MSU ’ LIBRARIES .n—L RETURNING MATERIALS: P1ace in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. PILLARED COFACIAL PORPHYRINS: SYNTHESIS, STRUCTURE, AND APPLICATION By Ismail Abdalmuhdi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1985 ABSTRACT PILLARRD COFACIAL PORPHYRINS: SYNTHESIS, STRUCTURE, AND APPLICATION By Ismail Abdalmuhdi A series of diporphyrin and monoporphyrin compounds has been synthesized and examined for the catalytic activity toward the four-electron reduction of dioxygen on graphite surfaces, a process that is very essential to the cathode reaction in the air-powered batteries (i.e., fuel cells). Stepwise synthetic methods were developed based on coupling of ethyl 3—ethyl-4-nethyl-2-pyrrolecarboxylate with aldehyde compounds to produce methine-substituted 5,5’-bis(ethylcarboxylate)dipyrrylmethanes which upon hydrolysis and decarboxylation gave the corresponding «- free dipyrrylmethanes. Porphyrins were obtained from the a-free dipyrrylmethanes upon coupling, in acidic media, with 5,5’-diformyl-3,3’-diethy1-4,4’-dimethyl-2,2’-dipyr— rylnethane. Meso di-substituted porphyrins were obtained in good yields by direct coupling of mono-aldehydes with 5,5’-unsubstituted 3,3’-diethyl-4,4’-dimethyl-2,2’- dipyrrylmethane. All the free bases of the porphyrin compounds were isolated by chromatography and further Abdalmuhdi, Ismail purified by recrystallization from CH30H/CH2C12 solution before the metal ions into the porphyrin macrocylic ligands. The catalytic activities were studied by the technique of rotating ring—disk electrode cyclic voltammetry. It was found that among all the examined compounds, the cofacially arranged, 1,8-anthracene-(DP-A) and 1,8—biphenylene-dipor- phyrins (DP—B) were the most effective and stable catalysts for the oxygen four-electron reduction process. Most other porphyrins were found to catalyze the two-electron reduction to hydrogen peroxide. X-ray crystallographic studies on (DP-A) and (DP—B) have shown a plane-to-plane separation of 3.88 A and 3.45 é, respectively. Although there is no clear connection between the catalytic activity (toward four- electron reduction) and the interplanar separation of the porphyrin rings, it was observed that the geometry of the macrocycles in these catalysts has a profound impact on their catalytic activities. TO MY FAMILY ACKNOWLEDGEMENTS I would like to express my sincere appreciation to Professor C.K. Chang for his inspiration, guidance, and support during the course of this work and the preparation of this dissertation. I would also like to thank all my colleagues for the fruitful discussions, and for their cooperation and encouragement. Financial support from the National Science Foundation, Dow Chemical Company in the form of a Summer Fellowship, and SOHIO Company in the form of a one year Industrial Fellowship are also acknowledged. Above all, I would like to thank my family for their faith and encouragement which made all of this possible. Special thanks are due to Ms. Margaret Lynch for her contribution by_typing this dissertation. -iii- TABLE OF CONTENTS PAGE LIST OF TABLES. . . . . . . . . . . . . . . . . . . . .vii LIST OF FIGURES. . . . . . . . . . . . . . . . . . . .viii LIST OF SCHEMES. . . . . . . . . . . . . . . . . . . .xiii LIST OF APPENDIX FIGURES. . . . . . . . . . . . . . . . xv. CHAPTER 1 INTRODUCTION. . . . . . . . . . . . . . . . . l A. Significance and Objectives of the present workO O O O O O O O O O O O O O O O O O O O O l B. Cytochrome Oxidase. . . . . . . . . . . . . . 2 C. The Chemistry of Dioxygen Reduction. . . . . 6 D. Fuel Cells. . . . . . . . . . . . . . . . . .10 CHAPTER 2 SYNTHESIS. . . . . . . . . . . . . . . . . . 13 A. Cofacial Porphyrins. . . . . . . . . . . . . 13 3. Mesa Mono-Substituted Porphyrins. . . . . . .25 C. Meso Di-Substituted Porphyrins. . . . . . . .30 D. Metal Insertion. . . . . . . . . . . . . . . 37 i. Insertion of Copper, Zinc, Nickel and Manganese. . . . . . . . . . . . . .38 ii. Insertion of Iron. . . . . . . . . . . .38 iii. Insertion of Cobalt. . . . . . . . . . .38 iv. Insertion of Mono-Metal in Diporphyrins. . . . . . . . . . . . . . 38 v. Insertion of Mixed-Metal in Diporphyrins. . . . . . . . . . . . . . 38 -iv- PAGE CHAPTER 3 ELECTROCHEMISTRY. . . . . . . . . . . . . . .40 A. Introduction. . . . . . . . . . . . . . . . 40 B. Rotating Ring-Disk Voltammetry. . . . . . . 46 C. Data and Results. . . . . . . . . . . . . . 49 i. Volta-metric Responses of the Diporphyrins in the Absence of Oxygen. . . . . . . . . . . . . . . . .49 ii. Catalysis of the Reduction of Oxygen. . . . . . . . . . . . . . . . .55 iii. pH Dependence of Oxygen Reduc- tion. . . . . . . . . . . . . . . . . .63 iv. Catalysis of the Reduction of H202. . . . . . . . . . . . . . . . . .65 v. pH Dependence of 8202 Reduction. . . . 69 D. Discussion. . . . . . . . . . . . . . . . . .71 E. Conclusion. . . . . . . . . . . . . . . . . .79 CHAPTER 4 MOLECULAR STRUCTURE AND PROPERTIES. . . . . .80 A. Electronic Spectroscopy. . . . . . . . . . .81 B. Nuclear Magnetic Resonance Spectroscopy. . . 84 C. Electron Spin Resonance Spectroscopy. . . . .87 i. Interaction of Oxygen with Dimetallic Diporphyrins. . . . . . . . .88 ii. EPR Spectroscopy of Dicopper Diporphyrins in Frozen Solutions. . . . 94 D. X-Ray Crystal Structure of Diporphyrins. . . . . . . . . . . . . . . .108 EXPERIMENTAL. . . . . . . . . . . . . . . . . . . . . .117 PAGE A. Reagents and Solvents. . . . . . . . . . . .117 B. Physical and Spectroscopic Methods. . . . . 117 C. Experimental Procedures. . . . . . . . . . .118 APPENDIX. . . . . . . . . . . . . . . . . . . . . . . .177 REFERENCES. . . . . . . . . . . . . . . . . . . . . . .202 -yi- TABLE LIST OF TABLES PAGE Distances and angles obtained for dicopper diporphyrins. (a) Copper-copper distances obtained from the relative intensities of the half-field transitions. (b) Copper-copper distances obtained by simulation of the allowed transitions. (c) Angles between the z-axes of the copper g and A tensors and the interspin vector. (d) Separation between the two parallel porphyrin planes. (e) Percentage of formation of H202 evaluated from 3 H202 = 2/[1+(ni )/(iR)]’ where i and i are ring and disR limiting currents, respectively, and N(= 0.182) is the collection coefficient; these data were measured by using dicobalt diporphy- rins coated on the graphite of a ring-disk electrode immersed in Oz-saturated 0.5 M aqueous trifluoroacetate acid. . . . . . . . . . 107 Comparison of distances and angles obtained by X-ray with those obtained by EPR for cofacial porphyrins. . . . . . . . . . . . . . . 115 -vii- FIGURE 1 LIST OF FIGURES PAGE A schematic representation of the gross structure of the redox centers in a membranous cytochrome oxidase. . . . . . . . . . .3 (a) A schematic representation of the asymmetric reduction of cytochrome oxidase by cytochrome g. In this case cytochrome g is shown introducing electrons into only one of the iron—copper couples with the reduction of dioxygen to water induced by electrons that originate on only one of the iron atoms. (b) A schematic view of proton translocation by cytochrome oxidase. The point of 02 reduction has been drawn near the C side of the membrane, but may also be located closer to the M side. Uptake of H’ from the M side and release on the C side show the net observed stoichiometry (per transferred electron). If the ”substrate" protons required to reduce 02 to H20 are taken from side C, the proton pump must translocate 2R’/e‘ across the membrane in order to preserve the overall stoichiometry observed . . . . . . . . . . . . . . . . . . . . .5 (a) A single cell in the General Electric ion exchange-membrane H2/02 fuel cell. (b) Assembly of a single cell in the General Electric Hz/Oz fuel cell. . . . . . . . .11 Standard reduction potentials of oxygen, in volts versus SHE. . . . . . . . . . . . . . . 41 A proposed reaction of 02 with cofacial binary metalloporphyrins. B is an axial ligand too bulky to fit in the cavity. Ovals represent porphyrin rings. . . . . . . . . 42 Structures of porphyrins previously examined for electrocatalytic activity. . . . . . . . . . 44 -viii- FIGURE 7 10 ll 12 PAGE Structures of porphyrins examined in this study. . . . . . . . . . . . . . . . . . . . . . 45 A schematic depiction of a rotating ring- disk electrode. . . . . . . . . . . . . . . . . .47 Cyclic voltammograms for 1.2 x 10'9 mol-cm'2 of complex 22 adsorbed on graphite elec- trodues: dashed line, electrode coated with cobalt free diporphyrin, §; dotted line, background current at an uncoated electrode. Supporting electrolyte: l M CF3000H saturated with argon. Scan rate: 100 mV s‘1. . . . . . . . . . . . . . . . . . . .51 Cyclic voltammograms for 1.37 mM g; (A), §b (B), and § (C) in dichloromethane, using polished glassy carbon electrode (0.34 cmz). Supporting electrolyte: tetrabutylammonium perchlorate. Scan rate: 100 mV s‘l. . . . . . .54 Current-potential curves for the reduction of 02 at the rotating graphite disk-platinum ring electrode. The polished pyrolytic graphite disk was coated with 1.2 x 10‘9 mol-cm'2 of (A) complex §a or (B) complex §b. Ring potential: 0.9 V. Rotation rate: 100 rpm. Supporting electrolyte 1 M cracooa saturated with 02. The disk potential was scanned at 10 mV s'l. The dashed curves are the disk and ring currents obtained under the same conditions from coatings with the active amide—linked cofacial porphyrin CozDP-4 . . . . . . . . . . . 57 Levich (A) and Koutecky-Levich (8) plots of the plateau current for the reduction of oxygen at graphite electrodes coated with 1.2 x 10'9 mol-cm‘2 of complex §a [ I ] or complex §b [ O ]. Supporting electr- lyte: l M CFacOOH saturated with air. The dashed lines are the calculated responses for the correction-diffusion limited reduction of 02 by four electrons, taking [02] = 0.24 mM and D = 1.8 x 105 cm2 s‘1..... ...°".’.............62 4x- FIGURE l3 l4 15 16 17 18 PAGE pH Dependence of plateau currents for the reduction of 02 at rotating graphite disk electrodes coated with 1.2 x 10'9 mol-cm'2 of complex 69 [ O ] or complex 66 [ I ]. Electrode rotation rate: 100 rpm. Other conditions as in Figure 12. . . . . . . . . . . .64 (A) Current-potential curves for the reduc- tion of 1 mM H202 at rotated graphite disk electrodes: uncoated electrode (1); elec- trode coated with 1.2 x 10‘9 mol-cm'2 of complex 66 (2); complex 66 (3); or complex 46 (4). Supporting electrolyte: -1 M CFaCOOH saturated with argon. (B) Levich plots for H202 reduction as catalyzed by complex 66 [ O ] or complex 66 [A ]. (C) The corresponding Koutecky-Levich plots. The dashed lines were calculated for the diffusion-convection limited reduction by two electrons. . . . . . . . . . . .68 pH Dependence of plateau currents for the reduction of 1 mM H202 at rotating graphite disk electrodes coated with 1.2 x 10'9 mol-cm‘z of complex 66 [I ] or complex 69 [ O ]. Supporting electrolytes at each pH as in Figure 10. Rotating rate: 400 rpm. . . . . . . . . . . . . . . . . . . . . 70 Plateau currents for the simultaneous reduction of 02 and H202 at a rotating graphite disk electrode coated with 1.2 x 10" mol-cm'2 of 6g. (A) reduction of H202 in the absence of 02; (B) repeat of (A) after the solutions were saturated with 02; (C) difference between the plateau currents in (B) and that for an 02- saturated solution in the absence of H202. Supporting electrolyte: 1 M CFaCOOH. Electrode rotating rate: 400 rpm. Plateau currents measured at -0.3 V. . . . . . . . . . . 74 A proposed mechanism for the four-electron pathway of the 02-reduction to water, catalyzed by dicobalt complexes of cofacial diporphyrins in acidic media. . . . . . . . . . .77 Hydrogen bonding in the dioxygen-monocobalt- porphyrin adducts. . . . . . . . . . . . . . . . 78 -x- FIGURE 19 20 21 22 23 24 25 PAGE Absorption spectra of anthracene monopor- phyrin (MP-A), anthracene diporphyrin (DP-A), and biphenylene diporphyrin (DP-E) in CH2012. . . . . . . . . . . . . . . . .82 250 MHz NMR of anthracene monoporphrin (a), anthracene diporphyrin (b), and biphenylene diporphyrin (c) in CD013. . . . . . . . . . . . .86 A schematic presentation of the interaction between 02 and dimetallodiporphyrins. . . . . . .89 EPR spectra of p-superoxo complexes of CozDP-A (A) and CozDP-D (B). Spectra were obtained by reacting the his Co(II) diners with dioxygen at 23'C in CHzclz containing 0.1 M N-tritylimidazole and a trace amount of iodine. . . . . . . . . . . . . . . . . . . . 92 X—Band EPR spectra of the allowed transitions for Cu2DP-4 (A), Cu2DP-5 (B) and CuzDP-H (C) at -180'C in 1:1 toluene/CHzClz solution. The spectra were obtained on 1 mM solutions with 1 mW microwave power and 4 G modulation amplitude. The peak marked "A" was attri- buted to aggregated material. The dotted lines indicate regions in which the calculated curves don’t overlay the experimental data. . . . . . . . . . . . . . . . 96 X-Band EPR spectra of the half-field transi- tions for Cu2DP-4 (A), Cu2DP-5 (D), and CuzDP-B (C) at -180'C in 1:1 toluene/CH2C12 solution. The spectra were obtained in 1 mM solutions with 20 mW microwave power at 16 G modulation amplitude. The overall amplification of the spectra in this figure is about 35 times that for Figure 23. The dotted lines indicate regions in which the calculated curves don’t overlay the experimental data. . . . . . . . . . . . . . 98 X-Band EPR spectra of the allowed transi— tions for Cu2DP-7 (A), CuzDP-A (B), and slipped Cu2DP-4 (C) at -180'C in 1:1 toluene/CHaClz solution. The spectra were obtained under conditions identical with those described in Figure 23. . . . . . . . . . 101 FIGURE _ PAGE 26 X-Band EPR spectra of the half-field transi- tions for Cu2DP-7 (A), Cu2DP-A (B), and slipped Cu2DP—4 (C) at ~180'C in 1:1 toluene/CH2C12 solution. The spectra were obtained in 1 mM solution with 200 mW micro- wave power and 16 G modulation amplitude. . . . 103 27 ORTEP Drawing of NizDP-A excluding the side groups. Views approximately mutually perpendicular: (a) parallel to C(m2)- C(m4) direction; (b) parallel to C(7m1)- (Cma) direction; (c) nearly perpendicular to porphyrin planes, ring 1 shaded. . . . . . . . .110 28 ORTEP Drawing of Cu2DP- B excluding side groups. Otherwise as in Figure 26. . . . . . . . . . .111 29 Perspective stereoview of NiaDP-A (a) and CuzDP-D (b). . . . . . . . . . . . . . . . . . .114 -xii- SCHEME 1 10 11 LIST OF SCHEMES General structure of amide-linked diporphyrins. . . . . . . . . . . . . Synthesis of 1,8-anthracene diporphyrin (6) by coupling of the a- -free dipyrryl- methane (4) with 5, 5’ -diformyl- -3, 3’ - diethyl- 4, 4’ -dimethy1- 2, 2’ -3ipyrryl- methane (5).. . . . . . . . . . Synthesis of 1.8-anthracene diporphyrin (6) by coupling the a-free dipyrryl- methane (4) with bis(methoxymethyl) dipyrrylmethene hydrobromide (7) or bis(bromomethy1) dipyrrylmethene hydrobromide (6). . . . . Synthesis of l, 8- diphenylene diporphyrin (14).. . . . . . . . . . . . . . . . Synthesis of "triple-deckered" triporphyrin (66). . . . . . . . . . . . . . . . . Approaches for the synthesis of the meso- dianthracene diporphyrin (64). . . Synthesia.of bis(2-pyridy1-B-ethyl)amino anthryl porphyrin (66) and crown- anthryl porphyrin (64). . . . . . . . . Synthesis of bis(2-pyridyl-B-ethy1)amino- naphthyl porphyrin (46). . . . . . . . . Synthesis of meso mono-substituted aryl porphyrins (44-42). . . . . . . . . . . . Synthesis of meso mono-phenyl mono-anthryl porphyrin (46). . . . . . . . . . . . . . Synthesis of meso bis(carboxyanthryl) porphyrin (66). . . . . . . . . . . . . -xiii- PAGE .13 15 .17 .19 .21 23 .28 29 .31 .32 .34 SCHEME PAGE 12 Synthesis of chiral dianthrylporphyrin derivative (61). . . . . . . . . . . . . . . . . 35 13 Synthesis of meso di-aryl porphyrins (66- 66). . . . . . . . . . . . . . . . . . 36 14 (Single and mixed) metal insertion in diporphyrins. . . . . . . . . . . . . . . . . . .39 15 Structure of the cofacial diporphyrins studies by EPR spectroscopy in frozen solutions. DP = diporphyrins, DP-A = 1,8- anthracene diporphyrin, DP-B = biphenylene diporphyrin. . . . . . . . . . . . .93 -xiv- FIGURE A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 LIST OF APPENDIX FIGURES 250 MHz 1H NMR spectrum of 1,8-bis{5,5’- bis(ethoxycarbonyl)-4,4’-diethyl-3,3’- dimethyl-Z,2’-dipyrryl]methy1} anthracene (6). . . . . . . . . . . 60 MHz 1H NMR spectrum of 1, 8- -bis[(4, 4’- diethy1-3, 3’ -dimethy1- -2, 2’ -dipyrryl) methyl]anthracene (4). . . 60 MHz 1H NMR spectrum of 8-methoxy- carbonyl-l-anthracene carboxaldehyde (61). 60 MHz 1 NMR spectrum of p-methoxy- a,a[5, 5 ’-bis(ethoxycarbony)- -4, 4’ -diethy1- 3, 3’ -dimethy1- -L 2’ -dipyrryl]toluene (67).. . . . . . . . 60 MHz 1H NMR spectrum of p- methoxy- a,a(5, 5’ -diformy1- 4, 4I -diethy1- -3, 3I - dimethyl- 2, 2 ’-dipyrry1)toluene (66). 250 MHz 1H NMR spectrum of l, 8- -bis[5- (2, 8, 13, 17- -tetraethyl- -3, 7, 12, 18- tetramethyl)porphy- rinyl]anthracene (6). 250 MHz 1H NMR spectrum of l, 8- -bis[5- (2, 8, 13, 17- -tetraethy1- -3, 7, 12, 18- tetramethyl)porphy- rinyllbiphenylene (14). . . . . . 250 MHz 1H NMR spectrum of 1-[5-(2,8,13,17- tetraethy1-3,7,12,18-tetramethyl)porphyrinyl] anthracene (46). 250 MHz 1H NMR spectrum of 5-(8-formy1-l- anthryl)-2,8,13,17-tetraethyl-3,7,12,18- tetramethylporphine (16). . . . . 