INFORMATION TO USERS

This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.

1.The sign or “target” for pages apparently lacking from the document photographed is “Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.

2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of “sectioning” the material has been followed. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again-beginning below the first row and continuing on until complete.

4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.

U niversi^ M io m lm s International 300 N. Zeeb Road Ann Arbor, Ml 48106

8305306

Caste, Maureen Lynch

TRANSITION METAL COMPLEXES OF POLYMER SUPPORTED MACROCYCLIC LIGANDS

The Ohio State University Ph.D. 1982

University Microfilms

I ntern atio n si 300 N. Zeeb Road, Ann Arbor, MI 48106

PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V

1. Glossy photographs or pages.

2. Colored illustrations, paper or print _____

3. Photographs with dark background _____

4. Illustrations are poor copy ______

5. Pages with black marks, not original copy.

6. Print shows through as there is text on both sides of page.

7. Indistinct, broken or small print on several p ag es.

8. Print exceeds margin requirements _____

9. Tightiy bound copy with print lost in spine ______

10. Computer printout pages with indistinct print.

11. Page(s) lacking when material received, and not available from school or author.

12. Page(s)______..seem to be missing in numbering only as text follows.

13. Two pages n u mbered ______. Text fol lows.

14. Curling and wrinkled pages ______

15. Other ______

University Microfilms International

TRANSITION METAL COMPLEXES OF POLYMER SUPPORTED

MACROCYCLIC LIGANDS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the

Degree Doctor of Philosophy in the Graduate School of

The Ohio State University

By

Maureen Lynch Caste, B.S., M.S.

*****

The Ohio State University

1982

Reading Committee:

Professor Daryle H. Busch

Professor Devon W. Meek

Professor Bruce E. Bursten Approved by:

Advisor Department of Chemistry To my Korn and Dad

ii ACKNOWLEDGMENTS

My sincere thanks are extended to all members of Dr. Busch's research group, past and present, whose assistance and encouragement aided in the accomplishment of this work. Dr. Jeffrey Church and

Dr. Judy Gallucci are specially noted for the solution of the crystal structure presented in this thesis and are thanked accordingly. As a long time friend and associate, I thank Dr. Randall J. Remmel who has been an Inspiration to me through his continuing encouragement and interest in my chemical career since I began my undergraduate degree.

To end on a personal note, I would like to express the appreciation that I feel toward Mr. Jeffery C. Bricker for his deep insight in the area of chemistry, his encouragement of my pursuits, and his special friendship. But, above all, the most especial thanks are given to

Dr. Daryle H. Busch for his support, both financial and emotional, his •j guidance in this work, and his friendship over the past three years.

ill VITA

November 30, 1954...... Born, Brooklyn, New York

August, 1978 ...... B.S., University of Alabama in Birmingham

Sept. 1978 - Dec. 1979 ...... Teaching Assistant, Department of Chemistry, The Ohio State Univ.

Jan. 1980 - Oct. 1981 .... Research Associate, Department of Chemistry, The Ohio State Univ.

October 1981 ...... M.S., The Ohio State University Columbus, Ohio

Nov. 1981 - Sept. 1982 ...... Research Associate, Department of Chemistry, The Ohio State Univ.

PUBLICATIONS

Maureen Lynch Caste and Daryle H. Busch, ''Transition Metal Complexes of Polymer Supported Macrocyclic Ligands — Oxygen Binding and Oxygenation of Organic Substrates,'' 14th ACS Central Regional Meeting, Midland, Michigan, June 1982, Abstract 120.

Sheldon G. Shore, Wen-Liang Hsu, Clemens R. Weisenberger, Maureen Lynch Caste, Melvyn Rowen Churchill, and Clifford Bueno, ''New Syntheses of Mixed Metal Clusters from HgOsa(CO)lo. Crystal and Molecular Structure of the Paramagnetic Cluster H 3 (n'-C5H 5 )CoOsa(CO), and Its Diamagnetic Structural Analogue H 3 (ri®-CsHs)Ni0 s3 (C0 ) 9 , " Organometalllcs. 1982, 1, 1405.

FIELD OF STUDY

Major Field: Chemistry

Specialization — Inorganic Coordination Chemistry, Professor Daryle H. Busch, Advisor

iv TABLE OF CONTENTS Page DEDICATION ...... ii

A C K N O W L E D G M E N T S ...... iii

VITA ...... iv

LIST OF T A B L E S ...... vii

LIST OF F I G U R E S ...... viii

Chapter

I. INTRODUCTION ...... 1

Polymer Supported Metals and Métal Complexes 1

Heme Protein Modeling 3

Polymer Supported Catalysts 17

Hydrogenations and Hydroformylations using Supported Catalysts 23 Selected Oxidation Reactions Using Supported Polymers 27

Complexes of Lacunar Ligands 43

Statement of the Problem 50

II. E X P E R I M E N T A L ...... 52

General Procedures 52

Physical Measurement 52

Synthesis of Nickel Complexes 54

Preparation of Ligand Salts 59

Synthesis of Cobalt Complexes 60

Preparation of the Supported Complexes 62

Adsorption of Cobalt(II) Complexes on Polymeric 6 8

Oxygen Exposure to the Polymer Samples 70 Polymer Reactions with Oxygen and Organic Substrates 71

V CONTENTS (CONT'D)

III. CRYSTAL STRUCTURE AND PROPERTIES OF A NOVEL COBALT(II)

COMPOUND CONTAINING FIVE COORDINATED NITROGENS,

[Co{ (MeNIMINOETHYL) (MeNHETHYLDMea [l6]TETRAENENs }] (PFg) 2 73

Chemical Evidence for Pentadentate Chelation to the Cobalt(II) Species 77

Crystal Structure 85

Reaction of the Cobalt(II) Complex with Oxygen 97

IV. RESULTS AND D I S C U S S I O N ...... 109

Model Studies 109

Chloromethylated Polystyrene 109 Poly(Vinylpyridine) 120 Controlled Pore Glass 122

Attachment of the Polymeric Supports 127

Covalent Attachment 127 Coordination Attachment 145 CFG Support 147

Reactions of the Polymer Supported Cobalt Complexes with Oxygen 152

Covalently Attached Complexes 152 Coordinately Attached Complexes 158 CPG Support 164

Reactions of Polymer Supported Complexes 166

Oxidation Catalysis 166 Gas Chromatographic Supports 171

APPENDIX A ...... 178

APPENDIX B ...... • 194

FOOTNOTES ......

VI LIST OF TABLES Page Table

1. Materials Used to Support Metal Complexes ...... 2

2. NMR Peak Shifts for ICo{(MeN Iminoethyl)(MeN

Ethi)Me2[16]-tetraene Ns}](PFe )2 in C D 3ON ......

3. Summary of Crystallographic Data for C0 C 2 2 N 7H 3 2 P 2F 12 89

4. Bond Distances (A) and Angles (Deg)(B) with their

Estimated Standard Deviations for C 0 C 2 2N 7H 3 2 P 3F 22 94

5. ‘h NMR Peak Shift for tCo{ (MeNIminoethyl)(MeNEthi)-

M e 2 Ï16]tetraene N 5}](PF«) 2 , After Exposure to Oxygen, CDsCN Solvent, Room Temperature ...... 99

6 . NÈR Peak Shifts for ICo{MeNIminoethyl)(MeNEthi)- M e 2 [16]tetraene N-JlCPF*): After Exposure to Oxygen, CDaCN.Solvent ...... 99

7. NMR Assignments for [Ni{(NBz)2 (CHa)eMea[16]tetraene

N 4 }](PFe) 2 in CDaCN, 80 MHz, 4 0 ° C ...... 118

8 . ^^C NMR Assignments for [Ni{(BzNMe)aMea[16]tetraene

n J ] ( P F 6 > 2 in CDaCN, 8 MHz, 4 0 ° C ...... 120

9. Complexes Which have been Covalently Attached to Chloromethylated Polystyrene...... 131

10. Analysis of Cobalt Complexes Covalently Attached to Chloromethylated Polystyrene ...... 132

11. Lacunar Complexes which are Coordinatively Attached to Polymer Through Interaction with Pyridine Function 146

vii LIST OP FIGURES Figure Page

1. Methods of Attaching Complexes to Polymers in Modeling of Natural Systems ...... 5

2. Structure of Iron(II) Protoporphyrin IX 6

3. Fuhrhop’s Polymer Study with Polymer and Porphyrin 8

4. Basalo's Imidazole Functionalized Silica Gel ..... 10

3. Allcock's Imidazole Functionalized Polyphosphazene P o l y m e r ...... 12

6 . Bayer and Holzbach's Peptide Synthesis Techniques Allow Incoporation of Both Imidazole and Heme Porphyrin into Polymer Backbone in a Water Soluble System .... 15

7. Collman's Picket Fence Porphyrin ...... 16

8 . Polymer Bound Diphenylphosphine with Rhodium Hydrogenation Catalyst ...... 23

9. Rigid Rhodium Polyyne Support ...... 24

10. Rhodium and Palladium on Modified Polystyrene Supports Show Selective Hydrogenation Catalysis ...... 26

11. Polymer Bound Rhodium Catalyst is Used in Hydro­ formylation Reactions ...... 26

12. Cycle Proposed for Copper(II) Catalyzed Autoxidation of Hydroquinone ...... 28

13. Polymeric Schiff-Base Compounds ...... 29

14. Rollman's Polymer Bound Metalloporphyrin which Contains both an Oxidation Promoter and Proton Acceptor’ 30

15. Drago's Method of Polymer Attachment of bis(Salicylal)- ethylenediimmine Cobalt (I I ) ...... 31

16. Mechanism Proposed for Oxidative Polymerization of XOH by Copper (II) Metal Ions...... 36

viii FIGURES (CONT'D) Figure Page

17. Complex of Poly(vinylpyridine) with Bis(Dimethyl- glyoximato)Cobalt(II) ...... 40

18. Model Proposed by Jones for Concerted Acid-Base Mediation of Catalase Enzyme...... 41

19. PshezhetsRi's Polymer Bound Catalase Activity .... 42

20. Synthesis of Lacunar Complexes Beginning with Neutral Nickel(II) Macrocycle ...... 44

21. Model of Myoglobin. The Essential Features Believed Necessary for Oxygen Binding Function are Highlighted 46

22. The Most Promising Iron(II) Dioxygen Carrier. . . . 47

23. Reaction Vessel for Polymer Reactions with Cobalt(II) Macrocyclic Complexes and Chloromethylated Polystyrene 65

24. Soxhlet Extractor for Removing Absorbed Cobalt(II)

Complexes from Covalently Bound Polymeric Supports 66

25. ESR Spectrum of [Co{MeNIminoethyl)(MeNHEthi)Me2 [1 6 ]- tetraene N,}] (PF6 )2 * Frozen Glass in Acetone, -196®C 74

26. Internal Rearrangement of Peripheral Secondary Amine Allows the Ligand to Coordinate to Cobalt(II) Through Five Nitrogen Donors ...... 76

27. Infrared Spectrum of [Co{(MeNIminoethyl)(MeNEthi)

Meg[l 6 ]tetraene N, }(PF«)a, Nujol Mull ...... 79

28. Cyclic Voltammograms for [Co{(MeNIminoethyl)(MeNEthi)

Meg 116]tetraene Ng}(PFe ) 2 in Acetonitrile Solvent (0.1 M BuNBFfc Supporting Electrolyte, vs AgNOg (0.1 M ) , Ft Disk Electrode)...... 81

29. NMR Spectrum of [Co{(MeN Iminoethyl)(MeNH Ethi) M e g [16]tetraene Ng}]•CIl3CN(PF 6 )s in CD 3 CN, room Temperature ...... 83

ix FIGURES (CONT'D) Figure Page

30. NMR Spectrum of ICo{(MeNIminoethyl)(MeNHEthi)

Me* 116]tetraene N, } ] >'CH3 CN(PF6 )3 in CD 3 CN, 40°C, Proton Decoupled ......

31. ORTEP Diagram of Pentacoordinate Cobalt(II) Showing the Rearrangement of the Peripheral Ligand ...... 90

32. ORTEP Diagram of Pentacoordinate Cobalt(II) Showing the Orientation of Acetonitrile Solvent Molecule . . 91

33. Stereo Packing Diagram of [Co{(MeNIminoethyl) (MeNHEthi)M0 2 [16]tetraene N,}] (PFg)2 ‘CHaCN ...... 92

34. Numbering Scheme for the Atoms ...... 93

35. Spectral Changes for [Co{(MeNIminoethyl)(MeNHEthi)Me2 - [16]tetraene Ns }] (PFs)% upon Exposure to Dioxygen in Acetone Solution at Room Temperature ...... 98

36. '’H NMR Spectrum in CD 3CN After Exposure of [Co{ (MeNIminoethyl) (MeNHEthi)Me2 [16] tetraene Ns).]

.(PF6 ) 2 to Oxygen at Room Temperature ...... 100

37. ^^C NMR Spectrum in CD 3CN (^H Decoupled) After Exposure of [Co{(MeNIminoethyl)(MeNHEthi)Me2 [16] tetraene

Ns)] (PF6 ) 2 to Oxygen at Room Temperature ...... 101

38. Spectral Changes Accompanying Successive Exposure of

[Co{(MeNIminoethyl)(MeNHEthi)Me2 [16]tetraene Ns) ](PFe)2 - to Nitrogen (Ultra High Purity) in an Acetone Solution, - 4 0 ° C ...... 102

39. Reversible Spectral Changes Accompanying the Cycling Between Oxygen and Nitrogen Atmospheres for the Penta­ coordinate Cobalt(II) Complex at 40.0“C in Acetone . . . 104

40. ESR Spectrum (Frozen Glass) Produced by Exposure of the Pentacoordinate Cobalt(II) Complex to Oxygen at -40.0°C in an Acetone Solution ...... 106 FIGURES (CONT'D) Figure Page

41. ESR Spectrum (Room Temperature) Produced by Adding Nitrosobenzene to a Solution of [Co{(MeNIminoethyl) (MeNHEthi)Me2 {16] tetraene N 5}](PFe): in Acetone Solution Saturated with 0% at 2 5 ® C ...... 107

42. Model Reaction of Macrocycles with Benzyl Chloride 111

43. NMR Spectrum of [Ni{(NBz)2 9CH 2 )sMea116]tetraene

N j ] ( P F a )2 in CDsCN, 25“C ...... 113

44. NMR Spectrum of [Ni{(NMeBz)aMea[16]tetraene Nj ]

(PFs) 2 in CDsCN, 25°C ...... 114

45. A) Coupled and B) Decoupled ^®C Spectra pf lNi{(NBz)a(CHa)«Mea[16]tetraene n J](PFg)a in CDsCN, 40°C ...... 116

46. A) Coupled and B) Decoupled ^®C NMR Spectra of

[Ni{(BzNMe) aMea[16] tetraene N 4 ] (PFs) 2 in CD 3CN, 40°C 117

47. ORTEP Drawings of [Fe (NHEthi)aCsMea[16]tetraene N»

(CO)(PY)](PF 6 )a'CHsOH ...... 122

48. GPG Propylamine May Coordinate to Co(II) as an Axial B a s e ...... 123

49. CPG Propylamine May be used to Covalently Attach Macrocyclic Ligands ...... 125

50. A) Spectrum of Co(II) Lacunar Complex in DMF, Frozen Glass, -196°C, B) Spectrum of Co(II) Lacunar Complex in DMF with Propylamine, Frozen Glass, -196°C .... 126

51. Comparison of Infrared Stretching Frequencies of Substrates Related to Chloromethylated Polystyrene 128

52. Infrared Spectrum of Chloromethylated Polystyrene After Treatment with a) Copper(II) and b) Cobalt(II) Lacunar C o m p l e x e s ...... 129

53. ESR Spectrum of [Cu{(NHa) 2 (CHa)«Me:,[16]tetraene Ni,}] (PF«)a in Acetone Solution, Frozen Glass, -196°C . . 134

xi FIGURES (CONT'D) Figure Page

54. ESR Spectrum of Polymer I, No Solvent, Frozen, - 1 9 6 ° C ...... 135

55. ESR Spectrum of Polymer II, No Solvent, Frozen, - 1 9 6 ° C ...... 137

56. ESR Spectrum of Polymer III, No Solvent, Frozen, - 1 9 6 ° C ...... 138

57. Kida Type 11^*'. The Unpaired Electron in the dyg Orbital. Ligands have Less Developed ir System, Electron Density is not Extensively Delocalized Over the Ligand S y s t e m ...... 140

58. Kida Type I^®’.. This Spectrum is Commonly Observed for Cobalt (II) Complexes with Four Nitrogen Donors Having Well Developed tr Systems...... 141

59. ESR Spectrum of Polymer III, No Solvent, -196°C. . . 142

60. ESR Spectrum of Polymer V, No Solvent, Frozen, -196°C 144

61. ESR Spectrum of Polymer VI, No Solvent, Frozen, -196°C 145

62. Typical ESR Spectrum Observed with Cobalt(II) Complexes Coordinatively Attached to PORAPAK [Poly(vinyl pyridine)], Frozen, No Solvent, -196°C ...... 148

63. ESR Spectrum of CPG Propylamine Polymer with

[Co{(MeNEthi)2 (CH2 )sMe2 [16]tetraene N^}](PF«)2 , No Solvent, Frozen, -196°C ...... 150

64. ESR Spectrum of CPG Propylamine with [Co{(HNethi)2 - (M-xylyl)Me2 [16] tetraene N^,}] (PF«) 2 , No Solvent, Frozen, -196°C ...... 151

65. ESR Spectrum of Polymer II After Exposure to O 2 for 45 Minutes, No Solvent, Frozen, -196°C ...... 155

6 6 . Decay of Oxygen Adduct of Polymer II, No Solvent, Frozen, -196° C ...... 156

xii FIGURES (CONT'D) Figure Page

67. ESR Spectrum of Less Concentrated Sample of Polymer III After Exposure to 0% for 30 Minutes, No Solvent, Frozen, -196“C ...... 157

68. ESR Spectrum of Oxygen Adduct of ICo{(MeNEthi)%-

(CHz)gMea[16]tetraene Nz,}](PF6 ) 2 Attached to Porapak S, Polymer XII ...... 159

69. Treatment of Sample Shown in Figure 68 to Dynamic Vacuum at 60°C, ESR Spectrum, No Solvent, -196°C . . 160

70. ESR Spectrum of CPG Propylamine with [Co {(MeNEthi)a- (CHa) fiMea[16] tetraene Nj,}](PF6)a Absorbed to Surface After Exposure to Oxygen, Frozen, -196°C ...... 165

71. Comparison of Product Yields in Two Types of Solvent for Polymer I I ...... 170 i 72. Polymer II Used as a Gas Chromatographic Column for Separation of Oa and Na ...... 172

73. Oa/Na Separation with Polymer II ...... 173

74. Separation of Oa/Na Was Achieved at -78°C Using a Sample of Polymer X I ...... 176

75. Compounds Used as Spin Traps ...... 179

76. ESR Spectrum (Solution Cell) of a Solution of

[Co{(Me2NEthi)aMea[16]tetraene N*}](PF 6 )a and 5,5'-Dimethylpyrroline N-oxide in CHsCN After Exposure to Oa, Room Temperature ...... 181

77. ESR Spectrum (Solution Cell) Observed for the Penta­ coordinate Cobalt(II) Compound with Pyrroline N-oxide in CHsCN Solution After Exposure to Oa at Room Temperature ...... 183

78. ESR Spectrum (Solution Cell) for a Solution of [Co{(MeNIminoethyl)(MeNHEthi)Mea[16]tetraene Ns) ]- (PFs)a and Nitrosobenzene in Water/Acetone (1 1) After Exposure to Oxygen at Room Temperature ...... 184

xiii FIGURES (CONT'D) Figure Page

79. ESR Spectrum of 2,6-Dimethylphenol, ICo{(MeNImino­ ethyl) (MeNEthi)Me2 [16]tetraene Ns}](PFg)2 » and N-Tertbutyl a-Phenylnitrone in CHaCN After Exposure to Oxygen, Room Temperature ...... 185

80. ESR Spectrum of [Co{(MeNIminoethyl)(MeNEthi)Mea[16]-

tetraene Ns}](PFe )2 and N-tertbutyl a-Phenylnitrone in CHsCN After Exposure to Oxygen, Room Temperature . 186

81. l-Oxo-2,6 -dimethylphenyl Adduct of Nitrosobenzene . . 187

82. NMR Spectrum of l-Oxo-2,6 -dimethylphenyl Adduct of Nitrosobenzene in CeDs, Room Temperature .... 188

83. Cobalt-peroxy Adduct of a Substituted Phenol that was Isolated by Nishinaga ...... 189

84. Mechanism Proposed by Nishinaga for the Oxidation of P h e n o l s ...... 190

85. Mechanism Proposed by Drago for the Oxidation of Phenols ...... 192

XIV CHAPTER 1: INTRODUCTION

Polymer Supported Metals and Metal Complexes

In recent years there has been a great research emphasis on the

attachment of homogeneous transition metal complexes to polymeric

supports.The polymer represents a specific structure whose enormous

chain surrounds the central metal ion or metal complex. The polymer metal complex can show important characteristics different from the

simple metal complex including enhanced stability and activity of the

catalyst. In nature large protein molecules surround metal ions; the

proteins have definite three dimensional structure and can give metal

ions unusual coordination number or oxidation state through the arrange­ ment of sterospecific coordination sites or by distorting the coordination

sphere.

Polymer supported metals have been of interest in modeling the

natural proteins and at the same time have been found to be highly i efficient catalysts. For example, the modeling of hemoglobin and myoglobin

in the reversible binding of dioxygen by iron(II) porphyrins supported on

polymers has been an active area of interest.'*’® Insoluble polymers have

a special application in catalysis since an immobilized metal is easily

recovered from the reaction mixture through simple filtration. In

addition, sometimes both an enhancement of catalytic activity and greater

selectivity have been observed over the transition metal complex in

solution. The supports used may be derived from organic or inorganic

materials, naturally there are inherent advantages and disadvantages in

1 2 the choice of the support (Table 1 ). The primary requirement for a useful organic polymer support is the incorporation of functional groups which are potential ligands for transition metals. The range of organic supports is enormous, with polymers extending from polystyrenes and polyamides to cellulose; and often groups can serve as an initial point for further modification and attachment of ligands, limited only by the chemistry of the polymers.^'*" The rediscovery of polymers as tools for synthesis was made by Merrifield in 1963 when he introduced his "solid- phase" technique for peptide synthesis using chloromethylated polystyrene now known as "Merrifield Resin." The insoluble crosslinked macromolecule was used as a protecting group, simultaneously providing a facile method for isolating and purifying the product of each condensation step during protein synthesis. ® Since that time functionalized polymers have found widespread application. They have been employed as stoichiometric reagents, as catalysts,®’® as protecting groups,^ in analytical chemistry,’”^® in ion exchange, in the detection of reaction intermediates, in chromato­ graphy^’^ and in the immobilization of enzymes and cells.

TABLE 1

MATERIALS USED TO SUPPORT METAL COMPLEXES^

INORGANIC ORGANIC

Silica Polystyrene Polyamino acids Zeolites Polyamines Urethanes Glass Silicates Acrylic Polymers Clay Polyvinyls Cellulose Metal Oxides such Polyallyls Cross-linked dextrans as Alumina Polybutadiene Agarose ' 3

The rigid structure of inorganic supports is an important advantage

relative to organic polymers. Often polymer flexibility leads to de­

activation through intramolecular interactions. The upper limit of

thermal stability of organic polymer supports is 433 K, but stability

with inorganic metal carbonyls has been observed above 573 K. Polymer

swelling under variable temperature and solution conditions makes practical

control of diffusion variables difficult. Inorganic supports can be pre­

selected for stable diffusional characteristics at most reaction conditions.

Many inorganic oxide supports possess active surface groups that have been

used for immobilizing metal carbonyls in various ways. Surface silanol

groups and siloxane bridges can react directly with carbonyls to immobilize

them through direct surface bonding and, like organic polymers, they can

react to produce a support with attached anchoring ligands for catalyst

immobilization. However, undesirable side reactions on the support prove

to be a frequent disadvantage of oxide-immobilized catalysts.®

Heme Protein Modeling

The study of myoglobin and hemoglobin, the proteins responsible for

the reversible binding of dioxygen, has been extensive. From the pooled

resources of many researchers there is now a clear idea of the struture

of myoglobin,*® and there are logical models concerning the interaction

of the tetramer units of the protein hemoglobin.*^ In oxygen transport,

the role of the protein in preventing irreversible oxidation of iron(II) is

twofold. The primary function of the protein is to prevent- p-oxo dimer

formation by isolating the individual heme units. A secondary role occurs

through the arrangement of amino acid residues which protect the oxygen 4 binding site from acids which would promote oxidation.* In 1958 Wang was the first to recognize and demonstrate that a synthetic polymer could serve these two roles of the protein. Collman, Halpert, and Suslick have written a review* which provides an excellent summary of the synthetic attempts to model the natural system. Tsuchida has also authored several reviews in the polymer area which summarize the literature. His work on the insoluble and soluble polymer systems with porphyrin has been extensive.®

Studies using polymers to model the reversible binding of oxygen by heme units typically have been divided into three general categories (Fig.

1). The first of these involves the absorption of both a porphyrin and an axial base onto a solid support as was the method used by Wang.^*’^®

Wang and co-workers prepared the complex containing polymer by reducing a benzene solution of iron(III) protoporphyrin IX diethyl ester (Fig. 2) and excess 1 -(2 -phenylethyl)imidazole with excess dithionite under carbon monoxide, and then adding polystyrene to the organic layer, followed by evaporating the benzene under carbon monoxide to leave a film containing

Fe(II)(CO)(PEIM). The CO was removed by flushing with nitrogen and an oxygenated hemoglobin was obtained by subsequent reaction with air or oxygen. Spectral studies showed that the oxygenation could be reversed by flushing with nitrogen. This technique was used more recently by Chang and Traylor^® with a heme containing an appended imidazole, thereby requiring no addition of base. Reversible oxygenation was detected spectro- photometrically.

The second approach to polymer supports in the study of natural systems involves the attachment of the axial base to the polymer through 1: Porphyrin and axial base absorbed on polymer

2; Base covalently attached to polymer

3: Heme or heme and base covalently attached to polymer

—B

Figure 1; Method s of Attaching Complexes to Polymers in Modeling Studies of the Natural Systems CH=CHg

ù

//

CH=CH,

R = CHgCHgCOOH

Figure 2; Structure of Iron(II) Protoporphyrin IX 7 a covalent bond. This represents a somewhat refined model for hemoglobin since an imidazole function from a proximal histidine residue is able to coordinate to the heme unit providing a fifth ligand for iron(II) in the natural products. Collman and Reed^^”^ ®claim to be first to prepare a polystyrene resin containing imidazole units which combined with iron(II) tetraphenylporphyrin. The reaction was carried out in a benzene solution and gave predominately a six coordinate bisimidazole complex with the porphyrin. When exposed to oxygen the p-oxo dimer was formed rapidly, suggesting that the solvent swollen polymer matrix was not rigid enough to prevent interaction of two heme units. Another research effort utili­ zing this approach was undertaken by Tschida and co-workers^ «-z s have studied the interaction of 0% with Fe(II) protoporphyrin IX supported on poly(A-vinylpyridine), poly(4-vinylpyridine) partially quatemized with benzyl chloride (20%), and poly(N-vinyl-2-methylimidazole). There is some disagreement about the data Tsuchida reports of the heme studies with O 2 .**

The problem centers on a one hundred fold excess of sodium dithionite which was used to reduce} the iron(III) complex. The dithionite remains in the polymer solution or film throughout subsequent manipulations and it is uncertain how Tsuchida determined the oxygen binding capabilities in his systems. The stoichiometry of oxygen uptake is reported to be .5 0 2 /heme, although Na 2 S 2 0 * can react rapidly with 0 % in solution and is present in excess in the system studied.

In an enlightening report on reversible oxidation in imidazole polymer- heme systems, Fuhrhop and co-workers varied the percentages of imidazole and substituted imidazoles with styrene using a variety of transition metal porphyrins, this is illustrated in Figure 3.“'* These important results Modified Protoporphyrin IX was used in the Study \/

Either Cft=GH2 was used or a saturated CHaCHs was used on the porphyrin periphery

Ra “ CH 2 CH 3

R 3 “ CH—CHa

The Porphyrin was polymerized with substituted Imidazoles and various ratios of styrene -H— C H 2

Ri A H

B CH 3 C CsHs

Iron(II), Iron(III), Co(II) were varied as the metal Center

[-Ha C-ÇHj^^CHa-ÇH-] Cf'

Ri H H H H CHa CsHj Monomer Mole Ratio 1 1 1 0 1 0 0 2 J L Styrene 2 1 1 1 1 1

Figure 3; Fuhrhop's Polymer Study with Imidazole and Porphyrins 9 have resolved many of the ambiguities reported in the literature up to that time. Where the concentration of imidazole in the polymer is greater than 10%, the heme is six coordinate and low spin, while it is five coord­ inate and high spin at lower imidazole concentrations, or if the polymer is prepared from the bulky Imidazoles, 2-methyl and 2-phenyl Imidazole.

The low spin hemes in the polymer matrix did not undergo change in the presence or absence of oxygen; but the high spin complexes of coordination number five showed reversible spectral changes when oxygen was added or removed. These oxygenated iron(II) porphyrin systems were unique in that oxygen was added at room temperature and the oxygen adduct was stable for weeks, with no oxidation to iron(III) porphyrins. The disadvantage found for this system was the slow adsorption and desorption of gaseous oxygen in the solid state. The Soret band reported for the oxygenated species does not correspond to the 402 nm band of the oxygen complex reported by

Tsuchida.*® Instead, the bands attributed to reversible oxygenation by

Tsuchida are postulated to be an iron(III) species; the seeming reversibility in the system could be explained by the reduction of the iron(III) complex back to the iron(II) complex by the excess sodium dithionite used in the system.*‘* Fuhrhop also showed reversible oxygen adduct formation with co­ balt (II) porphyrins in the solid state for a variety of imidazole function- alized polymers.*"

Basolo and co-workers*® used rigid functionalized silica supports to prevent oxidation via the dimerization route (Fig. 4). The porphyrins used in their study were solid well characterized ferrous compounds which avoided the need for reducing agents in the system. Initially six coordinate com­ pounds were formed on the polymer but heating to 25 °C under flowing helium 10

Figure 4: Basolo*s Imidazole Functionalized Silica Gel 11 removed one coordinated base to leave a five coordinate Fe(II) porphyrin.

The resulting silica gel was found to chemisorb approximately one mole of oxygen per iron at -127"C and to release oxygen quantitatively at room temperature. Determination of the stoichiometry of the dioxygen uptake was complicated by background due to absorption of comparatively large amounts of dioxygen by the silica gel. This accounted for two thirds of the total volume of oxygen and forced a large correction factor in the calculations. After the corrections, the amount of oxygen absorbed by the heme on silica was very low. The oxygen affinity observed is a weak one; at 0 P, is 230 torr and at -46"C it is 4 torr. Myoglobin ■1 has a much stronger oxygen affinity than does this modified silica gel.

Data for human myoglobin extrapolated to 0"C give Pj as .14 torr and for sperm whale myoglobin, 0.8 torr.Nevertheless, the stoichiometry of the reaction shows one oxygen molecule per atom of iron, and the adduct is not destroyed by exposure to higher temperatures. The resistance to dimer formation appears to result from the motional hindrance associated by attachment of the iron porphyrin to the rigid surface of silica.

A slightly different approach to base polymers was explored by

Allcock and co-workers.*® An imidazole functionalized polyphosphazene

(Fig. 5) was used to support iron(III) protoporphyrin IX. Excess aqueous dithionite was used to reduce iron, and Mossbauer and electronic spectra indicated that a six coordinate bisimidazole heme was formed. Exposure to oxygen in aqueous solution led to rapid and irreversible oxidation of the iron, although when the films were prepared by evaporating the watsr from the ferrous heme polymer, partial reversibility for oxygen was ob­ served spectrophotometrically on cycling between an oxygen atmosphere and 12

1 0 X

Figure 5: Allcock's Imidazole Functionalzied Polyphosphazene Polymer 13 vacuum. However Mossbauer spectra revealed that oxidation was also occuring and that the resulting iron(III) species was being reduced either by the polymer backbone itself or by excess dithionite.

A third general class of heme polymer systems occurs when the heme itself is covalently attached to the polymer. Fuhrhop and co-workers** were the first to adopt this approach by preparing a set of terpolymers through free radical polymerization of styrene, various

1-vinylimidazoles and iron protoporphyrin IX dimethylesters. The vinyl side chains of the polymer were thus incorporated into the backbone of the polymer. The results obtained here were identical to earlier results described by Fuhrhop with the functionalized imidazole polymers.

Solid terpolymers with greater than 10% imidazole units contain six coordinate bisimidazole heme which does not react with dioxygen while those with less than 10% imidazole contain five coordinate heme which reacts reversibly with dioxygen. Similar approaches have been reported by Tsuchida;** porphyrins with vinyl functions were copolymerized with styrene, and terpolymers were prepared by including 1-vinylimidazole in the reaction mixture as well. The iron(III) containing polymers were reduced in nonaqueous solutions with Cr(acac)z and the reaction with Oz in solution was monitored by visible spectroscopy. Spectra were consistent with the formation of dioxygen complexes, having decomposition half lives at 20 "C on the order of 4 to 40 minutes.

Ledon and Brigandat** prepared macroporous copolymers of styrene and

4-aminostyrene crosslinked with divinylbenzene. The porphyrin was attached by treatment of the polymer with tetra(p-chlorocarboxyphenyl)porphyrin and iron(II) was inserted directly with ferrous chloride in dimethylfor- 14

amide. The troublesome reduction of iron(III) with dithionite was there­ by avoided. The solid 20% crosslinked divinyl benzene polymer was ex­

posed to air and irreversible oxidation occurred with a characteristic p-oxo stretching mode appearing in the Infrared at 860 cm"*. However with 30% crosslinked polymer, reversible oxygenation was characterized

through visible spectroscopy and the absence of the oxo stretching band

in the infrared spectrum.

Bayer and Holzbach functionalized the two polymers shown in Figure

6 with imidazole and porphyrin units using peptide synthesis techniques.^®

This polymer system was revolutionary since its water solubility permitted

high concentrations of dioxygen carrier; the structure of the functionalized

polymer provided hindrance to irreversible oxidation of the complex through

the dimerization route; and finally, both proximal and distal imidazole

functions were imitated in the complex. After synthesis, the iron is

present in the +3 oxidation state, and it is reduced with sodium dithionite.

The excess reducing agent was removed by chromatography on Sephadex

supports. Oxygen was reversibly removed from the dioxygen-iron complexes

in vacuo (.01 to .04 torr, 5 minutes); this process could be repeated

several times. Carbon monoxide adducts are also possible in this iron

polymer complex, and the dioxygen adduct was obtained from the carbon

monoxide adduct after prolonged passage of gaseous oxygen through the system.

After several cycles of reversible oxygenation, the complex shows increasing

signs of irreversible oxidation to the iron(III) polymer, which was de­

tected by the unstructured spectrum in the visible region, and the shifts

of the bands in the ultraviolet region. The P. for oxygen binding is .20

torr in close approximation to the values for myoglobin. At room temper- 15

V *iî

V? - PE G j oo(T^”C - N —( C H 2 ) 2~N-( Cl 12) 2“ —C -0“ PE G 1000'" h h

Cl,, “y r Crt2 C'Ha 'Cf

f 1

k s « = " ' \ CH- :K - '

, CM>

Figure 6: Bayer and Holzbach's Peptide Synthesis Techniques Allow Incorporation of Both Imidazole and Heme Porphyrin into Polymer Backbone in a Water Soluble System, 16

ature, a graph of percentage oxygenated vs. partial pressure of O 2 is

sigmoidal, suggesting a cooperative effect in oxygen binding.

Tsuchida*® reports on the binding of the iron(II) picket fence porphyrins to amphillic block copolymers in water. The polymers were based either on the N-ethyl imidazole or styrene-N-vinyllmidazole mono­ mers. The block copolymers (ABA) were formed by combining A poly(ethyl-

eneoxide) with B poly(styrene-N-vinylimidazole) to give a hydrophopic polymer which was capable of forming micelles. The iron picket fence porphyrins were incorporated into the micelle forming copolymers in water; reversible oxygen binding due to the hydrophobic field around the

oxygenated complex was observed (Fig. 7).

CH3

CO

DCO*

Figure 7: Collman's Picket Fence Porphyrin 17

Polymer Supported Catalysts

For many years most of the known catalysts were heterogeneous and the catalyzed reactions were promoted by adsorption of reactants on catalyst surfaces. Such systems were necessarily composed of two phases. More recently, homogeneous catalysts have been developed in which the catalyst is a specific chemical compound that has been introduced into the reaction solution as a dissolved molecular entity. In this case, catalysis occurs through the formation of soluble intermediate compounds that can sometimes be detected and identified. These systems are especially amenable to mechanistic studies. A third classification involves supported complex catalysts, and this category includes some of the best features of both of the other types of catalyst. In order to appreciate polymer supported catalysts, it is necessary to look at the advantages and disadvantages of heterogeneous and homogeneous catalysts.^

A major disadvantage of homogeneous catalysts is the problem of separating the often very expensive catalyst from the products at the end of the reaction.With heterogeneous catalysts this can be achieved through filtration, while with homogeneous catalysts an efficient dis­ tillation or ion exchange process is usually required. Distillation is normally an endothermie process and is therefore expensive; furthermore, unless it is efficient, it may result in catalyst loss which could result in contamination of the product or render the process uneconomic.

