL). 50 mg of OEP was recovered as insoluble. The washing solution was reduced to 10 mL and charged onto a 5.5 cm X 10 cm silica gel column with 0.5% ethyl acetate in hexanes. The first green band was collected.* After the solvent was removed under vacuum, 58 mg of green solid were obtained. It was charged on the T L C , and run by developing solvent: 0.1% ethyl acetate in hexanes. Two green bands were collected and washed with THF. After removal of the solvent, two green solids corresponding to 179 (47 mg, 16%) and 180 (44 mg, 15%) were obtained. Characterization of Band 1 (179) ' H N M R (400 MHz, CDC13): 9.96 (s, IH, meso-CH), 9.87 (s, IH, meso-CH), 9.78 (s, IH, meso-CH), 8.89 (s, IH, meso-CH), 6.82 (m, IH, CH) 6.68 (m, IH, Ar-H), 6.25 (m, 5H, * After chromotography, a small amount of green band other than main products was also obtained. It was identified as compound 181. 205 Ar-H), 5.67 (s, 4H, Ar-H), 4.50-2.50 (m, 16H, 8 X C H 2 C H 3 ) , 2.40-1.40 (m, 21H^ 7 X CH 2C7f 3) 0.00 (t, J = 7.5 Hz, 3H, CH 2 C# 3 ) , -2.56 (s, IH, NH), -2.78 (s, IH, NH); l 3 C N M R (75 MHz, CDC1 3): 162.17, 157.29, 151.12, 149.70, 143.72, 142.74, 140.66, 139.48, 139.05, 137.41, 136.44, 136.39, 136.16, 135.95, 133.83, 132.34, 130.35,128.71, 128.32, 126.95, 126.92, 125.62, 101.87, 100.12, 98.02, 95.72, 95.12, 92.30, 81.03, 29.39, 26.64, 19.76, 19.73, 19.58, 19.47, 19.09, 18.74, 18.66, 18.44, 18.20, 18.16, 18.00, 10.53, 7.37; UV-visible (CH 2C1 2) X m a x (log e): 392 (5.79), 484 (4.25), 560 (4.18), 644 (5.08); LR-MS (LSEVIS) m/z = 796 (M+,100 %); HR-MS (LSIMS) m/z Calcd. for C49H57O3N5S: 795.4182; found: 795.4184; Anal. Calcd for C 1 3 H 1 1 N O 3 S : C, 59.77; H, 4.21; N , 5.36; found: C, 60.03; H , 4.16; N , 5.39. Characterization of Band 2 (180) [ H N M R ( C D C I 3 ) : 9.80 (s, IH, meso-CH), 9.72 (s, IH, meso-CH), 9.38 (s, IH, meso-CH), 8.82 (s, IH, meso-CH), 7.83 (m, 2H, Ar-H), 7.49 (m, 3H, Av-H), 6.79 (s, IH, CH), 4.90 (m, 2H, Ar-H), 4.20-3.70 (m, 16H, 8 X Ctf 2 CH 3 ) , 2.60-1.75 (m, 21H, 7 X CH 2 C# 3 ) 0.8 (t, J = 7.2 Hz, 3H, C H 2 C H 3 ) , -2.91 (s, IH, NH), -3.18 (s, IH, NH)\ ' 3 C N M R ( C D C I 3 ) : 168.44, 158.80, 154.76, 150.95, 149.96, 143.49, 142.77, 139.46, 139.23, 139.19, 139.08, 137.07, 136.00, 135.92, 135.77, 135.72, 135.72, 135.65, 133.20, 132.34, 129.49, 128.34, 127.43, 123.80, 111.42, 102.84, 98.48, 97.28, 94.96, 90.30, 78.67, 27.90, 26.82, 19.73, 19.51, 19.44, 19.22, 18.68, 18.47, 18.35, 18.21, 11.24, 7.60; UV-visible (CH 2C1 2) KM (log e): 392 (5.79), 484 (3.89), 560 (4.16), 644 (5.16); LR-MS (LSIMS) m/z = 796 ( M + , 70 %); HR-MS (LSEVIS) m/z Calcd. for C49H57O3N5S: 795.4182; found, 206 795.4184; Anal. Calcd for Ci 3 Hi,N0 3 S: C, 59.77; H, 4.21; N, 5.36; found: C, 59.89, H, 4.11; N, 5.53. Bacteriochlorin 181 (Oxidation of compound 179 with oxaziridine) Compound 179 (15 mg, 0.019 mmol) was dissolved in anhydrous methylene chloride (15 mL). After addition of 2-benzylsulfonyl-3-phenyloxaziridine (5.5 mg, 0.021 mmol, 1.1 eq.), the reaction mixture was allowed to stir in the dark. Inspection by UV-visible spectroscopy showed a peak at 700 nm appeared after 3 h and it reached a maximum after 20 h of stirring. The solvent was removed in vacuo. The residue was charged onto preparative TLC and run by eluting with a solution of methylene chloride and