250 MHz 1H NMR spectrum of 5-(8-formyl-1- naphthyl)-2,8,l3,17-tetraethy1-3,7,12,18- tetramethylporphine (61). . . 250 MHz 1H NMR spectrum of 1,8-bis{5- [2,8,12,18-tetraethyl-3,7,l3,17-tetramethyl- l5-(p-methoxyphenyl)porphyrinyl}anthracene (199)............... PAGE .177 178 179 .180 181 .182 .183 184 .185 .186 187 FIGURE PAGE A12 250 MHz 1H NMR spectrum of 5-[8-(hydroxy- methy1)-1-anthryl]-2,8,13,l7-tetraethy1- 3,7,12,lB-tetramethylporphine (16). . . . . . .188 A13 250 MHz 1H NMR spectrum of 5-{8-[(methane- sulfonate)methyl]- -l- -anthry1}- -2, L 13,17- tetraethyl-—3,17,12,lB-tetramethylporphine (60).. . . . . . . . . . . . . .189 A14 250 MHz 1H NMR spectrum of trans-5,15-bis{8- [5-(2,8,13,17-tetraethyl-3,7,12,18-tetra- methylporphyrinyl]-1-anthryl}-2,8,12,18- tetraethyl-3,7,13,l7-tetramethylporphine (66). . . . . . . . . . . . . . . . . . . . . .190 A15 250 MHz 1H NMR spectrum of 5-{[8-{bis(2- pyridyl- B-ethyl)amine}methyl]- -1- -anthryl}- 2, 8, 13, 17- -tetraethy1- -3, 7, 12, 18- tetramethyl- porphine (62). . . . . . . 191 A16 250 MHz 1H NMR spectrum of 5- -pheny1- -2, L 13,17- ’ tetraethyl- -3, 7, 12, 18- tetramethylporphine (44). . . . . . . . . . . . . .192 A17 250 MHz 1H NMR spectrum of 5- (L carboxyl- l- anthry1)-L L 13, l7-tetraethy1- 3, 7, 12, 18- tetramethylporphine (82). . . . . . . . .193 A18 250 MHz 1H NMR spectrum of 1-[5(2,8, 13,17- tetraethyl-—3,7,12,18-tetramethyl)porphy- rinyl]naphthalene (46). . . . . . . .194 A19 250 MHz 1H NMR spectrum of trans-5,15-bis(l- naphthy1)-2,8,12,18-tetraethyl-3,7,13,17- tetramethylporphine (66). . . . . . . . . . . .195 A20 250 MHz 1H NMR spectrum of 5,15-bis(4- methoxyphenyl)- L 8,12,18-tetraethy1-3,7,13, 17- tetramethylporphine (66). . . . . . . . . 196 A21 250 MHz 1H NMR spectrum of trans-5,15- bis[8-methoxycarbonyl-1-anthryl]-2,8,12,18- tetraethyl- -3, 7, 13, 17- -tetramethylporphine (66). . . . . . . . . . . . . . . . . . . .197 -xv¢- FIGURE A22 A23 A24 A25 250 MHz 1H NMR spectrum of trans-5,15- bis[8-hydroxymethy1-l-anthryl]-2,8,12,18- tetrsethy1-3,7,13,17-tetramethylporphin (99)................ 250 MHz 1H NMR spectrum of cis-5,15-bis[8- hydroxymethyl-l-anthryl]2,8,12,18-tetra- ethyl-3,7,13,17-tetramethy1porphine (66). 250 MHz 18 NMR spectrum of trans-5,15- bis[8-nitro-1-naphthy1]-2,8,12,18- tetraethy1-3,7,l3,17-tetramethy1porphine (gg)................ 250 MHz 1H NMR spectrum of cis-5,15-bis[8- nitro-l-naphthyl]-2,8,12,18-tetraethy1- 3,7,13,17-tetramethy1porphine (666). «adi- PAGE .198 .199 .200 . 201 CHAPTER 1 INTRODUCT ION CHAPTER ONE INTRODUCTION A. Significance and Objectives of the Present Work A major part of the effort in this work was focused on the synthesis and study of binuclear metal complexes of covalently linked cofacial diporphyrins. The two porphyrin rings in these diporphyrins were held in a face—to-face configuration by rigid spacers of different lengths. These models are considered of great significance. Syntheses of these novel compounds were themselves challenging, since they required the application of various organic and inorganic synthesis techniques. In addition they have the unusual capability of placing two metal ions at selected distances and thus can display interesting pro- perties arising from metal-metal interactions. Further- more, and from a biochemical point of view, these compounds represent a class of elaborately designed bioinorganic models for many essential biological systems, such as the chlorophyll "special pair" photosynthetic units [1,2]; chlorophyll aggregate models for studying excitation energy and charge transfer proceses [3]; cytochrome g oxidase models that are capable of catalyzing the multielectron reduction of oxygen; and monooxygenases models by which -2- molecular oxygen can be activated. In particular, metal complexes of cofacial diporphyrins have recently received considerable attention [4-13]. When adsorbed on graphite, the dicobalt derivatives of these dimers were able to catalyze the four-electron reduction of oxygen to water at ca. +0.68 V vs NHE at low pH [13]. Therefore, these electrocatalysts are very promising, since they may constitute a major contribution in fuel cell tech- nlogy by replacing the expensive platinum in the elec- trodes. It is also hoped that the the availability of these compounds will offer a better understanding of the mechanism of dioxygen reduction. B. Cytochrome Oxidase Cytochrome g oxidase, the terminal enzyme in the respiratory metabolism of all aerobic organisms, i.e., animals, plants, yeasts, algae, and some bacteria is responsible for catalyzing the reduction of dioxygen to water. This enzyme is bound to the mitochondrial inner membrane of the living cell (Fig. l), and it is considered very essential for the oxidation of foodstuff and genera- tion of energy [14-16]. The catalytic activity of cyto- chrome g oxidase toward the reduction of oxygen is demon- strated by its ability to transfer reducing equivalents from ferro-cytochrome g to molecular oxygen, as shown by -3- Equation (1). The electrons are provided by reduced cyto- chrome c: . 02 + 4Cyt (:23 4H+—» 2H20 + 4Cyt c3+ <1) The free energy generated in the process is usually used to promote oxidative phosphorylation and consequently becomes available as ATP, to satisfy the energy require- ments of the living cell. For instance, it is estimated that 908 of the energy for heart muscle contraction is provided through aerobic metabolism via cytochrome Figure 1 A schematic representation of the gross structure of the redox centers in a membranous cytochrome oxidase [16b]. -4- oxidase [17,18]. The essential nature of cytochrome oxi- dase is exemplified by the fact that it is considered res- ponsible for more than 90% of the oxygen consumption by living organisms on earth [19]. Biologically, the general significance of cytochrome g oxidase is believed to exceed that of hemoglobin, since the latter is only an auxiliary in the cell respiration process and its function is solely to carry oxygen to the tissues via the bloodstream. This is essential only in bulky ani- mals, where oxygen diffusion through the surface is usually not sufficient [14a]. It is well established that the active unit of the en- zyme contains two heme groups (heme g and 23), and two protein-bound copper ions (CuA and Gun). The major difference between heme g and heme 33 is that the first is usually of low spin and doesn’t bind ligands, while heme 53, on the other hand, is of high spin and has the ability to bind various ligands, such as 02 and CO in the ferrous state and HON, H S and RN 2 3 Although the structural relationship between iron and in the ferric state [20]. copper of the enzyme is far from established, the magnitude of the antiferromagnetic coupling observed for the oxidized Fe3+-Cu2+ suggests that the iron and copper atoms are sepa- rated by no more than a few atoms [21,22]. The functioning enzyme which is provided with electrons from the electron transport chain by cytochrome 2, uses -5- these electrons to reduce dioxygen bound at the active site, then communicates the energy released in this reduc— tion to the site of oxidative phosphorylation, (Fig. 2). Out In (C) (M) Why/WW; ,, \\ 3 Cu A IH‘ § 0 F' ' IlinO %02 Figure 2 (a) A schematic representation of the asymmetric reduction of’cytochrome oxidase by cytochrome 95 In this case cytochrome g.is shown introducing electrons into only one of the iron—copper couples with the reduction of'dioxygen to water induced by electrons that originate on only one of the iron atoms. (b) A schematic view of’proton translocation by cytochrome oxidase. The point of 02 reduction has been drawn near the 0 side of the membrane, but may also be located closer to the M side. Uptake of'H+ from the M'side and release on the C side show the net observed stoichiometry (per transferred electron). .Lf the "substrate" protons required to reduce 03 to 320 are taken from side 0, the proton pump must translocate 2H+/e across the membrane in ordgr to preserve the overall stoichiometry observed [25 ]. -5- C. The Chemistry of Dioxygen Reduction The chemical inertness of dioxygen at first seems sur- prising because the transformation to water is so strongly thermodynamically favorable (580 kcal/mole) [23,24]. However, on the basis of the standard redox potentials, the simplest reduction step, the one—electron step to super- oxide, is thermodynamically highly unfavorable [25]. Hence reactions involving dioxygen must either have enormous driving energies to go through the superoxide or have access to a two-electron step to peroxide. It is most likely that oxygen reduction by a low energy pathway proceeds via the two-electron reduction to peroxide as the first recognizable product [26]. The other property of dioxygen that contributes to the slowness of its reactions is its electronic structure. In common with most stable molecules, dioxygen has an even number of electrons. Uncommonly, though, the molecule is paramagnetic with two unpaired electrons in the two highest occupied molecular orbitals. Since both peroxide and oxide are completely spin paired, reactions involving dioxygen must involve spin reversal and are therefore spin forbidden and slow. The forbiddance can be removed if dioxygen can interact with a paramagnetic center to participate in exchange coupling. The transition metal ions frequently have unpaired electrons and turn out to be excellent cata- -7- lysts for dioxygen reduction [27]. Despite the long history of transition metal ion induced reduction of dioxygen, surprisingly little is known about the mechanism of the reaction [28]. Perhaps the most studied of these metal ion oxygenation reactions is that between the nitrogen ligand complexes of Co(II) and dioxy- gen. Upon oxygenation, a cobalt(III) complex is formed. It is now known that the first step in this reaction is the formation of an unstable CoII -O2 [29], and upon standing, a second Co(II) ion is added to produce the p-peroxo bridged complex (reactions 2 and 3). In the presence of a second bridging ligand, such as amido- or hydroxo—, the u-peroxo complex is stabilized and can be isolated (reaction 4). In the absence of the second bridging ligand, however, further an E: Co (2) E E Con-02‘ + can ——-> Cdn /"‘Com (3) G / 3, ENE RE! I 0 “MI am am am Com /O\Co _L_, Com\/ L\/C0m (4) E°E EO-OE moa Co / ‘Co -——> 2Co— OH (5) E 0 E \\\\\\ -3- oxidation by a series of, as yet, not completely understood steps and the cobalt(III) product results (reaction 5) [30]. Of special relevance to cytochrome g oxidase, the reac- tions of hemes [31,32] as well as simple aquo ferrous ions [33] with dioxygen seems to proceed through two-electron reduction of bridged intermediates. Thus, unprotected heme models react cleanly with dioxygen to form p-oxobishemins irreversibly, as shown by reactions (6) through (9). This indeed represents a major problem in trying to mimic the 02 or CO binding of hemoglobin or myoglobin by synthetic models. II II Py—Fe—Py ————’- Py_Fe + Py (6) II II H' II PY-Fe"02+ Py—Fe ———->- Py-Fen-Of FgPy (8) El HI II IE Py-Fe-oz-Fe-Py “>- Fe-O-Fe + zPy (9) When the solvent medium is able to provide protons, solvolysis to produce H202 would likely follow the formatin -9- of the p-peroxobishemin complex and the sequence (6)-(8). However, where solvolysis cannot occur, as in .O n: . Fen: / \Fem ___>2Fe-O (IO) 0 Fg—O. + Py—FéI1 _, Fg-o—Fén+ Py U I) Iv aprotic solvents, formation of the ferryl [Fe -0] inter- mediate, as shown in reactions (10) and (11), is reasonable [32]. But in no instance does our current knowledge about the mechanism of the reduction reaction extend beyond the bridged dimer. Apparently, the peroxide formed in the ini- tial reaction is a kinetically inert (fully spin-paired) molecule that does not readily accept additional electrons despite the favorable thermodynamics for reduction to water. The organization of cytochrome g oxidase to carry out the reduction of dioxygen to water must reflect the mecha- nism by which electrons are supplied, the thermodynamic requirements of the oxidation states accessible to oxygen, and the chemical transformation of oxygen enroute to water. Many mechanistic approaches for this process have been suggested although none of them has been fully documented and thus remains speculative [34]. However, we suggested a reasonable mechanism for the reduction of dioxygen to water by dicobalt cofacial diporphyrin catalysts, cf. Chapter III (Electrochemistry). -10- 0. Fuel Cells Among the large variety of fuel cells, the oxygen- hydrogen cell is considered the most advanced [35,36]. The basic design of the cell is shown in Pig. 3, which consists of a solid electrolyte ion-exchanger membrane, electrocata- lysts, current collectors, coolant tubes, water wicks, and gas-feed tubes. The membrane is nonpermeable to the reactant gases, hydrogen and oxygen, which thus prevent them from coming into contact, but permeable to hydrogen ions, which are the current carriers in the electrolyte. The two electrodes in the fuel cell, which consist of the electrocatalyst (usually finely divided platinum metal) and a plastic material for water-proofing, are in the form of finely metallic wire screens. Oxygen, either pure or in air, is as important to elec- trochemical conversion as it is to chemical combustion and to life [37]. Thus, particularly all earth based fuel cells use 02 asflthe cathodic reactant. Oxygen is reduced at the cathode, in acid solution, to water through a four- electron process according to Equation (12) 02 + 4H*+4e- —» enzo' (I2) A node - +Cathode Gos chombers203 yield. Surprisingly, when the cyclization was carried out by the standard MacDonald procedure [58] (HI-HOAc, at room temperature) no evidence at all for the desired diporpyrin was observed. The bis(methoxymethyl)dipyrrylmethene, 6, was found to react with the unsubstituted dipyrrylmethane 4 in boiling benzene and subsequent oxidation produced the anthracene- diporphyrin 6 in 7-108 yield (Scheme 3). Bis(methyoxy- methyl)dipyrrylmethenes have been employed to construct b- bilenes [59-61] previously, but the condensation with di~ pyrrylmethane in such a direct manner to form porphyrin, to our knowledge, has not been reported. This route may be a useful alternative for the general (2+2) porphyrin cycliza- tions. The reaction is carried out in the absence of -17- m MIMIOm (DI .m -18- excess protons; thus there should be little danger of scrambling of substituents. Indeed, the sharp singlet NMR peaks of the meso protons, as well as that of the methyl groups in the diporphyrin, clearly ruled out any other sub- stitution patterns. we have applied this approach success- fully to prepare meso monophenyletio porphyrin and octa- ethylporphyrin from appropriate precursors [42]. The cyclization was also possible (although with lower yield) when bis(bromomethyl)dipyrrylmethene hydrobromide 6, was used instead of bis(methoxymethyl)dipyrrylmethene hydrobromide, as shown in (Scheme 3). The dipyrrylmethene hydrobromide, 6, was obtained by treatment of ethyl 4,5-dimethyl-3-ethyl-2-pyrrolecarboxy- late [62,63] with 48% hydrogen bromide in formic acid and reacting the resultant orange crystals with BrZ/HOAc. Bis- (methoxymethyl)dipyrrylmethene hydrobromide, I, was obtained from 6 by refluxing in dry methanol. In yet another attempt, trace amounts of the diporphy- rin, 6, was obtained by direct cyclization of 1.8-anthra- cene-dicarboxaldehyde with two equivalents of 1,19-dideoxy- tetraethyltetramethylbiladiene-ac dihydrobromide, 6, [49] in methanol and in the presence of HBr/HOAc, cf. (Scheme 2). Similarly 1,8-biphenylene diporphyrin 64 was obtained in 98 yield by condensation of l,8-biphenylenedicarboxalde- hyde 66 [64-67] and the a-free pyrrole 6, followed by saponification, decaboxylation, and coupling with 5,5’-di- NH NH '2 ‘— l-b HN MN -19- / I I \ \ / 'i CHO HO I I\ \ — / -* O__O 1.9. :2 ‘4 "\CO,C,H, H R / / \ H~ .. O ._ R -.. _ «‘— R HN \ / R \ O ,Léa R= co,c,H, bR=H SCHEME 4 -20- formyldipyrrylmethane 6 in 0.43 HClO4/methanol solution, (Scheme 4). The l,8-biphenylenedicarboxaldehyde 46 was prepared by DMSO oxidation of the dimethyl precursor 46 [65]. Methylbenzyne was condensed in situ [64] to yield a mixture of 1,8-46 and 1,5-dimethylbiphenylene 44. The separation of the two isomers was found to be much easier if the mixture was converted first to the dialdehydes and then the isomers isolated by chromatography. Attempts to condense 46 with a mixture of pyrrole and benzaldehyde invariably produced only tetraphenylporphyrin and an untractable material. The "triple-deckered triporphyrin" 66 (Scheme 5) [44], was synthesized by using the dipyrrylmethane aldehyde 46, as the key intermediate. 