In a heterogeneous catalyst, the reaction of the catalyst takes place on the surface so that all molecules of the catalyst not present on the surface remain unused. By contrast, all molecules in a homogeneous reaction

are theoretically available as reaction centers; this potentially more 18 efficient use may decrease the amount of catalyst needed in a given reaction. Another advantage of the homogeneous catalyst is the repro­ ducibility of reaction conditions because of the definite structure and stoichiometry of the catalyst and derived intermediates. On the other hand, the surface of the heterogeneous catalyst is dependent on the method of preparation and subsequent handling, thereby often making reproducibility difficult.^’*

Normally homogeneous catalysts have only one type of active site; therefore, they tend to be more specific than the heterogeneous catalyst where several types of active site are possible due to variations in the surface structure. Furthermore, the homogeneous catalyst can often be systematically modified by altering ligands in such a way as to change the electronic nature or the steric requirements of the active site; such changes may fine tune the catalyst and impart specificity in reactions.

The thermal stability of heterogeneous catalysts is often much higher than that of homogeneous catalysts; because the rates of many reactions increase with temperature, a high operating temperature may be an advantage.

However, high operating temperatures can be a disadvantage for some reactions, and may be expensive to maintain. Finally, whereas the range of suitable solvents is limited by the solubility characteristics of the homogeneous catalyst, this clearly present no problem for the heterogeneous catalyst.^

At first glance, it would seem that the balance of the advantages would lie with the homogeneous catalyst. However, the problem of catalyst separation and recovery is so important that chemical processes based on homogeneous catalysts are relatively rare. For illustration, two commercial 19 processes which are homogeneous are the Wacker process®'* for the oxidation of ethylene to acetaldehyde and the Monsanto process®* for the carbonyl” ation of methanol to yield acetic acid. Both of these methods depend on the relatively low boiling points of the product (acetaldehyde, 20.8°C and acetic acid, 117.9°C) for separation.

Recently considerable effort has been expended in developing catalysts that combine the possible advantages of both kinds of catalysts. One approach had led to supported complex catalysts, which essentially involves the binding of a compound of a metal to some sort of support.*

A functionalized polymer is a synthetic macromolecule to which are chemically bound functional groups that can be utilized as reagents, cata­ lysts, or protecting groups. The macromolecule can be a linear species capable of forming a solution in suitable solvents; or alternatively, a crosslinked species, a resin, which although solvated by suitable solvents remains macroscopically insoluble. The use of resins has become more widespread because of the practical advantages accruing from their insolu­ bility. The ease èf chemical modification of a resin and the level of success in applications depends substantially on the physical properties of the resin itself, Functionalized polymeric supports must possess a structure which permits adequate diffusion of reagents into the active site, a phenomenon which depends on the extent of swelling or solvation, the effective pore size and pore volume, and the chemical and mechanical stab­ ility of the resins under the conditions of a particular chemical reaction or reaction sequence.'*’*^

Three main types of resins are commonly used; they are microporous or gel-type resins, macroporous resins, and macroreticular resins.' Micro- 20 porous species are prepared from a vinyl monomer and a difunctional vinyl comonomer in the absence of any additional solvating media. In the dry state they are microporous, with the polymer side chains being separated by typical solid state intermolecular distances. On contact with suitable solvents a gel network is formed with generation of considerable porosity depending on the degree of crosslinking. When crosslinking is less than 1% (i.e. % of difunctional comonomer used in polymerization), swollen resins have low mechanical stability and readily fragment. Increasing the amount of comonomer to 8% gives a resin which is mechanically stable, but problems arising from diffusional limitations result in slow and often incomplete reactions. Commercially prepared gel resins are usually 2% crosslinked which provides a compromise between rapid diffusion and mechanical stability to give ease in handling. The macroporous polymer is prepared in the same way as the above, but with the inclusion of an inert solvent. The solvent chosen solvates both the monomer and polymer; a fully expanded network is formed with a considerable degree of porosity. Removal of the solvent causes a reversible collapse of the matrix, and in the dry state these materials are similar to micro­ porous resins. To achieve mechanical stability, for ease of handling in the solvent swollen state, it is common to use large amounts of difunctional comonomer in the preparation (up to 20%). The macroreticular resin is formed when the solvent used dissolves the monomer but allows the polymer to precipitate. The resulting material is a highly porous rigid material which retains its shape w h m the precipitant is removed. In these resins a higher percentage of difunctional comonomer is used, thereby allowing materials to withstand high pressures such as is required for HPLC con- 21 ditions. The structure of these resins is very different from the two formerly mentioned resins. Macroreticular resins have a large and permanent pore volume, and reaction sites may be regarded as being located on permanent interior surfaces of the resin.

In order to determine the optimum polymer supported catalyst for a particular reaction there are a number of parameters which should be considered. The first of these is the nature of the support. For rhodium(I), silica supported catalysts may be more active in hydrogen­ ation* because the metal is bound to the outside of the surface and therefore is more readily available to the reactants than in the poly­ styrene based catalyst where the complex may lie deep within the polymer bead. The polystyrene based complexes,*® however, show more selectivity in reactions, while the silica-based catalysts prove valuable when rapid reaction is needed. A further consideration involves the amount of crosslinking present in the organic support. In polystyrene polymers, variation in crosslinking is achieved by varying the amounts of divinyl benzene in the polymerization process. The influence of crosslinking can be profound. For example, using a polystyrene phosphine compound based on chlorotris(triphenylphosphine)rhodium(I) Grubbs** found that the supported catalyst was only .06 times as active as the homogeneous catalyst, for the case of a 2% crosslinked polymer with large bead size (74-149 pm diameter). However, using smaller beads (37-74 pm) and 1% cross-linking,

Pittman^* found that the supported catalyst was .8 times as active as the monomer complex, but 13 times more reactive than the polymer system first tested. In addition to altering the activity, varying the degree of crosslinking alters the specificity. Higher crosslinking makes the 22 polymer chain more rigid, and this makes it more difficult for any but

the smallest molecule to penetrate the polymer.^”

It has also been established that solvent polarity affects the

swelling of organic polymers. For polystyrene polymers, an increase

in solvent polarity decreases the swelling of the polymer. When this was investigated the simple relationships predicted did not hold. In

addition to the change in pore size, an increase in polarity led to an

increase in the polar gradient between the bulk solvent and the local

environment around the catalytic site, which could promote diffusion of nonpolar olefins to the active size but would impede the diffusion of polar olefins. Nonpolar olefins all showed initial rate enhancements

as the percentage of ethanol was increased, until at high concentrations, the pore size decreased and the size effects of the substrate again

became important. Polar olefins showed steady rate reductions with

increasing solvent polarity.Clearly the influence of solvent is an

important factor which can be varied to control activity and selectivity

of a particular supported catalyst. It is also obvious that the nature

of the supported metal complex can have a profound influence on catalytic

activity. However, a less obvious effect shows that a difference in

rate is often observed for the same supported complexes prepared in

different ways."®^’®®

Activity of the supported catalyst compared to the free monomeric

compounds is an important issue. In some cases involving hydrogenations,

the reported activity^®”®* is lower, which is concluded to be due to the

unavailability of some of the metal centers. However a special situation

arises if the active homogeneous catalyst readily dimerizes, because di- 23 merization is prevented by isolation of the monomeric units in the polymer, and the reported catalytic activity is larger In the polymer.^®

Nevertheless, polymer metal complexes are generally useful as immobilized catalysts, and are sometimes more active than the monomer complexes due to the effects of the large ligand molecule, the polymer. Polymeric material can be used as insoluble carriers for catalysts with the envir­ onment surrounding the metal or the boundary of the surface providing a specific catalytic site. As stated earlier, a reliable advantage of the supported catalyst lies in its easy separation from the reaction mixture and the potential for repeated use of the catalyst.

Hydrogenation and Hydroformylation Using Supported Catalysts. —

Hydrogenations using supported metal catalysts have also been studied by many investigators. Most often the polymer used was chloromethylated polystyrene crpsslinked by divinyl benzene. This starting material will react with diphenyl phosphine and the product will complex with rhodium, ruthenium, or nickel to give active hydrogenation catalysts. Allum et al.^* report that the carbonylacetylacetonatorhodium(I) complex (Fig. 8) prepared in such a manner can catalyze the hydrogenation and hydroformyl­ ation of 1-hexene and and cyclodimerization of butadiene. The solid phase

o CH, "3 Figure 8; Polymer Bound Diphenylphosphine with Rhodium Hydroformylation Catalyst. 24 catalyst was easily separated from the products and could be used repeatedly. According to Bruner and Bailar (1973) and Kaneda® ®

(1975) the bischloropalladium(II) complex of the same type of phos- phinated polystyrene ligand showed a remarkably high hydrogenation efficiency for the methyl ester of soybean fatty acid than did the monomer. The high activity was considered to result from the coordinate unsaturation of the polymer complex. And, cyclooctadiene was more efficiently hydrogenated by the supported polymeric iridium(I) complexes of chlocarbonyltris(triphenylphosphine) to give cyclooctane and cyclo- octaene as hydrogenation products. The most important identified feature of an active hydrogenation catalyst is the presence of a coordinately unsaturated metal atom having one or more orbitals where facile substrate coordination or adsorption can occur. This may be inherent or created through ligand dissociation. Where dimerization of a metal is a cause in the loss of catalytic activity, polymer supported complexes are ad­ vantageous.®

The rhodium complex of a rigid polyyne shows high hydrogenation activity for 1,3-cyclohexadiene and acrylonitrile.®® This polymer (Fig. 9)' prevents the formation of a chelating complex and results in unsaturation of the metal which leads to a higher activity.

i f !

Figure 9; Rigid Rhodium Polyyne Support 25

Further reports of increased rate of hydrogenation due to lack of dimeric metal complex formation came from Bonds and Chandrasekaran,*® and Grubbs and Brubaker** who studied a polystyrene complex of titanocene dichloride.

The catalytic hydrogenation of olefins was 15-120 times greater for the polymer complex than for the monomeric titanocene dichloride.

Selective hydrogenation of olefins were reported by Pittman et al.^* and by Nakamura et al.**""^ with rhodium and platinum on modified poly­

styrene polymers. The polymers are illustrated in Figure 10. These poly­ mers were shown to catalyze the selective hydrogenation of octadiene to give cyclooctene. If the palladium complex in Figure 10 was replaced with a rhodium derivative, a high selectivity for branched olefins was ob­

served.*®

BurSten and Bergbreiter** have developed the use of ligand absorbing

polymers to increase the rates of alkene hydrogenation. Macroreticular

silver(I)-exchanged sulfonated polystyrene was found to selectively remove nonvolatile phosphines from solutions where chlorotris(triphenylphosphoniinn)-

rhodium(I) was us^d to catalyze hydrogenations of alkenes. The polymer was

believed to enhance the rates of hydrogenation by increasing the concen­

tration of the chlorobis(triphenylphosphonium)rhodium(I) intermediate.

Hydroformylation is the reaction of olefins, carbon monoxide, and

hydrogen to produce saturated aldehydes or alcohols. Most of the hydro­

genation catalysts previously mentioned are also effective as hydroformyl­

ation catalysts.® ChlorodicarbonyIrhodium(I) dimer bound to a crosslinked

polystyrene polymer containing bis(triphenylphosphine) or dimethylamine

(Fig. 11) catalyze the hydroformylation of 1-hexane. Interestingly, the

polymeric phosphine rhodium complex shows selectivity for aldehyde synthesis 26

or Rh / ^

figure 10; Rhodium and Palladium on Modified Polystyrene Supports Show Selective Hydrogenation Catalysis. 9

Figure 11: Polymer Bound Rhodium Catalyst is Used in Hydroformylation Reactions. 27 while the polymeric amine Rh complex shows selectivity for the alcohol

synthesis and Rh(C0 )2 Cl bound to phosphinated silica gel coated with polystyrene was a stable heterogeneous catalyst for the hydroformylation

of ethylene. Selectivity in a polymer has been correlated to the ratio

of ligand to metal in the complexes studied. When a high ligand to metal ratio exists the distribution of the products usually has a higher normal

to iso ratio.®*

More recently, Pittman has studied the polystyrene bound hydrido-

carbonyltris(triphenylphosphine)rhodium(I) as a catalyst in the hydro­ formylation of methylmethyacrylate and allyl alcohol.*®"*® The resins were prepared in order to examine the normal/branched (N/B) product

selectivity accompanying variations in phosphlne/rhodium (P/R) loading of the resins. In methylmethacrylate hydroformylation, the bound cata­

lysts were less active but more selective for branched product formation

than their homogeneous counterparts at comparable temperatures, pressures,

and P/R ratios. The phosphine loading; i.e., the percentage of the poly­ mer's phenyl rings which have been substituted with diphenylphosphine

groups, was also found to be an important factor. At similar P/R ratios,

catalysts with higher phosphine loading gave more branched products. In

the hydroformylation of allyl alcohol, the normal/branched ratio was

relatively insensitive to P/R ratio changes as well as pressure for both

homogeneous and bound catalyst; increases in temperature favored normal

aldehyde production. Anchoring the complex favored hydroformylation over

hydrogenation at high Rg/CO ratios.

Selected Oxidation Reactions Using Supported Polymers.— Over several

years Pecht and Levitzki*^”*® have investigated the properties of a water 28 soluble polymer, poly-L-histidine copper(II), as a model for the copper containing enzymes. The copper-catalyzed oxidation by molecular oxygen of ascorbate and hydroquinone as well as tetramethyl-p-phenylenediammonium and l-(2,5-dihydroxyphenyl)isopropyl ammonium was studied in the presence of poly(histidine) (Fig. 12). In certain oxidation reactions by di-

Og/S ^ C u llll COMPLEXv y 1/2 H g Q HYDROQUINONE

Og/ \C U III complex)^ ^1/2 Q QUINONE

Figure 12: Cycle Proposed for Copper(II) Catalyzed Âutoxidatlon of Hydroquinone

oxygen, the copper(II)poly-L-histidine complex was found to exhibit a higher catalytic activity than hydrated cupric ions, vdiile imidazole

and histidine coordination complexes showed no catalytic effect at all.

The poly(histidine) enhances the catalytic efficiency of copper(II)

toward negatively charged and neutral substrates, but inhibits it toward

positively charged substrates. Hydrogen peroxide was found to be formed

stoichiometrically in the reaction. The mechanism proposed was:

Cu^"^ + HzA "*• CuAH*" + h ”^

k 3 CuAH^ Cu'*’ + AH 29

+ . _ fast 2 + Cu + O 2 Cu" ' + O2

Cu'*' + O 2 " + 2H^ Cu^"^ + H 2 O 2

Vengerova’® showed that the copper complex of polyvinylpyridine also

catalyzed the oxidation of ascorbic acid with the polymer complex showing

200-150 times the effectiveness of free cupric ions. Sato’^'"®“ showed that the poly(1-vinylimidazole) complex of copper showed a catalytic activity twenty times higher than the monomeric analog in oxidation of hydroquinone. It was suggested that the polymer ligand accelerates the

reoxidation of Cu(I) to Cu(ll) by oxygen.

• « • • •

« •

Figure 13: Polymeric Schiff-Base Compounds

Kurusu, Storch, and Manecke” report the autoxidation of cumene

using polymeric Schiff base compounds (Fig. 13). The activity decreased

in the order Mn>CO>Ni>Cu>Zn and also decreased with increasing molecular

weight of the Schiff base. Most likely because of steric reasons, the

polymer complex showed lower activity than the monomer complexes. 30

Polymer complexes which contain more than one type of reactive center

at the active site may be very active. Rollman^'* synthesized polymer

bound metalloporphyrins which contained an oxidation promoter (Co(II)) and

a proton acceptor (amine), these polymers were effective catalysts for

the oxidation of thiols to disulfides (Fig. 14).

Drago and Gaul®®’®* have developed a synthetic procedure for attaching multidentate chelating amines to polystyrene (Fig. 15). These amines will react with aldehydes and ketones to form Schiff bases. The polymer

bound bis(salicylal)ethylenediimine cobalt(II) was investigated as an

oxidation catalyst for the oxidation of 2,6-dimethylphenol to 2 ,6-dimethyl- benzoquinone and a dimer species (Equation 1). In studying both gel and macroreticular resins, it was found that the rate of 0% absorption and

consequently the catalytic activity of the bound catalyst was reduced over

that observed for the homogeneous catalysis by the parent complex under

comparable conditions. At high concentrations of catalyst, regardless of

substrate concentration, the homogeneous system favors both high conversion

of substrate (-90-100%) and a high selectivity for production of 2,6-

dimethylbenzoquinone; whereas at lower concentration of catalyst, regard­

less of phenol concentration, catalysis resulted in lower conversion and

a greater production of the dimer coupled product. The bis(salicylal)-

ethylenediiminecobalt(II) bound to the macroreticular resin showed the

same general trends as those observed for the solution catalysis. The

polymeric compounds showed that doubling the resin loading of Co(II)

doubled the substrate conversion but decreased the selectibity for benzo-

quinone by a factor of 25. 31

Figure 14: Rollman’s Polymer Bound Metalloporphyrln Which Contains Both an Oxidation Promoter and Proton Acceptor. 32

HNlCHgCHgCNIg

^-^-CHgNlCHgCHgCNIg

BHgTHF

0^KCH2N|(CH2)3NH2l2

( C H g ^ — M 0 - Q - C H , U Co

*H 2W O

Figure 15; Drago’s Method of Polymer Attachment of Bis(salicylal)- ethylenedilmmine cobalt(II). 33

OH

Equation 1; Oxidation of 2,6-dimethylphenol leads to the Formation of Benzoquinone (substituted) and the Diphenoquinone Dimer (substituted).

Polymer metal complexes initiate radical polymerization and often

show higher activity than the monomeric compound. In 1959 Hay®^ showed that 2,5-dimethylphenol can be oxidatively polymerized by copper in the presence of oxygen to give poly(2,6-dimethylphenyleneoxide). The copper

catalyzed reactions are commonly described as free radical processes, although some researchers feel that organocopper intermediates are in­ volved in the mechanistic details. Tsuchida, Kaneko and Nishide*® studied the polymerization mechanism of phenols with polymer mounted catalysts in order to clarify the effect of the polymer complex since the polymerization mechanism was well established in solution. Other organic compounds containing mobile hydrogens such as aromatic amines or acetylenes, also undergo oxidative polymerization (Equations 2, 3, 4). 34

) 4-QM=N4%^ + 2 n H ^ 3

H C 5 C - ^ ^ î CH ----- ^ ^c5C-^-CïC^

The copper(I)-catalyzed oxidations of phenols show considerable ortho-para specificity. When the ortho positions are blocked by alkyl or halo substituents, reaction occurs at the para position. The oxidation of 2,6-dimethylphenol produces a para-phenyleneoxide polymer by coupling an oxygen of one phenol molecule to the para carbon of another. This polymer is a high melting plastic that is very resistant to heat and water.

The course of oxidation is sensitive to the aminezcopper ratio in the catalyst. High ratios of amine to copper produce the polyphenylene oxide polymer. However, at low ratios the major product is a benzoquinone diner.

The basis for the divergence in the reaction pathway lies in the intereaction of the phenol with the copper complex.**

The reaction of copper(l) chloride with oxygen in pyridine gives a polymeric copper(II) oxide, [Cu(Py)^0]^. The mechanism for the oxidative 35

polymerlzation by copper(II) amine complexes involves phenoxy anion

coordination to the copper(II) complex, followed by transfer of one

electron from the coordinated phenoxy anion to the cupric ion (Equation

5):

^-0-Cu"ci ^ Cu'ci ^.^^O-Cu'ci s

It seems likely that the 0-Cu association inhibits 0-C interactions between two radicals, but leaves a clear pathway for the C-C coupling of the para-quinoid form to give the observed dimer.In the presence of

excess pyridine the head-to-tail polymer forms. This reaction is most

easily explained by assuming that pyridine coordinates to copper and

displaces the phenolate ligand. The phenolate ligand has radical proper­

ties. The simplest chain growth mechanism is the coupling of a monomeric

phenolate radical with a similar radical generated from a polymeric

chain (Equation 6). Addition of the polymeric radical to the para position

of a monomeric radical forms a coupling product with a cyclohexadienone

end group. Tautomerization of the end group to the phenol structure gives 36 the enlarged polymer. The phenol end group can react with copper(II) again to form another polymeric radical that can undergo chain growth by a similar mechanism.The features of the overall mechanism are illustrated in Figure 16. The cuprous complex formed is oxidized by

oxygen to the original cupric complex which cycles again as the catalyst.

From the limiting value of O 2 consumption, it appears that one mole

of oxygen oxidizes four moles of cuprous complex. The role of oxygen is

to oxidize the copper complex which was reduced while acting as an oxi­

dizing agent; i.e., to recycle the catalyst, Tsuchida’ “ believes

that a cupric-dioxygen complex is involved earlier in the reaction, probably

at the electron transfer step. He concludes that oxygen promotes the

electron transfer step from the coordinated monomer to the metal ion, or

the dissociation reaction of the activated monomer, probably by forming

an oxygen metal complex. The reaction only works in the presence of oxygen.

The rate of the polymer complex catalyzed polymerization is much

higher than that of the monomer catalyzed polymerization. A somewhat dif­

ferent mechanism is suggested for the quartemized poly (vinylpyr id in e) -

copper(II) catalyst, since the rate was especially high. The polymer

complex has a higher value (Fig. 16) than the monomer complex. The

kg value increases with an increase in the percentage quart em i z at ion in

the polymer, and is reduced by the addition of neutral salt, idiile k^

decreases with the increase of quartemization. A separate experiment

showed that the increase in percentage of quartemization increases the

viscosity of the QPVP-Cu complex from 6,23 dl/g to 0.84 dl/g upon complex

formation with copper(II) ion. This is interpreted to result from con­

traction of the polymer chain into a compact formation due to intermolecular 37

L = amine ligand: aliphatic or pyridine

X = anion: chloro or hydroxyl

> ( •

Phenoxy Radical Coupling

/ ^ x /

Electron Transfer

'"P XX ' ^ ,X\X I ' dissociation

Radical Coupling X Recycle of the Catalyst

Figure 16; Mechanism Proposed for Oxidative Polymerization of XOH by Copper(II) Metal Ions. ' 38 coordination. Oxidation and reduction of the copper ion can lead to an expansion and contraction of the polymer chain.°

The activation energy of an electron transfer reaction is due

in part to the energy required to rearrange the structure of thé complex.*®

For electron transfer to take place between coordinated substrate and the cupric ion, the bond lengths between the metal ion and the ligand must lengthen (Franck Condon restrictions). The polymer ligands contain great amounts of strain caused by electrostatic repulsions by cations on the polymer chain and steric bulkiness of the polymer ligand.' which brings about easy lengthening of the bonds between copper and the ligands. The dynamic fluctuations of the polymer complex are an important aspect of

the catalysis due to the polymer.®

The chemical environment around the metal complex also affects the

catalysis. The polymerization rate varies as much as 100 times depending

on the composition of the mixed solvent. Benzene has the highest effect,

presumably because the hydrophobicity which favors the electron transfer

reaction. The composition of hydrophobic and polar groups in the polymer

backbone can give a similar arrangement in conformation around the

central metal and thereby accelerate electron transfer. The catalytic

activity of the copper(II) complex of the copolymer of styrene and 4-

vinylpyridine was found to be higher than that of other ligand polymers,

and the activity further increased with an increase in the styrene content

of the polymeric ligand.®

The decomposition of hydrogen peroxide by metal complexes has been

studied in order to model the enzyme catalase. Polymer metal complexes

have shown higher catalytic decomposition than the analogous monomeric 39 compounds.*®**® The high activity of the polymeric compounds was attributed to weaker coordinate bonds between the polymeric ligand groups and the metal ions. A variety of polymer complexes have been studied and increasing the molecular weight of the polymer gave higher

H 2 O 2 decomposition activity.

The complex of poly(vinylpyridine) with bls(dimethylglyoximato)- cobalt(II) catalyzed the decomposition of H 2 O 2 .** The activity increases with a decrease in the strength of the coordinate bond between cobalt and pyridine and also increases with a decrease in the basicity of the coordinated nitrogen of the pyridine ring (cf pKa of poly(vinyIpyridine) =

5.03 and pyridine = 3.44). The following order of activity was found:

CoCl(DH)2 'PVP>CoCL,DHo2 'Py>CoCN,DHo 2 PVP>CoCN,DHo 2 'Py (Fig. 17). The catalytic activity also depended on the axial ligand coordinated in the fifth position, the ions studied were cyanide and chloride.

Jones and Saggett postulated that hydrogen peroxide is activated prior to decomposition by the enzyme catalase through participation of one carboxyl group and one amino group (acidic and basic sites) of the apo- enzyme.*® These groups can hydrogen bond with one molecule of peroxide thereby activating it; a second IÎ2 0 2 can then associate with the first to form a complex and to cause fission of the 0-0 bond. The mechanism of Jones is shown in Figure 18. In order to model this mechanism,

Pshezhetski*® et. al. prepared a complex of iron(III) with poly(acrylic acid) which was partially amidated by diethylenetriamine (dien). Kinetic studies showed the dien group to be a cofactor and to activate E 2 O:. The copolymer was prepared and the catalytic complex was made by introducing the polymer to a hydrogen perchlorate solution at pH 2.15 and an iron(III) 40

(ÇH-CHgIn

X = CI , C N t^cA A ch, 6 h 6

Figure 17; Complex of Poly(yinylpyrldine) with Bis(dimethylglyoximato) Cobalt(II) 41

+ P I 1 ” \ - r H + r r > 'h* / F e + 3 H i 1 3 ^ f| it F e t • +3 H H

O T + A " H ^ O _ F e + a i A

0—0

Figure 18; Model Proposed by Jones for Concerted Acid-Base Mediation of Catalase Enzyme. 42

ÇOaH -CH2—ÇH-CH2—CH- CONH-(CHa) a-N H -(C H a) zNHa

complex at the same pH. The acid base behavior of the polymer suggested a spatial interaction of the functional groups in the polymer, simulating a protein. The mechanism he proposes is listed below (Fig. 19):

■ ’0“ 0 - — - - i o : -

) 4 __

- - f — rO 0

Figure 19: Pshezhetski's Mechanism for Polymer Bound Iron with Catalase Activity.

Substrate in the form of HOa ion reacts with active catalyst, i.e., the metal-polymer complex, E , to form an intermediate complex, EHOa , which

interacts with the second molecule of HaOa to form the products. Other

iron(III) complexes supported on polymers containing nitrogen donor sites 43

catalyze the decomposition of EgOg. It was suggested that when one or

two molecules of HaOa are coordinated, a distorted structure'of the metal catalyst would result in facile coordination of HaOa to the central metal and bring about high activity.®^

Complexes Of Lacunar Ligands

The research undertaken by Busch and associates involves the extensive

investigation of totally synthetic heme protein models which show revers­

ible binding of dioxygen.These systems are derived from a neutral

precursor macrocycle, structure (I), which was first reported by Jaeger.®*

The reaction scheme shown in Figure 20, for the 16 membered ring, shows

that the periphery of the macrocyclic ligand is modified to develop the

ligand superstructure. The term lacunar is used to describe this type

of complex because of the unusual cavity formed by the ligand. The

rigidity imparted by the protecting unit in the macrocycle is pronounced,

structure (II), leading to a small permanent void®®"** in the vicinity of

the coordination site.

A unique feature of this model system lies in the flexibility it

exhibits toward strucutral variation. Although most of the work has

focused on the 16 membered ring, the size of the chelating macrocyclic

ring has been varied by changing X and Y, as has the nature and size of

the second ring (Ri) which determines the volume of the protected void.

Further, variations in Rg, the substituent on a nitrogen atom of the

bridge, and in Rs, the blocking group guarding the entrance to the cavity,

have also been reported.**”*^ This tremendous flexibility has proven 44 OCH,

I

Figure 20; Synthesis of Lacunar Complexes Beginning with Neutral Nickel(II) Macrocycle, 45 valuable as attempts have been made to fine tune the system both

sterically and electronically to mimic those properties observed in the natural systems. Electrochemical studies on the nickel complex

show that changing Ri does not change the potential at the metal ion, while changes in Rg and Rs exert substantial electronic effects on the metal ion. Because tuning of the oxidation potential of the metal is believed to be necessary to mimic the natural myoglobin system, the potential for ligand variations in the macrobicyclic strucutre offers great promise.

Work aimed at the attainment of myoglobin's behavior in a totally

synthetic model is a continuing activity in the Busch research group.

In order to be an effective oxygen carrier, the equilibrium reaction of the metal with oxygen must be reversible under physiological conditions.

However, even the natural systems slowly oxidize to oxygen insensitive

forms, met-myoglobin and met-hemoglobin, but therein lies the advantage

of nature, since reductase enzymes function to reduce iron(III) back to

the iron(II) state, where the iron can react with dioxygen once again to form the oxygen complex.* The achievement of this goal had been elusive due mainly to the instability of dioxygen adducts at higher

temperatures but our research efforts have prevailed. Looking toward

the features present in myoglobin, it is generally agreed that the protein unit, and its resultant hydrophobicity, the oxygen binding heme unit, and

the proximal and distal imidazole units probably exert the major influ­

ences with respect to the oxygen binding function of myoglobin. The

functions provided by these features appear to constitute the minimum

requirements for any good model of the myoglobin system* (Figure 21). 46

h i s t i d in e

h e m e u n it

GIOBIN or PROTEIN

Figure 21; Model of Myoglobin. The Essential Features Believed Necessary for Oxygen Binding Function are Highlighted. 47

In the lacunar complexes the void Is protected sterically, and most studies have been limited to the study of complexes with excess

base in the solvent system. The base protects the metal site trans to

the lacunar area; the sixth site of the metal. The site present in

the "pocket" is available for oxygen binding or of other small molecules

such as SCN or CO. Zimmer,’® Herron,"^ and Cameron’® have worked

extensively with the iron(II) lacunar complexes. Although autoxidation

continues to be an extreme problem for iron compounds, these compounds

have recently shown reversible dioxygen binding at temperatures of 20“C

and above, in a variety of solvents. The most promising iron complex

Figure 22; The Most Promising Iron(II) Dioxygen Carrier. Ri = meta-xylene, Ra = benzyl, Rs = phenyl. 48 is illustrated in Figure 22; it contains large bulky substituents in the Rj, Rg and Ra position (Ri = meta-xylyl, Rg = benzyl, and R, = phenyl). The desired stability toward autoxidation has been achieved; however, these most durable iron complexes have Og affinities some^diat lower than those of the high affinity heme proteins.

The cobalt complexes of the macrobicycle have been extensively studied by Stevens,^® Kojima,®* Jackson,®* Evans,®® and Busch. Their observations may be summarized as follows:

1 ) as bridge size increases, the oxygenation affinity increases

up to the limit tdien no protection is afforded i.e., Ri is

(methyl)g, is present and the lacunar void is no longer

sterically restricted by the bridge; (the total range of Og

affinities extends over at least five orders of magnitude);

2) substituents at Rg and Rg exert electronic effects that

can change Og affinity by a factor of 100 for a given bridge

size;

3) smaller rings (15 member vs. 16 member) enhance Og affinity;

4) the ligands Impart abnormally large Og affinities to metal

ions ( 1 0 0 times greater than porphyrins);

5) the Og adducts are measureably protected from autoxidation

even in water at room temperature.

These variations have opened large research areas because the potential uses for these compounds include among others, cytochrome models, oxi­ dation catalysts and oxygen transport and scavenging systems.' 49

Recent unpublished work on the cobalt system by Evans'^ has verified the involvement of axial base in the autoxidation reaction of the macrobicycle system, with the decomposition rate law containing a base dependent term. While the actual mode of base promoted decom­ position has not been determined, the 1 0 , 0 0 0 to 1 0 0 , 0 0 0 fold excess amounts of free base which Stevens found to be necessary to saturate the axial base equilibrium is probably deleterious to the autoxidative stability of the cobalt complexes. Tweedy^' has incorporated an appended base into the totally synthetic macrobicyclic lacunar structure.

He observed a five coordinate cobalt compound showing an appended pyridine coordinated to the metal center. The development of this type of compound eliminates the necessity for larger excesses of external base.

In addition to the study of the reversible binding of dioxygen,

Evans,*® Herron*'* and Busch have recently explored the capability of the cobalt(II) and iron(II) lacunar complexes to promote the autoxidation of organic substrates. The lacunar complexes complexes were found to catalyze the conversion of alkylated phenols to alkylated quinones.

Evans has found that cobalt lacunar complexes catalyze the formation of the single product, quinone; while Herron has found that the iron lacunar complexes yield both quinone and a dimer coupling product. The cobalt complex gives maximum yields near ambient temperature, suggesting the possibility of the development of a low energy process. Oxidation of the phenols was rapid in all cases; however, the major problem of these systems is competing catalyst decomposition, and the number of turnovers of catalyst with respect to product is generally modest at best. 50

Statement Of The Problem

In order to seek the advantages attributed to polymer supported complexes, for example, as catalysts. Investigations were directed toward covalent attachment of lacunar complexes to a polymeric support.

After successful incorporation into the polymer was established, the

absorption of dioxygen by cobalt(II) lacunar supports was examined to evaluate some of the parameters that affect oxygen binding in polymers.

In particular, the question of oxygen adduct stability was addressed.

We sought to discover different methods of utilizing cobalt lacunar

complexes on supported polymers; these areas have included oxidation

catalysis, and as a support in gas chromatography in conjunction with

Dr. Robert Siever's group at the University of Colorado in Boulder.

This broadly defined project implied the need to synthesize model

compounds to establish the chemistry necessary to combine the lacunar

complexes with the polymer. It was decided to attach the complex to

preformed polymeric supports, rather than redesign synthetic procedures

to include the polymer early in the ligand synthesis. This method

establishes a distinction between this thesis research and other studies using supported complexes. The covalent linkage to the polymer in the

lacunar complexes is through nucleophilic displacement of a halide ion

in the polymer by a nitrogen atom found in the ligand superstructure.

Normally in the supported iron porphyrins, covalent attachment of the

complex is initiated by copolymerization of a divinyl group,*®

with substituent groups in the porphyrin. In the supported complexes

of organometallic compounds, a functional group is synthesized into the 51

polymer which is common to the organometallic compound; attachment normally occurs via substitution for the ligand on the metal.*’® For

the lacunar complexes, the transition metal ion was incorporated into

the macrocyclic structure before reaction with the polymer was attempted.

The polymeric support used most extensively in this research was

chloromethylated polystyrene. Dow Chemical Company generously supplied

us with all the 6 % crosslinked macroreticular resin used in this

research. Commercially available chloromethylated polystyrene was also

used to obtain a variation on the amount cf crosslinking and in bead

size. Some of the properties of the complexes bound to a macroreticular

poly(vinylpyridine) in an absorption study were elucidated to compare

the behavior of cobalt lacunar complexes covalently linked to a support

with the behavior exhibited when the complex was absorbed onto the

support by binding to an axial ligand.

The attachment of two types of complexes was emphasized; these were

the lacunar complexes which contain a permanent void, and the unbridged

complexes which are structurally open. We were interested in the

differences between the oxygen binding behavior and the catalytic ability

of these supported compounds with their monomeric solution analogs. One

result of these efforts led to the synthesis of a novel unbridged cobalt(II)

complex which had undergone ligand rearrangement. This cobalt complex has

some unusual properties in its reaction with dioxygen, and some of the

chemistry of this compound was investigated both in solution and on a

polymer. CHAPTER II: EXPERIMENTAL

General Procedures

All reactions and syntheses of nickel(II) complexes were carried out in the open atmosphere (except if the exclusion of water vapor was required, in which case a nitrogen blanket was used). All synthetic procedures involving cobalt(II) complexes were performed in a vacuum atmospheres glove box under an atmosphere of dry nitrogen gas containing less than 5 ppm of oxygen to prevent oxidation of air sensitive materials. Solvents used in the synthesis of cobalt(II) complexes were dried using standard techniques, distilled under nitrogen and degassed under vacuum prior to use.

Reactions of polymeric materials with nickel(II) or copper(II) macrocycles were done in the open in vessels topped with a drying tube. Any reaction of the polymer with cobalt(II) macrocycles involved the preparation of the reaction in the glove box. The reactions were removed from the box using glassware designed to keep a nitrogen atmos­ phere intact and exclude oxygen. Nine mm "Solv-seal" joints and Teflon

stopcocks were used in the preparation of this glassware and were obtained from Kontes glassware. Inc.

Physical Measurements

Elemental analyses were performed by Galbraith Laboratories, Inc., lùioxville, Tennessee. Infrared spectra were obtained using a Perkin

Elmer 283-B Infrared Spectrometer which operated in the region from

4000 cm~^ to 250 cm“ ^. Spectra were obtained as neat samples or in

52 53 the solid state as nujol mulls pressed between potassixm bromide plates or as pressed potassium bromide pellets containing the sample of interest. Visible spectra in the region 800-400 nm were measured using a Carey 17D Recording Spectrophotometer equipped with a variable temerature cell holder for 1 cm quartz cells.