methanol (99.5:0.5). Two green bands were collected. The small amount of first band was identified as starting material 179 and the second band, whose R f (CH2CI2), MS and UV-visible spectra are same as those of 181, was obtained (12 mg, 62 %). ' l l NMR: It seemed that the second bond was a mixture of diastereomers. NMR spectrum is not available because of impurity. UV-visible (CH2C12) X m a x (log e): 364 (6.09), 384 (6.26), 474 (5.12), 504 (5.32), 700 (5.88); LR-MS (LSIMS) m/z = 1056 (M+,100%); HR-MS (LSEVIS) m/z Calcd. for C 62H68N 60 6S 2: 1056.4640, Observed: 1056.4646; Anal. Calcd for Q2H68N6O6S2: C, 70.43; H, 6.44; N, 7.95; Observed: C, 70.23; H, 6.53; N, 8.05. 207 3.2 S U P P L E M E N T A L A N A L Y T I C A L D A T A F O R C O M P O U N D 179 in —• r - v m o cv o r v rs. r- uo r< rs. M M , I .1 ,t , It • tf • •„• .illliiJ • ..H • ii 11 ii in i ni II i( | . . • . • < < . | • i 1 • • | —;—t— 1 % . i 1 » ——•—-—1—i • — r — - | -PP» 150 100 50 0 Figure 3-15. C N M R spectrum of compound 179. 208 Figure 3-15. Detailed H Q M C Spectra (1) of Compound 179. 209 Figure 3-16. Detailed H Q M C Spectra (2) of Compound 179. Figure 3-17. Detailed H Q M C Spectra (3) of Compound 179. 2 1 1 Figure 3-18. Detailed H Q M C Spectra (4) of Compound 179. 212 J Figure 3-19. Detailed HQMC Spectra (5) of Compound 179. 213 Figure 3-20. Detailed H Q M C Spectra (6) of Compound 179. 214 Figure 3-21. Detailed H Q M C Spectra (7) of Compound 179. 215 References: 1 a) The porphyrins Dolphin, D. Ed.; Academic Press, 1978, Vols. 1-8. b) Porphyrins and metalloporphyrins Smith, K. M . Ed.; Elsevier: Amsterdam, 1976. 2 Svec, W. A. In Chlorophylls Scheer, H. Ed.; CRC Press: Boca Raton, 1991, 89. 3 Battersby, A. R. Acc. Chem. Res. 1993, 26, 15. 4 Gouterman, M . In The Porphyrins Dolphin, D. Eds.; Academic Press Inc. 1978 Vol . I l l , Chapter 1. 5 a) Gouterman, M . J. Chem. Phys. 1959, 30, 1139; b) Piatt, J. R. Radiation Biology, Ed.; McGraw-Hill, New York, 1956, Vol . I l l , Chapter 2. 6 Eaton, W. E.; Hochstrasser, R. M . J. Chem. Phys. 1967, 46, 2533. 7 Sternberg, E.; Dolphin, D. In Photodynamic Therapy and Biomedical Lasers; Spinelli, P.; Dal Fante, M . ; Marchesini, R. Ed.; Exerpta Medica, Amsterdam, 1992,470. 8 Edelson, M . F. Sci. Am. 1988, 259, 68. 9 Raab, O. Infusorin. Z. Biol. 1900, 39, 524. 10 Policard, A. C. R. Hebd. Soc. Biol. 1925, 91, 1422. 11 Auler, H . ; Banzer, G. Z. Krebsforsch 1942, 53, 65. 12 Lipson, R. L. In Proceedings of 9th International Cancer Congress, Tokyo, Japan, 1966, 393. 13 Morgan, A. R.; Garbo, G. M . ; Keck, R. G.; Skalkos, D.; Selman, S. H. Photchem. Photobiol. 1990,51,589. 14 Bonnett, R.; White, R. D.; Winfield, U-J.; Berenbaum, M , C. Biochem. J. 1989, 261,277. 15 Ritchter, A . M . ; Kelly, B.; Chow, J.; Liu, D. J.; Towers, G. H. N . ; Dolphin, D.; Levy, J. G. J. Natl. Cancer Inst. 1987, 79, 1327. 216 16 Van Lier, J. E. In Photodynamic Therapy of Neoplastic Dease; D. Kessel, Ed.; CRC Press: Boca Raton, 1990, Vol . 1, 279-291. 17 Turro, N . Molecular Photochemistry; W. A. Benjamin, Ed.; New York, 1965. 18 The Science of Photobiology Smith, K. C. Ed.; Plenum Press: New York, 1989. 19 Megda, D.; Wright, M . ; Miller, R. A. ; Sessler, J. L. Sansom, P. I. J. Am. Chem. Soc. 1995, 117, 3365; Micheali, A. ; Feitson, J. Photochem. Photbiol. 1994, 59, 284. 