6 Was obtained in 73S yield by reacting 2 equivalents of pyrrole 6 and 1.8-anthracenedi- carboxaldehyde 4. The formyl group in 46 had to be converted into a nonreactive form before the dipyrryl- methane could be manipulated further. Thus 46 was reduced to the alcohol 46, which after saponification and decar- boxylation, was-cyclized into meso-substituted porphyrin 46 by using the modified MacDonald procedure [43] with an overall yield of more than 403. The aldehyde functional group was then restored by oxidation. The aldehyde porphy- rin 46, behaving analogously to a substituted benzaldehyde, was allowed to react with equivalent amounts of a-free dipyrrylmethane 46 to yield the triporphyrin 66. This approach has been used for preparing 5,15-diphenyl-porphy- -21.. (01 at I .szu \ :2 £6.00 m alum—Om :z@ -22- rin [68,69]. In principle, atropisomers can result from this reaction, but in this present case, because of the steric bulk, the alternative cis arrangement of the two anthryl groups about the center porphyrin was not possible. Although the conversion yield from 46 to 66 was not high, the unreacted aldehyde 46 could be recovered for recycling. We found that the most efficient and convenient way to isolate the cofacial porphyrins 6, 44, and 66 from the reaction mixture was to isolate their metal-complexes (usually Zn-complexes) by chromatography, since they have very different Rf-values from the impurities. The free- base diporphyrins were then regenerated from the pure zinc complexes by treatment with a dilute solution of hydro- chloric acid. Various attempts to prepare the rigid dianthryl-dipor- hyrin 64 have been made (Scheme 6). 1.8-Antracenedicarbox- aldehyde 4 was treated variably with 2 equivalents of 3,3’- diethyl-4,4’-dimethyl-2,2’-dipyrrylmethane 46 in methanol and in the presence of p-toluenesulfonic acid (Gunter’s method) [72], no diporphyrin was formed and the only por- phyrin obtained from this reaction was found to be trace amounts of etioporphyrin II in addition to the dark brown untractable polymeric material. To increase the possibility of the formation of 64, we believed that coupling of cis dianthrylporphyrin dialdehyde 64 (which was obtained by a stepwise method) with 2 equiva- lents of the a-free dipyrrylmethane 46 would lead to the -23- m.mlmfl0m -24- formation of 64. After stirring the reactants in metha— nol/methylene chloride (70:30) solution containing cata- lytic amounts of p-TsOH, for 20 h, the unreacted dianthryl- porphyrin 64 was recovered and no diporphyrin was formed. In another attempt, the bis(biladiene-ac) 66 was let to react with l,8-anthracenedicarboxaldehyde 4 in acetic acid. It produced a highly colored polymeric material, with no indication for the formation of 64. Similar results were also obtained by coupling bis(a-free dipyrrylmethyl)anthra- cene 4 and bis(5,5’-diformyl dipyrrylmethyl)anthracene 66, in acidic methanol. We suspected that the failure of formation of 64 may be caused by the improper orientation of the pyrrole rings and the high steric hindrance in the dipyrrylmethane intermediates. In an attempt to vary the distance between the porphy— rin rings in the dimers, we considered the possibility of using different aromatic connectors, such as 1,8-fluorene- dicarboxaldehyde 66 and 1,8-naphtha1ene dicarboxaldehyde hydrate, 66 (prepared according to literature) [70,71]. Because of the short distance between the aldehyde groups in these two species, they tend to form hemiacetal and ace- tal derivatives, especially when the reactions are carried out in acidic alcohol media. However, when the reaction between these dialdehydes and ethyl 3-ethyl-4-methyl-2- pyrrolecarboxylate 6 was carried out in non-alcoholic sol- vents such as benzene and in the presence of p-toluene- -25- sulfonic acid, no formation of dipyrrylmethane was achieved. HO 0 /OH OHC CHO \HC/ \CH 31 2.9. The difficulties in preparing the dipyrrylmethane intermediates from 66 and 66 could be mainly due to steric factors. B. Meso Mono-substituted Porphyrins Of major interest is the selectively controlled reac- tion between dialdehyde species (such as 1.8-anthracene and 1.8-diphenylene dicarboxaldehydes) with only two equiva- lents of the a-free ethyl 3-ethyl-4-methyl-2-pyrrole- carboxylate 6, in which one of the aldehyde groups is reacted while the other is left intact. Our major concern, however, was to apply this process in the synthesis of a series of important mono- and di-meso substituted porphyrins. -26- When l,8-anthracenedicarboxaldehyde 4 was reacted in ethanol with 2 equivalents of a-free pyrrole 6, it produced the dipyrrylmethylanthracene aldehyde 46, as a yellow crystalline solid, cf. (Scheme 5). To protect the aldehyde group, it was reduced to the alcohol. The decarboxylated a,¢-unsubstituted dipyrrylmethane 66 was coupled with 5,5’- diformyldipyrrylmethane 6 to produce the mesa substituted porphyrin 46. In addition, 66 was also used to prepare the meso disubstituted porphyrins, cf. Section C of this chapter. The anthracene porphyrin-alcohol 46 was converted into methanesulfonate derivative 66 by heating at reflux with excess amounts of methanesulfonyl chloride in dry dichloro- methane. The reaction progress was followed by thin layer chromatography (tlc), since the sulfonate ester product has a much higher Rf value than the porphyrin-alcohol starting material. The reaction was complete after ~48 h. It was observed that when amine bases such as pyridine or tri- ethylamine were added to the reaction, in an attempt to reduce the accumulation of the hydrogen chloride that evolved from the reaction, a very dark green solution was usually obtained and the reaction became messy, and lower overall conversion was obtained. However, it seemed that addition of such bases was not necessary since the porphy- rin’s nitrogens can trap the evolved HCl. The resultant sulfonate ester-porphyrin 66 could be isolated by chroma- tography, although in many cases, it was not necessary and -27- not recommended. After total removal of all solvents, the crude residue was used in the next step without any problems. To introduce the bis(2-pyridyl-B-ethyl)amine ligand, the crude sulfonate ester 66 was treated with relatively excess amounts of the amine 64 [72] in dry dichloromethane, (Scheme 7). The excess amine was removed by washing the reaction solution with dilute hydrochloric acid (~53) taking advantage of the difference in solubility of the bis(2-pyridyl-B-ethyl)ammonium chloride and the ammonium salt of 66 in water. Under similar reaction conditions, sulfonate ester-por- phyrin 66 was used to prepare porphyrin-crown 64 by coupling with Kryptofix-22, 66. 1.8-Napthalene dialdehyde or its hemiacetal derivative 66 failed to react with a-free pyrrole after several varia- tions in reaction conditions. Hydrolysis of the hemiacetal linkage in acidic solution before the reaction also gave a negative result. To avoid this obstacle, acenaphthenequinone 66, was reacted smoothly [73] with 2 equivalents of ethyl 3-ethyl- 4-methyl-2-pyrrole-carboxylate 6 in ethanol to give the di- pyrrylmethane derivative 66 in a very good yield (Scheme 8). Under basic hydrolysis conditions, the carbonyl group cleaved to form the acid dipyrrylmethane 66, after neutralization. Decarboxylation of 66 afforded the a,“- unsubstituted dipyrrylmethane 66 in quantitative yield -28- SCHEME 7 -29- a Minnow “as; ~zz-~:u~=u-oz N1 ml n\ .. © :l/ wan fiéu A] . :08 © 3:3. 2:8. Sn : 338 = 2 32: N + .@ o . -30- which upon coupling with 5,5’-diformyl-3,3’-diethyl-4,4’- dimethyl-2,2’-dipyrrylmethane 6 in methanol gave the naphthalene porphyrin-acid 66. Esterification was possible by refluxing the acid chloride derivative of 66 in metha- nol. Reduction of the naphthyl porphyrin methyl ester 46 by LiAlH4/THF gave the corresponding alcohol 44 which was used to prepare the bis(2-pyridyl-B-ethyl)amine derivative, 46, through a procedure very similar to that of the anthra- cene analogue. Various other meso substituted monoarylporphyrin com— pounds were synthesized by a straightforward stepwise approach (Scheme 9). C. Meso Disubstituted Porphyrins Various disubstituted monoporphyrins were synthesized by direct coupling of the aldehyde compounds with 3,3’-di- ethyl-4,4’-dimethyl-2,2’-dipyrrylmethane according to Gunters procedure [68]. This was found to be the most con- venient method for preparing symmetrically meso disubsti- tuted porphyrins and gave the highest yield. However, when unsymmetrically disubstituted porphyrins are required, we found that the best way is to use the step—by-step approach, using species such as the a,u’-unsubstituted di— pyrrylmethane 66 which upon coupling with ¢,a’-diformyldi- pyrrylmethanes, with a substituent at the the methine carbon, gives the disubstituted porphyrins (Scheme 10). :2 .II / a Minnow 4>zwzm>xozhmZ|L 4>mzpz<-fi 4>zsxazmza .. 54‘ It)! (at Fl l e-z q»! V'l V! q- .4“ OH \ @ CH OH/H+ @ 3 © / OHC HN Hu——- NH NH + H H / OHC cm <14 ml. dd on GR -32- a @@@ OHC SCHEME lO -33- To synthesize the disubstituted porphyrin 66, we employed two methods. 8-Methoxycarboxylanthracenecarbox- aldehyde 64 was obtained by partial reduction of the cor- responding dimethylester 66 with lithium aluminum hydride, followed by oxidation of the crude alcohol. Separation of 64 from the 1.8-anthracene dicarboxaldehyde (the complete reduction product) was achieved by column chromatography. Direct coupling of 64 with the a,u’-unsubstituted dipyrryl— methane 46 in methanol/p-toluenesulfonic acid gave a fairly good yield of the dimethyl-ester dianthrylporphyrins 66 and 66 (~l:l isomeric mixture). Hydrolysis of the diester 66 in acid medium (hydrochloric acid/formic acid, 1:4 v/v) afforded the corresponding diacid porphyrin, 66 (Scheme 11). Although the overall yield of this method starting with 64 was good and the work-up was convenient, the separation and purification of 64, which was very laborious. Another approach to synthesize the diacid porphyrin 66 was also employed, starting with the a,¢’-free dipyrryl methane anthracene aldehyde 64. Upon cyclization of 64 and a,a’-diformyl-dipyrrylmethane 66 (obtained by Vilsmeier’s formylation of 64) and oxidation of the formed porphyrino- gen by tetrachloro-o-benzoqinone (o-chloranil) gave an isomeric mixture of trans and cis dialdehyde-porphyrins 66 and 66. Oxidation of the aldehyde groups was affected by treatment with Jone’s reagent in acetone at low tempera- ture; if the oxidation is carried out at room temperature -34.. a .n HIM—mom am: mosh -35_ as nausea -36- n.— Mluuum s>zczazmza>xoxemz-a n>mzsz<-fi 4>stazwxa NI yr)! gr? In? (pl (02 (D? (D? (DZ (02 .44 5" -36- MA almuom 4>zczazmza>xozsmz-a 4>mxsz<-a 4>Ihzazmza .uhlx N? at)! q-l [02 (pl cozcoz “N 032(01 .44 ._< -37- or higher, it may lead to the insertion of chromium in the porphyrin. The diacid porphyrin 66 was used as a precursor to introduce asymmetric substituents and to synthesize the chiral porphyrin, 64, by treatment of the diacid chloride derivative, 66, with a relatively excess amount of (-)-cis myrtanylamine 66 in dry CHZCl2 (Scheme 12). Various other meso disubstituted porphyrins were synthesized by direct coupling of the aldehyde species with 3,3’-diethyl-4,4’-dimethyl-2,2’-dipyrrylmethane (Scheme 13). D. Metal Insertion Metals were inserted into the free-base porphyrins according to general methods depending on the nature of the metal and the solubility of the porphyrin compounds. In choosing these methods, we were concerned in minimi- zing the solubility problems, problems associated with the acidity or basisity (donor strength) of the solvent, problems caused by the intrinsic stability of the metal carrier, and problems arising from lability of the metallo- porphyrin moiety under the reaction conditions. Insertion of Copper, Zinc, Nickel and.Mbngsnese: The complexes of these metals were prepared by addition of a saturated methanolic solution of metal(II) acetate to dichloromethane solution of the porphyrin. The product was isolated by filtration of the methanolic solution after -38- partial evaporation of dichloromethane. Insertion of Iron: Iron was inserted by the ferrous sulfate method, by heating a solution of the free-base in pyridine/acetic acid together with a saturated aqueous solution of iron(II) sulfate under argon. Insertion of Cobalt: Cobalt complexes were prepared by heating a saturated methanolic solution of cobaltous chloride together with a dichloromethane solution of the free-base porphyrin con- taining a trace amount of sodium acetate. Co(II) complexes were precipitated by graduate evaporation of dichloro- methane. Ihsertjon of’Mbno-Metal in Diporphyrins: Insertion of a single metal ion into one of the two porphyrin rings of the dimers was achieved by titrating a methylene chloride solution of the free-base diporphyrin with 1 equivalent of zinc acetate in methanol. The insertion progress was monitored by tlc. The resultant mono zinc diporphyrin was purified by chromatography. Single metal ions, other than zinc, were then inserted by metalation of the mono zinc-free base dimer and removal of zinc, by treatment with dilute hydrochloric acid solution. Insertion of Mixed-Metals in Diporphyrins: In the case of mixed metal dimers, the second metal ion -39- complexes. The sequence of metal insertion is shown in Scheme 14. fl A / M2+ £1. 0 SCHEME 14 CHAPTER 3 E LEG TROCHEMI S TRY CHAPTER 3 BLECTROCHEMISTHY A. Introduction The idea of developing metal complexe based on electro- catalysts that can efficiently mediate the multi-electron reduction of dioxygen on graphite electrodes has a great significance as such catalysts are very essential to the cathode reaction in the air-batteries (fuel cells). Stepwise (one- or two-electron) reduction of molecular oxygen is highly unfavorable because of the involvement of relatively unstable intermediates (Figure 4). Reduction of dioxygen, which has a triplet ground state, requires over— coming the conservation of spin and orbital angular momen- tum. This problem can be circumvented by using transition metal complexes with low-lying paramagnetic states. Upon searching for efficient catalysts for the reduc— tion of oxygen, several monometallic macrocyclic complexes have been tested [74]. The most effective macrocycles were found to contain four nitrogen donor atoms. However, these studies have revealed that none of such complexes was able to catalyze the reduction of 02 to water, and that these single-metal complexes were capable only of reduction of -40- -41.. Midpoint Reduction Potentials of Oz Figure 4 Standard reduction potentials of oxygen, in volts versus SHE. -42- .mmsms swsmxauom “sameness oNch .mswsco mm» :m «wk on mesa cos nzcmmn Novas so a» m .osms umxmsomoNNcnoE madame Ncmonkoo mums Ne_ke essences mmmomosm a w ossmmm -43- oxygen to hydrogen peroxide [75,76]. The two-electron reduction is the first step for the stepwise reduction of oxygen molecules to water, Equations (13) and (14) + 2n+ + 2c. a n o °2 2 2 0.44 V v SCE (13) + '— 0 H202 + 2n + 2c * ZHZO E 1.54 v 1;. son (14) However, it is the direct four-electron reduction of oxygen molecules to water that has the greatest practical applica- tion in fuel cell technology, Equation (15) 02 + 43* + 4e" a 2n 0 n' = 0.99 v 1;. so: (15) 2 Since the equilibrium concentration of hydrogen peroxide at the desirable operational potential of an Oz/H2 fuel cell (+0.99 V vs. SCE) is relatively low ($.10...18 M [13]), an ex- tremely high reaction rate (i.e. driving force) is required to subsequently reduce the hydrogen peroxide to water. Therefore, more attention was focused on binuclear metal complexes in which the two metal centers may act in concert to bind and reduce oxygen by simultaneously transferring two electrons from each metal, Figure 5. In this work, a number of cobalt monoporphyrins and diporphyrins (Figures 6 and 7) have been examined for electrochemical catalytic activity toward oxygen reduction. -44- .muwswuoc 03338950on 60-. mos-means mNmzomsmsm sesame-anon Re nonsense-em. m 23me taroNxomxoqm-.czoo~:o~:o um undo Nxouxoemezoouxouzo um mien Nzomxoqmézoouzo um mun-o ode , 7.. g m on _ -45- Figure 7 Structures of’porphyrins examined in this study. -46- B. Rotating Ring-Disk Voltammetry In studying the catalytic activity of the metal com— plexes of the cofacial diporpyrins toward the electroreduc- tion of oxygen, we have made extensive use of the rotating ring-disk voltametry technique [77,78], by which it is pos- sible to measure quantitatively the (unwanted) hydrogen peroxide production. Figure 8 shows a schematic assembly of the ring~disk electrode which consists of a pyrolytic graphite disk with a concentric platinum ring. The porphyrin to be tested is usually applied to the graphite disk by immersing the elec- trode in a dilute solution of the catalyst in dichloro— methane [79,80]. When the electrode is rotated, fresh electrolyte (saturated with oxygen) is drawn vertically toward the disk surface and then ejected radially across the disk to the ring. The disk potential is usually controlled by a potentiostat and the current-potential pro- file of the disk records the redox reaction. Meanwhile, the ring is held at a constant potential where any hydrogen peroxide reaching it is rapidly oxidized to dioxygen but no other electrode reactions take place. The ring current response thus monitors hydrogen peroxide production, and the ratio of (-iR/iDN), where iR and iD are diffusion-limi- ted currents at the ring and the disk respectively, and N is the collection efficiency of the electrode [74], can measure the relative contributions of the four-electron and Figure 8 -47- . . 