Electron Spin Resonance Spectra (ESR) were recorded using a

Varian Model 102 ESR Spectrometer on samples prepared either as frozen glasses at -196®C using a round quartz sample cell, or at ambient temperatures in a flat EPR aqueous cell with a capillary tip obtained from Wilmad (WG-814). Signal assignments were made relative to a pitch sample of 2.0036 g. Proton Nuclear Magnetic Resonance Spectra were recorded using either a Varian 360-L spectrometer which operated at

60 MHz or a Varian 390-L model which operated to 90 MHz. NMR spectra were recorded on a Brucker V7P-80 Fourier transform instrument which operated at 20.1 MHz. Both broad band proton decoupled and off resonance decoupling were routinely obtained for NMR samples. Chemical shifts in the and NMR spectra were assigned relative to a IMS value of

0 . 0 ppm, usually by direct resonance assignments in the deuterated solvents.

High performance liquid chromatography was performed using a

DuPont Instrument Model 841 unit with a 25 cm x 4.6 mm Zorbax CN normal phase column operating at 1000 psi. Products of phenol oxidations were loaded as ten microliter cyclohexane solutions and eluted with a solution of 2% methanol/98% cyclohexane or with mixtures of isopropanol/ cyclohexane. Detection was at 254 nm using a xenon lamp. Chromatograms 54 were recorded on a Varian Instruments Model A-25 strip chart recorder.

Standards were run to identify unknowns.

Electrochemical analyses were performed by Mr. Madhav Chavan using a Princeton Applied Research Corp. Potentiostat/Galvanostat

Model 173 which was equipped with Model 179 Digital Coulometer and

Model 175 Linear Programmer. All electrochemical studies were conducted in a Vacuum Atmospheres glove box containing oxygen free nitrogen and current versus potential curves were recorded on a Houston Instruments

Model 2000 X-Y Recorder. The working electrode for voltammograms was a stationary platinum disk which could be rotated in order to obtain information concerning half wave potentials (E 1/2) by measuring the potential at one-half the maximum current of the rotating platinum electrode signal. The solvent used for electrochemical studies was acetonitrile containing tetra N-butyl ammonium fluoroborate (TBAB) as supporting electrolyte. Values are reported versus Ag/AgNOa (0.1 M) standard.

Synthesis of Nickel Complexes

(3,11-Diacety 1~4,10-dimethyl-l,5,9,13-tetraaz acyclohexadeca-

1,3,9,11-tetraeneatoN&)nickel(II), [Ni(AczMe2[16]tetraeneatoN<.)].

This compound was synthesized using the procedure previously published by Dennis Riley."*

(2.12-Dimethyl-3,ll-bis[1-methyoxyethylidene]-1,5.9,13-tetraaza- cyclohexadeca-l,4.9,12-tetraeneN<.)nickel(II) Hexafluorophosphate,

fNi{(MeOEthi) aMe:[16]tetraeneNu})(PFeja-

(2,12-Dimethy1-3,11-bis[1-(dimethylamino)ethylidene]-1,5,9,13- tetraazacyclohexadeca-l,4,9,12-tetraeneN<.)nickel(II) Hexafluorophosphate, 55

[Ni{(Me2 NEthl)aMe 2 [16]tetraeneN&}](PFe)a. These two compounds were synthesized using the procedure of Schammel.^^

(2,12-Dimethyl-3,11-bis[l-(methylamino)ethylidene]-1,5,9,13- tetraazacyclohexdeca-1,4.9.12-tetraeneN&)nickel(II) Hexafluorophosphate.

[Ni{ (MeNHEthi) zMeg [163tetraeneNi,}] (PFç) 2 . This methylamine substitution product of the methylated 16 manbered ring Jaeger complex could be prepared according to either the method of Schammel^* or Zimmer.

(2,13-Dimethyl-3.11.bis[1-(piperaz ine)ethylidene]-1.5.9,13- tetraazacyclohexadeca-1,4,9,12-tetraeneN&)nickel(II) Hexafluorophosphate,

[Ni{(piperazineEthi)gMeg116]tetraeneN&}](PFs)a. The piperazine sub­ stitution product of the methylated 16-membered ring Jaeger complex was prepared according to Takeuchi.^“

(2,12-Dimethyl-3,11-bis[1-(N-benzylpiperaz ino)ethylidene]-1,5,9,13- tetraazacyclohexadeca-1,4,9,12-tetraeneN<.)nickel(II) Hexafluorophosphate,

[Ni{ (N-BenzylpiperazineEthPgNe, f 161 tetraeneN^ }] (PFs) 2 . The benzyl piperazine substitution product of methylated 16 membered ring Jaeger complex could be prepared as described by Takeuchi.” An alternative method of preparation was developed.

A solution was prepared containing 2.0 g (2.4 mmole) of [Ni{(piper- azineEthi)jMe2 [16]tetraeneNj,}l(PF« ) 2 dissolved in 250 ml of acetonitrile.

Two equivalents of benzyl chloride (.55 ml, 4.8 mmole) was added to this solution, followed by vigorous stirring with reflux for twelve hours.

A drying tube was placed at the top of the condenser. The orange colored solution became dark orange as the reaction progressed. Triethylamine was added at the end of the reaction and a precipitate formed which 56 redissolved when 10 ml of CHsCN was added. The reaction mixture was placed on a rotary evaporator to reduce the volume of the solution to 4 ml. This solution was chrcmatographed on neutral alumina with acetonitrile as elutant. The first yellow band was collected and the product was crystallized from acetonitrile/ethanol. Yield 1.4 gram

(58%).

Anal. Calcd. NiNgC^oHzePgFig: C(48.16), H(5.66), N(11.23) Found; C(48.06), H(5.73), N(ll.ll)

(2,11,20,26-Tetramethy1-3,10,14.18,21,25-hexaazabicyclo[10.7.7]- hexacosa-1,11,13,18,20,25-hexaeneN&)nickel(II) Hexafluorophosphate,

[Ni{(NH)2 (GHz)«Meg116]tetraeneN4}](FFe)a. This macrobicycle was prepared according to the method of Stevens.^®

(3,lO-Dibenzyl-2,1-,20,2 6-tetramethyl-3,10,14,18,21,25-hexaaza- bicyclo[10.7.7]hexacosa-l,ll,13,18,20,25-hexaeneN»)nickel(II) Hexa- fluorophosphate, [Ni{(NBz):(GHz)sMez[l6]tetraeneN&}](PF6 )2 . A solution containing 2.3 g (3 mmoles) of [Ni{ (NH)2 (CHz) eMez[ 16]tetraeneN,.}] (PFs) 2 was dissolved in 75 ml of dry acetonitrile and deprotonated with sodium ethoxide. Sodium ethoxide was made by dissolving .15 g (over 2 equiv­ alents) of sodium in a small amount of ethanol. The NaOEt was added to the nickel solution whereupon the solution color turned red and the deprotonated nickel solution was placed in a dropping funnel, A solution containing 0.69 ml (6 nmole) of benzyl chloride was dissolved in dry acetone and placed in a dropping funnel. A three arm round bottom flask was set up containing the two addition funnels and a condenser equipped with a nitrogen inlet. The flask bottom held 100 ml of acetonitrile in which 50 mg of sodium iodide had been dissolved. This solution was 57

stirred and brought to reflux under nitrogen before dripping of the reagents was initiated. The addition time was carried out oVer a four hour time period with the benzyl chloride solution dripping at a slightly faster rate than the deprotonated nickel solution. After the addition of the reactants was complete the reaction solution was

stirred for twelve hours with reflux, during which time a \Aiite pre­ cipitate formed. The reaction solution was concentrated on a rotary evaporator and excess triethylamine was added; this precipitated a \diite material which was filtered away. The remaining mixture was

chromatographed on neutral alumina with acetonitrile as elutant. One broad yellow band was collected and reduced in volume to 5 ml. The microcrystalline product could be precipitated by addition of ethanol to the cloud point of the acetonitrile solution. Yield 1.5 g (50%).

Anal. Calcd. NiNgCaBHgzPgFe: C(48.48), H(5.57), N(8.92) Found; 0(48.50), H(5.53), N( 8 .8 l)

(2,3,10,11,20,26-Hexamethy1-3,10,14,18,21,25-hexaazabicyclo[10.7.7]-

hexacosa-1,11,13,18,20,25-hexaeneN&)nickel(II) Hexafluorophosphate,

lNi{(NMe)a(GHz)eMea[16]tetraeneNA}](PFs)2 .

(2,3,9,10,19,25-Hexamethyl-3,9,13,17,20,24-hexaazabicycloI9.7.7]-

pentacosa-1,10,12,17,19,24-hexaeneN&)nickel(II) Hexafluorophosphate,

[Ni{(NMe)2 (CH2 ) sMea[16]tetraeneN»}](PFg)a. These two bridged compounds were prepared by the method of Stevens^* using a peristaltic pump for

dilute addition of reagents.

(2,12-Dimethyl-3,11-bis[1-(benzylmethylamino)ethylidene]-1,5,9,13-

tetraazacyclohexadeca-1,4,9,12-tetraeneN&)nickel(II) Hexafluorophosphate,

[Ni{(BzNMeEthi)zMe2 [16]tetraeneNt.}](PFe)a. A solution containing 2.5 g 58

(3.5 mmole) of [Ni{ (MeNHEthi)2Meg[16]tetraeneN*}] (PFg) 2 in 100 ml acetonitrile was deprotonated with two equivalents of soditm.ethoxide and placed in an addition funnel. This solution was simultaneously dripped with a solution containing .8 ml (7.0 mmole) of benzyl chloride dissolved in 100 ml of acetonitrile into a 500 ml three arm round bottom flask equipped with a condenser and a drying tube.

The flask containing acetonitrile was brought to reflux before dripping of reagents was initiated. After addition of reagents was completed, the refluxing solution was stirred for 20 hours during which time a white precipitate formed which was filtered off. The color of the solution gradually changed from dark red to yellow brown during this time. The solution volume was reduced and excess triethylamine was added. The mixture was chromatographed on neutral alumina with ace­ tonitrile as elutant. A fast moving yellow band was collected and concentrated in volume and the product was crystallized from an ace­ tonitrile/ethanol mixture. Yield 2.0 g (64%).

Anal. Calcd. NiN 6 C 3 ^Hfc6 P 2 F i 2 : C(46.02), H(5.23), N(9.47) Found: C(45.90), H(5.31), N(9.47)

(2,12,14,2O-Tetramethyl-3,11,15,19,22,26-hexaazatricyclo[11.7.7.1^’*] • octacosa-1.5.7.9(28),12,14,19,21,26-novaeneN<.)nickel(II) Hexafluorophos­

phate, lNi{ (NHEthi) 2 (M-xylyl)Me2 [IbJtetraeneNj.}] (PFs) g. This compound was generously donated by Naiomi Hoshino for further work described in this thesis. The monomer fraction was separated from the dimer fraction by Sephadex CM chromatography and eluted with .1 M NazSgO,,. 59

Preparation of Ligand Salts

All ligand salts synthesized for further derivatization.with copper or cobalt metal were prepared with a slight variation in the methods initially established by Schammel^^ and subsequently adapted for use by later members of this research group.All of the nickel(II) hexafluorophosphate complexes were dissolved in acetonitrile and bubbled with a stream of HCl gas for 20-30 minutes. The color of the solution changed from deep yellow to royal blue. Sometimes an

intermediate precipitate was noted, but this material would normally redissolve with further bubbling of the gas. It was not necessary to prepare an intermediate tetrachlorozincate complex before methathesis to the hexafluorophosphate salt. Instead the acetonitrile was stripped off by rotary evaporation, leaving a gummy white residue which was redissolved in 100 ml of water with 10 ml of ethanol. A solution of aranonium hexafluorophosphate in water was dripped in slowly to pre­ cipitate the ligand as the hexafluorophosphate salt. Usually complexes

containing a secondary amine group in the periphery of the macrocycle were precipitated at ice bath temperatures. In all cases, this method of preparation of the ligand resulted in comparable yield as when the

intermediate zincate complex was formed. The advantage of this method

includes the loss of one step, but it usually leads to higher incorporation

of the cobalt metal in the next step.

List of ligands prepared in this way:

Hs(NH) 2 (CHa)6 Mea[l 6 ]tetraeneNfc

Hs(NH)a(m-xylyl)Mea[16]tetraeneN&

Hs (NHMe) aMea [ 16 ] tetraeneUi, 60

Hs (NMe) 2 (CHz) «Meg [16]tetraeneNi,

Hs (NMeg) gMeg [l6 ]tetraeneN<,

Hs(piperazine)gMez[16]tetraeneN »

Hs(NBz)a(CHg)sMeg[16]tetraeneN&

Hs(NMe)a(CHg)sMeg[16]t etraeneN4

Synthesis of Cobalt Complexes

ICo{(NHEthi)g(CHg)Meg[16]tetraeneN^}](PFs)g

[Co{(NMeEthi)a(CHg)«Meg[16]tetraeneH4>](PF*)a

[Co{(NMegEthi)a Meg[16]tetraeneN*}](PFe)g

[Co{(NMegEthi)g(CHg)sMeg[16]tetraeneN*>](PFs):

[Co{ (NHEthi) a (M-xylyl)Meg [16]tetraeneN<,}] (PF*) g

These cobalt(II) compounds were prepared by the method of Stevens and Wilkins®’ in a Vacuum Atmospheres inert atmosphere enclosure with an atmosphere of pure nitrogem. A slurry of the ligand salt in methanol was brought to a reflux. Two equivalents of cobalt acetate and sodium acetate in refluxing methanol was added at once to the ligand slurry.

An instantaneous orange color was obtained for the solution. The reaction was heated gently with stirring for an additional hour. Upon cooling, a crystalline cobalt(II) compound was obtained. Preparations have been scaled up to 1 0 grams of ligand complex, and the procedure was found to work as well as on preparations on a smaller scale. Usually a 90% yield was obtained.

[(3-l-(methylaime)ethylldane)-ll-(l-methylimine)ethyl)-2,12- dimethyl-1,5,9,13-tetraazacyclohexadeca-1,3,9,ll-tetraeneNs]cobalt(II)

Hexafluorophosphate, [Co{(MeNIminoethyl)(NHMeEthi)Meg[IbltetraeneNg)](PFg)g■ 61

In a typical reaction 4.75 g of the ligand (HallSlMeaCNHMe)a

tetraeneN^jCKPFe): which was prepared by direct reaction with HCl

gas, followed by metathesis to the hexafluorophosphate salt, was

slurried in 2 0 ml of refluxing methanol in an inert atmosphere glove

box. Two equivalents of cobalt acetate (2.49 g) and sodium acetate

(.82 g) were dissolved in 2 0 ml refluxing methanol and added to the

ligand slurry. The color of the solution was initially orange but

quickly became dark green. All of the ligand salt material dissolved

in solution as stirring was continued for 30 minutes. Upon cooling

green crystalline material was obtained. The green solid was filtered

and dried by washing with diethyl ether. The material could be

recrystallized frcm an acetonitrile/ethanol mixture. Yield: 4.0 g

(76%).

Anal. Calcd. CoN«C 2 oH3 <.PaFx2 : Co(8.33), C(33.96), H(4.84), N(11.88) Found: Co(8.19), C(33.94), H(4.83), N(12.00)

If the ligand had been prepared using the tetrachlorozincate inter­ mediate prior to metathesis to the hexafluorophosphate salt, no reaction

took place.

[Co{(MeNIminoethyl)(NHMeEthi)Me2 [16]tetraeneNs}](PFs)3 *CH3 CN

.2 g (.28 mmole) of [Co{(MeNIminoethyl)(Me'NHEthi)Me2[16]tetraeneN5}]-

(PFe) 2 was dissolved in 5 ml of acetonitrile and 20 ml of dry methanol

in an inert atmosphere dry box. One equivalent of eerie ammonium nitrate

(.163 g (.28 mmole)) was dissolved in 10 ml of methanol and dripped

slowly into the stirring cobalt solution through a glass frit. During

the two hour addition process, the color of the cobalt solution turned

frcm green to dark pink. The solution was filtered in the dry box and 62

then removed to the open atmosphere. Ammonium hexaf luorophosphate in methanol was added slowly to the solution. The solution was'allowed to stand overnight and a purple microcrystalline material was obtained.

Yield: .20 g (83%).

Anal Calcd. CoCzzHayKyPaFie: 0(29.5), H(4.18), «(10.97), Co(6.59) Found: 0(29.74), H(4.14), «(11.29), Co(6.24)

Aquo «,«*-(l,l,2,2-tetramethylenne)bis(3-methoxy-salicylideniminato)

cobalt(II). Oo(3-CMe-SALTME«).

This schiff base complex was prepared according to literature

procedures®® by Mr. Lyndell Dickerson and was generously donated for work described in this thesis.

Preparation of the Supported Complexes

Chloromethylated Polystyrene.— Commercially prepared chloromethylated

polystyrene beads were available from Dow and Pierce Chemical. Use of

this polymer required a series of washings to remove any impurities which

remain after the emulsion polymerization process. Hartley^ recommends

the following procedure:

Successive washing in 1 N NaOH (60°C), IN HCl (60“C), IN «aOH (60“C),

1 N HCl (60°C), HaO (25°C), DMF (40°C), 1 N HCl (60“C), HzO (60"C),

CHsOH (20°C), 3:2 (v/v) C H 3 0 H:CH 2 Cla, 1:3 (v/v) CH 3 0 H:CH 2 C l a , 1:9 (v/v)

CH 3 OH:CHgCla, pure CHzClg, followed by drying to constant weight at 100"C

in vacuo.

The resins from Dow are yellow before this procedure and idiite

afterwards. The Pierce polymer appeared to be in a more pure state when

initially received. 63

lodomethylated polystyrene.— In a typical reaction, 0.5 g (1.5 meq) of chloromethylated polystyrene in a mixed solvent of (1 :1 ) (v/v) methylene chloride:acetone vas stirred gently. Five hundred milligrams of sodium iodide was dissolved in 2 0 ml of acetone and added to the flask. The flask was equipped with a condenser and a dry tube. The solution was allowed to stir at 60®C for three days. A ^ i t e precipitate was noticed at the end of 12 hours. Infrared red analysis showed the loss of the stretching frequency at 670 cm"^ and the appearance of a new stretching frequency at 580 cm“* indicating an iodide had replaced the chloride.

I2.10-Dipolystyrene-2,11,20,26-tetramethy1-3,10,14.18,21,25- hexaazabicyclo [10.7.7]hexacosa-1,11,13,18,20,25-hexaeneNt)cobalt(II) hexafluorophosphate, ICo{(PSNEthi)2 (CHa)gMeg[16]tetraeneNz,}](PFe)z.

In a typical reaction, .002 mole (1.53 g) of [Co{(HNEthi)a(CHz)eMej[16]- tetraeneN*}](PFe): was dissolved in 30 ml of aceonitrile in an inert atmosphere glove box. To this solution was added sodium ethoxide solution prepared by dissolving .004 mole (0.09 g) of sodium in 3 ml of dry ethanol. The color of the solution changed from orange to deep red. After stirring the solution for 10 minutes, the solution was added to a 100 ml round bottom flask equipped with a 9 mm Solv-Seal joint (Fig. 23). The flask contained a mini "flea" stir bar and 0.5 gram

(1.5 meq) of chloromethylated polystyrene and 15 ml of acetonitrile.

The condenser was attached to the round bottom flask with a teflon seal

and held in place with plastic- couplings. A top piece of glass was

attached to the top of the condenser through a similar seal. This glass 64

piece could adapt to a vacuum line, if desired and contained a Kontes

stopcock (See Fig. 23) to close off the atmosphere.

The reaction assembly was brought outside the box after the

Kontes valve had been closed. The polymer was stirred gently with

the deprotonated macrobicyclic solution at 60®C for seven days. The

color of the reaction solution slowly changed from bright red to brown

as the reaction progressed and a white precipitate coated the glassware.

At the end of seven days, the reaction apparatus was brought into the

inert atmosphere box. The treated polymer was then placed in a

soxhlet extractor equipped with 9 mm Solv Seal joints (held in place

with teflon seals and plastic couplings) and a Kontes stopcock (see

Fig. 24). The extractor was closed and brought out of the box. A

refluxing acetonitrile solution was used to extract the polymer beads

for four days. Little color change was noticed in the solution at the

end of two days. When the extraction was complete, the soxhlet was

brought back into the box and the treated polymer sample was removed.

The sample was dried under vacuum overnight by pumping in the port

chamber of the box.

6 %— Analysis of three separately prepared batches of polymer

show the cobalt concentration in the sample to maximize at .3 meq/gram

at the end of extended reaction times after the extraction procedures

were completed. Less time shows considerably less incorporation of cobalt

lacunar complex into the polymer. The nitrogen/cobalt ratio obtained

from analysis averages over 6 / . 1 and is in good agreement with the expected

ratio from the parent macrocycle, if acetonitrile is considered as a

coordinating, nitrogen donor solvent. 65

Teflon Stopcock

9 m m Solv Seal Joint Plastic Coupling Teflon "0" King

9 mm Solv Seal Joint Plastic Coupling Teflon "0" Ring

Figure 23; Reaction Vessel for Polymer Reactions with Cobalt(II) Macrocyclic Conq>lexes and Chloromethylated Polystyrene 66

I

Teflon Stopcock

Î

Figure 24; Soxhlet Extractor for Removing Absorbed Cobalt (II) Cooçlexes From Covalently Bound Polymeric Supports 67

1%— Analysis of two batches show the cobalt concentration in the polymer to average .2 meq/gram at the end of seven days.' Less reaction time shows much less incorporation of metal into the polymer samples. The nitrogen/cobalt ratio averages 7/1.

The color of the polymer samples after treatment with the lacunar complex was dark brown. The sample looked deep red when a light was shone on the sample. Aged samples that had been exposed to air look somewhat greenish. In one case when the extractor developed a leak during the extraction procedure, the sample was found to be bright green and exhibited no ESR signal.

(3.11-Dipolys t yr en e-2,12,14,20-tetramethyl-3,11,15,19,22,26- hexaazatricyclolll.7.7.1°’^]octocosa-l,5,9,12.1A,19.21,26-novaeneNa) cobalt (II) hexaf luorophosphate, [Co{ (PSNEthi)g (m-xylyl)Me2 [16]novaeneN^.}]-

(PF6)2.

In the manner described for the typical polymer reactions with the hexamethylene bridged lacunar cobalt(II) complex, .002 moles

(1.56 g) of ICo{(HNEthi) 2 (in-xylyl)Me2 [16]novaeneNi,}] (PFe) 2 was deprotonated with two equivalents of sodium ethoxide in an inert atmosphere. The reaction assembly described above (Fig. 23) was used to conduct the reaction outside the dry box. The polymer sample and the deprotonated macrocycle was stirred at 60®C for seven days. After one week, the reaction vessel was brought inside the box and opened to place the treated polymer sample in a soxhlet extractor (Fig. 24). Acetonitrile was used as the solvent for the extraction. The polymer beads were extracted for four days and brought back into the box. The treated 68 polymer was removed from the extraction thimble and dried by. pumping overnight in the port chamber of the inert atmosphere box.

[(3-l-(Methylpolystyreneamlno)Ethylidiene)-ll-(l-methylimine)- ethyl)-2.1]-dimethyl-1.5.9,13tetraazacyclohexadeca-l,3.9.ll-tetraene­

Ns] cobalt (II)Hexaf luorophosphate, [Co{(PSNMeEthi)(MeNIminoethyl)Mea-

[16]tetraeneNs}](PFe):.

In the manner described for the typical polymer reaction with the hexamethylene bridged lacunar complex, .002 mole (1.4 g) of [Co{(NHMe-

Ethi)(MeNIminoethyl)Me2 [l6]tetraeneNs}](PF6 ) 2 was dissolved in 30 ml of acetonitrile in an inert atmosphere box. Slightly over one equivalent of sodium ethoxide was used to deprotonate the macrocycle (.002 mole,

.045 g). The color of the solution turned from green to red. The solution was added to a 100 ml round bottom flask equipped with a

Solv-seal joint. A flea stir bar and 0.5 g (1.5 m/eq) of polymer was added and the reaction vessel was sealed off (Fig. 23) and brought out of the box. The reaction was stirred at 60°C for seven days. The reaction assembly was brought back into the box and the treated polymer was placed in an extraction thimble in a soxhlet extractor (Fig. 24).

Acetonitrile was used to extract any absorbed material from the polymer for four days. When the extraction procedure was complete, the polymer was brought into the box and dried overnight by pumping in the port chamber of the box.

Adsorption of Cobalt(II) Complexes on Polymeric Supports

PORAPAK S — Polyvinylpyridine— Waters Associates, Supelco, Inc.

CFG amino propyl— Controlled Pore Glass— Pierce Chemicals, Inc. 69

Both of these resins are commercially available and used without further treatment. The method of adsorbing the cobalt lacunàr complexes was identical for these two polymers. All handling of these reactions was done in an inert atmosphere glove box.

In a typical experiment, 0.80 g of the polymer resin was stirred with 10 ml of acetone using a mini "flea" stir bar in a 50 ml

Erlenmeyer flask containing a side arm. Normally between 0.4 and 0.6 mmole of cobalt complex (approximately 0.3 and 0.7 g of complex depending

on molecular weight) was dissolved in the minimum amount of acetone and added to the stirring polymer. The solution was gently warmed and stirred for two hours. A vacuum was applied to the flask and the volume of acetone was reduced to near dryness. The polymer sample which was then orange in color (coated with the lacunar complex) was then placed in the port chamber of the glove box and pumped overnight.

An alternate procedure was used initially which allowed the slow

evaporation of the concentrated solutions containing the lacunar

complexes. This was abandoned for two reasons. First it is not good

for the catalyst to absorb large amounts of solvent and necessitates more frequent regeneration. Second, some of the complexes tended to

cocrystallize on the polymer surface during the slow evaporation process.

Complexes absorbed on PORAPAK S:

ICo{ (NMeEthi) % (GHz ) eMe, [16]tetraeneNi,}] (PF«) g

[Co{ (NHEthi) a(CH2 ) «Mes [l6 ]tetraeneNi.}] (PF«) a

[Co{(NMeEthi)a(CHa)5Mea[16]tetraeneNi.}] (PF«)a

[Co{ (NHEthi) a(m-xylyl)Mea [ 16] tetraeneN^ }] (PFe) a 70

tCo{(NMeaEthi) 2M 6 2 [16]tetraeneN*}](PFe)a

ICo(3-CMe-SALTMEN]

See Table XI, Chapter IV for structures of these complexes

Complexes absorbed on CFG amino propyl:

[Co { (NHEthi) a (CHa ) eMea [ 16] t etraeneNi.} ] (PF e) a

[Co{(NHEthi)a(m-xylyl)Mea[16]tetraeneN*}](PFe)a

Cobalt complexes containing a proton (Ra=H) in the bridge tend to cocrystallize on the polymer surface if the solution becomes too

concentrated during the reaction. More acetone is added to redissolve the excess and reconcentrate. Methylated substitutents (Ra=CHs) do not have this problem and absorb easily onto the polymer.

This procedure was general for all of the lacunar-type complexes.

For Co(3-0Me-SALTMEN) the procedure was as described above except the

solvent used for absorption is toluene. Solubility of this complex is a problem and the cobalt complex tends to cocrystallize onto the poly­ mer surface, regardless of the method used for absorption.

Oxygen Exposure to the Polymer Samples.— Since both the unoxygenated

cobalt(II) complexes and the oxygen adducts are ESR active, examination

of the ESR spectrum of a series of polymer supported complexes under

similar conditions can give a measure of the relative magnitude of the

equilibrium constant with that series of complexes. Cobalt concentrations

on the polymers studied were of comparable magnitude. The polymers which

had been treated with the cobalt(II) complexes were transferred to ESR

tubes equipped with ground glass stopcocks to prevent entry of air, and 71

were removed from the box. The ESR spectra of these samples were obtained

frozen at -196°C, and then the samples were warmed to room temperature.

The top of the ESR tube was removed and air was blown through the ESR

tube for 10 minutes with a long steel syringe needle. The top of the

ESR cell was replaced and the samples were again frozen, and the new

spectra recorded at -196°C. The nitrogen stream cannot be too strong

or the polymer sample (dry or wet) will blow out of the ESR tube. The

tubes were rotated and gently shaken to redistribute the treated poly­ mer samples.

Polymer Reactions With Oxygen and Organic Substrates.— Typical oxidation

reactions were run with 2 0 0 mg of polymer which had been treated with the cobalt(II) complex. The typical concentration of cobalt averaged

.3 meq/gram but it could be varied to give less cobalt incorporation.

The substrates used in the oxidation reactions were typically present in

a 30-50 fold excess. The reaction vessel was a large test tube, 16 x

2.5 cm, equipped with a micro stir bar. Normally 20 ml of acetonitrile

or 20 ml 3:3:4 mixture (v:v:v) of acetonitrile/cyclohexane/acetone was

solvent for the covalently attached cobalt complexes. Toluene or benzene

was the solvent for the polymers which contained the cobalt(II) complex

absorbed to the surface. The polymer was mixed with the.organic substrate

and pure oxygen was introduced to the sample by bubbling with a long steel

syringe needle for one hour. The sample was stoppered and allowed to

stir for various time periods. Samples were withdrawn for analysis by

HPLC in 2 ml portions using a syringe. The aliquots were blown to dryness

by passing a stream of dry nitrogen over the samples until only a dry 72 residue remained. The residue vas redissolved in the solvent that would be elutant for the chromatography experiment. Ten pi of the reaction solution was placed on either a normal phase or reverse phase column. Oxygen was allowed to bubble through the samples gently for

one half hour after each sample was taken. CHAPTER III: CRYSTAL STRUCTURE AND PROPERTIES OF A NOVEL COBALT(II) COMPOUND CONTAINING FIVE COORDINATED NITROGENS, [Co{ (MeN TMlNOETHYL)(MeNH ETHYLI)Me2 tl 6 ]-

TETRAENENs}](PF 6 > 2 .

The major portion of the work described in this thesis involves the reactions of various macrocyclic transition metal complexes that produce polymer supported derivatives (Chapter IV). One major goal has been to prepare the cobalt(II) complex of an unbridged macrocycle, analogous to the lacunar ligands, vdiich would react with a polymer support through a covalent linkage. This seemed desirable in light of results obtained by Evans using such unbridged cobalt(II) complexes for the oxidation of phenols. His studies had suggested that unbridged complexes are the most promising catalysts for oxidation studies.

The appropriate starting complex [Ni{(MeNH Ethi)aMe2[16]tetraeneNft}](PF6)z was first synthesized by Schammel,^* and is a well characterized intermediate in the synthesis’ of the more complex lacunar macrobicycles.

The hexafluorophosphate salt of the protonated ligand was obtained as described in the experimental section; it was mixed with a slight excess of cobalt(II) acetate and two equivalents of sodium acetate, following the general method described by Stevens.’® The resulting cobalt(II) complex had a green color rather than the familiar orange

color of the related lacunar complexes. This immediately suggested that a different moiety was present instead of the expected [Co{ (MeNEthi)2 -

M e 2 [IbJtetraeneNiUCPFs) 2 . In addition the ESR spectrum of the

complex in acetone (Fig. 25) suggested a five coordinate structure.

In fact, all the chemical data for the compound suggested that the ligand

superstructure had rearranged. This was unprecedented in the cobalt(II)

73 1/ V

H —>

3zlx) 2 0 0 G

Figure 25: ESR Spectrum of [Co{(MeNIminoethyl)(MeNHEthi) Mea[16]tetraeneN,}](PF«),, Frozen Glass In Acetone, -196 C 3! 75 chemistry of this family of ligands, although the rearrangement of the ligand superstructure was known, and indeed well demonstrated, for iron(II) macrocycles as previously reported by Herron et al.®^

The crystal structure of this novel cobalt(II) compound was determined by Church and Gallucci and this confirmed the rearrangement of one of the peripheral nitrogen-containing functions in the ligand to give a pentadentate ligand within the coordination sphere for the cobalt(II) ion.

The interaction between the preferred coordination number of .a divalent metal ion and the structure and/or mode of chelation of a flexible ligand species is demonstrated when one considers a series of metal complexes with this unusual ligand. The nickel(II) complex is a tetradentate species (Structure III), Iron(II) Induces rearrangement of both peripheral secondary amine groups to give imine donors and a six coordinate compound (Structure IV). The preference for five coordination by low spin cobalt(II) is shown clearly by the spontaneous rearrangement of only one end of the ligand (Fig. 26) to produce a pentadentate ligand.

AT

Structure III Structure IV 76

spontaneous ^ \_J MeOH.heat ^

CH.

Figure 26 ; Internai Rearrangement of Peripheral Secondary Amine Allows the Ligand to Coordinate to Cobalt(II) Through Five Nitrogen Donors 77

Chemical Evidence for Pentadentate Chelation to the Cobalt(II)

Species. — It has been suggested above that the rearrangement of a peripheral secondary amine group of the ligand to produce a penta- coordlnate species could be deduced purely from chemical data. The

ESR spectrum shown In Figure 25 was recorded as a frozen acetone glass at -196"C. This is clearly an axial spectrum of the usual sort for five coordinate cobalt(II) complexes. The superhyperfine splitting of the parallel signal Indicates the presence of a nitrogeneous base coordinated In the axial site. Low spin cobalt (II) ion has a d^ electron system with one unpaired electron, assignable to the dz® orbital of the metal in five coordinate complexes. According to Kida,®® Basolo,®* and

Drago,®° tetradentate ligands produce compounds which have approximate

Civ symmetry in noncoordinating solvents. Their spectra have patterns that consist of an eight line g pattern resulting from the Interaction of the nuclear spin of cobalt (I = 7/2) with the unpaired electron.

However, when nitrogeneous ligands are coordinated axially to the metal,

coupling of the nitrogen nuclear spin (I = 1) to the unpaired electron

of the cobalt occurs. This Interaction Is observed as superhyperflne

(shf) coupling which splits each of the eight cobalt hyperfine signals

Into a triplet (g ) .

The spectrum In Figure 25 shows superhyperflne coupling on several

of the hyperfine signals which indicates coordination of an axial nitrogen

to the cobalt atom. Spectra of [Co{ (MeM Imlnoethyl) (MeNHEthi)Meail6 ]tetra-

eneNa}] obtained directly from, the reaction solution, before recrystall-

Izatlon from an acetonltrlle/ethanol mixture clearly show this structure. 78

The only nitrogen source capable of donating an axial base must come from the chelating ligand.

The infrared spectrum (Fig. 27) shows a sharp N-H stretching absorption at 3395 cm"' due to the uncoordinated amine and also shows several absorptions between 1600 and 1700 cm”' which are typical of the rearranged imine structure.®' Zimmer reports a similar pattern in the infrared spectrum for the CO adduct of an iron(II) complex which also contains a rearranged pentadentate ligand (Structure v).'**“^

r

Structure V 4(j00

-1 CM

Figure 27: Infrared Spectrum of [Cb{(MeNIminoethyl)(MeNEthi)Mea[16]tetraeneNs}](PFa)a » Nujol Mull •*4 80

Electrochemical studies were performed on this compound by

Chavan" using a protected sample, immediately upon removal from an

inert atmosphere glove box (Fig. 28). This work revealed the E 1 / 2

of the Co(II)/Co(III) couple to be -.40 volts in acetonitrile solution

containing tetra-N-butylaramonium fluoroborate (TMB) as supporting

electrolyte, versus Ag/AgNOs (0.1 M) standard. This potential is the most negative observed for cobalt(II) complexes in this ligand family;

in fact, the potential lies near the normal range for the unrearranged

lacunar iron(II) complexes. For comparison, the Ei/g of [Co{(MegNEthi)

Me2[16JtetraeneN*}](PFs ) 2 is -.145 volts under the same conditions.

E 1 / 2 for the pentadentate derivative of cobalt(II) is 250mV more negative

than that for the other unbridged cobalt compound in this ligand family.

This negative potential suggested that the coordination of the fifth

nitrogen placed more electron density on the metal atom, thereby allowing

easier oxidation. This result is supported by the qualitative obser­

vation that solutions of this green complex rapidly turn red when exposed

to air. This will be discussed in more detail later.

The shift to more negative potentials of the half potential values

of the Co(II)/Co(III) couple in pentadentate cobalt complexes was also

observed by Tweedy in his appended tail base complex.^® Basolo and

coworkers have postulated a correlation between more negative values

for El / 2 and higher values. It is rationalized that more electron

density at the metal should result in higher oxygen binding affinities.®^

In contrast, the sexadentate iron(II) complex, with coordinate saturation

of the metal by six Imine groups, contains a highly positive shift of the 81

î * -0.465 volts P

‘p - ^1/2 "

-.40 .07 -.375 .065 Ox -.465 .065

1.22 .08 1.12 .065

=E^y2 measured from rotating

Platinum disk electrode

1) 50 mv/sec 2) 100 mv/eec 3) 150 mv/sec

Figure 28; Cyclic Voltaramograms for [Cb{(MeNIminoethyl) (MeNEth±)Me2[16] tetraeneN,}] (PF*): in Acetonitrile Solvent (0.1 M Fu<,NBFt, Supporting Electrolyte, vs AgNOs <0.1 M), Pt Disk Electrode). 82

Fe(II)/Fe(III) electrode potential, and preserves the iron(II) in an essentially unreactive state. Unlike the cobalt species, the unrearranged tetradentate complex, is a coordinatively unsaturated species having an electrode potential some 0.8 V more negative. The iron complexes of the sexadentate ligands are stable indefinitely in air, whereas the parent iron complex is highly oxygen sensitive.

The pentadentate compelxes of iron, obtained from CO coordination, also exhibit more positive potentials, with values falling between the fully rearranged complex and the unrearranged complex. Consequently the latter complex is somewhat stable in air in the solid state, but is still unstable towards oxygen in solution.