20 Gomer, C. J.; Dougherty, T. J. Cancer Res. 1979, 39, 146. 21 Bonnett, R. Chem. Soc. Rev. 1995, 19. 22 Wilson, P. C. Photosensitizing Compounds: Their Chemistry, Biology and Clinical Use; Wiley Interscience: Chichster, 1989, 73. 23 Wainwright, M . Chem. Soc. Rev. 1996, 351; Lown, J. W. Can. J. Chem. 1997, 75, 99. 24 Reddi, E.; Cernuschi, S.; Bioli, R.G. Laser in Medical Science 1990, 339; Woodburn, K.; Chang, C. K.; Sangwan, L. ; Henderson, B.; Kessel, D. Photochem. Photobiol. 1994, 60, 154. 25 Shine, H . J.; Padilla, A . G.; Wu, S. M . ; J. Org. Chem. 1979, 44, 4069. 26 Rothemund, P. J. Am. Chem. Soc. 1936, 61, 2912; Eisner, U . J. Chem. Soc. 1959, 854; Tabushi, I; Sakai, K.; Yamumura, K. Tetrahedron Lettt. 1978, 1821. 27 Inhoffen, H . H. ; Nolte, W. Justus Liebigs Ann. Chem. 1969, 725, 167 28 Inhoffen, H. H. ; Brockmann, Jr.H.; Bliesenser, K . - M . LiebigAnn. Chem. 1969, 730, 173; Inhoffen, H. H.; Bliesenser, K . - M . ; Brockmann, Jr.H. Tetrahedron Lett. 1966, 3779. 29 Fischer, H.; Sckoldt, H . Liebigs Ann. Chem. 1940, 543, 138; Fischer, H. ; Pfeiffer, H . LeibigsAnn. Chem. 1944, 556, 131. 30 Chang, C. K.; Soririou, C ; Weishih, W. J. J. Chem. Soc, Chem. Commun. 1986, 1213. 31 Gobel, T.; Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1993, 32, 1329. 217 32 Berrisford, D. J.; Bolm, C ; Sharpless, K. B. Angew. Chem., Int. Ed. Engl. 1995, 34, 1059. 33 Schroder, M . Chem. Rev. 1980, 80, 187. 34 Bruckner, C. Ph. D.'s Thesis, 1995. 35 Sternberg, E. D.; Bruckner, C ; Dolphin, D. Tetrhedron 1998, 54, 4151. 36 Bonnett, R.; Nizhnik, A. N . ; Berebaum, M . C. J. Chem. Soc, Chem. Comm. 1989, 1822. 37 Krimm. H. , British Patent 1953, 743, 940; Chem. Abstr. 1957, 265. 38 Knippe, A.C. ; McAuley, I. E.; Kansal, V . K. Tetrahedron Lett. 1984, 25, 1411. 39 Emmons, W. D. Chem. Heterocycl. Compd. 1964, 19, 624. 40 Hata, Y . ; Watanabe, M . / . Am. Chem. Soc 1979, 101, 6671. 41 Davis, F. A. ; Harakal, M . E.; Awad, S. B. J. Am. Chem. Soc. 1983, 105, 3123. 42 Davis, F. A. ; Billmer, J. M . J. Org. Chem. 1983, 48, 2672. 43 Davis, F. A. ; Stringer, O. D.; McCaulley, J. P. Tetrahedron 1985, 41, 4747. 44 Albrecht, R.; Kresze, G.; Mlakar, B. Chem. Ber. 1964, 97, 32. 45 Davis, F. A. ; Lamendola, J., Jr.; Nadir, U . ; Kluger, E. W.; Sedergran, T.; Panuto, T. W.; Jenkins, R.; Terci, I. J.; Watson, W. H. ; Chen, J. S.; Kimura, M . J. Am. Chem. Soc 1980, 102, 2000. 46 Davis, F. A. ; Award, S. B.; Jenkens, R. H; Billmer, J. M . ; Jenkins, L . A . J. Org. Chem. 1982, 47, 1774. 47 Davis, F. A. ; Jenkins, R. H. ; Awad, B. S.; Stringer O. D.; Watson, W. H. ; Galloy, J. J. Am. Chem. Soc. 1982, 104, 5412. 48 Davis, F. A. ; Mancinelli, P. A. ; Balasubraminian, K.; Nadir, U . K. J. Am. Chem. Soc. 1979, 101, 1044. 49 Davis, F. A. ; Vishwakarma, L. C ; Billmer, J. M . ; Finn, J. / . Org. Chem. 1984, 49, 3241. 50 a) Davis, F. A. ; Sheppard, A. C. Tetrahedron 1989, 45, 5703. b) Davis, F. A. ; 218 Mcaulye, J. P; Chattopadhyay, S.; Harakal, M . E.; Towson, J. C ; Watson, W. H. ; J. Am. Chem. Soc. 1987, 109, 3370. c) Davis, F. A. ; Billmers, J. M . ; Gosiniak, D. J.; Towson, J. C.; Bach, R. D. J. Org. Chem. 1986, 51, 4240. 51 Hata, Y . ; Watanabe, J. M . J. Org. Chem. 1981, 46, 610. 52 Fleischer, E. B. Acc. Chem. Res. 1970, 3, 105. 219 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items