2 Pyrolytlc graphite disk, 0.46am (potential varied) Platinum ring, 023mm wude (potential constant +t.4vl + - +2H +2e , H202 At disk : 02‘ (I D) ’ ZHZO +4H++4e. At ring: r120z ——-. o2 +2H++2€ (IR) A schematic depiction of a rotating ring—disk electrode. ~48- two-electron reduction process. Moreover, possible contri- butions to the disk current arising from subsequent reac- tions of H202 (reduction to water or disproportionation to dioxygen and water) may be evaluated by examining the de- pendence of the current ratio an electrode rotation rate. At higher rotation rates, hydrogen peroxide is removed from the disk surface before further reaction can take place, resulting in an increase in the ring current and a decrease in the disk current. Invariance of the current ratio with rotation rate indicates that H202 is formed only as a parallel product in the dioxygen reduction, and not as an intermediate [81]. The average number of electrons involved in the oxygen reduction at the disk can be obtained from the following relation -i n = 4-2[iT!] (16) av 1D It is clear that if no ring current is detected, iR = 0, then nav = 4. If oxygen reduction at the disk entirely follows two electron pathways (i.e. to H202), -iR = iDN, then na§ = 2. The values of n can also be directly obtained from the limiting current using the Levich rela- tion [77] or the corresponding slope of the Levich plot (i1 vs. (w)1/2. -49- C. Data and Results“ j) Voltammetric Responses of the Diporphyrins in the Absence of Oxygen. The solid line in Figure 9 is a cyclic voltammogram obtained with a graphite electrode coated with dicobalt di- porphyrin 66 in the absence of dioxygen. Compounds 66 and 44g gave similar responses. The dashed lines resulted when the electrode was coated with the cobalt-free dipor- phyrin, 6, and the dotted curve represents the response of the uncoated graphite electrode. The current peaks obtained with the bare graphite electrode arise from the reduction and oxidation of quinone-like functional groups present on the graphite surface [82] which become sharper and somewhat larger in the presence of the adsorbed dipor- phyrin. The prominent pair of cathodic and anodic peaks centered at ca. +0.3 V in Figure 9 probably arise from a cobalt(III/II) couple, and the less prominent pair of peaks near +0.7 V may represent the second cobalt in complex 66. However, the poorly resolved response at the more positive potential appears only on the first scan with a freshly *This work was done in cooperation with Professor F.C. Anson and Dr. H.-Y. Liu of the Califernia Institute of Technology. -50- Figure 9 cyclic voltammograms fer 1.2 x10-9 molocm-Z of’complex ,Qa,adsorbed on graphite electrodes: _dashed line, electrode coated with cobalt free diporphyrin,,§; dotted line, background current at an uncoated electrode. Supporting electrolyte: 1 M CF3COOH saturated with argon. Scan rate: 100 mV 3‘ . -51- ! l ' l l l 0.8 0.6 0.4 0.2 0 E vs. SCE, Volt -52- polished and coated electrode while the response near +0.3 V is much more persistent. The transitory nature of the response near +0.7 V makes its assignment uncertain. Cyclic voltammograms for a solution of compound §2 in dichloromethane (Figure 10A) exhibit two peaks in the potential range where the cobalt(III) centers are expected to be reduced. (Additional peaks appear at more positive potentials that presumably arise from ligand oxidation processes were not examined in detail.) The presence of two peaks in Figure 10A is the principle reason for our suggesting that the small peak near +0.7 V in Figure 9 makes the point at which the first cobalt (III) center in adsorbed compound §2 is reduced. The area between the solid and the dashed curves at the prominent peak near +0.3 V in Figure 9 corresponds to approximately 0.8 electron per molecule of porphyrin initially deposited on the electrode. It therefore seems unlikely that this wave corresponds to the reduction of more than one cobalt center in the molecule. Two better formed and separate peaks are also presented in voltammograms of the analogous amide-linked dicobalt di- porphyrin adsorbed on graphite electrodes. The separation in peak potentials for the two identical cobalt centers in this complex has been attributed to electronic interactions that cause the formal potentials of the two metal centers to differ [12]. -53_ Figure 10 Cyclic voltammograms for 1.37 mM fig (A), Q (B), and ,6 (C) in dichloromethane, using polished glassy carbon electrode (0.34 cm2). Supporting electrolyte: tetrabutylammonium perchlorate. Scan rate: 100 mV 3’]. -54- I> IKDle B Is» C IfipA L l L l l 0.8 0.6 0.4 0.2 0 E vs. SCE, Volt -55- Complex 66, with only one cobalt center, yields a single prominent pair of voltammetric peaks centered near +0.35 V when adsorbed on the electrode surface. When dis- solved in dichloromethane, compound 66 exhibits only poorly formed voltammograms such as that in Figure 10B. Two anodic waves are evident. The more positive wave is believed to correspond to the oxidation of the metal-free base porphyrin ring in 66 because it appears at potentials similar to that of the first oxidation wave for the metal free dimer 6. However the electrochemical responses exhibited by 6 (Figure 100) and its various cobalt, copper, zinc, and mixed metal derivatives in dichloromethane are difficult to interpret unambiguously because the waves for the porphyrin rings are not clearly separated from those for the metal centers. No clear pattern could be discerned as to the effect of monometalation on the formal potentials for the oxidation of the metalated and unmetalated porphy- rin rings. It appears that linking the two porphyrin rings by an anthracene molecule produces rather complex coupling between the two rings and metal ions present in them. ii) Catalysis of the Reduction of Oxygen: Current-potential responses for the reduction of 02 at a rotating graphite disk-platinum ring electrode are shown in Figure 11. The solid curves in Figure 11A were obtained when the graphite disk was coated with 66 and the platinum ring was held at a potential where any H202 (formed at the disk and reaching the ring) would be reoxidized to O The 2. Figure 11 -56- Current-potential curves for the reduction of 02 at the rotating graphite disk-platinum ring electrode. The polished gyrolyticz graphite disk was coated with 1.2 X10 mol-cmz of’(A) complex,§g or (B) complex 6b. Ring potential: 0. 9 V. Rotation rate: 100 rpm. Supporting electrolyte 1 M CF COOH saturated with 02. The disk potential was scanne at 10 mV 3 . The dashed curves are the disk and ring currents obtained under the same conditions from coatings of an active amide-linked cofacial porphyrin CoZDP-4. -57- . 55 50> mom 9 0.0 N.o ¢.o 0.0 m.o _ _ a m 0.0 N.o ¢.o m.o m.o A 0' [I’ll I a. _ d — ozE+ <18--- 1. ozE+ -53- small maximum in the disk current response was not examined in detail. Similar behavior was also observed in the case of amide-linked cofacial dicobalt porphyrins (Figure 6) and was found to be strongly dependent on the polishing proce- dures employed to prepare the graphite surfaces before the porphyrins were adsorbed [91]. The same was true here, the maximum became less pronounced on successive scans with the same electrode coating as the limiting current at poten- tials more negative than the current maxi-um diminished. The magnitude of the current maxi-us was a function of the rate at which the potential was scanned; this is not present under true steady-state conditions. These features suggest that the current maximum may result from a small fraction of the adsorbed porphyrin catalysts that ten- porarily exhibits a higher activity toward the reduction of 02. That the polishing procedure used to prepare the gra- phite surface strongly influences the prominence of the maximum indicates that interactions of the cobalt centers with functional groups present on the graphite surface may alter the activity of the catalyst. The most noteworthy feature of the disk and ring currents in Figure 11A is their demonstration that the 02 reduction yields very little H202 at potentials near +0.45 V, and even at +0.2 V the ratio of ring to disk currents indicates that relatively little of the 02 is reduced to H202. Thus the dineric cobalt porphyrin, gs, provides a four-electron-reduction pathway for oxygen. -59- We were surprised to find that compound §b, with only one cobalt ion present in the diaeric porphyrin ligand, is also capable of catalyzing the four-electron pathway for the reduction of 02. The rotating ring-disk current-poten- tial curves for 02 reduction with this catalyst are shown as the solid curves in Figure 11B. The disk current is as large as that obtained with the dicobalt catalyst, and there is very little ring current. The major differences between the responses obtained with electrodes coated with §g or §b are the less positive potential at which the 02 reduction commences for catalyst fig and the slightly more gradual approach to the limiting current plateau. For comparison, the dashed curves in Figure 11 show the responses that result when the most active form of the amide—linked cofacial dicobalt catalyst described [83] previously was applied to the same polished graphite disk electrode. The reduction commences at slightly more posi- tive potentials than with §5 and significantly more posi- tive than with gb, but the li-iting disk currents are about the sane for all three catalysts and exceed substantially the value corresponding to the two-electron reduction of 02. Levich [84] and Koutecky-Levich [85] plots of the pla- teau currents and electrode rotation rates for the reduc— tion of 02 at electrodes coated with fig and §b are shown in Figure 12. As the electrode rotation rate increases, catalyst gg, with only a single cobalt center, is able to -50- sustain notably larger plateau currents than catalyst §2° The nonlinearity of the Levich plots (Figure 12A) with increasing curvature at higher rotation rates signals the likely presence of a chemical step that precedes the elec- tron transfer and limits the current to values below the convection diffusion limit. Previous studies [12] have assigned the current limiting step to the formation of a cobalt(II)-02 adduct that is the reducible species. The Koutecky-Levich plots in Figure 128, while linear, have slopes that differ somewhat from that of the dashed line calculated for the four-electron reduction of oxygen. The difference in slopes is in the direction expected if 02 were reduced to a mixture of H20 and 8202 at the catalyst- coated electrodes. This is also consistent with the magni- tudes of the anodic ring currents at potentials on the pla- teaus of the disk current-potential curves in Figure 11. The presence of a mixed reaction pathway complicates the interpretation of the intercept of the Koutecky-Levich plots in Figure 118. However, the ring-disk curves in Figure llA indicate that the major pathway must involve four electrons. On this basis, it is possible to obtain an approximate value of the rate constant, k, governing the current-limiting chemical reaction from Equation 17 [12] i k = F (17) 4FEat c02 -61- Figure 12 Levich (A) and.Koutecky-Levich (8) plots of the plateau current for the reduction of 0 gen at 95 aphite electrodes coated with 1. 2 X10' mol cm 5of complex 6a [ I ] or complex 6b [ Q ]. Supporting electrolyte: 1 M CF COOH saturated with air. The dashed lines are the calculated responses for the correction-diffusion limited reduction of’Og by four electrons, taking [02] =0.24 mM and 002 =1.8 x105 c7712 s" -52- I0. I O \ \- \.\\ .— \I \O \\ o \ \ . . fi.\\\; O. O. Q M N "' , (3M0 vw) ‘.-(‘“”!) \ .- I I I \ \ 20 0J4/2;(UPN1YJAZ (“I/2) ( rpm)1/2 -63- where iF is the reciprocal intercept of the Koutecky-Levich plot, F‘ is Faraday’s constant, Eat is the quantity of catalyst adsorbed on the electrode, and C02 is the concen- tration of 02 in the solution. The values of k obtained from the intercept of the lines in Figure 128 are 2 x 104 4 M-1 s.1 for catalysts §g and §b, respectively. 5 and 5 x 10 These values are somewhat smaller than the value of 3 x 10 M-1 s—1 measured previously for the amide-linked cofacial dicobalt porphyrin [12]. Thus the rate of reaction between 02 and complex §g may be somewhat slower than it is with the doubly linked analogue despite the greater cavity accessibility suggested by the more open structure of the former complex. However, the uncertainties in the values of rcat and the presence of mixed reaction pathways render this conclusion very tentative. 1‘1’1’) pi! Dependence of 02 Reduction: The course of the catalyzed reduction is influenced significantly by changes in the pH of the supporting elec- trolyte. Disk plateau currents at electrodes coated with gs or §b are plotted in Figure 13 as a function of pH. At electrodes coated with §g the current decreases somewhat between pH 0 and 2 but then remains essentially constant up to pH 12 before decreasing to about one-half of its initial value at pH 14. The currents obtained up to pH 12 are sig- nificantly larger than the two-electron diffusion-convec- tion limited value so that §g continues to provide a four- electron reduction of 02. As the pH is raised, an -64- l.2 . _ 3 \‘\m\. a__m\ “3 0'8 h— ‘\‘ 5 ‘0‘ E N. ----- s-""".\‘ ' 55-. ‘0, - ‘ . 0.4 _ “ \ \ l e 1 l l l l l 1 2 4 6 8 l0 l2 I4 pH Figure 13 pH Dependence of plateau currents for the reduction of 02 at rota ing grap ite disk electrodes coated .with2 1. 2 X10 mol-cm’ of complex gd [0 ] or complex ~b [ I ]. Electrode rotation rate: 100 rpm. Other conditions as in Figure 12. -6S- increasing fraction of O2 is reduced to H202 instead of H20. iv) Catalysis of tbe.Reduction of’EéOé: If the monocobalt diporphyrin, gg, were a catalyst for the reduction of H202 to H20 at potentials close to those where it catalyzes the reduction of 02, an explanation for the unexpectedly large limiting current in Figure 118 would be at hand. This possibility would be consistent with the appearance of a small ring current in Figure 118 if §b catalyzed the reduction to 8202 at a much lower rate than it catalyzes the reduction of 02. The slight rise in disk current and corresponding decrease in ring current at potentials less positive than ca. +0.1 V in Figure 11A suggest that §g may also function as a catalyst for the electroreduction of hydrogen peroxide to water [45]. Since most previous studies have not reported pronounced cata- lytic activity of cobalt porphyrins toward H202 reduction [86,87], we examined this point in some detail. Shown in Figure 14A are current-potential curves for the reduction of H202 at a rotated graphite disk electrode before (curve 1) and after it was coated with §g (curve 2) or §b (curve 3). The potential of the electrode was held at values no more positive than +0.45 V between scans to avoid the formation of 02 by oxidation of the 8202. It is clear from curves 1, 2 and 3 in Figure 14A that both §3 and §b are catalysts for the reduction of H202. However, even with the low rotation rates employed, the plateau currents -66- are much smaller than the calculated diffusion-convection limited Levich current [84] for a two-electron reduction process. This is evident from the Levich and Koutecky- Levich plots shown in Figure 14 B and C, respectively. Thus, a slow chemical step preceding the electron transfer reactions apparently limits the magnitude of the plateau currents for the reduction of 8202 as well as 02. That the rate of the preceding chemical step is much slower in the case of H202 is evident from the large intercepts of the lines in Figure 140. It is conceivable that Q; and gb catalyze the disproportionation of H202 so that the elec- trode reaction proceeding during its reduction involves only the reduction of 02 as in Figure 11. We regard this possibility as unlikely, however, because the potentials where the catalyzed reduction of H202 commences in Figure 14A are significantly less positive than those where 02 is reduced (Figure 11). If the totally irreversible reduction of H202 occurred by its prior disproportionation to O2 and H20, the reduction would be expected to commence at about the same potential where 02 is reduced. Only a potential dependence of the disproportionation reaction could alter this conclusion, and there is no reason to invoke such a potential dependence at potentials that are removed from those where the cobalt centers in the adsorbed porphyrins exhibit their redox activity. The lack of significant catalysis of the dispropor- tionation of H202 by complex §g and §b adsorbed on the Figure 14 -67- (A) Current-potential curves for the reduction of 1 mM H202 at rotated graphite disk electrodes: uncoated electrode (1); electrode coated with 1.2 X10'9 mol-cm"2 of complex €2.