Oxidation of the paramagnetic cobalt(II) compound with eerie ammonium nitrate gave the diamagnetic cobalt(III) derivative, [Co{MeN-

Iminoethyl)(MeNHEthi)Me2 tl6 ]tetraeneNs}](PF6 )s»CH 3 CN, \diich analyzed well. Further characterization by infrared and and ^®C M R spectros­ copy gave strong evidence for the rearrangement. The cobalt(III) complex

showed the N-H stretching absorption at 3395 cm~^, showing that further rearrangement to a six coordinate species did not take place. A coordinated acetonitrile band at 2230 cm~^ shows that solvent is the

sixth ligand for the metal. Two sets of doublets are centered at 6.15 (H^) and 8.15 (H^) ppm in the NMR spectrum. Figure 29. (This indicates that one side of the molecule has rearranged in the manner described for

the sexadentate iron(II) complexes."^ A hydrogen atom has shifted from

an external nitrogen atom to a bridge head carbon atom with rehybrid­

ization and geometrical rearrangement to yield the pentadentate ligand

(Fig. 26). One vinyl proton resonance is retained as a singlet centered Figure 29; ^ NMR Spectrum of [Co{ (MeNImlnoethyl) (MeNHEthDMez [ 16 ] tetraeneNg} ] «CHgCN(PFg ) 3 in C D 3 CN, 25 C 84

at 7.7 (Hg,) ppm. The remaining complexity in the proton spectrum, especially in the methyl region, is due to the absence of a center of

symmetry in the molecule. The NMR spectrum of the pentadentate

complex is shown in Figure 30. The spectrum supports the proposed

structure, with the low symmetry producing twenty unique carbon atoms.

TABLE 2

“ C NMR PEAK SHIFTS FOR [Co{ (MeN Iminoethyl) (MeNEthi)Mea [16]tetraeneNs }] IN CDaCN

Decoupled 183.06, 178.30, 177.47, 171.70, 170.97, 162.91, 109.51, 68.00, 55.18, 52.85, 49.50, 47.61, 40.67, 32.51, 26.68, 25.42, 25.18, 24.16, 21.05, 17.84

80 MHz, 40°C

Cirystal Structure.— The data collection and analysis were done by Chureh-'and Gallucci. Dark reddish brown crystals of the complex

suitable for X-ray diffraction were prepared by slow crystallization

from acetonitrile/methanol solution. A long rectangular plate-like

crystal with dimensions 0.15 x 0.60 x 0.90 mm was mounted on a quartz

fiber with its ^-axis approximately along the spindle direction. The

crystal was coated with several thin layers of epoxy cement and examined

with a syntex ?i four-circle automated diffractometer. The unit cell

dimensions and their estimated standard deviations were determined by

a least-squares fit of the setting angles for 41 high angle reflections

having 17.0° 1 26 1 30.0° using MoK^ graphite monochromatized radiation

(XMoKq = 0.71069A) at 22°C. Systematic absences, hOl, h+l=’2n+l and OkO,

k= 2 n+l uniquely determine the space group as Pa /n ; the centric nature 6

4 M L . i I 11 1

' 180 120 eb 0

Figure 30: N M R Spectrum of [C6 { (MeNIminoetlgrl) (NeNHEthi)Mea[16]tetraeneN,}]«CHaCN(PF*), in CD'aCN, 40°C, Proton Decoupled

00 Ui 86 of the space group was subsequently verified by the E-statistics.

Intensity data were collected by the w-2G scan technique with variable scan rate between 4° and 24°/min. Six check reflections were monitored every 1 0 0 reflections during the course of data collection, with no significant changes being observed. A total of

5563 independent reflections was collected with 4 - 29 i 50°, (+h,+k,

+1). Of these 2922 had I > 3a(I). The data were corrected for Lorentz and polarization effects and a set of structure factors were obtained.

Standard deviations were calculated according to the equation

o'(F') = r*/Lp*[S + G* (Bi + Ba)] + (pi)'

where r is the scan rate, Lp is the Lorentz-polarization correction,

S, Bi and Bg are the scan and background counts, G is the ratio of scan time to total background counting time (2.0). I is the net intensity

(S - G(Bi + Bz)), and p is a factor, here chosen to be 0.02 on the basis of prior experience with this instrument. The pi term is presumed to represent that component of the total error expected to be proportional to the diffracted intensity.®* No correction for absorption was made because of the low linear absorption coefficient calculated as 7.4 cm“^.

Wilson's method was then used to put the F®'s on an absolute scale.

Even with the small linear absorption coefficient calculated, the overall crystal dimensions suggest that an absorption correction should be applied. It was felt that application of a good analytical correction was not justified as the epoxy cement coating of the crystal made it

Impossible to accurately index and measure the crystal. Use-of an empirical method based on psi scans resulted in a test correction which 87

showed that the difference between the minimum and maximum absorption

corrections was insignificant.

Using only the data having I > 3a, the structure was solved by the heavy atom method, and refined by the full-matrix least-squares method. In the least squares procedure, the function minimized was

Iw|Fg - F“1, where w = l/a“ (F®). The scattering factors for H, Co, N,

F, P and C were taken from International Tables for X-ray Crystallo- cov graphy Vol. IV.*® All the calculations were carried out on an Amdahl

470 computer in the Instructional and Research Computing Center of the Ohio State University using the CRTM package.*®

The coordinates of the cobalt atom were obtained from a three dimensional Patterson synthesis. Several cycles of structure factor and Fourier calculations revealed all of the non-hydrogen atoms. Refine­ ment with isotropic thermal parameters was carried out and the discrepancy

indices

Ra = l\iKFol - 1kFc]|/z|kFo1

and

R = Zm*(|KFo|“ - |Fc 1“)V Z u “ |KFo 1'‘

were calculated to be 12.4% and 6.9%, respectively. At this stage,

general planes difference Fourier syntheses were calculated for the methyl

groups and many of the hydrogen atoms were revealed. The positions for

the remainder of the methyl hydrogen atoms were calculated using a least-

squares fit of the positions of the observed atoms. Coordinates for all

non-methyl hydrogens were calculated using the structural parameters of

the connecting atoms. The positions and isotropic thermal parameters of 88 the hydrogen atoms were held constant during refinement. Â full matrix least-squares refinement with anisotropic thermal parameters for all non-hydrogen atoms yielded the final values of 6.0% and 1.3% for Ri and respectively. A final difference Fourier map showed no significant features, the maximum peak being 0.43 e/A^. Final positional parameters with their estimated standard deviations are given in Table 3. Final atomic coordinates and thermal parameters, as well as F^y^, F^^^^ data, are located in the appendix to this thesis.

The ORTEP drawing (Fig. 31) shows the coordination sphere of the cobalt(II) with the macrocycle bound equatorially through four nitrogen atoms and with the axial donor provided by the rearranged peripheral group. The average bond distance of the four coplanar nitrogen atoms

0 is 1.92 A with the axial nitrogen-cobalt distance being considerably longer at 2.14 A. The cobalt atom is found to deviate only 0.05 A from the place defined by the four coplanar nitrogen atoms (Ni,Nz,N&).

An ORTEP drawing (Fig. 32) shows the orientation of the acetonitrile

(from solvation) near the unrearranged peripheral group of the macrocycle.

The acetonitrile is no closer than 2.645 A to the cobalt atom, but it is oriented so that the nitrogen is closest to the cobalt. The presence of the solvent molecule gives a distorted octahedral coordination to the metal. A STEREO packing diagram is shown in Figure 33 and shows all the non-hydrogen atoms of the structure. The saturated carbons in the trimethylene chains have undergone a twist and deviate from the chair-boat conformation found in the parent nickel complex. The five C=N bonds O are localized with bond lengths averaging 1.26 A. Thus there are five electronically isolated imine donors in the coordination sphere. The numbering scheme for the atoms is given in Figure 34. 89

TABLE 3

SUMMARY OF CRYSTALLOGRAPHIC DATA FOR CoCzaNyHsgPaFi:

Space Group Pz^/n

a 18.186(2)

b 7.296(1)

c 25.279(3)

6 110.66(1)

V 3354.1

T 295 K

1.51 g cm“®

1.58 g cm“ ®

Z 4

y(MoKa) 7.4 cm"'

scan technique 10-28

Ri 6.0

1.7 \

GOF 2.2

397 90

Figure 31; ORTEP Diagram of the Pentacoordinate Cobalt(II) Complex Showing the Rearrsmgement of the Peripheral Ligand, 91

Figure 32; ORTEP Diagram of Pentacoordinate Cobalt(II) Showing the Orientation of the Acetonitrile of Solvation Figure 33; Stereo Packing Diagram of [Co{ (MeNImlnoethyl) (MeNHEthi)Me2[16]tetraeneN5}]*CH3CN(PF«)s

VO 93

Figure 34; Numbering Scheme for-the Atoms 94

TABLE 4:

BOND DISTANCES (A) AND ANGLES (B) (DEC) WITH THEIR ESTIMATED STANDARD DEVIATIONS FOR CoCasNyHaiPaFxa

A. Bond Distances

(a) Coordination Sphere

Co - N(l) 1.915(4) Co - N(4) 1.929(4) Co - N(2) 1.947(4) Co - N(6) 2.148(5) Co - N(3) 1.920(4)

(b) Imine N = C Double Bonds

N(l) - C(l) 1.259(7) N(4) - C(2) 1.287(7) N(2) - C(5) 1.258(8) N(6) - C(10) 1.252(8) N(3) - C(4) 1.266(8)

(c) Ligand

C(l) - C(3) 1.422(6) C(5) ■ C(6) 1.508(7) C(3) - C(7) 1.384(8) C(6) - C(10) 1.551(10) C(7) - C(8) 1.505(7) C(10) - C(17) 1.485(10) C(7) - N(5) 1.331(6) C(6) - C(4) 1.506(7) N(5) - C(9) 1.444(8) C(ll) - C(12) 1.522(9) C(3) - C(2) 1.479(6) N(6) • C(20) 1.464(8) C(2) - C(19) 1.494(8) C(12) - C(13) 1.500(10) C(16) - C(15) 1.515(11) C(15) - C(14) 1.516(10) C(5) - C(18) 1.484(8)

(d) PFe ions

P(l) - F (11) 1.541(5) P(2) F(21) 1.560(5) P(l) - F (12) 1.532(6) P(2) 7(22) 1.495(5) P(l) - F(13) 1.529(5) P(2) F(23) 1.545(5) P(l) - F (14) 1.579(7) P(2) F(24) 1.541(6) P(l) - F (15) 1.519(6) P(2) F(25) 1.522(6) P(l) - F (16) 1.545(5) P(2) F(26) 1.556(6)

(e) Acetonitrile of crystallization

N(7) - Co 2.654 95

TABLE 4 (continued)

B. Bond Angles

(a) Coordination Sphere

N(l)-Co-N(2) 176.58(2) N(2)-Co-N(6) 87.04(2) N(l)-Co-N(3) 90.43(1) N(2)-Co-N(4) 91.33(1) K(l)-Co-N(4) 89.75(1) N(3)-Co-N(4) 177.42(2) N(l)-Co-N(6) 96.08(2) N(3)-Co-N(6) 86.72(2) N(2)-Co-N(3) 88.35(1) N(4)-Co-N(6) 95.81(1)

(b) Ligand

C(l)-N(l)-Co 124.17(3) C(3)-C(l)-N(l) 127.01(4) C(13)-N(l)-Co 116.19(3) C(19)-C(2)-N(4) 121.49(4) C(13)-N(l)-C(l) 119.34(4) C(19)-C(2)-C(3) 117.80(5) C(5)-N(2)-Co 121.64(3) (3)-C(2)-N(4) 120.20(4)

C(14)-N(2)-Co 115.27(3) C(2)-C(3)-C(l) 117.70(5) C(14)-N(2)-C(5) 122.82(4) C(2)-C(3)-C(7) 122.71(4) C(4)-N(3)-Co 119.07(3) C(l)-C(3)-C(7) 119.57(4) C(4)-H(3)-C(ll) 120.62(4) C(6)-C(4)-N(3) 119.39(4)

C(2)-N(4)-Co 128.39(3) C(18)-C(5)-N(2) 127.69(5) C(2)-N(4)-C(16) 119.64(4) C(6)-C(5)-C(18) 116.21(5) C(9)-N(5)-C(7) 127.29(4) N(2)-C(5)-C(6) 116.10(5) C(10)-N(6)-Co 117.95(4) C(5)-C(6)-C(4) 110.43(5)

C(20)-N(6)-Co 121.40(4) C(10)-C(6)-C(4) 108.42(4) C(10)-N(6)-C(20) 120.63(5) C(5)-C(6)-C(10) 109.56(4) N(5)-C(7)-C(8) 122.68(4) C(8)-C(7)-C(3) 122.68(4) C(6)-C(10)-C(17) 115.33(5) N(6)-C(10)-C(17) 129.47(7)

N(6)-C(10)-C(6) 115.19(5) N(3)-C(ll)-C(12) 112.29(5) Cfll)-C(12)-C(13) 114.76(4) N(l)-C(13)-C(12) 112.89(5) N(2)-C(14)-C(15) 110.60(6) C(14)-C(15)-C(16) 114.41(4) N(4)-C(16)-C(15) 110.24(5)

(c) PFe Ions

F(12)-P(l)-F(ll) 88.56(3) F(22)-P(2)-F(21) 89.52(3) F(13)-P(l)-F(ll) 179.10(4) F(23)-P(2)-F(21) 86.34(3) F(14)-P(l)-F(ll) 90.79(3) F(24)-P(2)-F(21) 176.87(3) F(15)-P(1)-F(11) 93.16(3) F(25)-P(2)-F(21) 91.55(3)

F(16)-P(1)-F(11) 89.10(3) F(26)-P(2)-F(21) 88.42(3) F(13)-P(l)-F(12) 92.25(2) F(23)-P(2)-F(21) 175.70(3) F(14)-P(l)-F(12) 87.16(2) F(24)-P(2)-F(22) 92.97(3) F(15)-P(l)-F(12) 92.94(3) F(25)-P(2)-F(22) 92.36(4) 96

TABLE 4 (cont inued)

F(16)-P(l)-F(12) 176.66(3) F(26)-P(2)-F(22) 90.70(3) F(14)-P(l)-F(13) 88.32(3) F(24)-P(2)-F(23) 91.13(3) F(15)-P(l)-F(13) 87.72(3) F(25)-P(2)-F(23) 88.91(3) F(16)-P(l)-F(13) 91.05(3) F(26)-P(2)-F(23) 88.02(3)

F(15)-P(l)-F(14) 176.06(3) F(25)-P(2)-F(24) 90.25(3) F(16)-P(l)-F(14) 90.49(3) F(26)-P(2)-F(24) 89.65(3) F(16)-P(l)-F(15) 89.57(3) F(26)-P(2)-F(24) 176.93(3) 97

Reaction of the Cobalt(II) Complex with Oxygen. — The pentacoordinate

cobalt(II) complex is extremely reactive toward oxygen. In acetone the

color of the green solution quickly turns to deep red \rtien the solution

of the complex is exposed to the atmosphere. Exposure to oxygen at

room temperature results in the visible spectral changes shown in Figure

35. Monitoring these changes using ESR spectroscopy shows the loss of

the initial five coordinate cobalt(II) center to give an ESR silent

compound. Proton and NMR spectroscopy (Figs. 36 and 37) on the resulting diamagnetic complex show some similarity to those of the

chemically oxidized cobalt(III) complex (Figs. 29 and 30). Proton

data is tabulated in Table 5; carbon data is tabulated in Table 6,

The NMR spectrum shows two sets of doublets at 5.8 and 7.8 ppm and

a vinyl proton resonance centered at 7.5 ppm, indicating that the

rearranged periphery is intact. The NMR spectrum shows fewer carbon

resonances in the solution that was exposed to oxygen than it did for

the chemically oxidized sample.

The behavior of the pentacoordinate cobalt(II) in an acetone solution

is much different in the visible spectral region at lower temperatures.

Figure 38 shows the unusual behavior of the pentacoordinate cobalt(II)

complex with ultra high purity (UHP) Matheson nitrogen (minimum ^ pur it y

99.999%) at -40.0°C. Absorbances at 720 and 500 nm increase with

successive bubbling of nitrogen in 200 second incranents. ESR experiments

were carried out after each bubbling period and the intensity of the

signal due to the five coordinate species decreased each time. Also,

no new spectral features were observed. Initial Spectrum

After Exposure to Moxygen for 200 sec. (1 atm.) \ \ < s o c/5

I

400 500 600 700 800 NM WAVELENGTH

Figure 35: Spectral Changes for [Co{(MeNImlnoethyl)(MeNHEthl)Mea[16]tetraeneNa}](PFe): Upon Exposure to Dloxygen In Acetone Solution at Room Temperature 99

TABLE 5

NMR PEAK SHIFTS FOR [Co{ (MeN Iminoethyl)- (MeNHethi)Me2 [16] tetraeneNg }] (PF*): AFTER EXPOSURE TO OXYGEN, CDgCN SOLVENT, ROOM TEMPERATURE

*7.8 ppm (d), 7.5 ppm (s), 5.8 ppm (s) 3.6 ppm (br), 3.0 ppm (s), 2.85 (s) 2.4 (s) 2.25 (s)

7.8 ppm (d), 7.5 ppm (s), 5.8 ppm (s) 3.6 ppm (br), 3.0 ppm (s), 2.85 (s) 2.4 ppm (s), 2.25 (s)

*d = doublet s = singlet br - broad

TABLE 6

" C NMR PEAK SHIFTS FOR [Co{(MeNIminoethyl)- (MeNethi)Mea[16]tetraeneNs}](PFs)a AFTER EXPOSURE TO OXYGEN, CDgCN SOLVENT

Decoupled 169.47, 161.41, 109.4, 65.43, 54.60, 51.69, 48.92, 48.24, 45.38, 25.62, 24.21, 23.14, 22.85, 20.19

80 MHz, 40°C I I I « t I ,, I 9 8 6 5 4 3

Figure 36; NMR Spectrum in CDgCN After Exposure of [Co{(MeNImlnoethyl)(MeNHEthl)Meg[16]tetraeneNg}] (PFe)a to Dloxygen at Room Temperature § 60 0

Figure 37; NMR Spectrum in CDgCN decoupled) After Exposure of [C'o{(MeNImlnoethyl)(MeNHEthi)Me2 - tl6]tetraeneNs}](PFsOa to Dioxygen at Room Temperature FINAL

üJ

§ S \ \\ eno

INITIAL

I

ü O O 500 600 700 800 WAVELENGTH Figure 38: Spectral Changes Accompanying Successive Esposure of [Co{(MeNDnlnoethyl)(MeNHEthl)Mea[16]" tetraeneNg}](PFe)a to Nitrogen(Ultra High Purity) in an Acetone Solution, o to 103

Figure 39 shows the reversible behavior of the moiety formed by exposure to UHF nitrogen \dien the atmosphere was cycled between oxygen and UHP nitrogen. Spectrum 1 was obtained after bubbling UHP nitrogen through the solution for 1000 seconds. Spectra 2, 3, and 4 show the effects of increasing the partial pressures of oxygen. The absorbanaces around 720 nm decrease in intensity and the peak at 510 nm, which had been a shoulder, appears as a definite band. Spectra 5, 6, and 7 involve cycling between nitrogen and oxygen atmospheres and back to nitrogen again. The system exhibits good reversibility at -40.0°C. An ESR spectrum taken after exposure to 1 atmosphere O 2 for 200 seconds is shown in Figure 40 (-40.0°C). This spectrum is typical of a one-to-one oxygen adduct of cobalt(II). The spectrum is attributed to the presence of the spin density mainly on oxygen, effectively producing a cobalt(Ill)-superoxide species.®®»*®

A reasonable model for this unusual behavior is as follows. The redox couple for the cobalt(II/III) couple is most negative, suggesting that the cobalt(II) complex should have a high affinity for oxygen.*®

At -40.0°C the cobalt system may be so sensitive that it will react with any small amounts of oxygen which may be present in the nitrogen. This behavior seemed to occur even when attempts are made to eliminate oxygen from the nitrogen. New oxygen scrubber bugles (2) from L.C. Company, Inc. were replaced and the nitrogen was allowed to circulate. When the experiments described above were repeated, identical results were obtained. When the oxygen adduct is formed at very low partial pressures of oxygen, there is an excess of five coordinate cobalt(II), an increase in the partial Initial Spectrum (after Na)

• • 200 sec., 1.9 torr 0,

200 sec., 38 torr 0»

200 sec., 760 torr 0

1000 sec Na (Indicates partial reversibility)

\\

tu (_»

g o c/5 g

500 600 700 800 NM Figure 39: Reversible (partial) Spectral Changes Accompanying the Cycling Between Dloxygen and Dlnltrogen Atmospheres for the Pentacoordinate Cobalt(II) Complex at -40.0®C in Acetone Solution

o 0 105 CoLB ------C 0 LBO 2 7

CoLBOa + CoLB — ...... * CoLBOaCoLB 8

pressure of oxygen cleaves this peroxo bridged dimer to give the monomeric oxygen adduct (Fig. 40) by depleting the concentration of CoLB in accord with the reverse of equation 8, Bubbling nitrogen through the system reverses the monomer-dimer equilibrium to form the peroxo bridged species. Additional evidence to support the formation of a peroxo dimer species is shown in the room temperature ESR spectrum in

Figure 41. This spectrum was obtained from an acetone solution which has been exposed to oxygen with nitrosobenzene present. The spectrum observed is typical of y-superoxo dimers of cobalt(111).®^“®® For this species to be formed, the nitrosobenzene must oxidize the peroxodimer. More commonly used oxidizing agents such as ceric(IV), silver(I), and iodine have not given an analogous signal in the ESR.*^’®®”'^®® With these oxidizing agents either the solution totally decolorizes or a large amount of precipitation of a fluffy white solid occurs (most likely the ligand).

The complex behavior described above has been repeated at -20“C, but decomposition of both the oxygen adduct and the peroxodimer species is more rapid than at the lower temperature. Once the monomeric oxygen adduct is formed at low temperatures, and it can be detected in solutions that have been warmed to room temperature for as long as 14 hours.

The earliest known synthetic dioxygen complex is probably the (NH3 )5 -

Co-0a-Co(NH3)5 dimer first characterized by Werner. The bonding of dioxygen involves two metal centers, with the bound dioxygen being in a 106

100 G

1+ C

Figure 40; ESR Spectrum (Frozen glass) Produced by Exposure of the Pentacoordinate Cobalt(II) Complex to Dioxygen at ^ O^C in an Acetone Solution 107

3 3 8 0 , I 2 0 G 4

H

Figure 41: ESR Spectrum (Room Temperature) Produced by Adding Nltrosobenzene to a Solution of [Cb{(MeNIminoethyl)(MeNHEthi)- MesIlôltetraeneNV}](PF« ) 2 in Acetone Solution Saturated with 0% at 25®C 108 reduced peroxo state by a concomitant one-electron oxidation of both metal centers. The p-superoxo complexes can be easily prepared by the one electron oxidation of the peroxo complex. These paramagnetic complexes give ESR speactra at room temperature \diich are well defined isotropic spectra, consisting of 15 lines.The spectrum shown in Figure 40 is very similar to those reported in the literature for such superoxo-bridged dimers.*^

Although the existence of peroxo dimers is known for a variety of cobalt salicylaldéhyde derivatives, prior to this example there was no direct evidence for dimer formation in this family of ligands vAich are either unbridged or are often used as precursors to the lacunar com­ plexes. Experiments with the other unbridged cobalt compound, [Co{(MezN-

Ethi) 2 M e 2 [1 6 ]tetraeneNft}](PF6 ) 2 in acetonitrile at -40"C were performed by Hoshino.^®^ At lower temperatures, exposure to one atmosphere of oxygen shows the formation of a band at 520 nm i^ich can be attributed to the one-to-one oxygen adduct. This adduct can be totally reversed by bubbling nitrogen through the sample. There are no spectral changes which indicate the presence of a new species in solution as was the case in the cobalt(II) complex with the rearranged ligand superstructure. ESR experiments verify the presence of the monomeric oxygen adduct after ex­ posure to an oxygen atmosphere. More recently, Chavan has demonstrated the presence of a y-superoxo bridged species with a lacunar macrobicycle using electrochemical techniques®” (Ri=(CH2 )e, R 2 =CHs, R 3 =CHs). CHAPTER IV: RESULTS AND DISCUSSION. TRANSITION METAL CCMPLEXES OF POLTMER SUPPORTED MACROCYCLIC LIGANDS

MODEL STUDIES

Since surfaces and surface reactions are difficult to characterize, it was important to establish the feasibility of the proposed surface reactions by analogous solution reactions. The advantages of these model reactions include greater ease of characterization, but more importantly, they constitute minimum criteria for the feasibility of similar chemistry

Involving the polymer. Different kinds of functionalized polymers were used in this study. Chloromethylated polystyrene was chosen in order to attach the macrocyclic ligand by covalent bonds to the polymer support.

Poly(vinylpyridine) was chosen as representative of those polymers having ligating groups that can coordinate to the metal complexes, thereby simultaneously binding the complex to the polymer and providing an axial base, or fifth donor atom for the metal atom. A controlled pore glass support (CPG) with propyl amine appendages was chosen as an example of a rigid inorganic support. The lacunar complexes can be absorbed onto this support and the amine group may function as an axial base for the metal atom. Alternatively, the amine functionality can be further derivatized i to covalently attach complexes to the support. The modeling studies aid in the interpretation of the interaction of these various supports with the lacunar complexes.

Chloromethylated Polystyrene.— This polymer is a macroreticular resin that is available from Dow Chemical and Pierce Chemical company in varied degrees of crosslinking (Pol. I.). In this polymer, a benzyl chloride

109 110 functional group provides a site at which covalent linkages to the lacunar macrocycles may possibly be formed. Nucleophilic sites can be activated on the periphery of the macrocyclic ligands having structure

I, (Fig. 20) through deprotonation of the secondary amine groups (R2 ) that are parts of the bridging unit. These deprotonated nitrogens can act as nucleophiles, displacing the chloride in the polymer to give a covalently bound macrocycle. This linkage will anchor the transition O

Polymer I: Chloromethylated Polystyrene

metal complex and prevent its leaching from the polymer during any subsequent treatment. Therefore any solvent may be used with these systems since the macroreticular resin is insoluble in all solvents. ^

Benzyl chloride was chosen as the reagent to serve as a model for the polymer. As described in the experimental section, the deprotonated macrocycle was treated with benzyl chloride to give the benzylated product.

The reaction is illustrated in Figure 42 for the Ri=(CH 2 )« bridged lacunar macrobicycle. The benzylated product could also be obtained for unbridged complexes which contained a secondary amine group (both the precursor of Ill

vinyl

Nal

deprotonate N a O E t

REFLUX

H p

CH 2 m e th y le n e

Figure 42; Model Reaction of Macrocycles vith Benzyl Chloride 112 the lacunar complex and the piperazine complex, precursor for the vaulted complex). The reaction appears to be general in complexes which contain a nucleophilic nitrogen site for substitution. And, as described in the experimental section, the benzylated products gave the expected analyses.

The product of the reaction has a benzyl group as a substituent

(Ra) on a peripheral nitrogen. In the infrared region, the N-H stretching band of the reactant at 3410 cm”’ is not present in the spectrum of the product while a band has appeared at 699 cm”^ indicating the presence of the benzylic function. The ’h NMR spectrum (Fig. 43) integrates well for the expected benzylated product, with the ratio of the signal intensity due to the vinyl protons on the macrocycle to that for the aromatic protons of the phenyl ring (H^/H^) being (1/5). This data suggests that both peripheral nitrogens have accepted benzyl substituents, and that the plane of symmetry is intact in the derivative.

There is an interesting difference between the methylene region of the benzylic group in the unbridged complex (Fig. 44) and that for the bridged lacunar compound (Fig. 43). For the unbridged complex, the methylene protons of the benzylic group give a sharp singlet signal

(4.6 ppm), whose integrated intensity is twice that of the vinyl proton signal. On the other hand, the methylene protons of the bridged compound appear as an AB quartet centered at 4.9 ppm, which also has an integrated signal intensity twice that of the vinyl proton signal. This suggests that the benzyl group has free rotation in the unbridged compound, while the motion is more hindered in the bridged compound. Coupled and decoupled NMR for both the bridged and the unbridged derivatives are f

h - < ™ 2>6 OOÎ " 2 ■ -” 2 ' ^ - ^

& 8 I A PPM

2 Figure 43: ^ N M R Spectrum of tNi{ (NBzJia (CHj) eMë [16]tetraeneNz.}] (PFs) 2 in CD3 CN, 25®C Ri- CH, h ' -”2C-^3

I I I I I I I 1 I 9 8 7 6 5 4 3 2 0 PPM

Figure 44: ^ NMR Spectrum of [Ni{(MeNBz)2Me2[16]tetraeneNj,}](PFs) 2 in C D 3 CN, 25°C 115 illustrated in Figures 45 and 46 respectively. Assignments for the carbon data are made in Tables 7 and 8. The main features shown by the NMR are the retention of the mirror symmetry in the molecule and the appearance of signals due to the aromatic protons of the phenyl ring between 129 and 136 ppm.

The successful modeling of the reaction scheme proposed for the chloromethylated polystyrene polymer with benzyl chloride does suggest that the chemistry proposed for the polymer will work. One added result of this work is a new synthetic approach to the preparation of lacunar complexes. In this work the R: group, the substituent on the nitrogen atom of the bridge, is placed on the macrocycle after the bridging unit has already been incorporated. This may give some flex­ ibility in synthetic strategy for the elaborately substituted lacunar macrocycles.

Poly(vinylpyridine).— This support is a commercially available polymer (Pol. II) manufactured by Waters Associates, Milford, Mass. under the trademark PORAPAK S, and, in a highly purified form, is used as a stationary support in gas chromatography. The nitrogen atom provided by the pyridyl group could coordinate axially to the metal atom in the

Polymer II; Poly(yinylpyridine) a) H coupled

j b) ^ decoupled

180 120 60

Figure 45; A) Coupled and B) Decoupled NMR Spectrum of [NÏ{(NBz),(CHg)«Meg[16]tetraeneNA}]- m (PF«)a in CDgCN, AO.O®C K a) ^ coupled

b) ^ decoupled

180 120 60 Figure 46: A) ^ Coupled and B) Decoupled NMR Spectrum of IN1{ (RzNMe)2Me2[16]tetraeneN<.}] (PF«) 2 M in CDaCN, 40.0®C •Vj 118

TABLE 7

*^C HMR assignments for [Ni{(NBz)a(CH2 )«- Mes[16]tetraeneNA}](PF6)2 in CDgCN, 80 MHz 40°C

H;C

CH 3

Decoupled Coupled* Peak Assignment”

172.62 172.67 A 167.96 168.01 B

160.82 fl62.14 1159.61 '

136.16 136.21 1

129.99 /131.31 2 or 3 1128.73

129.66 J130.48 3 or 2 1128.05 '

129.22 129.172 4

111.64 111.67 D

56.35 57.08 E,F,G,H tr 56.46

55.38 55.43 tr 55.23

55.13 f53.77 <53.38 I5 2 . 9 5

51.39 51.40 119

TABLE 7 (Continued)

Decoupled Coupled^ Peak Assigiiment^

30.37 30.132 tr I,J 30.03

25.67 r25.714 Q 0,P 24.50 124.597 01

20.86 21.59 tr Q,R 20.57 01 19.79 20.28 tr 19.11

a. The following abbreviations have been used: tr, triplet; q, quartet; 01, overlap; d, doublet b. The letters identify the carbon atoms in structure______. 120

TABLE 8

NMR assignments for [Ni{(BzNMe)2Mea[16J- tetraeneNfc}](PF«)a in CDsCN, 80 MHz, 40"C

Decoupled Coupled^ Peak Assignment^

173.21 173.40 A 168.25 168.25 B

160.195 161.65 . C 159.13 °

135.24 135.44 1

129.95 131.45 , 2 or 3 128.88 “

129.32 130.77 , 2 or 3 128.29 “

128.88 130.29 4 127.86 °

111.88 112.08 D

59.26 59.45 tr E,F,G

56.25 57.99 tr

51.01 51.20 tr

42.08 r43.05 H 141.44 ^

30.47 30.57 I.J 30.18 30.56

21.68 K,L 20.81 20.32 19.31 18.72

a. The following abbreviations have been used: tr, triplet; q ol, overlap; d, doublet

b. The letters or numbers identify the carbon atom in structure 121 macrocyclic complex, acting as a fifth donor atom for the metal. Zimmer has solved a crystal structure for [Fe{NHEthi)a(CHz):Me:[16ItetraeneN*}-

(CO)(Py)](PFe) 2 (Fig. 48) which shows a pyridine base coordinated to the iron atom in a complex having a lacunar ligand. The site occupied by the relatively large pyridine is opposite the protecting bridge.

In Zimmer's case the smaller carbon monoxide was found in the cavity.

It is well demonstrated that nitrogen donors can coordinate to cobalt(II) and iron(II) in their complexes with macrocyclic ligands.*®

In addition to providing the fifth donor atom for the complex, the base places additional electron density on the metal and enhances its oxygen affinity. For cobalt compounds in the divalent oxidation state (+2),

ESR can be used to investigate the coordination sphere of the transition metal. In the case of the polymer, ESR is a most useful tool for the examination of the interaction of the complex with ligating atoms at the solvent-polymer interface. With the poly(vinylpyridine) polymer, the transition metal complexes will be absorbed at this interface. An equi­ librium is established for the reaction of the cobalt complex with base groups of the polymer. Therefore, the absorbed material may not be permanently attached to the polymer. In applications where solvent is used with the polymer, the choice of solvent will become import'ant since the complexes may be stripped away from the polymer in some solutions.

Controlled Pore Glass.— Controlled pore glass (CPG) supports are commercially available from Pierce Chemical Company. They are produced from a borosilicate base material which has been heated to separate the borates and the silicates. The borates are etched from the material leaving o

Figure 47; ORTEP Drawings of [Fe{(HNEthi),(CH,)=Me,[16]tetraeneN..}(CO)(?Y)(PF»)z'CH, OH

ro ho 123 the porous structure. CPG propylamine contains an extension arm on the

silicate ^ i c h provides a primary amine group that can be further used

for immobilization and derivization (Polymer III) in many ways.^®** In

Si-(CH2)^NH2 -V

Polymer III; CPG propylamine

the investigations with the lacunar molecules there are two possible ways

to make the best use of this support. The first method is to absorb a

lacunar complex on the support. The primary amine may or may not act as an

axial donor base to the lacunar canplexes (Fig. 48). However in a second application the amino extension arm of the polymer support could be allowed

to react with intermediates at the nickel(II) complexes to covalently attach

V I, - V a ( \

Figure 48; CPG Propylamine may Coordinate to Cobalt (II) as an Axial Base 124 the lacunar complex to the polymer. This approach is illustrated in

Figure 49. This type of reaction is analogous to the chemistry involved in the synthesis of the bridging group (Ri) into the ligand super­ structure.

The behavior of propyl amine in DMF solution with the lacunar complexes was detected using ESR techniques. As is the case with the

Porapak S support, if the nitrogen atom of the amine group coordinates to the cobalt atom, ESR techniques can detect the interaction through the appearance of an axial five coordinate spectrum and superhyperfine coupling from the axial nitrogen donor. A typical spectrum obtained for the solutions of cobalt(II) lacunar complexes in DMF with excess propyl amine is illustrated in Figure 50. The spectral parameters observed are consistent with axial symmetry in cobalt (II) and the g|| tensor is split into 8 components by the ®*Co nucleus. Each component is broadened by the quadrupole of the nucleus. The value of gj^ is 2.33 and g^| is 2.01. This spectrum is strikingly different from the normal DMF spectrum. There each branch of g is sharp, without any

* ‘‘N superhyperfine splitting. Normally three g values are found Cgi =

2.56, ga = 2.33, ga = 1.98). These parameters are consistent for many so-called square planar cobalt(II) macrocyclic complexes. Therefore in the polymer support the interaction of the lacunar complex with the surface may involve coordination of the nitrogen atom of the propyl amine group as an axial base. 12S

C»

O^i-ICHglg-NHg ♦ - 0 ^

H -ICHoU -Si—O -

Figure 49; CPG Propylamine may be Used to Covalently Attach the Macrocyclic Complexes, Analagous to Bridging Reactions 126

A)

3 2 0 0

B)

3 2 0 0

Figure 50; A) Spectnan of Co(II.) Lacunar Complex in DMF, Frozen Glass, -196®C B) Spectrum of Co (II) Lacunar complex in DMF with propyl­ amine, Frozen Glass, -196®C 127

ATTACHMENT OT THE POLYMERIC SUPPORTS

Covalent Attachment. — As with the model systems (Fig.'42), the

first step in the reaction with chloromethylated polystyrene polymer

involves the deprotonation of the macrocyclic complex with sodium

ethoxide. The acetonitrile solution of the deprotonated macrocyclic

complex is allowed to react with the polymer in situ, by stirring the

solution gently at 60"C for seven to eight days. Following that, the

treated polymer is extracted in a soxhlet for five days to remove any

absorbed material.

Figure 51 shows portions of the infrared spectra of several

compounds; a) shows benzyl chloride which is closely related to the

active function in the polymer, and b) shows the genuine chloromethylated

polystyrene. Bands appear at 699 cm~^ and 670 cm"‘ in both spectra.

When the chloromethylated polystyrene polymer was treated with sodium

iodide, spectrum c) resulted; this is an iodomethylated polystyrene.