(2)3 complex 8Q. (3); or complexlgg (4). Supporting electrolyte: 1 M CF COOH saturated with argon. (B) Levich plots gor H202 reduction as catalyzed by complex fig [0 ] or complex g2 [A ]. (C) The corresponding Koutecky-Levich plots. The dashed lines were calculated for the diffusion—convection limited reduction by two electrons. ilim o (“A (In-2 -68- E vs. SCE, V0" (n (hm) '. (mA cm’2)°| N .u \ *1 14 I I/ ‘/ a ’ ’ 1 l # 0.025 005 -69- graphite disk was also demonstrated by means of the rotating ring—disk electrode in a dioxygen-free solution of H202. The anode current measured at the platinum ring at +0.9 V was essentially the same when the graphite disk (not connected to the potentiostat) was coated with complex fig or fifi as when it was uncoated. This was true with rotation rates as low as 400 rpm. Disproportionation of the H at 202 the surface of the disk would have produced a decrease in ring current so that the equality of the ring currents in the two experiments indicates that the disproportionation reaction proceeds too slowly to be important in these experiments. V) pH'Dependence of'HéOz Reduction: The catalytic activities of fig and fig toward the reduc- tion of H202 show notably different pH dependences. Plateau currents at disk electrodes coated with each cata- lyst and rotated at low rate are shown in Figure 15. fig sustains small disk currents over a wide pH range while with fig a decline in current begins as early as pH 2. Both complexes become relatively inert for 8202 reduction above pH 12. The difference in pH dependences of the activities of the two catalysts toward the reduction of H202 helps to explain the corresponding differences in the overall rates with which they catalyze the reduction of 02 in the pH range from ca. 2 to 12 (Figure 13). Complex fifi loses its activity for the reduction of H202 in the same pH range where it yields diminished plateau currents for the reduc— -70- 0.3 .. Figure 15 pH Dependence of’plateau currents for the reduction of 1 ”M H202 at sotating graphite disk electrodes coated with 1.2 x10" 5-mol -"cm of complex ,QQ [I ] or complex Q12, [ I ]. Supporting electrolytes at each pH as in Figure 10. Rotating rate: 400 rpm. -71- tion of 02 (Figure 13). If the catalyzed reduction of 02 by complex fig involved the production of H202 as an inter- mediate one would expect less disk current than if the reaction proceeded directly to H202 (as with compound fig). H202 exhibits a preference to react with transition- metal reductants by inner-sphere pathways [88]. It seems likely, therefore, that coordination of H202 to the cobalt center in the porphyrin catalysts precedes its catalyzed reduction. The alternative, outer-sphere pathway in which the cobalt(II) porphyrin transfers an electron to an uncoordinated H202 molecule is incompatible with the poten- tials where the catalyzed reduction proceeds: In 1 M CF3COOH fig catalyzes H202 reduction at potentials signifi- cantly more negative than those where the first cobalt(III) center is reduced to cobalt(II) and fig also exhibits catalytic activity at potentials less positive than that where the cobalt(II) porphyrin is first generated (compare Figures 9 and 14A). The cobalt(II) center in the porphyrin is essential; the metal-free diporpyrin ligand and its di- copper(II) derivative are both inert toward the reduction of 8202. 0. Discussion Complexes fig and fig catalyze the electroreduction of H202 at a much lower rate than they catalyze the electrore- duction of 02. This is apparent from a comparison of the normalized intercepts of the Koutecky-Levich plots in -72- Figures 128 and 14C. Both the magnitude and the disk current for the reduction of 02 by complex fig in Figure 11A and the lack of ring current on the rising part of the reduction wave at the disk electrode require that the 02 be reduced to H20, not to 8202. It follows that the mechanism of the four-electron reduction of 02 by catalyst fig cannot involve uncoordinated H202 as an intermediate. Any H 0 2 2 that was released into the solution would be subsequently reduced at the disk too slowly to provide the high disk current observed in Figure 11A or to escape detection at the ring electrode. This assertion was supported by experiments where both 02 and H202 were reduced simultane- ously at electrode coated with catalyst fig (Figure 16). At the same point, e.g., -0.3 V, where the 02 present reduced primarily to H20 at a high rate, H202 present in the solution was reduced much more slowly and the simul- taneous reduction of 02 had virtually no effect on the rate of the reduction of 8202 (compare A and C in Figure 16). The two oxidants seem clearly to undergo catalytic reduc- tion by independent pathways. The catalysis of 02 reduction to H20 by complex fig could involve the coordination of both cobalt centers to the O2 molecule with the formation of p—peroxo intermediate (Figure 17). Such an intermediate is not likely to be formed if the source of oxygen is H202 instead of 02, espe- cially in acidic solutions. Accordingly, the catalyzed reduction of H202 probably involves its coordination to a -73- Figure 16 Plateau currents for the simultaneous reduction of . 02 and H 0 at a rotating graghite disk electrode coated wit 1. 2 X10 9mol cm of,§g, (A) reduction of'H202 in the absence of 02; (B) repeat of'(A) after the solutions were saturated with 02; (C) difference between the plateau currents in (B) and that for an Og-saturated solution in the absence of’H202. Supporting electrolyte: 1 M CF3C00H. Electrode rotating rate: 400 rpm. Plateau currents measured at -0.3 V. -74- |.5 2.0 IO 05 A B C / .. x/ .. 1/ _. /TL 1 I I... I 1 A _ _ _ K b _ / _ vf _ _ L 6 4. 2. 4. O. 6 2 6. 4. 2 O O O 2 2 .l O O 0 Eu <6 . .5... Eu 4E . E... .60 ga. 6:. .. . E: N was: ~ 303 . N9 N 19 $0.03: -73- Figure 16 Plateau currents for the simultaneous reduction of , 02 and H 0 at a rotgting graghite disk electrode coated wit 1.2 X10" mol'cm‘ ofggg, (A) reduction of H202 in the absence of 02; (B) repeat of’(A) after the solutions were saturated with 02; (C) difference between the plateau currents in (B) and that for an Og-saturated solution in the absence of H202. Supporting electrolyte: 1 M CF3COOH. Electrode rotating rate: 400 rpm. Plateau currents measured at -0.3 V. -74- |.5 2.0 |.O 0.5 _ E: 19. 10.03: -75- single cobalt center. This would be compatible with the comparable activities toward H202 exhibited by catalyst fig and fig. The smaller limiting currents obtained with fig and fig and fig for the reduction of 8202 compared to that of 02 would then reflect the lower rate of coordination of H202 to the cobalt centers in these catalysts. The extensive four-electron 02 reduction activity exhi- bited by fig was surprising because the analogous doubly amide-bridged cofacial porphyrin containing only a single cobalt center had been reported to serve only as a two- electron reduction catalyst [12]. However, more recent experiments with greater quantities of more thoroughly purified material have shown that the monocobalt cofacial porphyrin does support a four-electron reduction of 02 [89]. Its four-electron activity declines within a few minutes, and this was one reason why its capacity to cata- lyze the four-electron reduction was overlooked in the previous studies. The anthracene bridging group in fig is not the source of the enhanced-activity because the monomeric porphyrin gfig (Fig. 7), shows essentially the same behavior as other monomeric cobalt porphyrins in catalyzing the two-electron reduction of 02 to 8202 with a further reduction to H20 proceeding at a much lower rate (Figure 14A, curve 4). The higher activity of fig toward the four-electron reduction of 02 might arise from the proximity of the second porphyrin ring that should be protonated in the acidic medium -76- Figure 17 A proposed mechanism for the four-electron pathway of the 02-reduction to water, catalyzed by dicobalt complexes of cofacial diporphyrins in acidic media. fivmo “8 In I- o 1:90.. CUM \ a / mo: o :00 no EmEmzoos. nomoooi -73- Figure 18 Hydrogen bonding in the dioxygen-monocobalt-porphyrin adhwws. a79- employed (Figure 18). It is conceivable that these protons, juxtaposed to the coordinated 02, could prevent the premature dissociation of, as well as assist in proton transfer to, the partially reduced 02 coordinated to the cobalt center in the second porphyrin ring. If such speci- fic proton catalysis proves to be the case, it would suggest new directions for the design of catalysts for multielectron reduction. Kinetic measurements of oxygen binding to complexes such as fig and fifi have shown that groups capable of forming a hydrogen bond make a signifi— cant contribution to the stability of the oxygen adduct of these complexes, that was reflected by the huge decrease in the off-rate of the ligand binding, compared to that of complexes such as fifi and 29 [90]. However more elaborate studies should be conducted before any definite conclusion can be made in this matter. 3. Conclusion It is clear that the anthracene and biphenylene-bridged cobalt diporphyrins provide four-electron pathways for the catalysts reduction of 02 when adsorbed on graphite elec~ trodes. -This is true for both the dimetalated and mono- metalated derivatives. These compounds represent the first effective macrocyclic metal-complex electrocatalysts that do not depend upon the "fear-atom separation" demonstrated to be essential in the case of the diamide-bridged catalysts. CHAPTER 4- MOLECULAR STRUCTURE AND PROPERTIES CHAPTER 4 MOLECULAR STRUCTURE AND PROPERTIES It is believed that the geometry of the metal complexes of diporphyrin compounds has a strong effect on their behavior toward binding oxygen, and on their catalytic activity toward the multi-electron reduction of dioxygen. It has been observed by Chang [7] and Collman [9] that the activity of such diporphyrin complexes strongly depends on the metal-metal separation. For instance, in the case of diamide-diporphyrins, the only dimer that is catalytically active has four-atom connecting groups separating the por- phyrin rings (i.e. DP-4), while dimers with 5-, 6-, or 7— atom bridges (i.e., DP-5, DP-6, and DP-7 respectively, cf. Figure 6) behave, to a large extent, like monomeric porphyrins. Clearly, a better knowledge of the molecular structure and (chemical and physical) properties of such metallopor- phyrins is of great importance toward understanding their catlytic and biological functions. -30- -31- A. Electronic Spectroscopy Typical UV-visible spectra of porphyrins consist of an intense absorption band at approximately 400 nm, that arise from u-w' transition [92], known as the Soret band, and four satellite bands labeled as I, II, III, and IV, located in the region between 700 and 500 nm. The relative posi- tions and intensities of these bands depend upon the nature and location of the substituents on the porphyrin ring. when the macrocycle has six or more alkyl groups in the B- pyrrolic positions, the visible bands have an intensity pattern such that IV > III > 11 > I, a so-called etio-type spectrum. In a configuration where two or more porphyrin rings are held in close proximity, interaction between the systems can cause a shift in the peaks positions [93]. The cofacial diporphyrins have distinctly different electronic spectra compared with those for the monoporphyrins. The visible bands of the free-base dimers are generally shifted to a longer wavelength (red shift) and Soret band is shifted to a shorter wavelength (blue shift). The bands also appear to be broadened. The cofacial diporphyrins in acidic solutions also show blue-shifted Soret band and red- shifted visible bands. These spectral patterns have been diagnostically useful in determining which products are the ”cofacial” dimers when new dimer preparations have been tried for the first -32- 210 .Ch 0 l ABSORBANCE 0.5 Figure 19 Absorption spectra of’anthracene monoporphyrin nuqu), anthracene diporphyrin (DPAA), and biphenylene diporphyrin (DP-B) in CHzClz. -83- time, and provide one piece of evidence of the cofacial nature of those species. Blue-shifted Soret bands and red- shifted visible bands relative to monomeric porphyrin have previously been reported for ”face-to-face" porphyrins by Collman et al. [8], and Chang and co-workers [4,5]. Simi- lar spectral shifts were also anticipated for our new dimers, as shown in Figure 19. While the spectral shifts are quantitatively useful for the identification of dimer compounds, a vigorous explana- tion for the observed shifts has not been attempted. The problem is confused by the variety of spectral properties reported for other dimer systems. Whereas the singly and doubly linked mesoporphyrin-IX dimers of Ichimura [100], which are not rigidly constrained to a "cofacial” orienta- tion, display blue—shifted Soret band and red-shifted visible band, the B,B’—alkyl-linked porphyrin dimers reported by Pain et a1. [94] show no spectral shifts when the porphyrin units are separated by more than two methy- lene groups. The well-defined ”cofacial" dimer of Hagan et a1. [95], based on meso-tetraphenylporphyrin, exhibits a broadened but unshifted Soret band and red-shifted visible bands. The doubly linked "cofacial” chlorin system of Wasielewski et a1. [96] exhibits no spectral shifts at all. Chang et al. [5,97] have observed strong exciton [98] coupling -in their diporphyrin systems. A similar phenom- enon was observed in some porphyrin and chlorin aggregates, in particular the p-oxo-scandium(III) dimers of octaethyl- -84- porphyrin and meso-tetraphenylporphyrin [99]. 8. Nuclear Magnetic Resonance Spectroscopy NMR spectroscopy is a particularly useful technique for establishing the integrity of cofacial porphyrin dimers. If a porphyrin ring is positioned atop another, the ring current of this second porphyrin can cause additional shifts of the proton resonances [4,5,8], particularly the pyrrolic NH signals. Their resonances are at higher fields than for the monomers. The upfield shift arises from the additional shielding effect of the second porphyrin atop the first. With previous dimers of Chang et a1. and Collman and co-workers, an increase in the shielding effect occurs as the interporphyrin distance is shortened. We observed similar upfield shift in our new diporphyrins. Upfield shifts have also been observed for peripheral B- pyrrolic substituents. Wasielewski et al. [96] observed shifts of 0.1 to 0.4 ppm for peripheral substituents of a doubly linked cofacial chlorin dimer. Since all the substi- tuents are moderately shifted upfield, an approximately "face-to-face” or ”center-to-center" orientation was inferred. In studies of "special pair” chlorophyll models, dramatic shifts of specific peripheral substituents indi- cate that "offset" aggregates are present [100,101]. Moderate upfield shifts for all peripheral substituents have been observed for some free-base porphyrins [102] and -85- Figure 20 250 MHz NMR of anthracene monoporphyrin (a), anthracene diporphyrin (b), and biphenylene diporphyrin (c) in CDClg. -86-- ° {ease LL *5 _JJQL a MQLJULA .: M L. .n )ngULT. -37- metalloporphyrins [103] under conditions in which inter— molecular aggregation occurs. A mercury(II) octaethylpor- phyrin "sandwich" complex has shown similar effects [104]. The spectral data for this complex are consistent with the formation of dimer aggregate with "center-to-center" orien- tations of the porphyrin monomers. Figure 20 shows the NMR spectra of the anthracene-, bi- phenylene-diporphyrins, and anthryl monoporphyrin. In addition to the difference in the extent of the upfield shift of the NH peaks, the peripheral pyrrolic methyl groups closer to the bridging group in the anthracene porphyrins show upfield shifts compared to that of the biphenylene dimer. C. Electron Spin Resonance Spectroscopies Simple copper(II) and low spin cobalt(II) porphyrins are S = 1/2 systems with well known ESR characteristics [105]. If two such porphyrins are close enough that the metals interact, the bimetallic systems will have singlet (S = 0) and triplet (S = 1) states separated by the energy J [106]. If the rate of electron exchange between the metals is fast compared to the resonance frequency (1010 sec-1), then each electron experiences nuclear spin equal to the total nuclear spin of the two metals (I = 3 for Cu;+ and I = 7 for 003+) [107,108]. The resulting hyperfine splitting, A, is half that for a related mononuclear S = -33- 1/2 system. The ESR transitions will be further split by the zero field splitting D. D is composed of two terms, (from spin orbital coupling) and Ddd (from the Dpseudo electron-electron dipolar interaction). Ddd is related to the metal-metal (M—M) separation, r, by the equation 1 r = (0.65 gfi/ndd) (3 It has been shown that D can be ignored, and r can be pseudo calculated from D Ddd without having a significant effect on the calculated r [109]. A more complete description of the method of determining r is contained in references 107 and 108. Several examples of ESR involving two interacting S = 1/2 metals have appeared in the literature [107-115]. Calculated distances have been in agreement with the x-ray crystal structures [109,110]. Intergction of axyxgggwitg Digg£§£1ic Qiporphyrigg Oxygen was found to interact with metal complexes of the diporphyrins to form 1:1 or 2:1 (M:02), depending on the metal-metal separation [117], (Figure 21). When a bulky ligand, l-triphenylmethylimidazole, was mixed with Co(II)~Co(II) diporphyrin (DP-7) and exposed to oxygen, both visible and ESR spectra documented the forma- tion of a (1:1) Co-O2 complex. The oxygenation is reversi- ble, evacuation results in eliminating the superoxo complex and restoring the Co(II) signal. On the other hand, our new Co(II)-Co(II) diporphyrin, with a smaller gap between -39- N .mzwsmxmsomwmoNNcsmswm use 0 rumssma :cmuocsmuxw mxu.%c rowussxummsmma omucsmxom w mm msxmem Alls— .— g 415. .