The band present at 699 cm~^ remains intact while the band at 670 cm~^

disappears, and a band emerges at lower wavenumbers indicating the C-I

stretching mode. These changes indicate that the chloride has been

replaced by iodide.®^ For reactions between the polymer and deprotonated

lacunar complexes, the loss of intensity of the band at 670 cm”^ will

be used as an indication of the extent of substitution onto the support.

Figure 52 illustrates the infrared spectrum of the polymer that has been

treated with copper(II) and cobalt(II) lacunar complexes (R = (CHz)@

bridges). In both systems, the band at 670 cm~^ has decreased in intensity

relative to the band at 699 cm”^; however, the intensity is not entirely 128

a) Neat Sample Benzyl Chloride on NaCl plates

b) KBr Pellet Chlor onié tliÿ làted P olys tyrene

c) KBr Pellet After treatment with Nal

I I f I I •00 «00 CM~^ Figure 51; Comparison of Infrared Stretching Frequencies of Substances Related to Chloromethylated Polystyrene 129

a) Copper(II)

b) CobaU(II)

R 2 =

I I I \ 900 700

Figure 52; Infrared Spectrum of Chloromethylated Polystyrene .After Treatment w i t h A) Copper(II) and B) Cobalt (II) Lacunar Complexex (R* = (CHg)*). Itegions near 700 cm~* should be Compared with Original Polymer Sample (Fig. 51, B) 130

lost, as was the case upon the treatment with sodium iodide. It is

likely that some of the chloromethyl groups remain, even after extended

reaction times.

Table 9 contains a list of the lacunar complexes which have been

attached to the chloromethylated polystyrene support. For convenience

in the remaining section the complexes will be cited by complex number

(I-VI) and bridging unit (Ri) [see structure II (Fig. 20)]. The

polymer is attached to the lacunar complexes through a benzylic function

and occupies the R 2 position in the complexes. For all the complexes

used in this thesis research the R 3 function is a methyl group.

Table 10 lists the results of chemical analysis of several of the

chloromethylated polystyrene polymers with various cobalt(II) complexes.

Analyses of three separately prepared samples of complex II (Ri=(CH:)@) which have been treated with the polymer for seven days show

that not all the chloro groups have reacted. The polymer contains

3 meg/gram of the active chloro function. If one complex per functional

group were bound, the cobalt concentration would be expected to be

3 meg/gram. Similarly, if one molecule of complex were bound to two

chloromethyl groups (there are two active nitrogens in each macrocycle),

the expected cobalt concentration would be 1.5 meg/gram. As Table 10

shows, on the average the concentration of cobalt is only .3 meg/gram, a

value that is less than would be expected for complete reaction of the

complex with the polymer. The analyses also show the nitrogen/cobalt

ratio is slightly greater than 6 to 1. Extra nitrogen could result from

the coordination of a molecule of acetonitrile, which is the solvent

used in the reaction. When reaction times were less than the specified 131

TABLE 9

COMPLEXES WHICH HAVE BEEN COVALENTLY ATTACHED TO CHLOROMETHÏLATED

POLYSTYRENE

I

R3 = C H g I I

AXIAL CROSS­ BASE LINKING NUMBER M «1 — C u i C H g i e 6 % 1

C o I C H g lg CH3 CN 6 ,1 % 11,111

C o C H ^ 6% IV

C o * 6,1% V,VI “ 3

REARRANGED LIGAND, CHAPTER III 132

TABLE 10

ANALYSIS OF COBALT COMPLEXES COVALENTLY ATTACHED TO CHLOROMETHYLATED POLYSTYRENE

moles N moles Co Ratio Days Number %N %Co Cx 1 0 ") (x 10*) N/Co Reacted

II 3.00 1.94 2.14 3.29 6.5/1 7

II 2.82 1.85 2.01 3.14 6.4/1 7

II 3.04 2.05 2.17 3.48 6.]/l 7

II 2.99 1.8 2.13 3.05 7.2/1 14

II .815 .47 .58 .80 7.3/1 2

III 2.05 1.12 1.46 1.9 7.7/1 7

III .57 .34 .41 .57 7.2/1 2

V .714 .43 .51 .73 7.1/1 2

VI .910 .54 . 64 .929 7.3/1 2 133 seven days, the concentrations of absorbed cobalt were considerably less than .3 meq/gram. One reaction was allowed to run for fourteen days to see if a higher incorporation of cobalt could be obtained; however the concentrâtion level never exceeded .3 meq/gram. An excess of the cobalt lacunar complex (greater than three milliequivalents) was always used in the initial stoichiometry of the reaction mixture.

Polymers containing copper(II) (d®> and cobalt(II) (d') contain paramagnetic metal ions and can be further characterized by electron spin resonance spectroscopy. The examination of the ESR spectrum of complex I (Ri=(CH2 )«) of copper(II) lacunar complex is compared with the frozen glass spectrum of the copper(II) complex before deprotonation and reaction with the polymer (Figs. 53 and 54). As has been reported in other polymer research involving transition metal complexes, the spectral lines of the copper complex that has been treated with the polymer appear broad when compared with the spectral features observed for the frozen glass spectrum of the complex in a solution.®»” "®®

The examination of the ESR spectra of polymer supported cobalt is used

extensively in this thesis research. It is used to interpret the environment of the cobalt(II) center in the polymer environment. The broadened features of the cobalt signals can obscure the hyperfine and

superhyperfine splittings. The cause of the broadening of the signals

is thought to be due to a certain amount of polymer chain motion even at

low temperatures or to a lack of magnetic dilution in the samples.’®

As mentioned earlier, the chloromethylated polystyrene is available with the concentration of chloromethyl groups equal to 3 meq/gram. In +

A = 205 Gauss

15 Gauss

w Figure 53: ESR Spectrum of [Cu{(NH)2 (CHa)«Mea[16]tetraeneN,.}](??*): in Acetone Solution, Frozen Glass, -196°C 200 0

I 3 2 0 0

Figure 54: ESR Spectrum of Polymer I, No Solvent, Frozen, -196°C 136 order to provide variations on the amount of crosslinking and in the size of the polymer bead, the polymer has been studied in two forms; a 6% crosslinked macroreticular resin and a 1% crosslinked micro- reticular resin. Cobalt compounds which have been treated with chloromethylated polystyrene show differences in their ESR signals depending on the extent of crosslinking (6% or 1%). Figure 55 shows the ESR pattern obtained at -196°C for the 6% crosslinked polymer with the cobalt lacunar complex II (Ri=(CHa)e)• The overall pattern indicates the presence of a large percentage of five coordinate cobalt; however, the broadened peak around 2500 gauss suggests that some four coordinate cobalt is present. For the five coordinate species, gj_ occurs at g=2.3 and g|| (g=2.00) is split into an eight line pattern due to the inter­ action with the ’*Co nucleus. Further splitting is observed in the parallel region from superhyperfine coupling with a nitrogen donor (1=1);

some of the hyperfine peaks appear as triplets. Acetonitrile is the solvent for the reaction and it is a coordinating solvent for cobalt

complexes in this ligand family. Therefore it is likely that the nitrogen

species is coordinated in the axial position. As mentioned earlier, the

analyses showed a high nitrogen content, and the superhyperfine splittings were more pronounced in the spectra of samples having higher nitrogen

analyses. Similar signals were obtained in the polymer samples containing

lower concentrations of the cobalt lacunar complex.

Figure 56 shows the ESR spectrum of the cobalt lacunar complex III

(Ri=(CHa)e) which is bound to the 1% crosslinked chloromethylated poly­

styrene polymer after seven days of reaction and four days of extraction 137

3 2 0 0 G

Co"

R*

Figure 55: ESR Spectrum of Polymer II, No Solvent, Frozen, -196“C 138

3 2 0 0

Figure 56: ESR Spectrum of Polymer III, No Solvent, Frozen, -196"C 139 with acetonitrile. This pattern is strikingly different from the one given in Figure 55 for the 6% crosslinked polymer derivative (II),

Ri=(CHz)6. The spectral pattern is not analogous to spectra obtained in coordinating nitrogen donor solvents or in non-coordinating solvents.

However, it is somewhat similar to the signal pattern Jackson observed for [Co{(CHsNNeopentyl)zMez[16]tetraeneN 4 }](PFe)a. Jackson assigned the signal to the category described by Kida as type II (see Fig. 57).^*

In type II complexes the unpaired electron of cobalt(II) is in the d^^ orbital. Characteristically the corresponding ligands have less developed

II systems and the electron density is not extensively delocalized over the ligand by a backbonding interaction.The type I behavior, illustrated in Figure 58 is more commonly exhibited by the lacunar cobalt complexes. The type I signal is typical of compounds containing four nitrogen donors with well developed II systems.*’®®

The cobalt concentration in this 1% crosslinked polymer was .2 meq/gram. In Figure 59, the spectrum is shown for a less concentrated sample containing only 0.05 meq/gram of cobalt. This spectrum is similar, but not identical to the one obtained for the more concentrated sample.

This less concentrated cobalt complex appears to have some of the features normally seen in axial five coordinate cobalt(II) polymers. Thfe differences between the reactions of these two samples with oxygen will be discussed as a function of the cobalt concentration later.

The unbridged cobalt compound described in Chapter III which con­ tains a rearranged ligand structure to give a cobalt(II) ion with five nitrogen donor atoms was treated with the chloromethylated polystyrene 140

yz

dxz

1 _ L

xy

Figure 57: Kida Type The Unpaired Electron in the dyz Orbital. Ligands Have Less Developed 1Î System, Electron Density Is Not Extensively Delocalized Over the Ligand System. Only

Observed in [Co{(CH3NNeopentyI)2Me2[16]tetraeneN4}](FFe) 2 in Acetone Solution, 141

I I

xy

xz yz

Figure 58; Kida Type I*“® This Spectrum is Commonly Observed for Cobalt(II) Complexes with Four Nitrogen Donors Having Well Developed Systems. Typically Observed for Most Cobalt(II) Lacunar Complexes in Acetone, Frozen Class Spectrum. .— _) 200gauss

I 3 2 0 0 G

Figure 59; ESR Spectrum of Poljimer III, No Solvent, -196“C. This Sample is Leas Concentrated (in Cobalt Concentrsition) than that Illustrated in Figure 56 for the same Polymer N) 143 support. The reaction procedure is similar to the one followed with the bridged complexes. A solution containing the cobalt complex is deprotonated and allowed to react with the polymer in situ, by stirring the solution gently at 60°C for seven to eight days. Following that, the treated polymer is extracted in a soxhlet for four days to remove any absorbed material. The study of this supported complex also shows differences in the ESR signals obtained for the 5% crosslinked

(V) and the 1% cross linked (.VI) variety. Figures 60 and 61 represent the ESR spectra of the 6% (V) and the 1% (VI) systems, respectively.

Here the 6% crosslinked sample looks more like the type II species vdiile the 1% system looks like the five coordinate signal for the frozen glass of the solution analog as reported in Figure 25 of Chapter III. The absence of the typical five coordinate signal in the 6% crosslinked polymer may indicate that the ligand is no longer in the rearranged state, where the ligand donates five nitrogens to the coordination sphere of the cobalt(II). When the rearranged complex is attached to the 1% crosslinked polymer, its ESR spectrum is typical of an axial five coord­ inate cobalt(II) species. This suggests that the ligand is still in the rearranged structure that it had when it was the starting material for the reaction with the polymer. ^

Coordination Attachment.— Table 11 lists all of the lacunar cobalt(II) complexes which have been attached to poly(vinylpyridine) by allowing solutions of the lacunar complexes to evaporate onto the surface of the polymer. Structure VI illustrates the aquo N,N'(1,1,2,2- tetramethylenthlene)bis-(3-methoxysalicylideniminato)cobalt(II), ^200 gliuss

3200 G

Figure 60; ESR Spectrum of Polymer V, No Solvent, Frozen, -196“C 200 gauss

32Î)0G

Figure 61; ESR Spectrum of Polymer VI, No Solvent, Frozen, -196®C 4> Ln 146

TABLE 9

LACUNAR COMPLEXES WHICH ARE COORDIHATIVELY ATTACHED TO POLYMER

THROUGH INTERACTIF WITH PYRIDINE FUNCTION

AXIAL BASE = © - ( ^ I 1

R3- CHg

REL A T IV E PERCENTAGE OXYGENATED(25“Cl NUMBER

ICHglg CH3 1 0 0 VII

ICH2 I6 H 90 VIII

ICH2 I5 CH3 80 IX

CH3 CH3 1 0 0 X

H 0 XI 147

Co(3-0MeSALTMEN), which was also attached to PVP. Figure 62 illustrates

-h-K

C o '

Mel M e

Structure VI; N,N' (l,l,2,2-'-tetramethylethylene)bis-(3- methoxysalicylideniminato)cobalt(II) a typical ESR spectrum found for cobalt compounds bound to this polymer by this procedure. The ESR spectra of cobalt complexes bound to this polymer do not vary as much as was seen with the chloromethylated polystyrene polymers. All of the spectra display axial symmetry and the parallel component of the g tensor is split into eight components by the cobalt-59 nucleus. Each component is broadened by the quadrupole of the nitrogen-14 nucleus (1= 1) of the axial base. The values of ^ range from 2.26 to

2.32 and the values of g|j range from 2.00 to 2.02. The hyperfine coupling constant A|| varies from 85 to 100 gauss. Superhyperfine splitting was not detected in these polymer samples) however, the parallel components were broadened considerably. The molar concentration per gram of polymer can be estimated from the reaction conditions. Typically between .2 and .6 mmole of cobalt complex are loaded per gram of polymer resin.

CFG Support.— Two lacunar cobalt(II) complexes have been absorbed on the surface of CFG propylamine by allowing concentrated solutions of the cobalt complexes to evaporate onto the surface of the polymer. 3200 G H

Figure 62; Typical ESR Spectrum Observed with Cobalt(II) Complexes Coordinatlvely Attached to PORAPAK S (poly(vlnylpyrldlne)), Frozen, No Solvent, -196“C 149

Alternatively a solution of the cobalt complex is stirred with the polymer sample while a vacuum is applied to reduce the volume of the

solvent. Figure 63 illustrates the ESR spectrum observed for one bridged lacunar complex (Ri=(CHz)e, Ra=CH 3 , Rs=CH 3 ). The spectral parameters observed for this complex are unusual for a cobalt(II) com­ plex. The absorption centers near 3200 gauss corresponding to an approximate g value of 2; however, the width of the signal is unusually broad and spans over nearly 6000 gauss. The appearance of this signal

contrasts strongly with the signal observed for another lacunar cobalt(II)

complex (Ri=m-xylyl, Ra=H, R 3 =CH 3 ). Figure 64 shows that this spectrum

is more typical of signals usually observed for axial cobalt(II).

However there is no sign of the expected hyperfine splittings from the

®’Co nucleus in the parallel region of the spectra. Hoshino has observed

this type of spectral feature in work with the binuclear cobalt(II)

complexes of related compartmental ligands. Since the intial cobalt(II)

lacunar complex is a monomeric unit, this suggests that the cobalt(II)

centers lie close to one another on the polymer surface.Perhaps

there is some degree of metal-metal interaction through two separate monomeric units. H

1 0 0 0 G

Figure 63; ESR Spectrum of CFG Propylamine Polymer with [Co{(MeNEthl),(CHa)gMe2[16]CetraeneN&}](PF*)a, No Solvent, -196°C. Ln O H 200

3 ^ 0

Figure 64; ESR Spectrum of CPG Propylamine with [Cb{(HNEthl)a(m-xylyl)Me2[16]tetraeneN£,}](PFs) No Solvent, Frozen, -196*C ' 152

REACTIONS OF THE POLYMER SUPPORTED COBALT COMPLEXES WITH OXYGEN

Covalently Attached Complexes.— The supported cobalt complexes can be studied as the six coordinate oxygen adducts by blowing a stream of oxygen through the sample. The 6% crosslinked polymer (II) reacts with oxygen slowly, either as a dry solid or in the presence of solvent.

As oxygen is added, the intensity of the ESR signal assignable to the five coordinate cobalt(II) decreases in intensity while the signal due to the oxygen adduct increases in intensity (Fig. 65). The ESR spectrum seen for the oxygen adduct is similar to those observed for many cobalt-oxygen adducts.*^ The gjj component shifts from 2.01 to

2.09 and the hyperfine splitting (eight components), A|| , decreases from 85-100 to 16-20 gauss. The gj^ values shift to 2.01 and Aj_ has a value between 7 and 10 gauss. This indicates that the electron spin density has partially shifted from the cobalt(II) to the oxygen and

is consistent with the Co(III)Oa formulation proposed by Basolo.“®’“^

The coordinate signal decreases further when the exposure time for oxygen is increased and, at the end of long exposure times, most of the five coordinate cobalt has combined with oxygen. A broad signal a centered at g = 2.5 normally remains unchanged during exposure to

oxygen. This signal is attributed to the presence of some four coor­ dinated cobalt(II) complex which does not react with oxygen. The unreactive species may be protected from oxygen by the arrangement of

the polymer or its O 2 affinity may be very small because it does not

have an axial ligand. 153

The cobalt-oxygen adduct is stable for weeks if no solvent is present.. If solvent is present (normally acetonitrile or acetone con­ taining N-methylimidazole), the oxygen adduct signal is slowly lost over a period of several days to a week. In Figure 66, a typical spectrum for an aged, oxygenated sample is illustrated. An intense

signal for a five coordinate cobalt(II) species has reappeared and

an organic radical signal (g=2) has emerged. The radical signal remains

strong for several days. The long-term study of the decay of the ESR

signals for the lacunar complexes actually dissolved in liquid solution has never shown radical signals. Therefore, this phenomenon appears to be a unique feature of the polymer systems. It is possible that the activated oxygen could attack the polymer and leave stable radicals

isolated in the polymer matrix.’ Even in the presence of the radical,

the five coordinate cobalt that is present can be reoxygenated to give

the oxygen adduct. However, the spectra of the dioxygen adduct and

the organic radical both occur near g=2 and the overlapping signals

complicate interpretation.

The 1% crosslinked (III) microreticular resin was studied to see

if the smaller bead size and less extensive crosslinking could facilitate

a more rapid binding of oxygen in the bridged cobalt lacunar complex (III)

(Ri=(CHz)e). In the more concentrated cobalt sample (0.2 meq/gram),

opening of the ESR tube to the atmosphere at room temperature causes

rapid loss of the paramagnetism in the sample. If a coordinating solvent

is added to the polymer beads in the absence of oxygen (dry box), the

initial signal is that of a typical axial five coordinate species. When 154 this sample is exposed to oxygen, an oxygen adduct is NOT seen, even as a transient intermediate. Instead an organic radical signal appears, and when the sample warms to room temperature, it will rapidly decay. Pumping on the system to remove oxygen restores the original five coordinate cobalt(II) signal, but it is accompanied by a signal due to an organic radical. Therefore some interaction with oxygen is inferred, although the ESR spectrum of a one-to-one cobalt-oxygen adduct is not detected in these systems. In general, polymers containing only 1 % crosslinking are not very rigid in structure within the polymer chain. Little restriction of the macrobicyclic cobalt complex is expected since the molecular motion in the polymer may approach that which is present in solution. It is concluded that a binuclear cobalt-peroxo dimer has formed.

For the less concentrated samples (cobalt) involving 1% crosslinked

chloromethylated polystyrene polymer, an oxygen adduct signal can be detected when air is passed over the solid sample (Fig. 67). This

oxygen adduct signal is stable for at least a week when no solvent is present. Since the less concentrated sample gives the ESR signal of a

stable one-to-one oxygen adduct, this observation supports the explanation

given above for the absence of a signal in the more concentrated sample.

A dimerization process probably occurs when the concentration of the

cobalt complex in the flexible polymer is sufficiently great.

The cobalt complex discussed in Chapter III which contained the

rearranged pentadentate ligand, [Co[ (MeNImlnoethyl)(MeHNEthi)Me2 ll6 J-

tetraeneHs]](PFe)2 , when bound to the 6 % crosslinked polymer to form /

200 gauss

3200G H

Figure 65; ESR Spectrum of Polymer XI After Exposure to 0, for 45 Minutes, No Solvent, Frozen, -196 C

a 200 gauss

3200 G

Figure 66: Decay of Oxygen Adduct of Polymer II, No Solvent, Frozen, -196"C Ln ON 200

3 2 0 0 G

Figure 67; ESR Spectrum of Leas Concentrated Sample of Polymer III After Exposure to O 2 for 30 minutes. No Solvent, Frozen, -196°C U i 158

polymer complex V does not react very quickly with dioxygen. When

the dry solid polymer is exposed to Oa no oxygen adduct ESR signal

is obtained and the five coordinate cobalt(II) spectrum slowly decays over a period of one hour. However, when supported on the 1% polymer

the complex VI gives a small oxygen adduct signal before decaying into

an ESR silent state. This result is not unexpected since the parent

complex (unsupported) is so sensitive to oxygen at room temperature in

solution. A more intense oxygen adduct signal can be obtained if the

polymer supported cobalt complex is exposed to dioxygen at reduced

temperatures. The adduct remains stable if the sample is kept below

0°C.

Coordinately Attached Complexes.— A series of cobalt complexes

has been absorbed on poly(vinylpyridine) and studied as oxygen carriers.

The ESR spectrum for the oxygen adduct formed by [Co{(MeNEthi)2 (CHz)«-

Me[16]tetraeneN(,}] (PFe) 2 which has been absorbed on Porapak S to form

complex VII is illustrated in Figure 6 8 . There is no indication that

any five coordinate cobalt(II) remains in the sample after a brief

exposure to oxygen. The signal is clear and sharp with the hyperfine

splitting frctn the ®*Co nucleus evident in both the parallel and

perpendicular regions. The oxygen adduct of this compound is stable

for a few days as a dry solid at room temperature. The oxygen adduct

gives up only part of its oxygen under rather extreme conditions in­

volving pumping the sample while heating to 60°C (Fig. 69). Obviously

this cobalt compound has great Oa affinity at room temperature. For

the parent complex, is .155 torr”^ at 20.1°C in an N-methylimidazole/

acetonitrile solvent system. In order to compare the behavior of a strong 200

3 2 0 0 G .

Figure 6 8 ; ESR Spectrum of Oxygen Adduct of [C6{(MeNEthi)z(CH,)«Mcz[16]tetraeneN»}](PF»)% Attached to PORAPAK S. Polymer XII M Ui VO 200

3 2 0 0

Figure 69; Treatment of Sample Shown in Figure 6 8 to Dynamic Vacuum at 60®C, ESR Spectrum, No Solvent, -196®C o\ o 161

oxygen binder with a rather weak one, [Co{(HNEthi)2 (m-xylyl)Me2 [16]- tetraeneNi,}] (PF6 > 2 \diich has almost no affinity for dioxygen at room temperature, was absorbed on the poly(vinylpyridine) polymer to form

complex XI. The five coordinate axial cobalt(II) ESR signal is retained upon exposure to oxygen at room temperature. This complex probably can be prepared without using the inert atmosphere conditions needed for most of the cobalt(II) chemistry. Partial oxygen adduct formation

can be detected if oxygen is blown through the sample at -78°C. Table

XI lists the results of exposing a series of Porapak S supported cobalt complexes to air.

These supported cobalt(II) complexes were allowed to stand, sealed, at room temperature after the oxygen adduct formation had been established

by ESR. Observation of the ESR spectrum of [Co{(MeNEthi)2 (CH2)e[16]Me2-

tetraeneN*}](PFfi)2 , complex XII, after one week showed the formation

of an organic radical species. This was also found in studies on the

decomposition of the oxygen adducts of the cobalt(II) lacunar complexes

supported on the chloromethylated polystyrene polymer. Observation of

the [Co{(MeNEthi)2 (CH2 ) 5 [1 6 ]Me2 tetraeneN4 )](PFs)2 , polymer IX, after

standing capped in an ESR capillary on the benchtop at room temperature

for six weeks showed a slight signal from an organic radical, but

it also showed the presence of an intense signal for a five coordinate

cobalt(II). In fact, the intensity of this signal was so close to the

intensity of the initial spectrum that it was tempting to suggest that

the cobalt(II) species has been quantitatively recovered using ESR for

detection. The samples appeared to reoxygenate to the same extent on

the second exposure to oxygen, as they had initially. Other cobalt(II) 162 com plexes, [Co{(MezNEthi)zMez[16]tetraeneN&}](PF«)a, p o l y m e r X, a n d

[Co{ (H N Ethi)aC 6M e 2 [ 1 6 ]tetraeneN j,}] (PFe)2, polym er VII, showed only the axial five coordinate cobalt signal (com parable to the intensity of the original signal) w ith no evidence of the organic radical signal.

The behavior of the supported [Co(II)(3-OMe-SALTMEN)] complex

contrasts with the behavior described for the above samples. Instead

of the reappearance of the axial cobalt(II) signal, the signal of the

one-to-one oxygen adduct was retained in the same intensity. After

six weeks as a dry solid there was no indication of the loss of the

oxygen adduct and the subsequent growth of an organic radical signal

or of the reappearance of the axial five coordinate spectrum.

It should be emphasized that the behavior of the supported unbridged

complex, [Co{ (MeaN) aMea [16]tetraeneNi,}] (PFe) 2 , polymer X, is remarkable.

The oxygenation of the supported complex contrasts strongly with the

behavior of the complex in solution. Not only is a detectable one-to-one

oxygen adduct formed by exposure to oxygen at room temperature, but

the loss of that adduct does produce the original axial five coordinate

cobalt signal. The polymer sample could be reoxygenated and the entire

sequence of formation of the original signal repeated. The solution

behavior of this cobalt(II) complex shows the formation of a one-to-one

oxygen adduct only at lower temperatures (below 0“C).^* Furthermore,

exposure to oxygen at room temperature causes rapid decomposition of

the cobalt(II) species and no paramagnetic species can be detected using

ESR spectroscopy.

Clearly one of the best features of the Porapak S polymer system

of the original sample is the display of a single, recognizable, axial 163 five coordinate cobalt(II) ESR signal in the initial spectrum. This behavior is simpler than that due to the mixture of four and five coordinate species in the case of chloromethylated polystyrene samples.

The initial color of the cobalt(II) polymer is a bright orange and it becomes brown after exposure to oxygen. Although the signal from the polymer samples which have stood for weeks suggests a return to the initial axial cobalt(II) species, the color of the polymer sample remains brown. The samples appear to reoxygenate to the same extent after the second exposure to oxygen as they did the initial time, as evidenced by the intensity of the one-to-one oxygen adduct ESR signal.

The appearance of an organic radical species early in the decomposition of one of the lacunar complexes suggests that something more than a reversible oxygenation of cobalt(II) is occurring (Equation 7). Herron has shown that the ligand of the unbridged iron complex [Fe{(Me2N)aMej[16]-

Co(li) + Oa ^ Co(III)Oa" 8

tetraeneNfc}](PFfi)a is actually oxidized in the presence of oxygen and dimethylacetamide is found as a product.®^ Aside from the questions which remain concerning the decomposition of the oxygen adduct signal, the limitation of this polymer system involves the necessity to choose solvents carefully to avoid the stripping of the lacunar complexes from the support. Typical frozen glass spectra of cobalt(II) lacunar complexes

in acetone solution have been observed when acetone has been used to wet a dried Porapak supported cobalt(II) complex. Apparently the cobalt complex can become removed from the polymer surface in soluble solvents. 164

The ESR spectrinn of the axial cobalt(II) complex in acetone (Kida

Type I, Fig. 58) is very different than the signal from an axial

species containing five nitrogen donor ligands.

CFG Support.— Samples of CFG aminopropyl support which had been treated with cobalt(II) lacunar complexes were exposed to oxygen by blowing gas over the polymer sample. The ESR spectrum observed for the bridged complex (Ri=(CH2 )e, Rz=CH 3 , R 3=CHa) gave a broad but discernable oxygen adduct signal (Fig. 70). Therefore, although the initial spectrum obtained for the complex is very unusual for a cobalt(II) complex (Fig. 63), the complex retains the ability to coordinate oxygen despite the lack of an axial base to provide a five coordinate cobalt(II) species. The oxygen adduct signal was stable for several days at room temperature standing capped in a sealed NMR tube. However the other unbridged complex which was studied (Ri=m-xylyl,

Rz=CH 3 , Rs=CH 3) does not normally react with oxygen at room temperature.

When oxygen was blown over the sample the signal from the axial species decreased in intensity, but there was no simultaneous growth of pattern due to a one-to-one oxygen adduct. If the cobalt centers were close enough to interact with one another and could bind oxygen simultaneously to two centers, the resulting complex would not give an ESR signal for

the oxygen adduct. This might also produce an 0% affinity much greater

than that previously observed for the five coordinate derivative. 200

3^00

Figure 70; ESR Spectrum of CFG Propylamine with [Co{(MeNEthi),(GHz)«Me,[16]tetraeneN,.}](PF»): Absorbed to Surface After Exposure to Oxygen, Frozen, -196®C g REACTIONS OF POLYMER SUPPORTED COMPLEXES 166

Oxidation Catalysis.— In solution, cobalt(II) lacunar complexes have been found to catalyze the autoxidation phenols to qulnones.

The formation of specific qulnones would be useful since there are many applications for these compounds. Qulnones are used as monomer precursors for many Industrial polymers and are also Involved In the natural product synthesis to produce compounds of Insecticidal and herblcldal value. Industrial routes to qulnones are often nonspecific and the byproducts Include dlphenoqulnone and long chain polymers.

Evans has shown that under homogeneous conditions, oxidation of substituted phenols using lacunar cobalt(II) complexes as catalysts yields only the corresponding substituted quinone products (near ambient temperatures). The oxidation of the phenols occurs rapidly. However,

the main problem of these systems Is catalyst decomposition. Often

deactivation occurs so quickly that poor product yields are found.

This short catalyst lifetime provides partial motivation for the study

of the polymer supported lacunar complexes as oxidation catalysts.

Perhaps catalyst lifetime could be Influenced by supporting the catalyst

on a polymer.

The 1% and 6 % crosslinked chloromethylated polystyrene (CMPS)

polymers were used as a support In studies of oxidation catalysis with

various lacunar cobalt(II) complexes. The structure of the lacunar

complex (Str. II, Fig. 20) is varied In the Ri position. This provides

a study where the oxygen binding affinities in solution range from very 167

strong to very weak.^* In all of the complexes the polymer is believed to occupy the R% position. Table IX lists the lacunar

complexes used in this study. The phenols studied include unsub­

stituted phenol, monomethylated phenols, dimethylated phenols, and trimethylated phenols, and the best results with respect to oxidation were obtained from the more highly substituted phenols.

Oxidation experiments with the complex-polymer were carried

out in a test tube (16 x 2.5 cm) equipped with a micro stir bar.

A typical reaction involved 200 mg of polymer sample containing the catalyst. A thirty to one hundred fold excess of substrate was

dissolved in twenty milliliters of solvent (usually acetonitrile) and

stirred gently with the polymer beads. A large stir bar or very rapid

stirring was found to crush the polymer beads into a fine powder. A

steel syringe needle was placed in the test tube and pure oxygen was

allowed to bubble through the sample for one hour. In some experiments,

a pale yellow color was observed as early as fifteen minutes after

initiation of the reaction. When the one hour interval had elapsed,

the test tube was stoppered with an atmosphere of oxygen above the sample

and allowed to continue stirring gently. At specified time intervals

(12 hours, 24 hours, etc.) two ml of the reaction mixture was withdrawn

from the test tube. The aliquot was dried by passing nitrogen gas over

the sample until only a dry residue remained. Analysis of the oxidation

products was done immediately using a DuPont Model 841 High Performance

Liquid Chromatograph (HPLC). Both normal phase and reverse phase columns

have been used to separate the reaction products. Oxygen was allowed

to bubble through the samples gently for one half hour after each sample 168

was taken.

For the 6 % crosslinked CMPS, IV, no oxidation products were

detected by HPLC for any of the substrates. This is not unexpected

since there is no detectable binding of this compound to dioxygen

at room temperature. With the more active complex II, no oxidation

products were detected with phenol or the monomethylated phenols.

However, for the more highly substituted phenols oxidation products

were detected by HPLC; these were usually qulnones. 2 ,6 -dimethylphenol

and 2,3,6-trimethylphenol generally reacted to give a greater yield

of product than did 2,5-dimethylphenol or 2,3,5-trimethyIphenol. The more hindered nature of the substrate may be a factor in the low

yields for these cases.

For complex II in an acetonitrile solvent system, 2 ,6 -dimethyl-

quinone was usually detected as an oxidation product of 2 ,6 -dimethyl­

phenol. Over twenty-five separate oxidation experiments were run with

this 6 % crosslinked system of CMPS. On the average, the variations

in yields were slight for different batches of the catalyst for cases

where the concentration of cobalt was .3 meq/g. Yields were calculated

on the basis of the quinone produced (moles) per mole of cobalt complex.

In these terms, yields measured after three days of reaction ranged

from 11 to 15. In some experiments the oxidation products continued

to accumulate slowly for as long as five days. One anomalous experiment

showed a yield of 80 after five days. Although these results are

encouraging, the limited reproducibility of the experiments suggests

that some uncontrolled factors were involved. The complex supported on 169

1% CMPS (III) usually showed a maximum yield around 10 after 15 hours of reaction and leveled off to a constant value at that point. After the reaction was complete (quinone ratio did not continue to increase), the ESR spectrum of the polymer showed the presence of an organic radical. In the mixed solvent system, acetonitrile:cyclohexane:acetone,

5:3:3, the yields were as high as thirty after day 3, but many other products began to appear in the HPLC at this point and these could not be identified conclusively. The effect of solvent on the rate of quinone formation for the 6 % crosslinked CMPS of II is illustrated in

Figure 71. The mixed solvent system was abandoned due to the number of products seen after extended reaction times. 170

14..

12.. 0 u (0 Q) 1 10 .. I

I 6.. § • I I II CE

3 4

Time (Days)

Figure 71; Comparison of Product Yields in Two Types of Solvents For Polymer II 171

Gas chromatographic supports.— In conjunction with the research group of Dr. Robert Sievers at the University of Colorado in Boulder,

the cobalt complexes on polymeric supports are currently under

investigation as possible packings for chromatographic columns to be used in the separation and analysis of the constituents of other planetary atmospheres at ambient conditions. Mr. John Gillis is actively engaged in the analytical aspects of the project and is responsible for all the gas chromatography reported in this section of the thesis.^®® Gas chromatographic techniques are best suited for the

analysis of multicomponent mixtures of the volatile components of these atmospheres. Ideally a single chromatographic column might be used to separate all components with base line resolution and good peak shape.

However in practice, successful separation with a single column is

unlikely because of the similarities and differences in the properties

of atmospheric constituents, and the way they interact with chromato­

graphic column packings. Sievers' group is involved in a joint research

project with NASA-Ames Research Center to improve chromatographic

separations of suspected constituents present in extraterrestrial atmos­ pheres. One particular aspect of the work center around the improvement

of the separation of oxygen, argon and carbon monoxide with the systems

that have been already developed by the NASA scientists.

The cobalt lacunar complexes which had been incorporated into the

polymeric supports seemed ideal for this application. These complexes

show reversible binding of oxygen and for best results in a column

separation, oxygen should be retarded on the column but not irreversibly

bound. The important interaction would be the separation of oxygen 172

a) b]

a) No polymer used (room temperature) b) Test column of Polymer II used (room temperature) c) Test column of Polymer II at 100"c.

Figure 72; Polymer II Used as a Gas Chromatographic Column for Separation of Oxygen and Nitrogen 173

B) C)

A) Single column ( 153 mg of Polymer Sample)

B) Two Columns ( 300 mg of Polymer Sample)

C) Two Columns with Faster Flow Rate

Figure 73; O 2 and Nj Separation with Polymer II, Increasing the Column Length Aids In Separation of the Peaks, Faster Flow Rate Causes the Peaks to Coalesce 174 through a gas-solid interaction. The initial work with the supported polymers suggested that the solid phase polymers of the cobalt(II) lacunar complexes would be stable enough for this application. The initial sample studied in this application was the 6 % crosslinked

CKPS of compound II (see page ). To our great surprise and delight the initial test column, a 15 cm x 2 mm glass capillary tube packed with the polymer beads, showed partial resolution of oxygen and nitrogen at room temperature, with helium as the carrier gas. In Figure 72 examples of these initial gas chromatographs are illustrated. Best separations occurred at room temperature with the O 2 /N2 peaks coalescing to a single peak as temperatures were raised to 40°C. O 2 was found to be retained on the column as temperatures were lowered. Resolution of the peaks could be enhanced by adding a second 15 cm column of the same cobalt compound (Fig. 73).

After many separate experiments with gas separation, the polymer lost activity with respect to the separation of oxygen/nitrogen gases, and the color of the polymer changed from the initial brown-red to greenish. An ESR of the sample which no longer showed activity showed the presence of 4 and 5 coordinate cobalt(II), and an oxygen adduct, but no evidence for an organic radical. There was no apparent reason for the activity toward separation to stop, and the difficulty remains unsolved. Chemical treatments of the polymer (e.g., sodium dithionite) did not regenerate the original activity of the polymer sample.

The 6% crosslinked macroreticular resins contain a bead size that is relatively large compared to the size of the resins used in most 175

column supports. It vas thought that the smaller bead size found in

the 1 % crosslinked microreticular resin might be more advantageous.

However 1% CMPS samples of complex III showed no activity in the

resolution of oxygen and nitrogen. Samples which varied the concen­

trations of cobalt loading were studied. Several of the other chloro­ methylated polystyrene polymers listed in Table IX were studied. Attempts

to reproduce the original findings or show any separation failed with

these derivatized polymers.