— _90_ the two cobalt atoms, reacts in a completely different way. Addition of oxygen to the [¢3CIm-Co(II)]2 complex at room temperature instantanously produced a species consistent with the formation of 200/02 stoichiometry. This complex, written as p-peroxo [Co-Oz-Co] is diamagnetic and gives no EPR signals; however, when a trace amount of 12 was added to this solution, a well-defined isotropic spectrum was obtained, consisting of 15 lines, Figure 22. Such spectra could be expected if p-peroxo dicobalt became oxidized to a p-superoxo dicobalt complex in which the two equivalent 5900 nuclei would give a total of (2 X 2 X 7/2) + l = 15 lines [108]. There is little doubt that the ability of such dipor- phyrins to catalyze the oxygen reduction via a 4-electron process, is related to the structural features and special arrangements of the porphyrin rings in those dimers. We have estimated the interplanar separation from the value of 20 of the EPR spectra of frozen solutions of the copper- copper diporphyrins [43,83,118]. However, this method can only give the interspin- distance, r. To determine the interplanar separation, it is necessary to know the angle between the interspin vector and the normal to the porphy- rin planes (i.e., the slip angle) as well as r. Further- more, the determination of r from the value of 20 is not reliable if the dipolar interaction is of the same order of magnitude as the nuclear hyperfine splitting and the inter- spin vector does not coincide with a principal axis of the -91- Figure 22 EPR spectra of’u-superoxo complexes of CogDP-A (A) and CogDP—B (B). Spectra were obtained by reacting the bis CoI dimers with dioxygen at 23°C in CH2C22 containing 0.1 M N-tritylimidazole and a trace amount of’iodine. I! I'll-I [I'll 1(1)) [If _93- I op-r Rm HEXYL Y=CWHW8°fi “82": c“ suppeo 09-4 a: ream 94:204. Cu 09-5 a: HEXYL 7360-5146408 “3214,00 DP-4 RSPENTYL Ymflflg “121%th OP-ANTHRACENE waZH. Cu OP-BPHENYLEIE teau. co SCHEME 1-5 -94- nuclear hyperfine tensor [119]. The interspin distance can also be obtained from the intensity of the half-field transitions as well as computer simulation of EPR spectra [119]. The two methods were used here to probe the struc- tural features of the dicopper diporphyrins in frozen solutions [120]. ‘EPR Spectrogcogy of .Qicopper Diporggyrigg ig Fraggg Solutions‘ The EPR spectra of the allowed transitions for copper- copper diporphyrins in frozen solutions is shown in Figure 23. The dipolar splitting of both the copper parallel and perpendicular lines was well resolved. The simulated spec- trum shown in Figure 23A was obtained with r = 4.15 A, [J] = 0.5 cm-l, and O = 15'. The value of 20 read from the splitting of the copper parallel lines (800 G) corresponds to r = 4.2 A. The spectra of the half-field transitions are shown in Figure 24. The nuclear hyperfine splittings by the two copper ions were well-resolved as expected for copper porphyrin dimers [121]. In Figure 23A, the ratio of the intensity of the half-field transitions to the intensi- 3 ty of the allowed transitions for DP-4 was 4.6 x 10- , which corresponds to r = 4.04 A [120]. This value agrees *This work was done in cooperation with Professor S.S. Eaton of the University of Colorado at Denver. -95.. Figure 23 X—Band EPR spectra of the allowed transitions for Cu2DP-4 (A), CugDP-S (B) and CuZDP-B (C) at -180°C in 1:1 toluene/CHZCZZ solution. The spectra were obtained for 1 mM‘solutions with 1 mW’microwave power and 4 G modulation amplitude. The peak marked "A" was attributed to aggregated material. The dotted lines indicate regions in which the calculated curves don't overlay the experimental data. -96.. _97- Figure 24 X-Band EPR spectra of the half>field transitions for CugDP—4 (A), Cu2DP-5 (B), and CligDP-B (C) at -180°C in 1:1 toluene/CHZCZZ solution. The spectra were obtained in 1 mM solutions with 20 mW’microwave power at 16 G modulation amplitude. The overall amplification of the spectra in this figure is about 35 times that for Figure 23. The dotted lines indicate regions in which the calculated curves don't overlay the experimental data. _93- -99- well with the values obtained from 20 and by computer simulation of the allowed transitions. The simulated spec- trum was obtained with ¢ = 15'. Figures 23C and 240 show the allowed and half-field transitions, respectively, for the copper-copper diporphy- rin DP-B. The spectra were similar to that of DP-4 and DP- 5. The sharp lines between 3100 and 3200 G in Figure 230 were due to a small amount of monomeric copper porphyrin or to diporphyrin that had copper coordinated to only one of the porphyrins. The simulated spectra were obtained with r = 4.13 A, IJI = 0.5 cm—l, and 0 = 20'. The value of 20 obtained from copper parallel lines (820 G) corresponds to r = 4.2 A. The relative intensity of the half-field transitions was 4.0 x 10-3, which corresponds to r = 4.14 A. The X-ray structure of this complex (cf. section 0 of this chapter), gave a Cu-Cu distance of 3.81 A [47], which indicates substantial similarity between the structures in solution and the single crystal. The allowed and half-field transitions for copper- copper diporphyrin DP-7 are shown in Figures 25A and 26A, respectively. The smaller splittings of the allowed transitions for DP-7 than DP-4, DP-5, or DP-B indicated a longer interspin distance. Due to the weaker dipolar interaction, the copper parallel lines on the high-field side of the spectrum were poorly resolved, which precluded an estimate of the value of 20. The simulated spectrum was obtained with r = 4.95 A, [J] = 0.5 en‘l, and o = 40°. Figure 25 -100- XéBand EPR spectra of the allowed transitions for Cu2DP-7 (A), CugDP-A (B), and slipped CugDP-4 (C) at —180°C in 1:1 toluene/CHgCl2 solution. The spectra were obtained under conditions identical with those described in Figure 23. -101— -102- Figure 26 X—Band EPR spectra of the halfoield transitions for CuZDP-7 (A), Cug-DP-A (B), and slipped CugDP-4 (C) at -180°C in 1:1 toluene/CHgClg solution. The spectra were obtained in 1 mM solution with 200 mW microwave power and 16 C modulation amplitude. -103- -lO4- The major discrepancy between the observed and calculated spectrum in Figure 25A was at about 3100 G, which is at the position assigned to aggregated materials. This region of the spectrum was concentration dependent, which suggests that the discrepancy was due to aggregation. The relative intensity of the half-field transitions was 1.3 x 10-3, which corresponds to r = 5.0 A. The X-ray crystal struc- ture of DP-7 gave a copper-copper distance of 5.22 A and a slip angle of 46.6' [122]. Thus the EPR results indicate that the structure in frozen solution is similar to that observed in the single crystal. It should be noted that both the large copper-copper distances and the larger value of o are consistent with the slightly greater slip of one porphyrin plane relative to the other in crystal than in the frozen solution. The spectra of DP-A (Figures 250 and 260) were similar to that of DP-7. The simulated spectra were obtained with r = 4.95 A, IJI = 0.5 cm-1, and ¢ = 20'. The discrepancy between the observed and the calculated spectra at approxi- mately 3100 G was attributed to aggregation of a small amount of the sample. The relative intensity of the half- field transitions (1.4 x 10-3) gave r = 4.9 A. The X-ray structure of the nickel-nickel analogue of DP-A gave a metal-metal distance of 4.57 A and O = 20' [47]. The values of the interplanar spacings for these dipor- phyrins obtained from the values of r and o (d = r cos ¢) are included in Table I. The last column compares the O2 -105- electroreduction response of the corresponding cobalt derivatives. There appears to be no relationship between r, 0, or interplanar spacing and the effectiveness as a 4- electron catalyst. For example, while the three effective 4-electron catalysts DP-4, DP-B and DP-A (M = Co) have a small slip angle, so does DP-5. However, for the "pillar” type complexes and DP-A we have suggested [45] that the rings are flexible enough to achieve whatever structure is advantageous for the formation and cleavage of the cobalt peroxo intermediate, even to overcome the large discrepancy of r between DP-B and DP-A. The answer to this question may be found when mechanisms of the reduction of Co-Oz-Co species are fully understood. For all of the experimental spectra, it was only possi- ble to simulate the spectra if the absolute value of J was >0.3 cm-l. Thus, although there was no short bond pathway between the two copper ions, there was an exchange inter- action that was large on the X-band EPR scale. Such an interaction could occur via interaction between one copper and a nitrogen coordinated to the second copper or via w-n interaction between the two porphyrins. Table 1 -106- Distances and angles obtained for dicopper diporphyrins. (a) Copper-copper distances obtained from the relative intensities of the halfoield transitions. (b) Copper- copper distances obtained by simulation of the allowed transitions. (c) Angles between the z-axes of the copper g and A tensors and the interspin vector. (d) Separation between the two parallel porphyrin planes. (e) Percentage of formation of H202 evaluated from % H202:=2/[1+(niD)/(iR)], where iR and i0 are ring and disk limiting currents, respectively, and N(= 0.182) is the collection coefficient; these data were measured by using dicobalt diporphyrins coated on the graphite of a ring-disk electrode immersed in Og-saturated 0.5 M'aqueous trifluoroacetate acid. -lO7- .mam amen to use nanosecom 6 © e v on one. one on... <43 © e © 6 Q n . om om...” n3. Eé mica I 9 © om. me om.» one ohm Vlao SE n n. cos. n _ .v no.4 VI no a oe A om om.» mi. N _ .e mIn-o e om A on on.» m3. 8. n N... do Sign seamen .u 3- N N . III II]. n o 1.x. no um. um U a .- m:_.>za-oa_o :0 _o mo_mc< s moses-ma " _ 033. -108- D. X—Hay Crystal Structure of Diporphyrins"I While other methods of characterization (NMR, UV-vis, and EPR) support the identification of the meso-linked co- facial diporphyrins, X-ray crystal structure studies will add to our knowledge an important piece of information about the characteristic features and geometries of these catalysts. There are only two crystallographic structure studies of diporphyrins in the literature [122,124], but neither is for the active (4-electron reduction process) catalyst and neither is very accurate due to disorder leading to relatively poor diffraction quality. Figures 27 and 28 show three approximately mutally per- pendicular views of NizDP-A and CuZDP-B complexes, respec- tively. It is clear that porphyrin rings are not exactly stacked over one another but rather, have slipped with respect to each other as noted previously in other cofacial porphyrin structures [122-124], and that the porphyrin rings are markedly on-planar. This non-planarity is more significant in the case of NizDP-A complex. The magnitude of the slip is 2.40 A in Ni DP-A with a Ni-Ni distance of 2 *This work was done in cooperation with Professor A. Tulinsky of this department. -109- Figure 27 ORTEP Drawing of’NigDP-A excluding the side groups. Views approximately mutually perpendicular: (a) parallel to C(m2)-C(M4) direction; (b) parallel to C(m1)—C(m3) direction; (C) nearly perpendicular to porphyrin planes, ring 1 shaded. -llO- / \ /. I e O \ / \ / T \s e ”0'. s' 9 / ' 5.3”! mg- . ’ f.._.© ...‘ \/ ’49 I ~111- Figure 28 ORTEP Drawing of CugDP-B excluding side groups. Otherwise, as in Figure 26. -112- -113- 4.566 A. This corresponds to a slip angle of 31.7 degrees [125]. The corresponding values of CuZDP-B are: 1.60 A for slip, 3.807 A for Cu-Cu distance, and 24.9' for slip angle. The slip exhibited by NiZDP-A leads to an average porphyrin plane-to-plane distance of 3.88 A while that of CuzDP-B particularly corresponds to a van der Waals contact at 3.45 A. However, an examination of a space-filling representation of the structures with the interactive graphics program FRODO [126] shows that the "vacant” inter- planar space of both molecules is strikingly similar in both magnitude and extent indicating that the3.88 A inter- planar separation of NiZDP-A is only an apparent difference resulting from the greater degree of non-planarity. A per- spective stereoview of the molecules is shown in Figure 29. The fact that NiZDP-A doesn’t attain as close a ring contact as CuZDP-B suggests that either: a) further lateral translation to achieve this is offset by the loss of total number of van der Waals contacts, or (b) further rotation of the porphyrin rings about the connector bond leads to distortive repulsions between the aromatic connec- tor and methyl groups of the porphyrin ring which contri- bute to the buckling of the porphyrin, since the porphyrin core is much more flexible than the aromatic connector; (c) the degree of non-planarity can also be a factor, by inhibiting further slippage to avoid the development of un- favorable contacts. -ll4- ”0 00 \V ‘V/ v «y ) a l a) b Figure 29 Perspective stereoview of NiZDP—A (a) and CuZDP-B (b). -115- A summary of the results is shown in Table II, which also ’includes the EPR results, for comparison. More details of the X-ray crystal structures of these diporphy- rins are found in ref. 47. Table 2 Comparison of’distances and angles obtained by X-ray with those obtained by EPR fer cofacial porphyrins. O 0 mi d,A ch. deg X-raya EPRb X-ray EPR X-ray EPR Dp_A 4,55 4.90 3.88 4.60 3L7 20 “Ni-complex of DP-A was used for X-ray, Cu-complexes were used otherwise. br was calculated from the halfoield transition. The difference in metal-to-metal distances in the di- porphyrins is notgworthy. Prior to our study of these com- pounds it was thought that the metal-metal distance is the most crucial factor that dictates whether or not the cobalt dimer can serve as an effective four electron electrocata- lyst for the dioxygen reduction. This is borne out by the fact that among the ten or so amide chain-linked diporphy- rins that have been synthesized, only one compound with diametrical -CH2-CONH-CH2 connecting straps has been shown -116- to be active [11,91,83]. Increase or decrease in the num- ber of methylene units or transposition of the individual constituents in that series of diporphyrins would lead to near total loss of activity. This is not the case with DP- A and DP-B. As indicated above, the metal separations in the two dimers differ by 0.76 A yet it doesn’t seem to have much of an effect on their electrocatalytic performance. Although the use of such a distance obtained from non- cobalt complexes to discuss the behavior of the cobalt catalysts on a graphite surface is admittedly not direct, on the other hand, there is no evidence to prove that other metalloporphyrins would adopt a grossly different struc- tural configuration in another environment. In fact, both the metal-metal separation and ring-to—ring distance agree well with those obtained by EPR studies in frozen solution (cf. Table II and section C of this chapter). The results of the X-ray studies seems to reiterate the conclusion about the lack_ of a clear connection between inter-ring separation and a preference for 4-electron versus 2-elec- tron reduction pathways of oxygen. Further structural studies as well as the synthesis of other diporphyrins are obviously needed to more clarify the structure-function relationship of this very important class of catalysts. EXPERIMENTAL E XPERIMENTAL EXPERIMENTAL EXPERIMENTAL A. Reagents and Solvents All reagents and solvents were of reagent grade qualities unless otherwise mentioned, and were purchased. All solvents were distilled before use; methylene chloride, tetrahydrofurane, toluene, benzene, diethylether were distilled from LiAlH4; pyridine, triethyl amine, collidine were distilled from calcium hydride; methanol and ethanol were distilled from sodium. Silica gel for column chroma- tography (60-200 mesh) was purchased from J.T. Baker (3405); preparative silica gel plates were from Analtech, Inc.; for analytical TCL, Eastman 13181 chromatography sheets were used. B. Physical and Spectroscopic Methods: Melting points were obtained on an Electrothermal melting point apparatus and are uncorrected. UV-visible spectra were obtained on a Cary 219 spectrophotometer. The infrared spectra were recorded on a Perkin-Elmer Model 2378 spectrophotometer. PMR spectra were obtained on a Varian T-60 or_Brucker WM-250 MHz spectrometer, with chemical shifts reported in 8-units measured from tetramethylsilane as the internal standard. Mass spectra were obtained in a Hitachi Perkin-Elmer Instrument EMU-6 mass spectrometer and Finnigan 4000 GC/MS system using the direct inlet mode, at -117- -118- 70 eV ionization energy. High-resolution positive ion mass spectra were obtained on a Kratos MS-50 RF equipped with Ionteck FAB gun, operated at 8 kV. Elemental analyses were performed by Spang Microanalytical Laboratory, Eagle Harbor, Michigan; C, H and N analyses were within 30.40%. C. Experimental Procedures 1,seais{[5,5’—bis(etnoxycarbany1)-4,4’-d1ethy1—3,3’— dimetbyJ-Z,2’-dipyrry1]metby1} anthracene, g: 1,8-Anthracene dicarboxaldehyde (7.0 gm, 0.03 mole) and ethyl 3-ethy1-4-methyl-2-pyrrolecarboxylate g (21.7 gm, 0.12 mole) were heated in absolute ethanol (200 mL) containing 5 mL of concentrated hydrochloric acid. The solution was refluxed for 1 h and allowed to cool in an ice bath. The yellow crystals (25 gm, 91% yield) were filtered and dried; m.p. 183-185°C. NMR 8 1.10 (12H, t, Et), 