At this point the emphasis shifted to the study of the cobalt

lacunar complexes absorbed on Porapak S, poly(vinylpyridine) PVP (Table

XI). Since the gas phase interactions required for supports eliminates

the problems associated with solvent use in this polymer system, it

was ideal for further screening since compounds could be prepared for

sampling in a less time-consuming fashion. The first compound studied

on PVP was [Co{(NMe)2 CeMe2 [16]tetraeneNi,}], IV. Since only partial

reversibility was found for the oxygen adduct of this complex at 60°C,

it was not unexpected that this compound did not show separation of

oxygen at room temperature. It appeared to keep the first injection

volumes of oxygen, and not stop later volumes until the column had been

regenerated by flowing helium at 60°C for an hour or more. The strong

binding affinity of oxygen would saturate the oxygen binding sites in

the complex and remain affixed, and further oxygen would pass through

with the nitrogen as a single peak. An encouraging result was obtained

with the meta-xylyl bridged complex at -78°C (Fig. 74) which showed the

separation of oxygen and nitrogen. The peaks coalesced to a single

peak at 0°C, but separation was regained when the temperature was lowered 176

A) Room Temperature Separation of N 2 /O 2

B) -78®C Separation of Ng/Og

Figure 74: Separation of O 2 /N2 was achieved at -78"c Using a Sapiple of Polymer XI 177 to -78°C. Since the binding of the meta-xylyl cranplex is poor at room temperature, while the binding is too good with the hexa- methylene bridge at room temperature; ranges are established for possible complexes for study in the lacunar family. Samples are currently under study in Colorado and we are optimistic that a complex will be found that will reproduce the initial most exciting results. APPENDIX A: STUDIES USING COBALT(II) COMPLEXES WITH SPIN TRAPS

The use of spin traps for the identification of short lived radical species was developed extensively by Janzen.* The spin trapping technique combines diamagnetic compounds with free radicals.

This results in the formation of a more stable, complex radical which usually can be observed for longer time periods using ESR spectroscopy.

Nitroso and nitrone derivatives are frequently used as spin traps (Fig.

75). These radical scavengers are designed so that a nitroxide is produced upon reaction with a radical species (Equations 1 and 2).

R* + R'-N=0 R ' - N - R I O'

». f —C = N — Ra R — C — N - R@

Ra O'

The identification of the trapped radical requires analysis of the splitting of the resonance due to the unpaired electron by the magnetic moments of the nuclei associated with it. In the nitroxide traps, the unpaired electron resides on the oxygen atom and couples with the adjacent nitrogen atom (1=1) to give a three line signal. If a proton atom (1=1/2) is present on the carbon atom adjacent to the nitrogen, it will couple with the spin system and split each line of the triplet into a doublet.

A six line pattern will result. 5,5-Dimethyl pyrroline N-oxide is a sensitive trap and it can be used to distinguish between alkyl and aryl radicals on the basis of the magnitude of the coupling constants from the

178 179

6,5-dimethyl-1-pyrroline

N-oxide -O

nitrosobenzene I

I phenyl N-tert-butyl

/ + T ® nitrone hr ^"3

Figure 75; Compounds Used as Spin Traps 180 proton splitting.^®’ In the complex radical formed with nitrosobenzene, the unpaired electron is directly associated with the nitrogen atom.

Therefore, this spin trap may yield complex splitting because of the close interaction of the radical with the nitroxide. N-Tert-butyl-a- phenyl nitrone is an extensively used trap, but it is less sensitive to the aryl or alkyl nature of the trapped species. The application of

spin traps has been Important in recent years, especially in the

investigation of biological systems. Superoxide ion and hydroxyl radical have been confirmed as intermediates in certain biological processes

because of the use of spin traps.

In conjunction with our interest in dioxygen activation by lacunar

complexes, the three spin traps mentioned above have been used. Of

particular interest are the radicals observed in the autoxidation

reaction of the lacunar complexes and in their reaction with organic

substrates. Using 5,5-dimethy1-1-pyrroline N-oxide, Herron has identified

HOj« as a trapped species with iron(II) lacunar complexes at room temp­

erature in pyridine solvents saturated with one atmosphere of oxygen.

With the same trap, he has also detected an unusual radical with the

unbridged cobalt lacunar complex [Co{ (MeaNEthi) 2Me2[16]tetraeneNi,}] (PFs) a

(Fig. 7 6 ) at room temperature in an acetonitrile solution which has been

exposed to oxygen. This signal can be interpreted as a cobalt(III)-

superoxo radical which has been trapped with pyrroline N-oxide. Inter­

pretation of the ESR spectrum shows that each line of the nitroxide's

six line signal is further split into eight lines from coupling to the

cobalt nucleus (1=7/2). The splitting = 19.4G, = 16.06G, and

Ag^ = 6.36G. The cobalt(II) compound does not give the observed signal 3 3 8 0 I I I ( f ^ O G — » f I

Figure 76; ESR Spectrum (Solution Cell) of a Solution of [cb{(Me2NEthl)aMe2[16]tetraeneHi,}](PF*)a and 5,5'-dimethyl pyrroline N-oxlde in CH CN After Exposure to Oxygen, Room Temperature 182 with the spin trap unless dioxygen is present in the system."^

In addition, a genuine sample of pyrroline nitroxide radical does not give the same sort of signal when placed with a sample of the cobalt(II) complex. Both cobalt(II) and dioxygen seem to be necessary to observe this ESR signal at room temperature with the radical trap.

With the unbridged cobalt complex, (discussed in chapter III), which contains the rearranged ligand superstructure to produce a five coordinate cobalt(II) complex, [Co{(MeNHEthi) (MeNEthyl)Me%[16]tetraeneNs)](PFe):, a different signal is observed. The signal observed from a mixture of pyrroline N-oxide with an oxygen saturated acetonitrile solution appears to be a superimposition of two radicals (Fig. 77). The more intense signal is a triplet. This suggests that hydrogen abstraction has occurred on the carbon atom adjacent to the nitrogen which normally is responsible for the further splitting of the electron resonance. The weaker signal observed may be due to the same type of cobalt radical species as pre­ viously detected by Herron for the other unbridged cobalt complex (Ri=Ra=CH3 )

This complex radical signal is short lived and its signal is weaker in intensity than the triplet signal. The intense triplet signal can be ob­ served for a long time after the disappearance of the more complex signal.

In a water/acetone solvent (1:1) (v/v), nitrosobenzene produces a 45 line radical (Fig. 78). When nitrosobenzene reacts with the rearranged cobalt complex (Chapter III, Fig. 26) it was shown that a cobalt-superoxo dimer is formed in oxygenated acetone solvent. Perhaps this radical is a pro- tonated superoxo-dimer; the intensities of the signal vary in a 1 :2 : 1 pattern throughout the signal suggesting that two protons are coupled.

Spin traps have been used with the lacunar cobalt(II) complexes. 183

Figure 77; ESR Spectrum (Solution Cell) Observed for the Pentacoordinate Cobalt(II) Compound, Chapter III, with Pyrroline N-Oxide in Acetonitrile Solution After Exposure to Oxygen at Room Temperature 184

lit 1

* 3380 Gauss 10 Gauss

Figure 78: ESR Spectrum (Solution Cell) for a Solution of [Co{(MeNImino- ethyl)(MeNHEthi)Mea[16]tetraeneNs}](PF6)2 and nitrosobenzene in Water/Acetone (1:1) After Exposure to Oxygen 185

3 3 8 0 G

Figure 79; ESR Spectrum of 2,ô-dimethylphenol, [C6{(MeNIminoethyl)- (MeNHEthi)Meg[16]tetraeneN,}](PF,),, and N-tertbutyl o-phenyl nitrone in CHjCN After Exposure to Oxygen, Room Temperature 186

r

4 - 2 0 G M - »

3 3 7 0 G

Figure 80 ESR Spectrum of [ C o { (MeNIminoethyl)(MeNHEthi)Mes[16]tetraene- Hs}](PF6)a and N-tertbutylo-phenyl nitrone in CHaCN After Exposure to Oxygen, Room Temperature 187

The results do not show the formation of radicals in the cases where the ligand has been designed to contain a protective void. ESR spectra with the complicated cobalt splitting in the presence of the spin traps only occur with the cobalt compounds that have less sterically con­ strained ligands.

Spin traps have also been used to investigate the mechanism of oxidation of phenol, in the hope of detecting some of the intermediates involved in these processes. The use of N-tert-butyl-a-phenyl nitrone as a spin trap and 2,6-dimethylphenol (2,6-DMP) as substrate with unbridged cobalt(II) and oxygen resulted in a relatively simple pattern exhibiting a doublet of triplet (Fig. 79). This contrasts with the complicated pattern observed when 2,6-DMP is not present in these samples

(Fig. 80). This suggests that a radical from 2,6-DMP is trapped in these systems. V?hen nitrosobenzene was used instead of the o-phenyl nitrone traps, a red crystalline material was recovered from the aceto­ nitrile solution after standing overnight. This material analyzed well for the l-oxo-2,6-dimethylphenyl adduct of nitrosobenzene (Fig. 8l).

The NMR spectrum of this sample taken on a 300 MHz spectrometer

Figure 81; l-oxo-2,6-dimethylphenyl Adduct of Nitrosobenzene Figure 82: "H N M R Spectrum of the l - o x o - 2 , 6-dimethylphenyl Adduct or Hitrosooenzene 00 CgDe, Room Temperature 0 0 189 supports the structural assignment of a bismethyl-oxy-phenyl adduct of nitrosobenzene and suggests that the phenoxy radical added para to the initial hydroxyl group of the phenol (Fig. 82). The NMR spectral assignments imply that the phenyl ring of the nitrosobenzene is able to undergo free rotation while the rotation of the substituted ring

is hindered at room temperature in deuterated benzene solvent. The same phenoxy adduct of nitrosobenzene was found independently of the cobalt(II) catalyst used. Unbridged cobalt(II) complexes and lacunar

cobalt(II) complexes gave an adduct with the spin trap.

Nishinaga has isolated a para substituted complex as the product

of a reaction of trisubstituted phenol with a cobalt(II) complex and

dioxygen. This complex has been characterized as the one shown in

Figure 83 and contains a dioxygen-cobalt complex as an adduct to the

phenol substrate.^*®

Figure 83: Cobalt-peroxy adduct of a Substituted Phenol that was Isolated by Nishinaga 190

He has also used ESR to demonstrate that the monomeric cobalt-dioxygen complex can abstract a hydrogen atom from 2,4,6-triphenyl phenol. The disappearance of a cobalt(II)-dioxygen signal occured simultaneously with the growth of a phenoxy radical signal when the oxygen supply was limited. The simplest interpretation of these two results would suggest that the formation of the peroxy complex occurs directly from the combination of the phenoxy radical generated from hydrogen abstraction and the cobalt-dioxygen complex. However, Nishinago does not think that this is the mechanism for the reaction.Instead, he proposes that the formation of the peroxide complex is preceded by an equilibrium involving electron transfer from cobalt(II) to the phenoxy radical to give a cobalt(III)-phenolate complex. Dioxygen is then inserted into the complex to,form the peroxo species. This mechanism is illustrated in Figure 84.

L 5 C 0 - O 2 + ^ + L g C o d l ) + H0|

LsCo(XX) + 0* V

L 5Co(III)^p^ ^ O" O 2____ ^ l:sCO(III)-0— ^ =0

Nishinaga works with SDPT [bis (salicylideniminato-3-propyl)amine] and SMDPT [bis(salicylideniminato-3-propyl)methylamine] as the ligand for the cobalt(II) Metal'

Figure 84: Mechanism proposed by Nishinaga or oxidation of Substituted Phenols^ 191

In a contrasting mechanism. Drago proposed that the cobalt-dioxygen complex is involved in more steps than the initial hydrogen abstraction from phenol (Fig. 85). Using results (discussed earlier in the intro­ duction of this thesis) with the polymer-supported complexes of cobalt

(salen)-type (page 32), he reiterates that at high cobalt concentration in the polymer the ratio of the formation of benzoquinone to the formation of the benzoquinone dimer was high, while at lower cobalt con­ centrations the ratio of the products was substantially lower. The formation of the dimer was rationalized on the assumption that the phenoxy radical fails to coordinate with a cobalt center. In a separate experiment, a high concentration of phenoxy radicals were generated with Pb02 in the presence of dioxygen and bis(acetylacetonate) cobalt(II) (which does not form an oxygen adduct under the reaction conditions) and only the dibenzoquinone dimer was produced. This result suggested that the phenoxy radical needed to encounter a Co-Oj species in order for the quinone production to occur. The following steps proposed for c, d, and e are by analogy to the decomposition of an organic peroxide species, and by Drago's admission, must be viewed with caution since no experimental details suggest these intermediates.

The mechanisms proposed by these researchers are of great interest to our research interests since we are working with cobalt(II) complexes and similar oxidation substrates. If a similar chemistry can be assumed for the cobalt(II) metal ion, regardless of the ligands surrounding it, the chemistry we have observed in the spin trapping experiments may be useful. Several spin traps were used in experiments conducted in the 192

a) -CoClI) ----^ Co-0: Co(II) + HOa* + / \ ^ V

Co—Og b)

0-0—Co *— / \

\ if c) X — Co" OH * / \ / V

u

II d) Co""-OH + + + HaO

e) O-

Figure 85: Dragons Mechanism for the Oxidation of Substituted Phenols 193 presence of oxygen, 2,6-dimethyIphenol and a variety of cobalt(II) complexes and the results suggest several questions concerning both of the mechanisms proposed by the above mentioned researchers. In both mechanisms, the existence of H 02* and phenoxy radical are proposed early in the reaction scheme. The pyrroline N-oxide trap is an excellent trap for HOa* and has been used to identify this species in the chemistry of the iron(II) lacunar complexes. However, in none of the solvents investigated did we observe a similar species for cobalt(II) complexes.

When nitrosobenzene was used as the trapping reagent the adduct of nitrosobenzene and 2,6-dimethylphenol was isolated after twelve hours but the early signals observed in the ESR spectra do not indicate a phenoxy radical but rather a cobalt center (from the abundant number of lines in the signal). Furthermore, observation of the ESR signals of oxidation solutions after short time period and often more extended

periods (12 hours) never shows a signal from a five coordinate cobalt(II)

species once the oxygen adduct has been formed.

The conclusion that can be drawn from our spin trapping work is

that the formation of a phenolate radical does occur as an intermediate

in the formation of quinone products. A solid adduct has been isolated

with a nitrosobenzene and a simple organic radical was observed in the

ESR with N-tert-butyl-a-phenyl nitrone. The intermediate HOa« is

questionable since the pyrroline N-oxide trap fails to detect it in any

concentration in our experiments and is specific for HO*». The catalytic

regeneration of Co-OH or Co(00) are not in evidence in our ESR work. APPENDIX B

Table of Final F and F , Data obs calc

Table of Positional Parameters

Table of Thermal Parameters

194 TABLE OF FINAL F^^g AND ^CALC 0 0 1 3 0 0 3 6 12 194 - 191 10 3 1 4 3 3 - 1 4 5 6 10 10 3 5 2 3 1 8 10 1 169 - 1 4 3 16 e e L 6 3 7 2 3 6 4 6 10 139 148 14 4 2 0 0 8 - 2 0 9 6 2 0 1 3 L 16 2 8 7 . 2 7 0 1 1 2 2 7 8 - 2 7 5 11 7 4 3 0 - 4 4 7 5 5 6 0 0 6 4 3 9 17 21 1-223 13 3 153 158 18 4 7 0 9 - 7 9 0 0 il 5 2 9 - 0 4 3 6 0 7 L 6 801 3 0 2 10. ■20 2 0 0 150 15 18 180-123 14 G 157 - 1 6 4 18 6 2 3 6 5 - 2 4 1 9 17 9 3 1 5 - 3 1 4 6 7 7 4 9 - 7 6 2 10 •21 140 - 1 4 3 18 2 0 107 131 17 6 173 - 1 4 4 16 0 6011 6 1 2 5 10 106 130 10 7 1 42 - 1 0 0 13 8 1 1 0 9 - 1034 13 2 0 151 - 1 6 4 17 11 2 6 0 2 4 6 12 1 0 1 0 0 4 - 1013 9 12 4 0 4 - 3 7 6 6 9 134 122 15 9 199 - 1 7 3 II •III 231 -219 1 f 1 e1 L 9 144 - 1 2 9 19 14 2 0 9 - 2 1 0 6 14 5 3 3 521 7 10 7 8 7 0 3 4 12 17 2 8 5 - 2 6 8 10 13 451 - 4 4 6 6 13 3 5 3 3 6 0 7 0 U h 13 2 9 0 3 2 7 3 •10 2 5 3 2 3 8 10 - 1 4 3 8 9 - 3 9 4 11 1 8 L 22 201 221 14 16 3 3 3 - 3 5 7 6 14 5 0 7 5 2 9 9 ■13 3 7 7 - 4 0 6 9 - 1 3 3 1 2 - 3 1 2 11 13 3 5 7 3 4 0 7 0 19» - 2 0 6 16 16 129 - 1 3 6 14 •11 4 0 2 4 2 5 II - 1 2 331 3 2 8 lO - 4 137 129 2 0 0 1 L 2 0 155 - 1 3 5 12 1 162 127 13 17 190 2 0 3 12 •10 2 0 4 - 1 9 5 9 - 1 0 126 - 1 2 4 17 21 2 3 3 - 2 1 7 9 2 13» 122 10 111 2 1 7 197 1 I - 9 174 - 1 3 7 10 - 9 130 - 1 6 0 17 2 0 L 113111-1230 9 2 4 2 0 3 - 2 0 2 1 f 3 199 - 2 1 3 11 19 10» - 1 9 3 16 -71030 1022 14 -O 3 8 2 3 9 5 9 2 2 9 7 - 2 1 5 3 26 2 2 0 190 10 0 143 160 15 2 0 170 159 15 —6 6 4 1 6 1 7 9 - 7 120 123 17 - 2 0 196 - 1 7 7 13 3 7 3 9 7 4 9 6 6 184 139 12 21 1116 1611 15 - 5 8 3 4 - 8 6 6 12 - 6 4 2 8 - 4 3 4 9 - 1 8 4 7 8 - 4 4 0 9 4 1 6 1 3 - 1639 12 0 4, L 2 2 2 0 8 193 14 - 3 2 4 4 2 3 6 8 - 4 201 197 11 - 1 6 2 4 7 2 6 0 9 0 1 3 9 4 1436 10 1 0 L 2 3 151 153 19 - 2 3 7 5 3 8 6 7 - 3 4 8 6 4 5 7 9 - 1 4 135 157 12 7 2 0 5 177 6 0 2 4 3 2 9 3 0 2 4 2 0 6 - 2 6 2 12 -1 5 0 0 4 7 6 8 - 2 3 4 0 - 3 2 2 9 - 1 2 6 3 9 6 0 3 9 3 321 - 3 0 0 0 1 197 177 6 - 2 7 1 0 5 119 13 0 9 8 -OU 12 - 1 155 - 1 0 8 12 - 1 0 2 2 7 155 7 9 2 3 6 - 3 0 3 6 2 4 1 7 460 5 - 2 5 2 2 0 2 6 0 14 2 2 0 9 2 2 4 9 0 4 1 6 41 6 II -81300-1304 14 10 7 2 0 7 1 6 U 3 2 1 2 2 0 9 7 -22 005 5 7 2 I I » 9 1 2 - 9 1 1 12 2 5 5 8 - 5 5 8 9 - 0 4 5 3 4 3 9 7 1 1 -1 2 0 115 10 4 5 9 4 - 5 9 9 6 - I V 0 2 2 5 3 4 10 2 4 186 - 1 6 3 16 4 5 3 4 —56 6 9 3 2 9 2 -283 9 -42419 2374 2 4 12 435 -430 6 5 1 0 1 3 - 1 0 1 8 10 - 1 7 7 0 2 7 1 2 II ■23 2 9 2 - 3 0 0 12 0 5 1 8 490 9 4239 203 10 - 2 8 7 3 - 8 9 7 9 14 143 132 9 6 2 5 9 - 2 5 4 5 - 1 5 5 1 0 3 4 2 II ■22 160 2 1 1 1» 6 2 4 3 - 2 7 3 II 5 2 3 6 2 7 0 11 0 1 1 9 9 116 4 12 10 41 1 3 9 9 6 7 4 2 4 4 1 3 9 -13 120 - 3 0 13 ■21 143 120 18 7738-741 10 6 151 - 1 8 7 14 2 2 8 2 4 2 8 2 8 3 0 16 2 5 0 - 2 6 6 7 0 9 6 - 1 14 12 - I l 0 2 4 •290 7 20 3 0 3 - 2 9 1 11 8 137 - 1 0 1 12 7 150 8 7 13 4 9 5 8 -9311 11 17 102 -215 10 10530 5511 7 - 9 7 0 9 0 9 6 10 •17 4 4 2 4 5 3 9 9 3 3 9 3 6 4 II U 491 501 9 6 108 - 6 3 I I 19 4 4 4 4 3 6 7 11 5 7 3 5 9 7 7 - 7 I 0 « 204 9 •16 209 -187 12 M 4 2 0 - 4 3 2 il 10 3 1 9 - 2 9 6 9 8 8 9 2 4 0 4 7 2 0 - 1 3 0 - 4 5 14 12 139 - M i l 10 - 5 U O O 3 1 3 10 10 5 8 9 -U 6 8 9 12 2 2 3 2 0 9 10 II 2 7 4 - 2 7 4 11 10 661 - 6 4 4 9 21 4110 —130 7 13 3 0 9 - 3 7 8 7 -0 011 - 3 6 6 •14 to n 131 1 1 13 4 7 9 4611 9 12 361 3 4 0 10 14 6 0 3 - 6 3 0 10 2 2 341 31*3 0 14 3 3 3 3 3 7 7 3 0 0 9 9 7 6 10 •13 108 - 5 4 10 14 149 -101 13 14 361 - 3 7 3 10 16 3 3 7 3 3 3 II 2 3 2 7 0 2 7 6 9 16 9 1 4 - 3 2 3 7 5 161 - 9 4 9 •12 3 0 4 - 3 6 9 7 15 159 - 1 3 2 13 1 8 2 0 8 2 0 6 14 2 0 179 184 15 2 5 134 - 1 5 7 11 2 0 136 113 14 7 0 9 0 •013 10 -11 106 - 1 4 3 14 17 3 1 6 2 7 6 10 2 7 2 0 2 130 11 0 0110 7 3 0 1 I 10 3 1 » 33 5 8 19 199 - 2 2 7 13 1 6 L 2 1 L 0 0 L 13 un 194 10 -I l 2 9 3 -3 ::» 7 2 0 189 2 0 8 15 0 a 1 L 1 0 1 9 3 193 10 - 7 4118 4 7 0 U - 1 9 153 125 2» -27 167 179 20 1 3 4 2 - 3 4 4 6 1 7 4 3 7 5 0 2 10 - 6 2 1 0 20V 7 1 4 L - 1 7 170 - 1 6 » 17 -25 200 -273 18 0 6 1 2 3 3 9 9 2 2 9 4 2 1 » 6 1 9 0 0 5 5 6 4 10 - 0 196 195 7 -13 133 104 20 -24-26 1 16 13 1 3 4 7 3 1 3 4 4 421 - 4 1 2 6 2 1 : M 3 4 0 6 10 - 4 1021 -96(1 12 ■20 145 -16» 19 -13 255 - 2 1 5 13 - 2 3 195 187 14 2 3 5 121 10 0 4 7 0 4 6 4 6 2 3 2 1 5 2 1 3 13 2 1 0 2 2 0 7 1» 2 3 » 238 13 -11 249 241 11 -22-15» -16 17 3M O O 11 6 < 96 0 3 3 7 2 3 3 2 0 2 3 2 12 - 2 1061 É 0‘ 8 12 ■16 2 2 5 - 2 0 9 12 -7 275 304 12 -21 132 -2 :1 0 2 0 4 1 0 9 2 1651 9 9 24» 2 7 4 II 2 7 1 7 2 133 13 - 1 19 7 9 - •1954 2 0 15 2 1 5 2 1 7 13 - 6 127 151 19 - 1 9 6 9 0 6 9 4 12 5 6 5 3 - 3 7 0 5 10 2 4 0 1(15 U ( 7 0 » VU9 10 •14 4 8 9 531 9 -5 171 -115 14 -17 322 -333 8 6 4 7 3 - 4 3 3 6 1 I 4 7 2 - 1 5 9 7 1 1 L 1 1 3 9 9 -1 3 6 2 15 •13 157 -132 14 - 4 3 5 0 -357 10 —16 167 162 11 7 3 3 3 3 6 3 U 12 3 1 3 - 3 4 5 7 2 7 8 8 - 7 7 1 10 12 4 9 2 -4911 10 - 1 156 - 1 5 0 15 - 1 5 1 14 146 14 Il 3 9 9 4 1 4 5 14 2 2 0 2 1 7 9 - 1 9 2 9 0 3 2 6 31210 1185 14 10 2 1 7 2 0 4 10 0 2 1 9 199 11 - 1 4 2 2 2 - 2 5 1 9 9 0 3 9 - î ‘63 0 15 136 - 2 0 5 10 - 1 3 2 7 6 - 3 3 3 4 7 9 5 8 3 5 11 - 9 3 2 0 - 3 1 9 8 1 156 142 15 - 1 3 4 5 0 -4116 II 10 691 - ( 6 5 7 17 3 4 0 3411 II - 1 7 - 1 0 3 - 1 0 4 5 4 0 8 - 4 0 7 7 -U 3 0 8 - 3 1 8 9 3 157 121 15 -12 222 2 3 2 9 I l 101 135 11 19 124 - 1 14 16 - 1 6 3 1 I - 3 2 I «> 3 3 7 - 3 3 0 7 - 6 3 5 5 3 4 5 11 4 147 143 16 -11 3 2 4 3 1 4 7 12 149 174 U 2 0 .161 - 1 5 3 I I - 1 5 - 1 5 4 13 V 3 4 9 3 6 0 7 - 4 2 0 4 183 9 6 3 1 6 - 2 7 0 10 - 1 0 (6 9 - 1 5 3 9 13 109 - 169 7 21 154 135 1 4 -12 430 —44 6 n 189 180 9 0 155 — 166 12 7 14» 129 16 - 9 3 0 5 - 3 0 9 7 14 2 5 9 - 2 5 3 6 2 2 127 115 17 - 1 0 4 2 9 4 0 3 9 3 3 8 3 4 6 » 2 146 - 1 5 7 I I 9 3 6 0 - 3 3 9 10 - 7 141 171 9 15 0 3 4 5 0 9 6 - 9 2 5 0 - 2 4 4 10 3113 - 3 5 9 7 3 2 1 6 - 2 0 8 9 1 1 160 158 10 -6 391 360 7 16 110 -151 13 0 6 L -Il 5 6 5 - 5 5 4 11 135 152 12 4 2 3 3 - 2 2 0 9 13 2111 2 2 7 14 -51118-1050 12 111 2 1 7 196 3 - 7 9 3 9 9 3 2 12 4 6 7 4 3 9 U 0 189 - 1 7 5 10 19 168 132 19 - 4 2 6 3 - 2 8 3 7 2 0 109 153 12 0 4 0 8 - 4 1 6 10 - 6 1 1 4 103 12 13 139 166 12 6 4 7 4 - 4 7 2 9 -31431-137418 21 2 3 2 -2115 9 2 441 4 3 9 7 - 5 5 9 0 6 1 0 9 14 2 7 2 - 3 1 3 9 7 2 4 7 2 4 8 9 1 ? L -21130-1070 12 2 6 IIU - 1 2 3 m 3 152 175 11 - 4 7 7 5 - 7 6 3 10 17 4 8 3 4 6 6 10 8 3 6 7 4 0 2 8 -1653-6598 4 2 3 9 - 3 4 5 7 - 3 1 2 3 0 1 1 6 3 13 18 3 1 3 - 3 0 0 1 1 9 109 - 7 5 16 - 1 2 175 127 16 01459-1089 18 0 01 L 5 3 3 9 3 4 9 7 - 2 3 0 7 3 3 0 4 4 3 2 19 193 - 1 9 4 14 10 2 2 4 - 2 4 0 10 - 1 0 2 7 6 - 2 7 » 12 1 9 1 3 1155 10 6 2 1 0 1113 9 - 1 1 6 5 5 - 1674 17 2 0 3 1 6 3 3 2 1 1 1 1 155 151 13 -4 231 23 1 12 2 2 3 7 2 2 7 7 1 5 3 3 5 7 7 6 7 2114 - 2 9 5 U 0 1 9 0 7 - 1939 19 2 2 2 3 0 - 2 3 0 11 12 5 2 2 5 3 8 10 - 2 2 7 4 - 2 6 8 12 3 1160 8 3 4 10 a 3 0 1 299 5 B 3 6 2 - 3 6 4 7 t 3 3 4 - 3 6 9 6 2 4 2 5 2 221. 13 13 144 128 16 -I 1113 192 10 4 8611 5 7 6 9 4 3 0 4 3 0 3 0 10 4 2 0 4 4 7 7 2 2UI 136 7 2 6 148 - 1 0 0 2 0 14 0 9 3 - 5 9 9 11 0 4 2 8 4 2 4 10 6 2 0 6 2 0 3 7