1.30 (12H, t, OEt), 1.65 (12H, S, Me), 2.75 (8H, q, Et); 4.30 (8H, q, OEt), 5.88 (2H, s, CH), 6.96 (2H, d, 2 and 7H- anthryl), 7.40 (2H, t, 3 and 6H-anthry1), 8.00 (2H, d, 4 and 5H-anthry1), 8.38 (1H, s, 10H-anthryl), 8.50 (4H, broad, NH), 8.55 (1H, s, 9H-anthry1); MS, m/e 922 (39), 921 (39), 876 (42) 875 (100). Analysis calculated for H N O ' C, 72.86; H, 7.21; N, 6.07; found: C, 71.67; H, C56 66 4 8‘ 7.24, N, 6.08. -ll9- 1 , 8—Bis [ (4, 4’ -die thy] -3, 3’ dime thy] -2, 2’ -dipyrry1) - methyl]anthracene, g: The tetraester 3 (21.0 gm, 22.7 mmol) was dissolved in refluxing ethanol (95%, 250 m1) and hydrolyzed by addition of aqueous sodium hydroxide (5 gm in 30 ml water). The mixture was kept refluxing for 10 h before ethanol was removed as much as possible on a rotary evaporator. The residue was diluted with water (200 mL) and filtered. The filtrate was cooled down by adding ice (~50 gm) and then neutralized by adding glacial acetic acid; the precipitated tetracarboxylic acid was filtered and dried, 19.0 gm (93%), m.p. 170'C (decomposition). The tetracarboxylic acid was dissolved in ethanolamine (100 mL) and refluxed under nitrogen for 1 hr. The solu- tion, while still hot, was poured into ice-water (~600 mL). Golden crystals of g soon separated and were filtered and air dried (11.5 gm, 84% yield, no crystallization was necessary), m.p. 150-151'C. NMR 8 1.15 (12H, t, Et), 1.62 (12H, 3, Me), 2.43 (8H, q, St), 5.98 (2H, s, CH), 6.38 (4H, s, 5H-pyrrole), 7.05 (2H, d, 7H-anthry1), 8.45 (1H, s, 9H- anthryl), 8.70 (1H, s, lOH-anthryl). MS, m/e 635 (65), 634 (45), 525 (100). a,¢-[4,4—DietbyJ-5,5’-difonmy1-3,3’-dimetbyJ—2,2’- dipyrryl]toluene,1§§: Benzaldehyde (3.18 gm, 0.03 mole) and ethyl 3-ethyl-4- methy1-2-pyrrolecarboxylate g (10.86 gm, 0.06 mole) were -120- hedted in absolute ethanol (300 mL) containing concentrated HCl (5 mL). The solution was refluxed for 1 h and then allowed to stand in ice-water for 2 h. The white crystal- lized material formed was collected by filtration to give fiz, 12.5 gm (94X); m.p. 150-152'C. MS, m/e 450 (95), 270 (100). NMR 6 1.20-1.30 (12H, 2 triplets, Et), 1.70 (6H, 5, Me), 2.80 (4H, q, Et), 4.20 (4H, q, OEt), 5.60 (1H, s, CE), 7.3 (5H, m, phenyl), 8.30 (2H, broad, NH). The above diester was dissolved in hot ethanol (300 ml), to which sodium hydroxide solution (5.0 gm in 20 mL) was added, and the mixture was refluxed continually for 8 h. The solvent was then evaporated and the residue was dissolved in water (750 mL), cooled, and acidified with glacial acetic acid to give a white solid. After collection by filtration and vacuum drying, the diacid was decarboxylated by heating in ethanolamine (100 ml) for 1 h. The hot mixture was poured into ice- water and the product was extracted twice with dichloro- methane (2 x 150 m1). Evaporation of the organic layer afforded the corresponding a,a’-free dipyrrylmethane fifi, 7.5 gm (828), m.p. 95'C (decomposition). NMR 6 1.10 (6H, t, Et), 1.70 (6H, 3, Me), 2.85 (4H, q, Et), 5.60 (1H, s, CH), 6.40 (2H, s, 5H-pyrrole), 7.3 (5H, m, phenyl), 8.30 (2H, broad, NH). Phosphorus oxychloride (3 ml) was added dropwise, while stirring, to a cold solution of the a,a[4,4’-diethyl- 3,3’-dimethyl-2,2’-dipyrry1]toluene (5.0 gm) in DMF (50 m1) -121- at -5°C. The mixture was allowed to stir at low tempera- ture (-5'C to 0'0) under nitrogen for 4 h, then let stand at -15'C for 8 h. Water (200 mL) was added and the aqueous layer was washed several times with methylene chloride until the organic layer was almost colorless. The iminium salt was hydrolyzed by adding 10% NaOH solution dropwise to the aqueous layer until complete precipitation was achieved. The light brown crystalline material was collected by filtration and recrystallized from ethanol (95%) to give white needles fig, 3.5 gm (60%); m.p. 210- 212'C. Analysis calculated for 023H26N202: C, 76.21; H, 7.23; found: C, 75.95; H, 7.20. MS, m/e 362 (100), 226 (50. NMR 6 1.10 (6H, t, Et), 1.75 (6H, 3, Me), 2.60 (4H, q, Et), 5.45 (1H, s, CH), 7.10 (5H, m, phenyl), 9.25 (ZR, 3, CHO). 9.55 (2H, broad, NH). [3,3’-Dietby1-5,5’-bis(metboxymetby1)-4,4’-dimetby1-2,2’- dipyrryllmetbene Hydrobromide, Z: Ethyl 4,5-dimethy1—3-ethyl-2-pyrro1ecarboxylate, 29 (15.0 gm, 0.066 mole), suspended in formic acid (88%, 50 mL) was heated in a steam bath. HBr solution (48%, 20 ml) was added and heating was continued for 8 h. The solution was then allowed to stand at room temperature overnight. The orange precipitate was isolated by filtration, washed with acetic acid (20 mL) and ether (50 mL) and dried to give 13.0 gm (59%). —122- Bromine (6 mL) was added to a stirred suspension of the above dipyrrylmethene in HOAc (120 mL) and the mixture was kept at 80'C for 30 min. The solution was allowed to cool and the precipitate was filtered and washed with ether to give [5,5’-bis(bromomethyl)-4,4’-dimethyl-3,3’-diethyla 2,2’-dipyrry1]methene hydrobromide g (10.0 gm, 53% yield). NMR showed the 5,5’-methy1ene protons at 8 4.95, in contrast to the unbrominated methyl signals at 6 2.67. The bis(bromomethyl)dipyrrylmethene fi (5.0 gm) was re- fluxed in dry methanol (50 mL) for 30 min. The solution was then cooled to room temperature and diluted with ether (50 mL), and the precipitates were filtered to give (3.2 gm, 81% yield), m.p. 285'C dec; NMR 6 1.31 (6H, t, Et), 2.22 (6H, 3, Me), 2.77 (4H, q, Et), 3.55 (SH, 3, OMe), 5.13 (4H, s, CH2), 7.38 (18, s, methine). MS, m/e 316 (98, M‘-HBr), 287 (80), 255 (76), 242 (100). 1,8eBis[5-(2,8,13,17-tetraetbyl-3,7,12,18-tetrametby1)- porphyrinyijantbracene g: A) Frgg 5L5’-big(yethoxyggthyl)dipyrrylggthegg_ 1 con- W: A solution of 1,8-bis[(4,4’-diethyl-3,3’-dimethy1- 2,2’-dipyrryl)methyl]anthracene g (634 mg, 1 mmol) and [3,3’-diethy1-5,5’-bis(methoxymethyl)-4,4’-dimethyl-2,2’- dipyrry1]methene hydrobromide Z (794 mg, 2 mmol) in benzene (100 mL) was heated to reflux for l h. The solution was cooled to room temperature before tetrachloro-o-benzo- -123- quinone (1.0 gm) was added and stirred for 1/2 h. TLC (silica gel, CHC13) revealed the formation of two porphy- rins; etioporphyrin II, Rf 0.98 and the anthracene dipor- phyrin, Rf 0.3. To facilitate the separation of the dipor- phyrin from impurities with low Rf value, the residue, after removal of benzene, was heated in methylene chloride with methanolic solution of zinc acetate and sodium ace- tate. The zinc-diporphyrin complex, Rf 0.9 (silica gel, CHCl3) was separated by column chromatography. The free base diporphyrin was obtained by demetalation of the zinc complex upon washing the methylene chloride solution of the complex with 10% hydrochloric acid and purification by chromatography. Yield: 80 mg (7%) of anthracene diporphyrin fi and 150 mg of etioporphyrin II. NMR of the diporphyrin: 6 1.27 (12H, t, Et), 1.66 (12H, t, Et), 1.87 (12H, 3, Me), 3.17 (12H, s, Me), 3.38 (8H, q, Et), 3.90 (8H, 2q, Et), anthryl: 7.58 (2H, t, 2,7-H), 7.71 (2H, t, 3,6-H), 8.50 (2H, d, 4,5-H), 8.90 (1H, s, lO-H), 9.00 (18, s, 9-H), 8.95 (4H, s, meso-H), 9.34 (2H, s, meso-H), -4.98 (4H, broad, NH). MS, m/e 1131 (89), 1130 (100, 565 (81); high resolution MS, 1130.6670 (C78H82N8); UV-vis xmax nm (6."), 625 (4.0), 572 (9.0), 537 (10), 503 (22), 394 (232). Analysis calculated for C78H82N8: C, 82.79; H, 7.30; N, 9.90; found: C, 82.81, H, 7.41, N, 9.85. B) Frog5,5’-bingrogggethy1)dipyrry1!§thene fi conggg- sation: A solution of a—free bis(dipyrrylmethyl)anthracene g -124- (300 mg, 0.47 mmol) and [5,5’-bis(bromomethyl)-3,3’- diethyl-4,4’-dimethyl-2,2’-dipyrryl]methene hydrobromide g (470 mg, 0.94 mmol) in glacial acetic acid (50 mL) was refluxed for 2 h, after which time air was passed through the solution for 20 h. The solvent was removed by evapora- tion, and the residue was dissolved in chloroform (50 m1) and refluxed, together with 20 m1 of a saturated methanolic solution of Zn(OAc)2 and NaOAc, for 1/2 h. Work-up, as in part (A), gave 25 mg (4.6% yield) of the diporphyrin fi, in addition to 30 mg of etioporphyrin II. The diporphyrin is identical to the authentic sample prepared by method (A).' C) Frq_ Coggling of 1.8-gnthrgcene giggrboxgldehyde with l,lQ-dideogy-2,8,12,l8-tetragethylbilggiene-gc dihydro- brggide g: A suspension of 1.8-anthracene dicarboxaldehyde ; (117 mg, 0.5 mmol) and l,lQ-dideoxytetraethyltetra- mutylbiladiene-ac dihydrobromide 9 (625 mg, 1 mmol) in dry methanol (10 mL) containing 5 drops of a solution of glacial acetic acid saturated with hydrogen bromide, was refluxed under argon atmosphere for 18 h. The solution was then poured onto water (25 mL) and the products were extracted with CHCl3 (2 x 25 mL). Chromatography (silica gel, CHZCLZ) gave monoporphyrin anthracene carboxaldehyde lfi (30 mg, 9% yield) and 1,8- anthracene diporphyrin fi (32 mg, 5.6% yield), together with etioporphyrin II (23 mg, 10% yield). The products were identical in every respect with authentic samples -125- prepared previously by different methods. D) Frogg,1,19-dideoxybilggiene-ac g, condensgiion with gldehyde: A suspension of l,lQ-dideoxy-2,8,12,18-tetraethyl- 3,7,13,17-tetramethy1 biladiene-ac dihydrobromide g (70 mg, 0.11 mmol) and 5-(8-formyl-l-anthry1)-2,8,13,17-tetraethyl- 3,7,12,18-tetramethylporphyrine lg (76 mg, 0.11 mmol) in MeOH (10 mL) containing 4 drops of acetic acid solution saturated with HBr were heated under reflux for 24 h. The solution was then poured onto a saturated solution of sodium carbonate (50 mL) and the product was extracted with CHCl3 (3 x 25 mL). Column chromatography (silica gel, CHClg) was used to separate the diporphyrin. For further purification, zinc was inserted in the diporphyrin, chroma- tographed and then demetalated to give 15 mg (12% yield). The product is identical to the authenic sample. E) Eng! 5,5’—diformyldipyrrylgethgne 5: To a solution of a-free bis(dipyrrylmethane)anthracene g (634 mg, 1 mmol) and 5,5’-diformy1-3,3’-diethyl-4,4’-di- methyl-2,2’-dipyrry1methane fi (572 mg, 2 mmol) in methanol (80 mL) was added 70% perchloric acid (0.5 mL). The dark red solution was stirred under argon atmosphere for 12 h at room temperature in the dark, after which time a solution of NaOAc (0.5 gm) in methanol (10 ml) was added, followed by another solution of o-chloranil (200 mg) in methanol (10 mL). After stirring for l h, the solvent was removed -126- and the residue was taken up in CHZCI2 (20 mL) and a solu- tion of zinc acetate (200 mg) in methanol (10 ml) was added. After being stirred for 1 h, the solvents were evaporated and the residue separated by chromatography (silica gel, CHC13). The isolated Zn(II) diporphyrin was demetalated by washing with 10% HCl in CHZCl2 to give (238 mg, 21%) of the free base diporphyrin, 300 mg of etiopor- phyrin II was also obtained. The product obtained is similar to the authentic sample. 5-Pfieny]-2,8,13,17—tetraetby1-3,7,12,18-tetrametby1 porphyrin 55: A)‘ Frog, the 5,5’—bis(gethoxygethyl)dipyrrylgethene hydrobromide Z condegggtion: a~Free phenyldipyrrylmethane fifi (470 mg, 1.5 mmol) and [bis(methoxymethyl)dipyrryl]methene hydrobromide Z (650 mg, 1.55 mmol) were dissolved in benzene (80 mL) and heated to reflux for l h. Tetrachloro—o-benzoquinone (200 mg) was added to the cold solution and stirred at room temperature for 1/2 h. The products were separated by chromatography using silica gel and CH Clz/hexane (80:20) as eluent to 2 give 160 mg (17.8% yield) of the phenylporphyrin 24. A small amount of etioporphyrin II (30 mg) was obtained. NMR for phenylporphyrin, 6 1.81 (6H, t, Et), 1.90 (6H, t, Et), 2.45 (6H, 3, Me) 3.69 (SH, s, Me), 4.05 (8H, 2q, Et), 7.78 (3H, m, 3,4,5-H phenyl), 8.05 (2H, d, 2,6-H phenyl), 9.96 (1h, 3, meso), 10.15 (2H, s, meso) -3.30 (2H, broad, NH). -127- MS, m/e 554 (100), 277 (34); UV-vis, x (6 max mM)’ 628 (2.5), 559 (6.7, 534 (7.0), 501 (15.5), 402 (188). Analysis cal- culated for C38H42N4: C, 82.27; H, 7.63, N, 10.10; found: C, 82.14; H, 7.71; N, 10.17. B) Frogf5L5’(difoggyl)dipyrryl tolgene fig, condenggtion with a,af-free dipyrryl getgggg: a,«[5,5’-Diformyl-4,4’-diethyl-3,3’-dimethyl-2,2’- dipyrrylltoluene fig (362 mg, 1 mmol) and 4,4’-dimethy1- 2,2’-dipyrry1methane (230 mg, 1 mmol) were stirred in methanol (50 mL) under argon in the presence of perchloric acid (70%, 4 drops), for 18 h. Tetrachloro-o-benzoquinone (100 mg) in methanol (5 ml) was added and stirring was continued for one more hour. Chromatography, as in part (A), gave 182 mg (33% yield) of phenylporphyrin, identical to the authentic sample obtained above. Etioporphyrin II (50 mg) was also obtained. Octaetbylporpbyrin from the Condensation of [(methoxy— methy1)dipyrry1]methene Eydrobromide Z: [5,5’-Bis(bromomethyl)-3,3’,4,4’-tetraethyl-2,2’- dipyrryl]methene hydrobromide g (12.0 gm) in methanol (200 mL) was heated under reflux for 30 min. The mixture was evaporated to dryness and the residue was triturated with ether/methanol (5:1). The resultant [(methoxymethyl)- dipyrryl]methene hydrobromide 2 (7.5 gm) was used without purification to condense with (3,3’,4,4’-tetraethyl-2,2’- dipyrryl]methene hydrobromide in hot benzene. After oxida- -128- tion in o-chloranil, octaethylporphyrin was isolated in 28% yield: NMR 61.09 (24H, t, Et), 2.40 (16H, q, Et), 10.1 (48, s, meso-H), -3.88 (2H, broad, NH); m.p. 324'C (Lit. m.p. 324-325'0). This porphyrin was identical in every respect with an authentic samples of OEP. 1,8rBis[(4,4’-dietby1-5,5’-diformyI—3,3’-dimetby1-2,2’- dipyrryl]metbyllantbracene gfi: To a solution of 1,8-bis[(4,4’-diethyl-3,3’-dimethy1i . 2,2’-dipyrry1)methyl]anthracene g (2.00 gm, 3.1 mmol) in dimethylformamide (15 ml) was added phosphorus oxychloride (5.0 m1) dropwise, while cooling (-5' to 0'0). The solu- tion was then stirred at room temperature under nitrogen atmosphere for 5 h, after which time water (100 ml) was added. The dark brown solution which resulted was washed several times with ether, and the aqueous layer was treated with 10% NaOH solution until complete precipitation was assured. The light brown precipitate was collected by filtration, washed with water (100 m1) and air dried. The tetraaldehyde éfi was purified further by chromatography (silica gel, CH 012) to give 1.15 gm (50% yield); m.p. 221- 2 222°C. NMR, 6 1.20 (12H, t, Et), 1.70 (12H, t, Me), 2.65 (88, q, Et), 5.90 (2H, s, CH), 6.90 (2H, d, 2,7H-anthryl), 7.38 (2H, t, 3,6H-anthry1), 7.90 (2H, d, 4,5H-anthry1), 8.25 (1H, s, lOH-anthryl), 8.40 (1H, s, 9H-anthryl), 8.65 (4H, broad, NH), 9.20 (4H, s, CHO); MS, m/e 747 (20), 719 (15), 634 (28), 178 (100). -129- 1,8e8isl5-(2,8,12,18—tetraetby1-3,7,13,17-tetrametby1-15- pheny1)porpbyriny1]antbracene 21: The a-free bis(dipyrrylmethyl) anthracene g (500 mg, 0.78 mmol) and e,a-diformy1 dipyrryl toluene fig (572 mg, 1.56 mmol) were stirred in methanol (100 ml) for 15 min, then HClO4 (70%, 1 ml) was added. The reaction mixture was stirred for 24 h in the dark and under argon. Tetrachloro- o-benzoquinone (200 mg) in methanol (20 ml) was added, and stirring was continued for one additional hour. The sol- vent was then removed by a rotary evaporator. The residue was dissolved in CHCl3 (100 ml) and washed with a saturated solution of sodium carbonate, then with water, and the organic layer was dried over anhydrous sodium sulfate. To the organic layer was added a saturated solution of Zn(OAc)2/NaOAc in methanol (20 ml) and the solution was refluxed for l h. The solution was then washed with water. The organic layer was evaporated to dryness and the residue was dissolved in CHCl3 (5 ml) and chromatographed through a silica gel column (CH2012); all the red zinc diporphyrin eluate was taken. After concentrating the eluate to 50 ml, it was washed in a separatory funnel with 10% HCl solution (50 m1), a saturated Na2C03 solution (50 m1), then water (2 X 50 ml), successively. Further purification of the free base diporphyrin was carried out by column chromatography (silica gel, 1% CH30H/CH2C12) to give diphenyletioporphyrin II (80 mg, R 0.95) and diphenyl f -l30- diporphyrin anthracene Z} (100 mg, 9% yield); UV—vis, xnaxnm (emu) 624 (3.8), 570 (10), 535 (13), 502 (26), 393 (218). NMR, 6 1.20 (12H, t, Et), 1.70 (12H, t, Et), 1.85 (12H, 3, Me), 3.20 (12H, s, Me), 3.95 (16H, 2q, Et), 7.75 (5H, m, phenyl), anthryl: 7.5 (2H, t, 2,7-H), 8.10 (2H, t, 3, 6-H), 8.45 (2H, d, 4,5-H), 8.75 (1H, s, 10-H), 9.50 (1H, s, 9~H), 10.25 (4H, s, meso—H), -4.95 (4H, broad, NH). 1,8a8is{5-[2,8-dietby1-3,7,12,18—tetramethy1-13,17- bis(methylpropionate)Jtporpbyriny1}antbracene 2?: A procedure similar to that in the anthracene diporphyrin fi was followed using 1 equivalent of the a—free bis dipyrrylmethane anthracene and 2 equivalents of [5,5’- diformyl-4,4’-dimethy1-3,3’-bis(methylpropionate)-2,2’- dipyrryl]methane lg. Tetra(methylpropionate)porphyrin (10% yield) was obtained in addition to 9% yield of the desired diporphyrin product. NMR of the diporphyrin, 1.20 (12H, t, Ht), 1.60 (12H, 3, Me), 3.00 (8H, t, methylene), 3.20 (12H, 8, Me), 3.30 (8H, t, -0CH -), 3.50 (12H, 3, -OCH 4.10 2 3)! (8H, 2q, Et), anthryl: 7.40 (23, t, 2,7—H), 7.50 (23, 5, 3,6-H), 8.50 (2H, d, 4,5-H), 8.80 (1H, s, lO-H), 8.95 (4H, s, meso-H), 9.15 (1H, s, 9-H), 9.30 (2H, s, meso-H), -4.50 (4H, broad, NH). Uv-vis xm nm (EmM) 623 (4), 570 (9), 534 ax (11), 502 (25), 391 (220). -l3l- 1,Relis{[5,5’-bis(etbyoxycarbonyI)-4,4’—dietbyI-3,3’- dimetbyJ-Z,2’-dipyrry1]metby1 biphenylene ggg: Basically, a similar procedure to the one followed in reacting 1,8-anthracene dicarboxaldehyde with a-free pyr- role was carried out. The a-free pyrrole g (4 equivalents) and 1,8-bipheny1ene dicarboxaldehyde lg (1 equivalent) were refluxed in ethanol for l h, under nitrogen and in the presence of catalytic amounts of concentrated hydrochloric acid. Precipitation of the product was enhanced by adding a few pieces of ice to the cold reaction solution, the white solid obtained by filtration was dried and recrystal- lized from aqueous ethanol to give 90% yield of the 1,8- bis(dipyrrylmethyl)biphenylene £3, m.p. 123°-125'C. NMR, 6 1.05 (12H, t, Et), 1.25 (12H, t, -OEt), 1.85 (12H, 5, Me), 3.75 (8H, q, Et), 4.30 (8H, q, -OEt), 5.50 (2H, s, CH), 7.30-7,80 (6H, m, biphenylene), 8.10 (4H, broad, NH). MS, m/e 896 (15), 850 (22), 804 (30), 715 (100). 