vo Ln 5 0 6 9 - 5 4 2 1 - 4 2 0 7 1 0 9 2 0 6 n - U 1 0 6 6 7 1 4 - 9 4 41 1 4 1 7 9 - 2 7 3 3 2 - 3 1 4 1 2 1 5 1 » 9 B 5 6 3 - 5 7 1 tt - 7 1 1 7 1 3 6 1 3 -Il 21111 2 5 1 1 9 - 2 3 1 4 4 - l i a 1 » 2 L 2 1 2 2 9 5 1 3 - 4 4 4 5 4 2 5 A - 9 3 5 1 3 1 0 7 1 0 6 9 9 6 9 6 9 - 6 2 9 4 - 3 1 4 11 - 7 1 4 9 - 1 6 1 1 1 4 - 2 1 0 1 1 3 0 4 1 0 3 9 9 5 - 3 1 7 5 1 0 1 2 2 4 0 - 2 4 4 n - 5 3 3 3 3 5 3 7 - 6 1 5 5 - 1 0 7 1 3 - 1 9 1 9 3 - 2 1 0 1 9 -26 102 169 26 4 2 0 1 2 2 0 9 - 1 3111 - 3 7 0 9 3 3 4 3 2 6 1 9 2 1 0 2 1 5 1 0 - 4 3 6 7 - 3 1 4 a - 5 2 4 4 - 2 1 1 1 1 0 - 1 7 6 4 1 ) 6 2 » 1 1 -24 002 -274 11 r> 9 6 3 1 0 3 6 1 3 0 9 1 6 - 1 1 0 2 6 1 7 - 3 9 9 4 - 9 r 1 1 3 - 4 3 1 4 2 9 1 II - 1 5 4 9 4 - / / 9 9 -2 2 2911 304 11 6 2 5 7 2 6 5 a 1 4 1 1 4 3 5 9 1 7 7 1 3 - 6 ' > 3 1 2 - 2 1 7 3 - 1 / 7 9 - 3 5 2 6 - 4 9 5 9 -13 3116 3f U 1) -21 1711 193 15 7 3 5 1 - 3 3 3 a 4 1 9 7 1 9 9 11 in 3 2 7 2112 1 0 - 1 2 9 7 -21 6 7 - 2 2 9 9 - 3 0 1 1 1 0 - 9 0 3 7 s u a II -2 0 324 -319 10 9 3 4 5 3 3 » 0 0 2 3 0 2 2 0 11 21 3 2 0 - 3 1 6 11 0 2 2 1 212 11 0 1 6 7 - 1 1 ) 5 1 2 - 7 3 9 2 3 : 9 6 -19 223 -207 12 1 0 1 9 0 - 2 3 4 1 0 6 1 5 0 - 91 ) 1 4 2 3 3 3 0 3 0 9 11 1 5 4 7 5 4 2 9 1 3 01 1 2111 U - 5 3 1 2 2 / 0 1) -II) 213 191) 12 1 1 2 7 6 - 2 ( * 4 9 7 292 -2‘i4 lO 2 4 1 9 7 -212 1 6 3 2 1 3 - 2 1 4 11 2 2 6 7 2 9 4 9 - 3 5 0 6 - 5 3 1 1 0 -17 232 265 10 12342-332 9 U ISO 1 6 6 1 5 2 9 1 7 7 - 114 1 6 4 4 3 7 -412 11 3 3 1) 2 3 6 4 9 - 1 2 2 9 1 9 2 6 -16 223 -240 10 1 3 2 2 9 2 1 » 1 0 9 2 0 0 2 2 0 1 0 5 1 17 -102 1 3 4 2 0 6 - 1 7 6 11 1 6 4 9 o r o 9 -15 109 -100 16 1 0 2 3 9 - 2 4 5 11 1 0 2 7 1 - 2 9 1 II 2 2 L 6 2 7 4 2 7 6 0 6 1 9 0 1 6 4 11 3 5 91 1 - M 3 I 9 -1 4 3U6 400 a 16 140 145 16 11 2 9 6 - 3 ) 3 11 7 2 1 9 - 2 4 6 9 7 5 0 7 - 4 6 4 l O 0 2 4 9 * - 3 4 4 1 0 - -1 3 036 5(11 10 1 2 2 1 6 2 *2 1 2 - 2 2 2 1 0 2 1 0 1 3 1 0 5 7 9 5 7 3 9 1 2 2 6 3 - 2 4 3 1 2 7 3 1 3 - 3 1 1 7 -12 377 -363 7 3 4 L 1 4 1 5 3 - 1 2 7 1 7 - 2 0 1 9 9 - 2 0 4 1 3 12 3 3 6 - 3 0 4 9 1 3 3 1 1 1 3 3 4 1 0 9 1711 1 9 1 1 0 - I l 490 —444 a 1 5 1 3 5 - 4 1 1 9 - K l 2711 -26 1 10 1 3 1 19 -71 1 1 6 1 5 2 3 4 - 2 0 1 1 3 11 176 -11*4 9 -1 0 423 424 u - 2 6 1 6 3 1 5 7 2 0 1 7 1 0 7 1 5 6 1 0 - 1 6 1 2 9 - 1 0 7 1 4 1 4 5 0 9 4 0 6 1 0 19 1411 - 1 9 5 2 0 1 3 1 3 2 1 0 3 1 3 -9 i4n 16U 11 - 2 5 H I 7 - 1 6 » 1 7 - 1 9 3 4 4 - 3 4 9 0 1 6 4 4 6 - 1 2 9 9 2 0 1 4 2 - 1 2 0 2 1 I S 7 7 1 ) - 7 6 3 11 -Il 296 -303 7 - 2 4 M O -ISO 2 1 0 6 L - 1 2 1011 1 2 0 1 4 1 7 216 - ! 1 9 1 4 1 7 2 4 2 2 7 4 11 -7 429 -433 7 - 2 2 3 1 0 3 2 3 1 2 - 1 0 2 1 7 - 2 2 2 11 2 0 1 3 6 - 1 2 4 1 9 2 (> L 1 9 1 2 9 - 1 0 6 1 9 -6 466 431 7 - 2 1 1 4 6 1 0 9 1 » - 1 9 1 6 4 1 0 9 1 9 - 9 9 3 2 9 0 1 1 3 2 2 1 7 9 1 7 5 1 7 2 1 6 3 7 6 2 0 II -B 124 76 10 - 2 0 2 3 3 - 2 0 5 1 2 - 1 7 1113 - 2 1 2 17 - 0 9 0 1 1 176 1 2 2 4 1 6 3 - l U O 2 1 • 1 4 2 1 0 2 1 3 1 4 2 3 1 9 5 - 2 2 1 1 5 -4 S110 -532 II -19 403 390 10 -11 1 9 6 1116 1 4 -7 oau - 9 0 6 n • 1 2 1 6 7 - 1 4 1 1 6 -3 623 571 10 - l U 1 4 5 - 1 1 5 1 7 -Il 2 3 0 - 1 9 6 11 - 6 6 0 2 - 3 o 7 u 2! 4 L • 1 0 3 1 0 21)5 11 3 1 L -2 277 275 6 - 1 7 1 9 3 - 2 0 9 1 0 -Ü 1 6 » - l a i 1 4 - 9 6 6 1 6 6 0 1 0 - ü 3 2 5 - 3 3 2 1 0 -1 743 737 10 -16 32» -309 10 - 0 1 2 5 - M O III - 4 3119 - 3 9 1 1 6 - 2 1 2 0 7 2 1 1 1 4 - 4 2 114 - 2 4 7 1 0 - 2 3 1 0 9 1 5 3 1 7 01006 997 12 - 1 4 4 1 7 4 1 7 9 - 2 1116 1 7 9 13 - 3 3 0 0 -2114 7 -17 197 1 9 3 1 3 - 3 I 7 U - 1 6 1 ) 1 3 -22 105 -159 16 1 321 334 7 - 1 3 2 2 0 - 2 0 9 1 2 1 2 5 5 2 7 6 1 0 - 2 7 0 6 - 7 1 9 1 0 - 1 9 3 7 3 - 4 0 1 1 0 - 2 4 1 1 4 0 9 1 0 - 2 0 101) - 1 4 5 1 0 2 410 -410 a - 1 0 1 4 4 1 9 2 1 3 2 1 7 9 - 1 0 9 1 3 -1 2111 -227 7 -14 3 9 9 3 5 9 9 - 1 2 3 3 2 2 4 11 - 1 9 1 4 2 1 0 7 1 5 4 1(10 1110 a - 9 2 4 3 - 2 7 7 9 3 1 6 2 1 7 7 15 0 2 6 4 205 6 -13 2113 2 7 3 9 0 4 0 9 - 4 3 9 l O -III 2 4 7 2 6 7 9 6 350 -309 a - 7 2 2 1 1 9 6 9 0 1 2 6 - 1 3 2 1 0 1 6 9 2 - 6 9 7 1 0 - I l 3 9 B - 4 1 1 n 1 1211 -140 III - 1 7 3 0 3 - 3 0 9 U 7 060 -6211 10 - 6 3116 - 3 7 2 a 9 1 9 7 - 1 9 9 1 4 2 7 0 3 - 6 1 1 5 1 0 - 9 6 0 2 6 1 7 11 4 2 2 0 - 2 0 2 1 2 -14 265 271) 0 0 4 1 0 409 a - 4 0 5 3 - 3 5 5 II 3 1 0 0 1 1014 12 -0 499 - 4 9 1 1 9 6 3 7 4 3 7 5 11 -13 226 -221 B 9 113 1311 13 - 3 1 0 1 lai 1 0 3 7 L 4 6 9 3 7 1 0 1 0 - 7 3 4 0 - 3 3 1 1 il 7 1 6 5 - 1 6 1 1 6 -12 172 -21(1 II 10 372 -414 u - 2 1 0 » 7 6 1 4 5 3 5 2 -5114 11 —6 256 - 3 0 3 II O 2113 - 2 1 1 9 1 1 - 1 1 7 3 1 1 6 116 1 2 12 1 1 6 911 1 5 - 1 1 9 2 1 7 3 1 0 - 1 0 3 2 0 - 3 0 5 11 6 3 6 6 - 3 6 4 7 -5 7 7 6 7 7 7 1 1 1 0 2 7 4 2 113 1 2 -10 731 721 il 13 5 1 0 411» 9 0 3 7 9 3 9 3 a - n 2 2 1 2 1 9 1 4 7 4 1 7 4 0 4 7 - 4 1 9 2 1 5 7 1 0 11 1 6 3 1117 1 7 - 9 1 2 2 - < « 1 0 14 237 -212 II 1 3 6 1 ) 3 6 3 a - 6 1 3 6 - I I I 1 9 fl 4 3 0 4 1 9 fl - 3 4 6 2 - 4 0 5 9 1 2 2 0 7 - 2 0 1 1 4 -Il 1 9 7 - 2 6 3 (1 15 426 -435 9 2 2 0 4 2 3 1 lO - 4 3 3 2 3 0 7 1 0 9 2 111 - 2 3 2 II - 2 51)11 6 2 4 1 0 1 7 1 5 3 - 1 3 5 2 0 - 7 211 1 2 1 6 7 16 163 167 14 3 1 2 0 1 3 3 1 3 - 2 2116 - 2 1 1 5 1 2 1 0 4 11 1 - n u 11 - 1 5 1 2 - 5 1 0 a - 6 4 0 5 4 0 2 7 111 332 -3 2 » 10 4 161) 11)5 lO O 2 0 3 1 9 1 1 4 1 1 3 9 7 3 6 7 U 0 6 5 0 - 6 1 1 1 7 2 7 L - 5 4 2 1 - 3 9 4 7 19 271 -215 1 0 5 3 » 9 - 3 6 7 a 2 2 211 - 2 0 2 1 2 1 2 121 120 13 1662 - 7 1 4 1 0 - 4 1 123 1165 1 0 20 271 299 13 7 1 6 3 1311 11 3 1 7 3 1 2 6 1 5 1 3 260-214 9 2594 - 5 2 1 1 0 - 1 7 1 7 5 1 6 0 III -3 1200-12311 13 22 142 -160 20 1 0 5 4 2 - 5 2 3 1 0 4 2 4 2 2 3 1 1 2 1 3 1 2 7 1 0 2 1 5 3 5 2 1 1 4 41 1 9 II 1112 - 2 0 3 1 7 -21030 1013 11 24 210 203 10 1 2 3 3 2 3 3 4 9 II 2 0 5 1119 1 5 1 6 1 4 0 6 6 1 5 9 2 7 0 - 2 9 6 U - 9 1 5 7 1 6 5 l U -1 70(1 -6CII 9 1 3 2 6 3 - 2 3 0 1 1 14 144 -100 21 1 7 3 3 3 - 3 5 3 9 6 5 3 6 0 4 0 9 9 1 4 5 1 2 0 1 9 O 11112 - D O S 1 0 3 3 L 1 4 1 2 9 - 1 3 6 in l U IIKI 132 14 7 700 lit 3 11 1 1140 o::6 l O 1 0 1311 1211 1 » 0 U L 1 9 1 4 3 1 7 3 111 •• n 1 9 9 2 1 1 11 îi 0 L 2 2115 - 2 9 7 7 -21 337 -319 10 1 6 2 0 3 2 2 2 1 3 2 2 3 7 7 - 3 7 2 1 1 9 2 3 7 - 2 0 2 9 3 9114 1005 12 -20 191 -203 13 1 » 1 4 3 - 1 2 1 1 1 9 - 0 1 4 4 9 0 2 0 1 0 2 9 5 2 (1 1 9 - ü 1 4 6 - 1 5 0 2 1 4 2 0 9 - 2 2 1 7 -III 270 234 11 2 2 1 0 2 - 1 1 5 1 9 2 3 L 1 2 4 6 4 - 5 1 3 9 - 7 1 9 7 1 9 5 1 7 6 3 31 1 3 3 0 7 - 16 339 330 9 1 3 3 5 9 - 3 5 1 1 0 - 3 1 6 0 1 2 6 1 0 7 6 1 7 - 6 2 9 9 - 1 5 446 4 5 » a 5 L - 2 4 2 2 7 2 1 3 1 4 1 9 1 6 2 1 3 6 1 5 - 2 1 5 3 1 5 0 2 0 II 4 6 2 - 4 4 1 7 -1 3 299 -29» II - 2 0 4 9 0 5 1 2 9 - 2 2 170 - 1114 17 23 174 -ICO 1 0 - 1 1911 - l < 9 1 6 9 1 7 9 1 4 9 9 -12 166 - 1 5 5 11 - l U 1 6 7 - 1 7 4 1 6 - II I 4 9 4 - 4 6 5 9 - 2 0 2 2 0 221 1 1 2 0 1 9 1 - 1 9 6 1 6 1 0 1 10 1 0 5 1 3 - I l 397 396 9 - 1 6 3 0 9 2 9 3 1 1 - 1 6 3 9 1 3 6 3 lo - 1 9 2 1 3 1 7 3 1 2 2 0 L 1 1 6 1 1 4 1 1 0 11 1 9 1 1 6 1 1 0 - 9 7 7 2 - 7 9 1 10 - 1 5 1 6 1 1 3 9 16 - 1 4 9 1 1 7 - 1 0 6 6 1 4 -III 351 1 - 3 1 1 0 1 1 3 11)7 - 1 9 4 1 7 1 2 2 7 V - 2 6 5 II -Il 3117 39» 7 - 1 3 2 9 1 - 2 9 0 1 1 - 1 0 7 1 1 - 6 5 7 II - 1 7 1 3 4 1 16 1 4 - 2 3 1711 - 1 6 3 I V 4 1 4 2 - 1 3 2 2 1 1 3 1 2 2 1 1 6 1 4 - 7 7 1 0 61)9 10 - 1 2 2 3 4 2 4 3 11 -Il 1474 1461 1 6 - 1 0 4 2 2 4 3 9 9 - 2 1 1 4 1 1 7 5 2 1 5 1 U 3 1 6 0 1 7 1 4 1 5 9 1 6 2 1 2 - 5 1)112 -1)43 13 - I l 1 7 2 1113 1 4 - 6 I 4 0 S 1 3 3 7 1 5 - 1 9 1 3 4 1 5 7 1 3 - 2 0 11(1 1 13 1 9 1 6 2 01 1 1 4 7 1 0 - 4 246 -236 a - 1 » 21)9 - 2 7 5 1 0 - 4 1 7 4 1 2 2 9 - 1 2 * «36 « 2 4 1 1 - 1 7 U 1 4 1 6 7 1 5 3 0 L l Y 3 91 1 3 6 7 1 0 -31001 1603 9 -Il 1 2 0 - 1 2 5 1 6 - 2 1 1 2 6 2 - 3 1 9 3 3 4 - 1 1 1311 - 1 7 3 1 1 - 1 5 3 3 9 - 3 3 4 1 0 19 1 7 0 - 1 7 2 1 5 -1 6 1» -63» 9 - 7 4 6 2 4 6 2 I) 01473- 141)9 1 5 - l O 4 3 2 - 4 7 Ü U - 1 4 3 1 2 - 1 9 6 1 2 -39 207 130 10 2 0 l U U - 1 9 9 1 0 0 3U0 -39» a - 6 0 7 0 - 0 0 9 9 2 1 1 2 7 1 1 4 2 1 0 vo o> - 0 5 1 9 0 1 1 9 3 9 1 2 - 1 0 5 5 5 - 3 6 8 9 - 1 5 2 3 6 - 2 6 2 1 2 -4 I6(f -170 15 - 1 5 4 9 9 5 0 9 9 4 4 0 8 4 5 8 8 - 8 3 4 7 7 4 - 3 7 2 l ino 0 6 9 - 9 7 6 1 1 7 2 » 1 2 - 1 » :i9i 4 1 1 9 - 2 2 6 4 2 9 1 11 - 1 4 4 3 2 4 4 3 8 n 1 4 7 1 11 1 2 -r» 3 -^ 8 9 6 II 1 4 1 2 6 0 II - ü 2 2 7 2 1 7 II - 1 2 4 0 9 - 3 8 4 9 - 1 1 4 8 1 2 3 1 5 - 1 2 8 5 9 - 8 5 9 1 3 6 4 2 0 - 4 1 0 8 - 4 1 0 8 1 5 8 2 m i 3 1 4 2 9 8 u o n - 3 4 9 II - 7 1 1 7 3 - 1 1 4 1 1 4 - 1 0 1 8 4 1 6 7 1 0 0 2 7 » - 2 6 2 1 1 - 1 1 6 6 2 6 9 1 9 7 1 6 9 - 1 2 8 11 - 2 9 1 0 6 8 2 4 0 1 - 3 7 1 1 6 0 l O — 6 9 2 8 2 1 » - 9 1 8 6 2 0 1 11 1 1 9 » - 2 2 9 1 » - 1 0 1 4 3 1 1 3 7 9 II 9 6 6 0 1 0 1 8 1 2 I9CI - 4 2 8 1 1 9 1 4 1 1 6 - 1 3 1 1 3 - 5 1 2 7 2 1 2 6 7 1 4 -Il 1 4 0 1 5 0 i » 2 1 7 7 1 5 5 1 4 - 9 6 9 6 - 7 0 2 11 1 0 4 2 8 8 0 - 7 4 1 0 2 0 7 - 1 9 9 4 1 9 7 2 1 6 1 6 4 4 0 4 3 1 9 - 4 6 0 » 5 9 8 II - 7 8 1 5 - 8 2 6 11 4 2 0 0 - 2 0 9 1 » -Il 9 2 0 - 9 0 1 1 2 II 1 0 11 1 2 1 0 9 1 5 6 1 2 6 2 3 1 - 3 3 » l O 1 0 2 1 1 1 9 0 1 2 - 3 8 8 4 - 8 8 0 11 - 6 5 6 3 5 9 1 9 6 2 5 0 2 6 1 1 1 2 - 7 3 2 2 2 8 3 7 14 192-184 12 7 r.oi 2 1 1 11 2 2 1 0 3 - 1 1 4 1 8 - 2 9 7 0 - 9 16 1 2 - 5 f 0 5 6 7 7 1 0 7 1 6 9 1911 1 5 - 6 2 1 8 2 1 1 7 - 1 4 5 6 4 3 2 8 — 4 » 9 » - 3 8 1 8 II 25» -240 12 - n 8 0 2 - 8 5 8 11 1 3 3 7 1 - 3 6 4 9 1 0 3 1 9 - 3 2 7 1 0 2 7 1 2 6 6 1 2 1 2 1 3 6 1 3 7 1 8 4 1i L 0 2 7 7 2 5 8 6 - 3 3 9 8 - 4 1 0 8 l o 1 11 1 0 0 2 0 - 4 1 8 4 5 - 9 9 2 1 2 1 6 1 1 3 2 7 - 1 3 3 8 1 5 - 2 5 1 4 5 0 4 9 1 4 1 9 0 1 3 4 1 9 - 8 2 8 2 3 0 2 7 1 7 1 3 1 1 3 5 1 9 1 3 2 9 9 - 2 9 1 1 1 i n o - 1 0 7 i n 2 1 3 1 lo:! 11 - 1 2 3 1 1 9 4 9 - 2 1 8 5 1 8 8 8 1 8 1 9 9 - 2 0 9 1 4 1 5 2 1 3 2 1 » 1 4 - 2 6 2 1 0 2 1 9 1 6 9 2 0 2 J O » 8 0 8 7 7 2 11 0 5 7 0 - 5 9 7 9 4 7 1 - 1 2 2 2 2 4 3 7 20 229 221 1» 16 1 4 - 2 a 2 1 III» 1 7 2 1 6 1 8 1 9 4 - 2 0 7 1 6 - 2 1 1 0 1 - t o o 1 0 4 8 0 8 7 3 3 1 2 1 7 5 2 - 7 5 9 11 O 7 9 9 - 7 6 6 It 2 2 1 4 6 - 1 5 7 2 1 - 2 0 2 6 » 2 0 1 11 5 1 6 8 0 - 1 0 3 0 1 4 2 2 2 8 1 9 5 9 -15 228 -211 14 1 6 4 6 6 4 6 1 0 1 0 2 3 1611 - 1 1 9 1 8 tP Ci L -l'I 2 0 7 2 0 4 1 1 6 3 8 1 - 4 2 8 8 » 9 2 1 9 4 8 1 4 - I l 1 5 1 - 9 4 1 0 2 6 7 9 6 0 9 0 3 9 - 0 7 0 1 0 7 5 1 9 5 1 7 1 0 4 3 5 » 3 9 4 8 - 1 0 1 4 1 1 2 0 1 9 » 7 8 9 7 9 0 11 -lü 5 3 1, -22 202 -177 15 - 1 7 C16II -! L 1 3 4 4 7 - 4 3 4 0 - 5 2 31 1 1 9 2 8 - 4 1 4 8 - 1 2 7 1 » -1 1 7 2 - 1 6 7 9 - 1 9 3 2 2 - 3 2 1 9 0 3 1 8 2 7 5 - 2 6 8 8 - » 2 9 8 2 9 0 9 » 3 3 » - 3 3 3 7 - 1 8 2 7 1 2 4 » 1 0 9 4 6 1 4 9 8 9 1 4 u n » 9 -» 1 6 2 1 9 1 9 9 2 9 9 1 1 - 2 3 0 7 3 1 8 7 - 1 4 0 2 4 0 5 9 5 1 7 1 1 2 2 9 - 1 7 2 1 0 1 6 8 1 0 l e 1 8 9 - 1 5 6 11 - 2 1 - 1 6 1 1 0 - 2 0 1 4 5 - 9 6 2 0 3 3 9 - 3 2 0 1 1 0 6 4 9 1 4 1 2 3 3 2 5 8 l O 7 3 7 2 - 3 1 1 8 H - 1 0 3 3 0 - 3 1 0 O 1 1 4 7 1 - 4 : 1 3 9 17 ti 1 2 4 1 0 1 1 7 - 1 9 2 1 6 2 2 5 1 5 1 9 4 2 4 4 1 1 1 9 » 2 0 9 - 2 0 4 9 2 1 2 6 1 4 1 1 6 9 :176 3 6 6 8 - 1 5 2 5 7 2 4 7 9 1 2 3 1 6 3 0 0 1 0 -17 268 -25 V 12 2 0 - 2 1 9 1 6 4 4 3 6 - 4 4 3 » » 5 3 1 - 5 1 2 l O II 4 4 0 - 4 6 2 1 0 - 1 4 1 8 0 1 7 5 1 0 1 4 I9B 1 5 - 1 5 1 5 0 1 0 4 II) 2 1 1 6 6 - 1 6 7 1 6 5 f fO - 1 1 4 1 4 4 2 2 0 - 2 2 4 1 1 1» 4 4 9 4 6 0 8 - 1 » 5 0 6 4 9 7 9 1 7 1 8 1 1 8 4 6 6 0 2 5 8 4 7 5 3 0 8 - 2 9 0 9 1 5 3 9 7 - 3 7 4 9 - 1 2 8 4 0 - 8 8 8 1 3 1 0 1 3 6 - 1 2 3 2 0 -9 200 -210 13 2 3 1311 1 3 4 2 0 2 1 2 3 3 2 3 0 7 8 9 2 4 0 2 3 2 11 1 7 4 0 2 3 , 0 10 - I l 5 8 1 -5li 1 1 0 - 8 2 1 7 1 3 2 0 1 0 7 -100 2 0 7 - 7 1 7 8 2 5 6 - 2 7 8 9 1 3 :!22 2 4 1 f» 1 9 1 5 6 - 1 4 3 1 7 - 1 0 1 2 1 1 2 8 1 4 11 4 L 1 3 7 8 6 9 2 6 2 3 2 9 9 1 5 2 4 9 -256 1 » 2 1 2 0 5 3 0 6 1 4 - 9 4 7 2 4 1 » 0 0 1 6 0 - 1 5 0 1 6 2 L 1 1 7 8 - 1 8 1 1 4 1 0 4 1 7 4 2 2 9 1 8 1 4 2 1 0 6 2 0 2 » 2 9 4 - 2 8 2 1 2 - ü 1 1 5 1 2 5 1 2 -26 355 338 11 2 0 1 2 4:10 - 4 4 1 9 - 7 n o n - 4 9 6 9 2 4 2 5 2 - 2 7 5 1 3 4 2 0 9 1 7 2 1 » 2 6 1 4 » 1 3 3 3 0 6 1 2 1 4 2 - » 2 5 1 8 2 7 6 - 2 3 6 1 2 1 6 3 0 6 - 3 0 5 11 4

VO - 1 4 1 7 7 - 1 4 0 1 7 n 2111 - 1 9 5 9 - 4 4 2 6 4 3 9 9 0 171 -lue 14 3 3 3 5 3 1 7 n 9 1 3 3 - 1 0 7 1 7 - 1 2 2 2 5 2 1 3 1 4 9 4 4 7 - 4 5 6 9 - 3 2 0 1 1 9 9 1 0 7 1:14 1:14 17 . t L 5 6 4:1 6 2 7 1 0 1 0 1 3 0 - 1 3 0 i n - t e 2 1 9 -::o:i 14 10 f.îW 0 0 4 1 0 0 1 7 0 - 1 1 7 1 0 9 106 142 10 6 6 1 5 - 6 2 : 1 1 0 11 ( 0 3 1 6 7 1 4 - 9 1 4 1 C I O 2 0 i:i f,T0 - O U I 11 1 3 C 1 - 5 4 1 9 10 210 191 13 -îîîl 2 3 0 - 2 5 0 1 2 7 1 0 7 - 6 1 1 5 13 130 -1011 20 -Il 2116 : 9 0 C l 1 5 2 7f | 2 5 1 II Q 1 3 0 - 1 1 6 ( 2 11 343 -:I00 11 - 2 2 2 4 0 2 5 7 1 2 O 1 12 0 2 1 6 14 1*61 -256 12 - 6 ( 7 0 - 1 * 1 7 1 7 1 6 1 7 1 -lOII 1 4 3 1 9 0 1 7 4 i O 1 1 7 0 1 0 1 10 - 2 1 2:10 2 0 6 1 2 9 4 2 1 - 3 7 ( 1 9 1 6 2 9 2 2 9 4 C l - 2 2 6 5 -:.r»6 11 1 9 I 4 U 1 0 5 111 4 1 3 0 - 1 : 1 0 1 3 10 107 - Mil 1 9 - 2 0 2 2 9 - 2 3 4 l e l e 1 6 3 - 1 5 1 1 3 l ü 2 0 3 - 2 0 6 1 6 0 1113 1 7 4 15 21 2:111 • 2 2 7 1 4 6 3 4 3 3 4 4 II f« 1 0 1 - 1 0 7 1 9 - 1 9 1 0 2 1 0 2 1 4 11 4 0 9 - 3 7 5 9 4 2 3 7 2 3 0 1 3 2 2 1 4 ? I 5 U 2 0 7 1 0 1 1112 11 IV 214 209 10 - 1 7 151) 1 2 5 1 2 1 2 2 3 5 2 4 1 11 0 4 6 1 - 4 9 1 9 - 1 6 7 5 4 - 7 7 9 II 1 4 2 3 4 - 2 : 1 0 1 2 € fl L 2 L 9 26:1 - 1 9 2 9 6 h - I B 1 4 4 1110 12 1 5 2 6 2 2 4 1 1 2 - 2 0 2 1 6 2 1 4 1 4 11 2 2 0 - 1 9 5 11 - 1 4 21)1 2 7 0 II 1 6 3 7 1 3 7 6 1 0 - K l 1114 -IIKI 1 5 - 2 0 1 6 4 -MO 111 -23 132 114 19 1 2 C 1 2 - I l 1 1 7 • 1 6 1 3 7 - 1 1 1 3 2 t -1:1 4 4 9 - 4 : 9 II 1 7 2 3 9 2 4 1 1 3 - 1 7 1 6 1 - 1 6 2 1 6 - 2 4 1 4 7 - 1 2 5 1 7 -22 310 -207 10 1 3 1 6 3 7 6 1 4 “ 1 2 2 9 3 - 2 U 3 II - 1 2 5 2 6 - 5 0 0 II l O 2 6 9 - 2 4 9 1 1 - 1 6 1 5 0 - 1 6 5 17 -20 7 2 6 7 2 9 1 2 -21 2(0 -?2 0 13 1 4 4 1 4 :)i)7 1 0 -il 1 6 4 1 6 0 1 0 - 1 0 7 9 7 1 10 4 1 2 - 1 0 1(1(1 1114 1 4 -III 4 4 6 - 4 9 7 1 0 -20 706 /30 12 1 6 1 4 6 - 1 5 3 1 0 • 1 0 4 5 0 4 4 b 1 0 - 9 0 5 9 - 7 0 0 9 a L - 1 2 2 , ( 1 2 7 2 1 0 — 1 6 124 4 5 1 4 -III 271 292 10 1 0 3 1 9 ;ioii 1 1 -Il 2 7 9 -2112 12 - 7 3 7 6 3 11 2 7 - 9 41:0 - 4 : 1 9 1 0 - 1 4 7 7 5 - 7 7 2 II -17 2(12 776 9 2 0 2 0 3 - 2 5 1 1 2 -6 209 2111 12 - 0 7 7 1 1 7 3 2 U 2 3 2 3 0 2 4 1 1 3 -il < 117 4 7 3 9 - 1 2 1 3 7 7 1 3 6 9 1 6 -16 201 164 11 - 4 3 4 5 - 3 2 0 1 0 - 5 1 0 0 - 0 4 1 3 20 140 -140 17 - y 3 117 : I72 1 0 -10 956 -1190 1 3 -15 2(»9 -202 9 7 1 4 -:i io;i - n 12 1 6 1 2 0 - 1 4 7 1 7 - 0 1 4 4 - 1 4 9 1 0 - 6 1 16(1- 1 ( 5 5 : a -13 11)2 154 JJ - 2 2 2 1 9 2 1 1 1 4 2 3115 : 110 to -2 4 1 1 1 4 0 9 1 4 1 2 0 1 1 5 I S - 4 4(1(1 4 9 9 9 - 4 1 3 1 2 1 2 3 1 1 1 4 -12 145 109 1 1 - 1 7 1 3 6 1 3 4 111 4 3 7 9 -:09 10 -I 4 9 5 - 4 0 3 11 1 2 2 2 1 2 5 9 1 0 - 3 3 1 4 2 9 6 9 -21235- 1207 1 4 - I l <•02 -617 10 - 1 5 1 6 0 1 3 1 1 4 Il 1 9 1 - 1 5 4 I » e 2 3 3 - 2 2 4 7 11 1 6 1 1 7 0 II - 2 ni7 - 3 3 6 10 0 9 11 7 1 0 5 0 1 2 -lO : 27 -220 9 - 1 3 2 9 0 2 9 1 1 0 1 2 1 5 0 - 1 1 1 1 9 I 4 4 5 4 3 9 11 - 9 1112 - 1 6 7 1 0 -I 29:1 - 2 9 ( 1 1 0 2 1 11 2 - 7 3 4 11 -9 650 603 9 - I l 1 3 4 - 1 4 9 1 4 2 1 3 0 1 121 10 1 5 - 7 2 0 0 24:1 9 0 4 2 6 4 1 2 9 4 0 1 6 7(11 1 2 -7 5116 -620 9 - 1 0 1 7 1 1 9 4 11 7 L 0 6 2 3 - 6 4 1 1 9 - 6 4 6 5 4 3 2 11 1 2 2 6 2 5 3 11 6 1 U 3 1 - 1 1 0 9 1 6 -6 :(6b -361 II - 7 9 7 1 - 5 1 1 2 9 4 1 1 1 9 - 1 1 2 9 1 4 - 5 (111 - 0 2 5 II 2 4 4 - 2 / 5 1 2 U 3 6 1 31)0 i) -3 333 :151 7 — 6 1 5 5 o i 11 -17 11)5 193 10 G 0 4 4 0 3 1 10 - 3 O K I 1 117 il 3 1 3 4 - < 0 1 1 ( 7 III 4 2 6 - 1 2 . 2 9 -4 :uii 311 7 - 5 <» 19 6 0 9 9 -10 192 I :•? 16 6 3 7 3 3 0 2 II - 2 4 6 1 4 7 7 11 4 1 9 3 1 9 4 Cl 1 2 0 4 7 1170 1 2 -3 200 -179 0 - 4 9 4 9 - 9 4 5 1 4 4 1611 141 17 7 1 0 3 1 5 4 1 0 - 1 4 4 9 - 4 2 0 9 9 1 4 9 1:14 1 7 1 4 3 0 9 - 3 0 0 1 0 -2 20Ü -291) 7 - 3 5 0 5 - 6 0 9 9 7 177 -Ibb 17 Il 3 9 4 - 4 0 5 t 1 3 5 1 4 1 1 2 1 2 2 0 7 ( 9 2 14 1 6 2 0 7 - 2 3 1 1 2 -1 1171 1:67 12 - 2 5 5 0 5 9 2 9 Il 244 -2011 14 9 5 0 6 4(1 6 9 2 1 6 2 - 1 7 9 11 1 0 1 3 0 - 1 4 1 1 9 0 2 3 4 255 9 -! 7 9 4 ( H I II J U 0 9 3 7 0 4 11 :i 1 5 2 C I 2 II 7 1< L i 1)32 -0 3 13 0 2.19 9 0 0 1. 11 1 5 9 - I I U » 1 3 4 3 0 6 - 4 3 7 9 t 1 I. 3 Y05 61 0 10 1 5 6 0 - 5 2 1 9 1 2 5 0 7 - 4 9 ( 1 1 0 8 4 6 0 4 4 5 9 - 2 1 2 2 1 -.3 17 1 6 4 449 4111 0 2 4 2 0 4 1 ! 9 -10 160 101 19 1 3 2 4 1 1 - 2 5 6 11 6 3 9 7 - 3 1 1 1 11 - 1 9 2 7 5 3 5 9 CI - 2 9 -2«1 1 5 0 494 —51.6 0 4 :)2i - 3 3 3 II - 9 1 7 6 - M O III 1 5 1:13 - 1 1 1 K l 7 4 0 7 - 4 2 0 9 - 1 5 1 4 5 1 3 5 1 9 -2:1 1 1 9 1 6 6 321 - 3 :6 U 5 3 6 1 - : 1 6 2 9 1 1 7 1 M O 111 10 101 1 0 0 1 7 1 9 3 0 2 3 3 9 0 - 1 4 2(7 - 1 9 6 (4 -222 11 7 061 556 9 6 KM) - 2 0 5 1 1 3 1 0 I - 1 4 0 2 0 1 2 1:1:1 411 1 7 - I l 1 6 3 1 1 9 1 6 -2.1 - 2 6 4 C l a 110 (03 14 7 35:1 3 5 2 9 2 L 1 3 1 7 4 1 4 4 1 4 - 9 K*:i - 1 3 2 (6 -2 0 202 I f 9 494 - 4 - U i 9 II 2 4 2 - 2 4 4 11 e L 15 270 -25(1 11 -7 (77 (77 K l - 1 9 *139 9 11 239 215 l O 9 12'* - 1 1 2 1 7 - 2 0 1 7 7 - U K i 1 0 1 7 1 6 0 1 6 7 l U -6 221 -21 1 1 2 - K l -1.117 9 12 153 ! M 14 1 0 1 3 ) 1 CI ( 7 - 29 2 5 1 ) 261: 1 3 - 2 6 2 0 2 2 7 4 1 2 -5 1 3 0 9 9 1 9 - 1 6 5 9 7 II 14 459 -420 9 1 1 2 2 6 221» 11 2 1 4 1 3 A 0 9 l O - 2 4 2 3 0 - 2 6 9 1 2 4 L — 4 1 7 4 - 1 5 9 (0 - 1 5 1 0 4 9 1 6 7 0 1 5 17 145 -1:19 III 1 6 1 0 1 - 1 4 1 1 1 5 •1* 1 1 6 4 - 1 1 1 9 1 7 - 2 2 2(111 2 7 0 1 f 1 2 1 0 - 2 1 : 1 1 3 - 1 4 - : ' 2 4 II l ü 130 49 10 1 7 2 2 7 - 2 3 0 1 4 •IV 0 1 5 ( 1 0 11 - 2 1 ( 3 6 9 3 III -26 214 16 2 1 6 3 1 5 4 l<> - 1 2 r »67 1 0 K : i06 3 5 9 - 1 9 1 4 2 1 0 9 1 6 - 2 4 333 1 1 7 1 7 4 1311 1 6 - I l - . ; / 9 9 6 3 1. <» Qi L C 6 2 0 UtKl 1 0 - 1 7 1 0 6 1 9 9 1 2 -22 223 1 4 9 1 6 9 - ( 6 2 1 7 - 9 1 1 3 0 1 1 6 Il 1 1 6 9 - 1 4 5 0 1 7 - 1 6 5 0 0 - 5 1 6 9 -21 2 7 V 1 3 II 1 9 7 2 1 2 1 6 -Il - 1 9 6 II -2 0 174 176 1 9 - 2 2 2 9 9 - 2 * 1 0 1 3 - 9 4 4 3 4 3 1 7 - 1 3 5 7 2 - 5 4 1 9 - I I I 302 10 1 3 l Ü G - 1 6 0 1 7 -6 6 611 1 0 -26 372 -3115 12 - 2 1 1 9 0 16(1 1 6 - 7 10(1 47 13 -13 31)2 - 3 1 1 1 U -16 :>.73 11 -G -nii4 II -24 3 0 2 340 11 - 1 9 2 1 2 - 1 7 9 1 4 - 5 2 0 2 2 7 5 7 - 1 2 2 9 7 - 2 0 5 9 -10 200 12 7 îr L - 4 -303 6 - 2 2 310 -:I09 11 - 1 7 2 0 1 ( 9 6 1 4 - 3 7 0(1 6 7 3 1 1 - 1 0 1 5 9 1 9 2 1 0 -1 4 1:10 1 6 -3 -:*.'î4 6 -20 2 2 2 197 (2 - 1 6 2(1 0 2 6 2 1 1 - 1 6 2 7 6 4 7 9 - 9 7 3 5 7 4 9 1 0 - 1 3 i:>.i 1 7 - 1 2 2 (10 2 9 0 (2 -2 - 6 1 1 4 1 0 -19 IIUI -197 14 - 1 1 2 9 7 - 2 9 1 (O 1 5 0 3 5 6 0 9 -Il 4 1 1 0 4 6 7 0 -1 2 50(1 1 0 - 1 0 2 1:1 - 2 2 7 1 4 - 1 5 9 2 1 0 - 1 5 355 344 9 - 1 0 2 1 0 - 2 1 4 1 2 il 900 -1151 13 -7 564 - 5 0 7 9 - I l 233 11 -Il 1 3 7 1 3 6 2 0 0 9 5 1 12 -1 4 499 -409 9 - 7 2 6 9 - i7 * 1 0 » 2 5 4 2 3 4 II — 6 1 3 2 1 3 1 II - 7 3 1 4 0 - 6 1 4 1 - I V2 2 0 1 - 4 1 4 7 -13 177 -1511 11 - 5 6 1 2 1*1:1 1 0 7 5 0 4 - 3 4 6 O - 5 3 2 4 3 1 9 7 -6 516 9 0 2 7 5 2 V I I 1 2 2 - 1 4 4 10 -1 2 ;i67 :oi:i n - 4 1 4 5 H*9 1 5 V 2 9 5 2 ( 3 II - 4 1 0 0 1 4 1 3 - I 109 1 3 1 1 4 6 -115 ( 9 3 3 : 5 11 -II 237 253 9 - 3 176 ( 7 6 1 2 1 i 6 3 9 - 6 0 9 1 1 -:i *195 5(1 0 0 325 11 2 5 9 -23 1 1 2 4 -636 9 - 9 :)33 -334 II - 1 5 2 0 -r 1 0 1 3 0 2 4 0 3 4 1 2 - 2 2 3 6 19(1 11 2 :{I6 1 0 4 MO r t;i 2 1 r> 1 2 5 1 2 -Il 102 146 15 5 1 7 2 - 1 9 3 1:1 1 9 154 -115 (Il -1 30:1 -376 II 4 242 10 6 1 9 9 - 2 0 7 1 6 6 ■i\r 1 4 - 7 293 30 f (1 4 1 3 3 - ( 5 2 ( 6 2 1 ( 4 2 1 6 9 2 1 1 :1V 2 3 2 4 O 5 1611 1 3 ü 1 5 6 116 2 0 7 0 6 0 6 9 0 10 - 5 149 -102 1 1 3 1 6 9 1 0 3 1 4 2 3 1 7 1 - I Ü 4 1 9 2 H » 0 1 7 0 11 U 376 10 )-» VO 00 199

#5:^ ' R3&ë:!;sB^'3nÎ3R^=g " =é^;ëmM^#a^Ün§5=5ëg:ËS ] ;§=# s*"'* = S?ir.S£-S=TfT?®'*'^'==12 - SSrif?r.£2!:£-2=£fT'rTY'fT*'’==“'‘=^-222 “ §r?,2 I I I I I I I r . I I I I I I I i 1 I I 1 till e : < * ts®w®K-©r.*N---tt©-B<'^ NNCC*nNNN©‘^©-N-©-iîtcs^© ©®©g©»R 5 © e h . s# " ^ %u:g§g::2g:K|:s^5§g ^ EgH:g#;:ss.

' È:È####sgBÜü:à^^!È " :§!s:E^^#:&s%^ÜÊ:!gg ^ W : Ê § ! Ü k " 22 ' 5?s;;2=£227'rr??"''*222 ' S2S2ar«?'r'f?T?V7*""'"“*' '' S2=?f t I i 1 I 1 I I I I I M i < I I I

2S£2=2:«-»''>'Z“'*'®2£®'=®’'»'»'£2rt=£*££5 g2®22!:=±»=="':!:=££:"'®=£'^±£2

" gî;Siir£22S:£2T7>rf7r*-N"'»*''=®=2ï£=§ * 5££:S£:22=£T?'r'f?Tï7*-"‘"“*'-=® I 1 I i I i I I i I I I I I I t I I I i i

<- » «®t:-N©B=3'ao^tt©3NSe*c®®N© «ftce©S2Z®'®'S®^2®*'®-®“ *'£®'2®'£®'®‘2 - £ î 5 ^ 2 ": :ËË=;5m$=:#^Egs:gËÈ^.

^ ^ 7 ^ îî^?Sr.i:=-=î‘rT77“'“''^=22S gl5?!Sî2£222:77'r??T?77®"''^'"“’''‘'-ii2 11) 111(11 I ) I I I I I I ( I I e-NStf'C©©—©©CoSC©——©N© ©îî©wt5©©—NefJ©r-etïvt^^— © « —cwisSS-^—w©—— o ©

:^E;ië##ÜB2ÜʧEig " ;=E=ï=:ëiRm :È:gs ' |g2sÉg:g:^&îg^§ ^ g 2Ï”=7=7Î7T777-'''"''*£:2 “ ?i£t:£277'r?T"”“''"222S " 22£7?'rf?T?7®'''''* " I i t I I 1 I I I l i t

Mv5C-r<'©5*©©«©©S®SS©e©©NMrt RNN-f R NON© Mf'f:©-© ©sa© »©©=«--©-V© - gBsm:::B;SEm:gg:^ESg:;g#2g ^ gsB:

Sr22‘r7'rf7T?7~"”*‘*'*‘£22££ ® gSSSîî5lî=!:£2222Tî???7"'*"*''=’2ï£E^ §22L: 1 I I I I I I I I I I I I I I I I I I I I I

e©«‘©^©CMNS^-Cr*©'ro N©©N©©e©-*©@-e©©C@@N©0»5CCN*©©©©©;- ©^©f'R© ^ fiKiHpiPIpii ^ piii«?|îæf|îïiH|pîiî|sf ^ ip î ° iBI;”îtilÉ5ëili! ' S5SI«lilîî|!iiSSSiH-5giH»iïHiH " ISIsH ?l=£22=7fT?'*''*‘*22£ RK^§%;:^:22=£T?7f?:??7'-'"='"'''2:22% SS2=t:2 i I I t I I I I I I I I I I I I I I I I I I i t § CSJ fil Zlll nos ni 01 Oil 161 01 9 1 101- 921 iZ- fil 6 9 1 6 2 1 i 0 1 Z Zfi Zif i 6 OSV/- fini fil 0 1 QZi^ Z I - ZI C 6 Z - C 6 Z II 0 1 O i O - Ofifi M I I 1 0 . ; - o z n Z 11 n o s O llZ 11 Ol 2 f i « - 6 ,c II ZI utz lisa 1:1- C l llll c i z G ZI 9i;Z vZZ fil 1 !S Z I 6 1 « Z l - 01:1 z - 11 097 ; 9fiZ i G 0 6 i 2 « v G i'I Z4»l 9 1 Z i'I- G IZS- llfiZ fi (Il IIZI- O i l i l 0 1 0 9 Z 2 2 Z c - 0 1 Ü C 9 - V C 9 C ZI 6<:z o o z fi Il (UiiW I1IZ 01- 2 1 6 9 2 1 1 1 V I 2 I Z - O O Z fil' 9 1 z m 1 6 1 fil 1 i 61 n CM:V- o l 2 1 1 1 - 2 I Z Z fil « 1 - « n i - c Z l ?.tiz Ü fl Z 1.1 • (1 2 Z C l i e c - 6 1 0 6 1 - fiil 9 1 II fifiZ- 611Z C l 0 1 C f i Z - 0 2 Z 0 - 9 1 1 9 « 0 1 0 11 VCI (191 1 tll Z V I O U I III- 6 tIf'Z- Z i Z 2 - fil 6 0 Z 6 I Z 2 1 i l Z 9 1 9 0 1 I I U l 2ZI- ZCI 9- 01 6 9 1 C 6 l 1 - 6 V I Z - i Z Z C - II 120- 200 61- II U 9 I - z i: z 6- 21 Gfil fi9l (Il 11 2 9 Z - Z 6 Z 6 i l I Z Z « 6 1 «- « i z n - OIC e - 1 1 l l i i l - OVI 2 - 0 1 1 6 Î - U 2 i 1 1- 0 1 9Cfi o o n 2 6 1 6 1 1 - « C l G - fil 2 0 1 € 1 1 c - II fici: ficn G - 01 6 ( l / Z 9 6 Z 0 1 - T \k Z I 0 1 C 2 î;- C 6 Z i i l 0 2 1 - ZGI 0 1 - VI « 9 6 Z I V- SI € 6 1 e « i 1 1 - 91 GUI CCI fil- 6 6 I C - V G Z 1 C l K i Z ( l i s II- 11 « V I « 2 1 Ü- U UIV 9 l i G l- il 20Z OiZ n i 6 0 6 0 OiO 21- 21 2CI fini il Z I 1 ( 1 1 - « 9 1 1 - 2 1 9 2 1 - CCIVI- O l Z ‘>| - l i n 2 - G 2 I C - c m fil- 1:1 ti'J Z - !12Z 91 ZI U l M - fio;: 6 1 - i l Z V Z - O i Z Z I 6 w c 9 Ü C Z - C l 6.;: 6 9 Z 0 1 - G(,«(:- ICI: n - 0 1 o t z - li«Z 2 1 - fil OWf. OHZ 61 fil 2 9 1 - 6 2 1 t :z - n i fiOl V i l 0 1 fil Clîl- Oil 9- Ul îlii- Gfil 21- fil 2«l- V Z I 6 - Ol 2 C Z - Z2C 6 1 - Il 6un- ilio ZI 01 CIIO Cli fiS- 6 1 0 2 1 - llil (1 6 I K . i - 2ifl 6 - 2 1 9 6 1 - i 6 l S 2 - G 2 2 C - f iiC Z l - 11 6 IZ CVS IS- il 221 2111 Ol i l V Cf i 6 Zfi 9 6 2 C « - I 12 Z 0 1 - 6 2 9 C O it; « 1 - II Z 6 Z - Z2Z r z - 01 Ot.O 6 9 0 9 T 0 fil 9 1 1 1 2 1 - 0 9 1 i II 9 2 1 Z 6 I C l - *1 1B ]II 6 i:*i: O G C v l - 0 1 02Z zo.: fiZ- ZI 607:- Ii:z II 1 1 fiiC IfiZ Z G fini 9 Z i f il- 0 1 2 0 1 : - 2 6 Z 9 1 - ZI OIC- CZÜ 2 Z - 6 100- 6Z0 i I Z 0 1 1 - Ifil 2 - 0 1 u i ; s - O i Z 0 0 1 oui: c m : 9 1 - 9 1 9 2 1 fiGI ZI 9 1 1:21 «fit 2 1 - 6 1 i;;z 2 2 1 G Z - ZI 191 601 I 2 1 9 11- 9 S I z - fil 9 1 : 1 - ifil 2 1 - II 9 6 Z - I C C 9 0 1 v c c CCI: III- ZI 0..Z- 6IZ 0 1 2 S I O l 9 6 Z I 9 Z c - Z I Z i Z - o c z i r i - 2 1 Z O I - ICI 1 II ovz- 9iC G1 — 1 e II 01 l:ilZ llftZ I- 6 Cfifi- lifi û - 9 1 C i l - Zfil IZ- 0 1 C 6 9'tîl 0 6 1 f i Z l - ICI 0 2 - 01 001 OZI z- fil 06Z OÎZ Z 2 1 n o i 2 Z I 9 - 6 1 i O I - 01:1 z z - fil ofil 9€l s- fil C 2 I 6 9 1 IZ- I l 9HZ- fOZ 0 1 " 01 6IZ- 2UZ fi­ fil 1 6 1 - 2 0 % 0 0 1 I I Z 9 9 Z u - i l 6 9 1 1 6 1 c z - II 6 2 Z - C 9 Z 0 - fil fiiz C « l £<:- e l 061 i n i n - OS iiî;i - C i l 1 - fil G l l - 2 U I I I i l 6 V Z - 6 Z Z f iZ- Z I i i Z C V S 1 1 - II H i C - c c c C Z " T l ei 6 i:Or. 6 2 Z 6 - 6 1 6 f t l - (lil i - 9 1 v n i l i l ZI fil 0 6 1 V O S 2 Z - 6 1 i l C O i C V I - fil VOZ- « 6 1 v z fil 1:61- ZUI Bi­ 6 1 (III 6 0 1 fi- 6 1 Z 9 Z 0 2 Z fil 2 1 u n i 621 nz- 61 IVI- l i l 1 1 - 6 1 «111 C 9 I 9 Z - 91 Z f l l - 0111 i ll 161 261 II- 2 1 fiCZ M i l 9 - 0 1 6 6 f i - 961: i l fil 2 0 C - n e c 6 Z - 1 1 6 1 1: z z c 1 1 1 - 9 1 l ü Z - « O Z « c - 11 227, rif z Z 01 94,6- 1160 il- 9 1 GZK- (lin II- i l V G 1 O U I 9 1 6 1 9 « n - 9 G C 0 7 .- 7,1 OIZ- V9Z 0 II 60Z- 02Z fil- III II 1 fifii 0 1 - Z I « 2 Z - V 2 Z (Il T 1 Z I 9 1 2 6 , 9 6 1 ü S - T iR 1II Cl w.«: IVZK- 6 0110 Z26 91- i l 9 2 1 - IZZ Z l - 6 1 O l l - ?:g i 6 1 Cl V fZ 9 - C l 20.:- 261 21- (Il Ox »- Zfil 0 1 - 11 z z v V Gf i O Z II 2 1 C - GOfi 0 1 •1 1► 1II i l M i l 9 6 1 C l 61 G i l - «VI U- II 6ii;:- iHiz III- 9 1 z »,î;- f iOZ 9 1 - II ZOII- 2Zfi ZZ 6 f i 6 i — COfi 9 ZI 0 2 Z i ;c z 0 1 fil O c 2 - 9i>2 Z l Z I 1:1: : 6 6 S « Z - U l C O ï 9111 n i - VI fiiz «IZ iZ Z I G Ü Z iz<; i 6 1 CCI Gfil >1 ZI zi.z- liez u 91 G /l OUI i l I I Z 6 t î - 61:1: z z - ZZ C 9 I - C i l 9 Z II 2 I Z « O Z Z VI CM VOZ :i 6 |9 f i - . o n 2 III c u i - Gfil 91 01 1:61 m i l 6 Z - '1 9 Z I 1 1 01:9 V«J9O 61 VU 6 Z I Z I V .O Z M l Z 9 91 921- 261 9Z- T n Z I 0 1 0 2 Z - 2 i Z z - ZI l i i z - f ins 1 6 OfiC- 121: C » e i OZ l i l - 2 9 1 II 6 1 66VI -0«€|9- 0 1 ZOfl i Ofi û 6 Z lli 9 2 V Z 1 C 01 o;: 0 6 2 C I 9 m € 0 1 i 9 l fil fil 2IIBI B60M1- II 6 € S C G S i 6 C 6 i ’l'dv 1 IZ I t l - C il SI 91 2*1 C2I Z- 21 92.1 - 0 2 1 fil i l 0 9 1 - U Z I 0 1 - 6 1 fiCi»- l O i fi II € 0 2 - 9 9 9 0 61 ZM fiVI 6 ZI 06Z IVZ Zi 9 1 21:1 S 9 I fi- VI c o z - fiGI 6 I) i Z V Gfii Z l - ZI Ü 9 ! nr.f ô 11 9 6 1 O O Z 1 - «1 l i l - Zfil 2 ZI OUI- lOZ 01 VIZIZ-ZIZ 6 - ZI O iZ 6 9 Z 2 6 9 C C - 921: VI- 6 o a v 9 Z i' 1 n S I C « U 7 : z - Ol 111: OVI: C- 01 non 6I.Z 2 Z I I V Z 0 1 - 11 € i Z - 9 iZ fi 91 OZI- Cil III- fil 6 1 1 ZCl 0 G s o i C CI' c - Cl 2 t.u - Z IZ 2 - SI 061- Mil 9 2 1 f i z i - 2 9 1 I I - 9 1 9 C I 9 i l i Z I 0 C 2 - Z 2 2 I- « €61: V- 21 2 Î - 61:1 01 6 2 6 Z - UIIW fi ZI 21 z IIZZ Zl- Ol Ztifi OliU fi T (0 Z I 0 1 V 9 C - HVC z - G C tifi- ««fi fi- fil €1% c m ZI 01 26Z IfiZ i 2 1 0 ‘ 1 - Z 9 I V I - fil i l l l - 6 2 1 Z 0 1 C i 9 CU9 c - II ilOl fiV2 9 - i l 061 CGI Cl 6 ZM: 611: o 9 1 « 6 1 2111 2 1 - 6 C 6V - 9 l i I I Z 0 9 1 GVI 0 0 9CU- 26V €- 01 2 6 9 - V 2 9 «- OZ IVI- z v i U l- 6 9 Z i : - COI: z 9 1 ZWI- M i l I I I - O l m . ’fi fiifi 0 2 1 Gfi.:- « 6 1 S- 6 CiU U9f i 2 - « V 9 C - GVI: 6 - i l 261 ♦Z Z IZ 6 2 6 0 - 91 fi I OC fiOl Cfil I Z - 0 1 6«Jfi 9 ifi 1 - fil Z 2 Z fifiZ II- 6 c».« c o i : (1- II Z Z C 9

O -1 9 240 -2 0 4 13 -1 293 -2 9 2 12 -1 9 16? 193 m -1 3 iim -1 9 2 17 -2 0 107 - l » 4 19 -1 7 1311 127 17 17 S! L I 172 139 17 -1 7 139 - 103 III -1 3 211 249 10 -17 174 1112 17 -2 0 174 104 19 - l î l 107 139 10 3 240 -2 1 0 13 -13 221 213 13 -12 162 -1 3 7 19 - 13 im i -IVO 10 - 0 142 1 14 20 -1 1 137 130 17 -2 4 104 -1 0 3 20 G 1110 192 1» -7 19» -200 14 -1 1 130 -1:1» 20 -Ü 14:1 147 20 “ 9 290 293 12 -9 3 132 -1 13 19 -1 170 liiO 17 -1 0 224 243 13 -1 144 -141 21 2111 2 L -S 213 -2 0 0 13 -2 2 139 l’iii 19 17 4 L -9 IflÜ 143 10 7 177 -1119 lU -2 0 1/3 -1 0 3 17 lü 2 L -11 143 -9 2 20 19 3 L -IB 1113 - l i n 17 - |0 222 0 14 -17 170 -106 1» -7 3411 -3 3 9 12 -1 3 107 i<:i 17 17 1 L - 1 0 130 1:4 91 -1 7 17» 210 17 -5 204 191 13 -1 7 167 —163 19 - 1 2 2:i:î 293 14 -1 7 104 1113 III 17 5 1. -1 0 130 120 21 -1 3 l»9 13» 17 -9 103 140 1» 130 -1 3 3 in -1 3 1,3 -1 3 9 10 -13 233 -233 lü 19 9 L -1:1 249 -244 13 - 7 194 -1 9 4 10 -III 101 174 III -1 3 204 21:9 13 -1 3 133 -1 4 3 20 -1 3 m a 190 10 -1 1 2110 209 12 -0 207 -1 7 0 13 -1 7 237 12 - 1 2 107 - l o i 10 -1 2 142 131 21 -11) 223 -2 4 2 14 -1 9 213 -2 0 » IS - 9 200 -2:17 13 - 3 131 -1:15 :»o ~ I5 2:11 107 12 -Il 130 1 17 19 -10 241 -229 14 -9 190 2 ^M 10 -1 7 144 1:1:1 20 - 7 204 20» 1:1 - 2 133 -1 2 0 20 — 1 fOO 13 i fO - 7 211 -2 9 0 13 -Il 197 103 10 -Il 2111 tItH 14 - 9 171 - 1:1» 17 - 5 m a -2 1 9 17 -1 2 233 -r.;i4 11 2 133 -1 3 0 19 - 7 300 -3 1 4 H -1 14» -9 7 21 21 0 L -1 1 103 103 17 3 I3Ü -1211 20 ffl 0 L -5 29» 2117 1 1 19> 1 L - lO 292 3 10 12 -P 2111 -:*M3 13 20 0 L • -1 7UOU 324 14 -9 227 - :.I 9 12 17 at L -2 0 13:! 13» 19 -1 133 137 20 -2 2 1116 177 20 - 7 130 rv». 10 -1 3 102 MO 17 t 1110 197 in -111 201 249 in -2 0 274 2113 13 21 1 L 2#»» 2.70 12 -1 9 212 193 14 -2 27.3 -294 13 -1 0 1112 -1114 10 - 1» 373 -3 0 0 12 -Jï 204 -::o7 13 -1 3 330 331 12 203 13 lU 3 L -1 2 103 - 111:1 III -1 0 201 210 10 -1 0 133 130 21 - 4 221 -221 13 -13 / 4O - 4 t 0 10 4 300 203 10 -1 0 m o 194 10 -1 2 239 20 » 12 -1 4 195 211 17 - 3 100 130 10 -1 1 234 14 0 219 -177 in -2 9 233 203 14 -0 137 140 lit -1 0 10» -1 0 3 lit - 1:1 130 -1 3 7 20 - 2 143 174 19 -9 241 -2311 13 -13 146 1411 2 0 -4 2114 - 2 0 1 12 - 6 :i2 o - 2 »fi 12 -1 0 192 174 17 -I 201 - 13 1 14 -Il 141 140 19 1» 1 L -1 4 134 -1 3 3 19 -3 105 144 17 —4 2111 239 12 -Il 209 -2 4 0 1:1 e 340 - 3 3 3 1 1 - 7 40» 4911 1 1 - 9 147 -1 2 9 19 - 2 102 104 19 - 2 102 “ 109 141 - 7 1110 102 17 1 174 M3 10 -3 4 0 7 -371 #0 -2 3 103 130 19 0 lUü 1119 l ü —0 221 247 10 2 139 149 21 -3 :i'2 317 1 1 -3 3 130 -9 0 19 l ü 4 L 191 2 L 6 193 103 10 -2 102 -1 0 3 lU -21 200 -2 2 3 10 20 1 L

ro o to TABLE OF

POSITIONAL PARAMETERS

XYZ

Co 49557( 5) 92012( 13) 73948( 4) N1 44946( 28) 103054( 72) 78917( 21) N2 54233( 29) 79399( 73) 69120( 21) N3 39969( 29) 104170( 71) 70136( 22) N4 59359( 28) 104170( 72) 7 7 9 9 K 20)

N5 67788( 32) 101135( 80) 95178( 21) N6 45423( 31) 112158( 70) 67362( 22) N7 96045( 37) 114413(104) 68834( 27) Cl 48736( 37) 107428( 94) 83969( 26) C2 61895( 33) 109862( 90) 83123( 25)

C3 5 7 0 4 K 35) 107604( 93) 86701( 24) C4 37604( 38) 77397( 93) 64808( 29) C5 50698( 39) 78062( 92) 63867( 27) C6 42429( 40) 85491( 93) 61618( 26) C7 60265( 38) 105382( 91) 92507( 26)

C8 55417( 41) 106508(111) 96246( 28) C9 71823( 45) 99243(130) 101200( 30) CIO 42619( 38) 106622(103) 62356( 27) Cll 35269 ( 38) 71477(103) 73378( 29) C12 32273( 40) 86435(115) 7 6 3 0 K 29)

C13 36312( 36) 104603(104) 76733( 28) C14 62457( 41) 73282(110) 71968( 29) CIS 67967( 39) 89598(128) 73555( 28) C16 64136( 39) 107322(111) 74352( 26) C17 39435( 51) 117314(117) 57032( 33)

CIS 53717( 45) 70124(111) 59623( 30) C19 69217( 42) 121079(103) 85413( 28) C20 46006( 49) 131806( 99) 68635( 32) C21 91451C 49) 107064(112) 58343( 31) C22 94063( 41) 110989(106) 64275( 31)

X,Y,Z values have been multiplied by 10**5.

203 204

POSITIONAL PARAMETERS (CONT.)

X Y Z

PI 15338( 13) 99527( 36) 59511( 9) P2 33965( 13) 78398C 33) 42020( 9) Fll 18922( 41) 100933(117) 54834( 31) F12 8031( 34) 90721(108) 55182( 25) F13 11769( 40) 97790( 85) 64222( 24)

F14 18718( 39) 79494( 93) 61026( 30) F15 11848( 43) 118675( 87) 58375( 28) F16 23002( 32) 107658(103) 63751( 24) F21 30566( 40) 84863( 91) 46580( 24) F22 42064( 36) 82668(133) 46008( 29)

F23 25385( 36) 74056(107) 38281( 27) F24 36978( 44) 7 1 1 8 K 95) 37423( 26) F25 32961( 39) 97456( 82) 39402( 29) F26 34357( 43) 58669( 82) 44479( 30)

X,Y,Z values have been multiplied by 10**5. t a b l e o f

THERMAL PARAMETERS

ylT j22 ii33 vl2 %il 3 v23 Co 318( 5) 398( 5) 369( 5) 24( 6) 174( 4) -26( 5) N1 303(32) 466(40) 451(34) 5(28) 184(28) -68( 9) N2 422(35) 451(38) 385(34) 66(31) 201(28) -34(30) N3 392(35) 390(37) 493(37) 60(30) 189(30) 52 31) N4 380(33) 515(41) 380(32) -16(30) 221(28) -9(29) N5 498(38) 669(43) 355(34) 25(35) 176(30) -4(32) N6 570(39) 336(39) 481(36) 49(31) 259(31) 4(29) N7 689(48) 1200(68 ) 612(45) 52(47) 286(41) -114(47) Cl 459(42) 450(45) 461(41) -4(43) 282(35) -65(42) C2 276(35) 391(42) 446(40) -35(36) 152(31) 15(38) C3 388(39) 360(41) 366(37) -109(40) 181(31) -64(37) C4 415(45) 435(48) 574(49) 18(39) 119(39) -2(41) C5 526(48) 419(46) 402(42) -24(40) 229(38) -13(38) C6 498(47) 463(49) 376(43) -29(85) 116(36) -14(35) C7 462(42) 431(47) 419(41) -85(40) 238(34) -33(37) C8 624(50) 779(59) 551(46) -126(50) 381(40) -101(47) C9 633(55) 1119(73) 454(48) 13(55) 97(43) -39(49) CIO 467(45) 506(51) 449(44) 76(43) 168(36) 44(43) Cll 381(44) 614(55) 649(51) -103(42) 207(39) 124(44) C12 420(47) 891(69) 569(49) -28(47) 238(39) 15(46) C13 313(41) 686(60) 563(46) 108(41) 179(35) -96(42) C14 450(48) 767(61) 578(50) 228(46) 256(40) 46(46) CIS 404 46) 1140(76) 499(46) 173(53) 271(38) 64(52) CIS 503(46) 729(55) 458(43) -122(48) 298(36) 4(46) C17 950(69) 686(61) 627(56) 71(57) 154(50) 183(49) C18 770(59) 674(58) 581(51) 66(50) 371(45) -106(45) C19 602(52) 627(54) 547(48) -151(45) 309(41) -26(43) C20 966(67) 356(48) 716(56) 58(49) 304(51) 0(42) C21 1000(67) 545(53) 642(54) 133(55) 271(50) -43(49) C22 582(50) 594(57) 532(48) 64(45) 272(43) 33(46)

205 206

THERMAL PARAMETERS (CONT.)

yll u22 p33 yl2 yl3 u23 PI 579(15) 883(19) 533(14) 49(14) 179(12) P2 641(17) 668(17) 618(15) -39(14) 246(13) silisl Fll 2028(71) 2834 97) 1217(50) --1217(70) 1211(53) -810(57) F12 945(43) 2276(81) 1271(49) -602(53) 90(38) -481(54) FI 3 2082(66) 1316(54) 1253(47) 142(50) 1208(49) 49(41) F14 1568(62) 1315(56) 1865(65) 550(51) 727(53) -52(51) F15 2010(73) 1008(49) 1744(65) 520(52) 590(56) 671(47) F16 842(40) 2125(76) 1252(48) -122(50) -103(37) -576(53) F21 2051(71) 1471(61) 1270(50) 1(53) 1022(51) -361(44) F22 868(46) 3107(111) 1643(66) -471(63) -227(44) 497(71) F23 1136(51) 2074(80) 1384(55) -395(54) 61(43) -295(54) F24 2384(78) 1641(64) 1452(56) 769(61) 1453(58) 433(49) F25 1723(63) 1001(48) 1987(66) 208(46) 903(56) 688(47) F26 2152(74) 995(48) 2327(74) 494(53) 1627(65) 638(52) ylJ has been multiplied by 10**4. The form of the anisotropic temperature factor is

exp[-2ir^(h^a*^ pll + kfb*^ ^22 + £^c*^ y33 +

2hk a*b*yl2 + 2hl a*c*yl3 + 2k£ b*c*yl3)] Footnotes

1. Hartley, F.R.; Vezey, P.N. Adv. Organom. Chem. 1978, 15. 189.

2. Leznoff, C.C. Chem. Soc. R e v . 1974, 2» 65.

3. Kaneko, N.; Tsuchida, E.E. Macromol. Rev. 1981, 16, 397.

4. Collman, J.P.; Halpert, T.R.; Suslick, K.S. "Metal Activation of Dioxygen" Ed. Spiro, T.G., Wiley and Sons, NY., 1980.

5. Tsuchida, E. J. Macromol. Sci. - Chem. A 1979, 13(4), 545.

6. Bailey, D.C.; Langer, S.H. Chem. Rev. 1981, 109.

7. Sherrington, D.C.; Akelah, A. Chem. Rev. 1981, 557.

8. Merrifield, R.B. J. Am. Chem. Soc. 1963, 85, 2149. Merrifield, R.B. J. Am. Chem. Soc. 1965, 150, 178.

9. Tashiro, T.; Shimura, Y. J. Appl. Pol. Sci. 1981, "2J_y 747.

10. McVicker, G.B.; Collins, P.J.; Ziemiak, J.J. J. Cat. 1982, 74, 156.

11. Royer, G. A d v . Catal. 19,50, M , 197.

12. Phillips, S.E,V. Nature 1978, 773, 247.

13. Perutz, M.F. Sci. Amer. 1964, 211(5). 64.

14. Wang, J.H. J. Am. Chem. Soc. 1958, 3168.

15. Wang, J.H. Acc. Chem. Res. 1970, 2» 90.

16. Chang, O.K.; Traylor, T.G. Proc. Natl. Acad. Sci. USA. 1975, 72, 1166.

17. Collman, J.P. Acc. Chem. Res. 1968, 136"

18. Collman, J.P.; Reed, C.A. J. Am. Chem. Soc. 1973, 95, 2048.

19. Tsuchida, E.; Honda, K.; Sata, K. Biopolym. 1974, 13, 2147.

20. Tsuchida, E.; Honda, K.; Sata, K. Inorg. Chem. 1976, 1^, 352.

21. Honda, K . ; Hata, S.; Tsuchida, E. Bull. Chem. Soc. Japan 1976, 49, 868.

207 208

22. Tsuchida, E.; Hasegawa, E.; Honda, K. Biochan. Biophys. Res. Coma. 1975, 845.

23. Tsuchida, E.; Hasegawa, E.; Honda, K. Biochem. Biophys. Acta 1976, 520.

24. Fuhrhop, J.H. ; Besecke, S.; Vogt, W. ; Emst, J. ; Subramanian, J. Macromol. Chem. 1977, 78, 1621.

25. Leal, 0.; Anderson, D.L.; Bovman, R.G.; Basolo, F.; Burwell, R.L. J. Am. Chem. Soc. 1975, 9 7 , 5125.

26. Allcock, H.R.; Greiger, P.P.; Gardner, J.E.; Schmutz, J.E. J. Am. Chem. Soc. 1979, 101, 606.

27. Ledon, H.; Brigandat, Y. J. Organom. Chem. 1979, 165, C25.

28. Bayer, E.; Holzbach, G. Angew. Chem. Int. Ed. Eng. 1977, 16, 117.

29. Tsuchida, E.; Hasegawa, E.; Kanayama, T. Macromol. 1978, 11, 947.

30. Schmidt,, J. Chem. Ind. (London), 1962, 5 4 .

31. Roth, J.F.; Craddock, J.H.; Hershman, A.; Paulick, F.E. Chem. T ech. 19|71, 600.

32. Michalska, Z.M.; Webster, D.E. Plat. Metals Rev. 1974, 18, 65.

33. Grubbs, R.H.; Kroll, L.C. J. Am. Chem. Soc. 1971, 3062; Grubbs, R.H.; Kroll, L.C.; Sweet, E.M. J. Macromol. Sci. Chem. 1973, 1_, 1047; Grubbs, R.H.: Gibbons, C.; Kroll, L.C.; Bonds, W.D.; Brubaker, C.H. J. Am. Chem. Soc. 1973, 95, 2373.

34. Pittman, C.U . ; Smith, L.R.; Hanes, R.M. J. Am. Chem. Soc. 1975, 97, 1742; Pittman, C.U. ; Smith, L.R. ; Hanes, R.M. J. Am. Chem. Soc. 1976, 98, 5402.

35. Capka, M. ; Svoboda, P.; Cemy, M. ; Hetflejs, J. Tetrahedron Let. 1971, 4787.

36. Allum, K.G.; Hanock, R.D.; Howell, D.V.; Pitkethly, R.C.; Robinson, P.J. J. Organom. Chem. 1975, 8 7 , 189.

37. Bruner, H.; Bailar, J.C. Jr. Inorg. Chem. 1973, 12, 1465.

38. Kaneda, K. ; Terasawa, M.; Imanaka, T.; Teranishi, S. Chem. Lett. 1975, 1005. 209

39. Kiji, S.J.; Koda, S.; Furukawa, J. Angew. Makromol. Chem. 1967, 163.

40. Bonds, W.D.; Brubaker, C.H. Jr.; Chandrasekran, E.S.; Gibbons, E.S.; Grubbs, R.H.; Krolo, L.C. J. Am. Chem. Soc. 1975, 97, 2128.

41. Chandrasekaran, E.S.; Grubbs, R.H.; Brubaker, C.H. Jr. J. Organom. Chem. 1976, 120, 49.

42. Nakamura, Y.; Hirai, H. Chem. Lett. 1974, 643. Nakamura, Y.; Hirai, H. Chem. Lett. 1974, 809.

43. Nakamura, Y.; Hirai, H. Chem. Lett. 1975, 823. Nakamura, Y.; Hirai, H. Chem. Lett. 1976, 165.

44. Bursten, M.S. Ph.D. Thesis, Texas A&M University, 1980.

45. Pittman, C.U.; Honnick, W.D. J. Org. Chem. 1981, 45, 2132.

46. Pittman, C.U.; Honnick, W.D.; Yang, J.J. J. Org. Chem. 1981, 684.

47. Levitzki, A.; Pecht, I.; Anbar, M. Nature 1965, 207, 1386.

48. Levitzki, A.; Pecht, I.; Berger, A. J. Am. Chem. Soc. 1972, 6844.

49. Pecht, I.; Levitzki, A.; Anbar, M. J. Am. Chem. Soc. 1967, 89 1587.

50. Vengerova, N.A.; Lukashina, N.N.; Kirsh, Y.E.; Kabanav, V.A. Vysokomal. Sardin. Ser. A. 1973, 1 5 , 773.

51. Sato, M.; Rondo, K.; Takemato, K. J. Biol. Chem. 1978, 1979, 601.

52. Sato, M. ; Inaki, Y.; Kondao, K. ; Takemoto, K. J. Polym. Sci. 1977, 2059.

53. Kurusu, Y.; Storch, V5.; Manecke, G. Makromol. Chem. 1975, 176, 3185.

54. Rollman, L.D. J. Am. Chem. Soc. 1975, 2132.

55. Drago, R.S.; Gaul, J. J.C.S. Chem. Comm. 1979, 746.

56. Drago, R.S.; Gaul, J. ; Zombeck, A.; Straub, D.K. J. Am. Chem. Soc. 1980, 102, 1033.

57. Hay, A.S.; Blanchard, H.S.; Endres, G.F.; Eustance, J.W. J. Am. Chem. Soc. 1959, 6335; Hay, A.S. J. Org. Chem. 1960, 2 5 , 1275; Hay, A.S. J. Org. Chem. 1962, 2J_, 3320. 210

58. Tsuchida, E.; Kaneko, M.; Kurlmura, Y. Makromol. Chem. 1970, 132, 209.

59. Tsuchida, E.; Kaneko, M.; Nishide, H. Makromol. Chem. 1972, 155, 45.

60. Cotton, F.A.; Wilkinson, G. "Advanced Inorganic Chemistry," 1980, 4th ed., Wiley Interscience Publication, New York.

61. Parshall, G.W. "Homogeneous Catalysis," 1980, Wiley Interscience Publication, New York.

62. Nozawa, T.; Hatano, M. Makromol. Chem. 1971, 141, 21. Nozawa, T.; Hatano, M. Makromol. Chem. 1971, 141, 31.

63. Nose, Y.; Hatano, M. ; Kamboro, S. Makromol Chem. 1966, 136.

64. Nozawa, T.; Nose, Y. ; Hatano, M. ; Kamboro, S. Makromol. Chem. 1968, 112, 73.

65. Jones, P.; Saggett, A. Biochem. J . 1968, 110, 621.

66. Pshezhetsky, V.S.; Ikryannikov, S.G.; Kuznetsova, T.A.; Kabanov, V.A. J. Polym. Sci. A-1 1976, 14, 2595.

67. Barteri, M.; Farinello, M.; Pispisa, B. Biopolym. 1977, ]^, 2569.

68. Stevens, J.C.; Busch, D.H. J. Am. Chem. Soc. 1980, 102, 3283.

69. Stevens, J.C.; Jackson, P.J.; Schammel. W.P.; Christoph, G.G.; Busch, D.H. J. Am. Chem. Soc. 1980, 102, 3283.

70. Schammel, W.D.; Mertes, K.S.; Christoph, G.G.; Busch, D.H. J. Am. Chem. Soc. 1979, 101, 1622.

71. Schammel, W.D. Ph.D. Thesis, The Ohio State University, 1976.

72. Stevens, J.'. Ph.D. Thesis, The Ohio State University, 1979.

73. Zimmer, L.L. Ph.D. Thesis, The Ohio State University, 1979.

74. Jackson, P.J. Ph.D. Thesis, The Ohio State University, 1981.

75. Takeuchi, K.J.; Busch, D.H. J. Am. Chem. Soc. 1981, 103, 2421; Ph.D. Thesis, The Ohio State University, 1981.

76. Tweedy, H. Ph.D. Thesis, The Ohio State University, 1981.

77. Daszkiewcz, B.; Busch, D.H., unpublished results.

78. Cameron, J.; Busch, D.H., unpublished results. 211

79. Kojima, M.; Busch, D.H., unpublished results.

80. Busch, D.H.; Christoph, G.G.; Zimmer, L.L.; Jackels, S.C.; Grzybowski, J.J.; Callahan, R.C.; Kojima, M.; Bolter, K.A.; Mocak, J.; Herron, N.; Chavan, M.; Schammel, W.D. J. Am. Chem. Soc. 1981, 103, 5107.

81. Jaeger, E. Z. Anorg. Allg. Chem. 1977, 346, 76. Jaeger, E. Z. Chem. 1968, 30, 392.

82. Evans, W.; Busch, D.H., unpublished results.

83. Herron, N.; Busch, D.H., unpublished results.

84. Riley, D.P., Ph.D. Thesis, the Ohio State University, 1975.

85. Wilkens, R.G.; Busch, D.H., unpublished results.

86. Huie, B.T.; Leyden, R.M.; Schaefer, W.P. Inorganic Chemistry 1979, 1^, 125.

87. Herron, N.; Grzybowski, J.J.; Matsumoto, N.; Zimmer, L.L.; Christoph, G.C.; Busch, D.H. J. Chem. Soc. 1982, 104, 1999.

Riley, D.P.; Stone, J.A.; Busch, D.H. to. Chem. Soc. 1976, 98, 1752.

Riley, D.P.; Stone, J.A.; Busch, D.H. 2- Chem. Soc. 1977, 99, 767.

88. Kubokura, K.; Okawa, H.; Kida, S. Bull. Chem. Soc. Jpn. 1978, 51,, 2036.

89. Carter, M.J.; Rillema, D,P.; Basolo, F. J. to. Chem. Soc. 1974, 9^, 392.

90. Drago, R.S. "Physical Methods in Chemistry," W.B. Saunders Company, 1977.

91. Silverstein, R.M.; Bassler, G.C.; Morrill, T.G. Spectrometric Identification of Organic Compounds, 3rd Edition, Wiley and Sons, Inc., 1972.

92. Chavan, M.; Busch, D.H., unpublished results.

93. Jones, R.D.; Summerville, D.A.; Basolo, F. Chem. Rev. 1979, 79(2). 139.

94. Busing, W.E.; Levy, H.A. J. Chem. P h y s . 1957, 2 6 , 563. 212

95. "International Tables for X-Ray Crystallography," Kynoch Press, Birmingham, 1974, Vol. I.

96. DuChamp, D.J. Paper B-14 "Programs and Abstracts," American Crystallographic Association Meeting, Bozeman, Montana, 1964 as modified by G.C. Christoph at The Ohio State University.

97. Mari, M.; Weil, J.A. ; Kinnaird, J.K. J. Phys. Chem. 1967, 71, 103.

98. Sykes, A.G.; Weil, J.A. Prog. Inorg. Chem. 1970, 13, 1.

99. Vagt, L.H.; Faigenbaum, H.M.; Wiberly, S.C. Chem. Rev. 1963, 62, 269.

100. Wilkins, R.G. Ad v . Chem. Ser. 1971, 100, 111.

101. McClenden, G . ; Martell, A.C. Coord. Chem. Review 1976, 1 9 , 1.

102. Martell, A.C. A c c t s . Chem. Resh. 1982, 12, 155.

103. Haskins, N.; Busch, D.H., unpublished results.

104. Pierce Catalog, 1981-1982.

105. Nishida, Y.; Kida, S. Coord. Chem. Rev. 1979, 27, 275.

106. Gilles, J. & Sievers, R. Unpublished results at Univ. of Colorado, Boulder.

107. Janzen, L.G. A c c t s . Chem. Resh. 1969, 2» 279. Janzen, L.G. Accts. Chem. Resh. 1971, 4^, 31.

108. Janzen, L.G.; Evans, C.A.; Davis, E.R. ACS Symposium Series 1978, M , 433.

109. Evans, C.A. Aldrichimica Acta. 1979, 1 2 , 23.

110. Sawyer, D.T.; Valentine, J.S. Accts. Chem. Resh. 1981, 1 4 , 393.

111. Valentine, J.S., "Biochemical and Clinical Aspects of Oxygen," Caugley, W.S., ed.; Academic Press: New York, 1979; pp. 659-677.

112. Nishinago, A. ; Nishizawa, K. ; Tamita, H. ; Matsuu Ra, T. 2* Am. Chem. Soc. 1976, 9£, 1287.

113. Hishinago, A.; Tamita, H. J. Mol. Cat. 1980, T_, 179.

114. Zombeck, A.; Drago, R.S.; Carden, B.B.; Gaul, J.H. J. Chem. Soc. 1981, 703, 7580.