1, B-Bis[4, 4’ -dietby1-3, 3’ -dimetby1-2, 2’ -dipyrry1)metby1] — biphenylene £39: 1,8-Bis{[5,5’-bis(ethoxycarbonyl)-4,4’-diethyla 3,3’dimethyl—2,2’-dipyrryl]methy1}biphenylene 13g (700 mg, 0.78 mmol) was heated at reflux in ethanol (30 mL) contain- ing potassium hydroxide (200 mg) for 6 h under nitrogen atmosphere. Most of the ethanol was then removed under vacuum and the residue was dissolved in water (40 mL), cooled in ice and acidified by glacial acetic acid. The -l32- precipitated acid was separated by filtration and washed with excess water to give 500 mg (90% yield). The tetra carboxylic acid (500 mg, 0.70 mmol) was dissolved in ethanolamine (10 mL) and heated at 100°C for l h under argon. The dark hot solution was added to ice (~25 gm). The white precipitate was filtered and washed with water, 400 mg (93% yield), m.p. 110-113'C. MS, m/e 608 (15), 510 (21), 411 (18), 150 (100); NMR, 8 1.05 (12H, t, Et), 1.85 (12H, 3, Me), 3.75 (8H, q, Et), 5.50 (2H, s, CH), 6.50 (4H, broad, singlet, a-H pyrrole), 7.30-7.80 (6H, m, biphenylene), 8.15 (4H, broad, NH). l,8—Bis[5-(2,8,l3,17-tetraethyl-3,7,12,l8—tetramethy1)- porphyrinyllbiphenylene 15: A procedure similar to that of the anthracene dipor- phyrin analogue was followed. Biphenylene diporphyrin was obtained in 9% yield, starting with a-free bis(dipyrryl)- methylbiphenylene 13g and 2.0 equivalents of (5,5’~ diformy1)-3,3’—diethy1-4,4’-dimethyl-2,2’-dipyrry1)methane g. NMR, 8 1.45 (12H, t, Et), 1.7 (12H, t, Et), 2.90 (12H, 3, Me), 3.15 (12H, 8, Me), 3.55 (8H, q, Et), 4.00 (8H, 2q, Et), 6.90-7.25 (6H, m, biphenylene), 8.45 (4H, s, meso-H), 9.00 (2H, s, meso-H), -7.40 (2H, s, NH), -7.80 (2H, 3, NH); UV-vis, x xnm (emM), 630 (2.8), 580 (7.5), 540 (7.8), 510 ma (13.3), 380 (189.4). -133- 1([5,5’-Bis(etboxycarbony1)-4,4’-dietby1-3,3’-dimetby1- 2,2’-dipyrry1)metby1}antbracene Zfi: l—Anthracene carboxaldehyde 73 (2.06 gm, 10 mmol) and ethyl 3-ethyl-4-methy1-2-pyrrolecarboxylate g (3.62 gm, 20 mmol) were dissolved in ethanol (40 mL) containing 1 mL of concentrated HCl. The solution was refluxed for l h under 'argon and then cooled in an ice bath for 3 h. The yellow crystals which resulted were filtered and dried. Recrys- tallization from 95% ethanol gave 5.15 gm (94% yield); m.p. 130-132'C. NMR, 8 1.15 (6H, t, Et), 1.40 (6H, t, -0Et), 1.75 (6H, 5, Me), 2.80 (4H, q, Et), 4.30 (4H, q, -OEt), 5.80 (1H, s, CH), 7.20-7.60 (9H, m, anthryl), 8.45 (28, broad, NH); MS, m/e 550 (30), 477 (21), 404 (48), 331 (15), 259 (55), 178 (100). 1 -[4, 4’ -Die thy] -3, 3’ -dine thy] -2, 2’ -dipyrry1)metby1] - anthracene zfi: The hydrolysis was carried out by refluxing the di- ester Zfi (5.0 gm) in ethanol (100 mL) to which 2 gm of NaOH (in 10 mL water) was added, for 8 h. Ethanol was evapo— rated and water (100 mL) was added. Acidification of this aqueous solution by glacial acetic acid gave the diacid after extraction with ether. Decarboxylation was possible by heating the diacid solution in ethanolamine (100 mL) at 150'C for 1/2 h. Pouring the hot solution onto ice (300 gm) and filtration of the yellow solid gave 3.0 gm (81% yield); m.p. 75'-77'C (decomposition). MS, m/e 406 -134- (48), 297 (100), 282 (37), 203 (20); NMR, 6 1.20 (68, t, Et), 1.65 (68, 3, Me), 2.50 (48, q, Et), 5.90 (18, s, CH), 6.40 (28, s, u-H pyrrole), 7.20—7.60 (108, m, anthryl and NH). J-[5-(2,8,13,17-Tetraetby1-3,7,12,18—tetrametby1)- porphyrinyljanthracene jg: The procedure is basically the same as that used in the synthesis of the 1,8-anthracene diporphyrin fi. Column chromatography (silica gel CHZCIZ/hexane, 50:50) was used to purify the product; 1.30 gm (24% yield). NMR, 6 1.60 (68, t, Et), 1.85 (68, t, Et), 2.05 (68, 3, Me), 3.70 (68, 5, Me), 3.90 (4H, 9, Et), 4.05 (48, q, Et), anthryl: 7.00 (18, t, 2-8), 7.30 (18, t, 3-8), 7.60 (18, s, lO-H), 7.85 (1H, d, 5-8), 8.00 (18, t, 7-8), 8.10 (18, d, 4-H), 8.70 (18, a, 9—8), 9.95 (18, s, meso-H), 10.20 (28, s, meso-H), ~3.10 (28, d, NH); UV-vis, x nm (EmM) 623 (3.1), 570 max (7.7), 534 (9.2), 500 (70.8), 404 (194). 1, 5-Bis {[4, 4’ -dietl1y1-3, 3’ —dimethy1—2, 2’ - dipyrryl]metbyljantbracene 22: A similar procedure to that of 1,8-analogue was followed: refluxing of 1.5-anthracene dicarboxaldehyde and ethyl 3-ethyl-4-methy1-2-pyrrolecarboxylate in ethanol in the presence of catalytic amounts of concentrated HCl to give 91% yield of 1,5—bis{[5,5’-bis(ethoxycarbonyl)-4,4’- diethyl-3,3’-dimethyl-2,2’-dipyrry1]methy1}anthracene, -135- after crystallization from aqueous ethanol. M.p. 170°C (decomposition). MS, m/e 923 (15), 876 (100), 875 (75). NMR, 8 1.05 (128, t, Et), 1.30 (128, t, -OEt), 1.60 (128, 3, Me), 2.85 (88, q, ET), 4.25 (88, q, -0Et), 6.20 (28, s, methine-H), 8.25 (48, broad, NH), anthryl: 6.95 (28, d), 7.50 (28, t), 8.10 (28, d), 8.40 (18, s), 8.50 (18, s). The tetra ethyl ester was hydrolyzed by heating in ethanolic alkaline solution, and the resultant acid was decarboxylated by heating in ethanolamine to give the a“ free 1,5-bis(dipyrrylmethyl)anthracene in 95% yield. M.p. 120-123'C (decomposition); MS, m/e 634 (25), 525 (15), 416 (90), 94 (100). NMR, 8 1.15 (128, t, Et), 1.90 (128, 3, Me), 2.45 (88, q, Et), 6.15 (28, s, CH), 6.30 (4H, s, 58- pyrrole), 7.00-8.70 (128, m, anthryl and NH). I,5-Bis[5-(2,8,13,17-tetraetby1—3,7,12,18-tetrametby)- porphyrinyllantbracene fig: To synthesize l,5-bis(etioporphyriny1)anthracene, a similar procedure to that of the 1.8-analogue (method D) was applied. The a-free l,5-bis(dipyrrylmethyl)anthracene 4 (900 mg, 1.42 mmol) was dissolved in C8308/C82012 (450 mL, 50:50), then 3-ethy1-2-formyl-4-methyl pyrrole Zfi (857 mg, 5.67 mmol) in methanol (50 mL) was added in one portion. The solution was stirred for 11/2 h while bub- bling nitrogen through the solution, after which time 30% HBr/HOAc (3 mL) was added and the stirring was continued for 15 min longer. Diethyl ether (400 mL) was added to the -136- solution after 082C12 was removed on a rotavap, and the precipitated 1,5-bis(biladiene-ac)hydrobromide-anthracene 79 was collected by filtration and used without further purification; 1.00 gm (49% yield) of dark green prisms. The biladiene-ac derivative (1.00 gm) was dissolved in dry DMF (40 mL) containing anhydrous copper(II) chloride (2.5 gm), the solution was then stirred at 145°C under argon atmosphere for 10 min and poured onto 200 mL water. The diporphyrin copper complex was extracted from the aqueous layer with chloroform (3 X 50 mL), and passed through a silica gel column (using 082012 as an eluent). The dark red eluate was concentrated up to 50 mL and washed with concentrated sulfuric acid, saturated NaZCO3 solution and with water, successively. The free base diporphyrin thus obtained was further purified by TLC (silica gel, CHCl3 as an eluent) to give 70 mg (9% yield). UV-vis, xmaxnm (emM), 624 (3.5), 568 (7.8), 536 (8.7), 500 (18.0), 406 (163.6); NMR, 8 1.45 (128, t, Et), 1.75 (128, t, Et), 2.05 (128, 5, Me), 3.55 (128, 3, Me), 3.75 (88, q, Et), 4.05 (88, 2q, Et), anthryl: 7.75 (28, t, 2,6-8), 2.90 (28, t, 3,7-8), 8.20 (28, d, 4.8-8), 8.50 (28, s, 9,10-8), 10.3 (28, s, meso—H), 10.50 (48, s, meso-H), -3.15 (4H, broad singlet, NH). -137- 8—anmy1-1{[5,5’~bis(etboxycarbony1)-4,4’—dietby1-3,3’- .dimetbyI-Z,2’~dipyrry1]metby1}anthracene gg: To 1,8-anthracene dicarboxaldehyde ; (500 mg, 2.1 mmol) suspended in dry ethanol (30 mL) was added cone centrated hydrochloric acid (0.5 mL). the mixture was stirred at room temperature until all the dialdehyde dissolved. to this solution was added ethyl-3-ethyl-4- methyl-Z-pyrrolecarboxylate g (774 mg, 4.2 mmol) in ethanol (10 mL) in three portions over a period of 15 min. After the additions, the solution was allowed to reflux for 30 min under nitrogen. A yellow precipitate started to form soon after the completion of addition. The reaction mixture was cooled and the solid was filtered, washed with water and dried (750 mg). The filtrate was concentrated to one-half of the volume and cooled to give a second batch (250 mg) of the dipyrrylmethane. The solids were combined and recrystallized once from benzene to give yellow plates (900 mg, 73x; m.p. 217-218°C; us, m/e 578 (M+, 45), 505 (10), 459 (8), 398 (30), 352 (25), 324 (40), 178 (100), NMR, 8 1.10 (68, t, Et), 1.30 (68, t, -0Et), 2.00 (68, 5, Me), 2.80 (48, q, Et), 4.10 (48, q, -OEt), 6.50 (18, s, methine CH), 7.00-7.70 (78, m, anthryl), 8.20 (18, s, 9-H) anthryl), 8.40 (28, broad, NH), 10.10 (18, s, 080). Analysis calculated for C 8 N O ' C, 74.71; 8, 6.62; N, 36 38 2 5' 4.84; found: c, 74.58; a, 6.75; h, 4.46. -l38- BeardroxymetbyJ-I-{[5,5’-bis(etboxycarbony1)-4,4’-dietbyl- 3,3’-dimetby1-2,2’-dipyrry1]metby1}antbracene gfig: To the 8-formy1-l-dipyrrylmethyl anthracene ;fi (578 mg, 1.0 mmol) in ethanol (2.0 mL) was added sodium borohydride (30 mg in 0.1 mL of water), and the mixture was stirred at room temperature for 15 min. A solution of NaOH (6N, 0.4 mL) was added. The mixture was heated on a steam bath for 5 min and then poured onto ice. The product was extracted with CH Cl2 (3 X 20 mL) to give a white solid, 2 550 mg (95x yield); m.p. 128-130°C; MS, m/e 580 (M+, 10), 553 (12), 399 (12), 308 (25), 176 (100). NMR, 8 1.10 (68, t, Et), 1.30 (68, t, -OEt), 2.00 (68, s, Me), 2.80 (48, q, Et), 4.10 (48, q, -OEt), 5.00 (28, s, 0820-), 5.30 (18, s, OH), 6.50 (18, s, methine CH), 7.00-7.77 (78, m, anthryl), 8.2 (18, s, 9-H anthryl), 8.40 (28, broad, NH). This solid was used in the next step without further purification. 8—(Bydroxxmethy1)-1-{[4,4’-dietby1-3,3’-dimetby1-Z,2’— dipyrryl]metbyljanthracene gfig: The diester dipyrrylmethyl anthracene lfig (500 mg, 0.86 mmol) was saponified by refluxing for 8 h in ethanol (10 mL) containing sodium hydroxide (300 mg) and water (1 mL). After the hydrolysate was concentrated to remove ethanol, water (20 mL) was added and the solution was extracted with dichloromethane (20 mL). The aqueous layer was kept in an ice bath and neutralized with glacial acetic acid, the precipitated white solid was extracted into ether -139- (3 X 20 mL). After removal of solvent, the crude diacid was dissolved in ethanolamine (5 mL) and heated to a gentle reflux under nitrogen for l h. The dark hot solution was poured into ice water (50 mL); the resultant light yellow solid was collected by filtration. This material was chro- matographed on silica gel (CHZCIZ) to give pure a-free dipyrrylmethane (360 mg, 95% yield); m.p. 98-100'C; MS, m/e 436 (M+, 7), 327 (20), 298 (8), 229 (10), 176 (100). NMR, 8 1.10 (68, t, Et), 2.00 (68, 5, Me), 2.30 (48, q, Et), 5.00 (28, s, O-CH 5.20 (18, s, OH), 6.20 (18, s, methine 2). . CH), 6.30 (28, s, a-H pyrrole), 7.00-7.77 (78, m, anthryl), 8.20 (18, s, 98-anthry1), 8.40 (18, broad, NH). 5~[8-(Hydroxymetbyl)-1-antbry1]-2,8,13,17-tetraetby1- 3,7,12,IE-tetrametbylporphine £2: To a solution of the decarboxylated 8-(hydroxymethyl)- l-[(4,4’-diethy1-3,3’-dimethy1-2,2’-dipyrry1)methyl]i anthracene ;§§ (347 mg, 1.20 mmol) and the 5,5’-diformyl- 3,3’-diethyl-4,4’-dimethy1-2,2’-dipyrrylmethane § (347 mg, 1.20 mmol) in dry methanol (70 mL) was added 70% perchloric acid (0.5 mL). The dark red solution was stirred for 12 h at room temperature in the dark, after which a solution of sodium acetate (0.5 gm) in methanol (10 mL) was added, followed by another solution of o-chloranil (200 mg) in methanol (10 mL). After 1 h, the mixture was evaporated; the residue was taken up by CH Cl2 (2.X20 mL) and a zinc 2 acetate (200 mg) in methanol (10 mL) was added. After -140- being stirred for 1 h, the mixture was evaporated and the residue was separated by chromatography (silica gel, CH2C12)' The isolated Zn(II) porphyrin ycomplex was demetalated by washing with 10% hydrochloric acid in di- chloromethane: yield, 400 mg (48%). NMR, 8 -3.00 (28, d, NH), 1.70 (68, t, Et), 1.90 (68, t, Et), 2.10 (6H, 3, Me), 3.70 (68, 5, Me), 3.80 (28, 5, C820—), 3.90 (48, q, Et), 4.10 (48, q, Et), 10.00 (18, s, meso-H), 10.30 (28, s, meso-H), anthryl: 7.10 (18, d), 7.40 (18, t), 7.80 (28, m), 8.00 (18, d), 8.10 (18, d), 8.40 (18, d), 8.70 (18, s); UV- vis, x mm (emM) 624 (2.4), 569 (6.0), 535 (6.5), 502 max (13.0), 405 (129.0). Analysis calculated for H N 0: C47 48 4 C, 82.42; H, 7.06; N, 8.18; found: C, 82.33; 8, 7.15; N, 8.09. 5~(8-Formy1-1-antbry1)-2,8,13,17—tetraetbyJ-3,7,12,18- tetrametbylporpbine gg: Oxidation of the anthracene alcohol porphyrin 17 (280 mg, 0.4 mmol) was effected by addition of its solution in pyridine (30 mL) to a cold solution of chronic trioxide (350 mg) in pyridine (20 mL) at 0-5'C. After stirring for 15 min, the ice bath was removed and the solution was stirred at room temperature for 4 h and then poured into water (100 mL). The product was extracted into dichloro- methane (3 X 50 mL) and purified by chromatography (silica gel, CHZClz) to give the corresponding aldehyde $3 in quan- titative yield; NMR, 8 —3.10 (28, d, NH), 1.67 (68, t, Et), 1.85 (68, t, Et), 2.03 (68, 5, Me), 3.68 (68, 5, Me), 3.85 -l41- (48, q, Et), 4.03 (48, q, Et), 9.38 (18, s, CHO), 9.91 (18, s, meso-H), 10.15 (18, s, meso—H), anthryl: 7.40 (18, t), 7.67 (18, d), 7.85 (18, t), 8.10 (18, d), 8.28 (18, d), 8.46 (18, d), 8.75 (18, s), 9.00 (18, s). UV-vis xmaxnm (6mm) 624 (2.4), 569 (6.1), 535 (6.6), 502 (13.5), 404 (140.0); MS, m/e 682 (M+, 5), 655 (8), 654 (12), 178 (100). trans-5,15-Bis{8-[5-(2,8,I3,17—tetraetbyI—3,7,12,18—tetra- methylporpbyrinyIJ-I-antbry1}-2,8,12,18—tetraetby1- 3,7,13,17-tetrametby1porpbine g9: To the porphyrin aldehyde lfi (50 mg, 0.073 mmol) suspended in methanol (10 mL) was added first the a-free (3,3’-diethyl-4,4’-dimethyl-2,2’—dipyrry1)methane lg (17.3 mg, 0.073 mmol) and the solution was deaerated by bubbling argon for 15 min before p-toluenesulfonic acid (3.4 mg, 0.018 mmol) was added. The mixture was stirred at room temperature (under argon, in the dark) for 10 h before the solvent was pumped dry. The residue was dissolved in THF (10 mL), treated with a solution of o-chloranil (10 mg) in THF (5 mL), and stirred for 1 h, and the solvent was removed again by evaporation. This mixture contained a large amount of unreacted porphyrin aldehyde which can be recovered during the isolation of the trimer. The chroma— tography was carried out with a silica gel column, using 082C12 to elute the porphyrin aldehyde lfi and 5% MeOH-CHZCI2 for the triporphyrin 39. The trimer thus obtained was purified further by conversion and chromato- -142- graphy of the Zn complex, which moves much faster than the free base, to remove impurities of low Rf values. Pure free base triporphyrin was then derived by demetalation of zinc complex using 10% 801 solution: yield, 7.0 mg; UV— vis, xmaxnm (emM) 625 (3.9), 573 (9.3), 538 (8.3), 506 (18.0), 395 (169.0). NMR, 8 ‘4.5 (68, three singlets clustered together, NH), 0.95 (128, t, Et), 1.10 (128, t, 81), 1.45 (128, t, Et), 1.60 (128, 3, Me), 1.80 (128, 3, Me), 3.02 (128, 3, Me), 3.20 (88, q, Et), 3.41 (88, q, Et), 3.55 (88, q, Et), anthryl: 6.81 (28, d), 7.35 (28, d), 7.50 (28, t), 7.59 (28, d), 7.60 (28, s), 8.05 (28, s), 8.45 (28, t), 8.72 (28, s), meso: 8.95 (28, s), 9.25 (4H, s), 9.30 (28, 3). Analysis calculated for 8 N C, C124 126 12‘ 83.46; H, 7.12; N, 9.42; found: C, 83.75; 8, 7.30; N, 9.32. High resolution positive ion mass spectra of the triporphy- rin have been obtained on Kratos MS-SORF equipped with ion- teck FAB gun, operated at 8 kV. The sample was prepared in l-thioglycerol matrix containing trifluoroacetic acid. Calculated for monoprotonated trimer: 1784.0306; found, 1784.0140. 5-{8—[(Methanesulfbnate)metby1]-1-antbry1}-2,8,13,17- tetraetby1-3,7,12,IB-tetrametbyiporpbine g9: 5[8-(Hydroxymethy1)-1—anthryl]-2,8,13,l7—tetraethyl- 3,7,12,18-tetramethy1porphine 11 (345 mg, 0.5 mmol) and excess amounts of methanesulfonyl chloride (2 mL) were dissolved in dry dichloromethane (20 mL). The solution was -l43- heated at reflux until all the porphyrin alcohol reacted (~48 h). The reaction was monitored by TLC. When the reaction was completed, the solvent was pumped dry, and the residue was dissolved in CHZCI2 (3 mL) and purified by column chromatography, using silica gel and C82012 as an eluent. The methyl sulfonate product has a higher Rf value than the corresponding alcohol, 300 mg (80% yield); UV-vis, xmaxnm (emM) 625 (2.4), 570 (6.6), 535 (7.1), 500 (14.3), 405 (125.0). NMR, 8 1.55 (68, t, Et), 1.75 (68, t, Et), 1.90 (38, s, -SOzMe), 2.00 (68, 3, Me), 3.55 (68, 3, Me), 3.75 (28, s, C82803-), 3.85 (48, q, Et), 4.00 (48, q, Et), 9.90 (18, s, meso-H), 10.15 (28, meso-H), -3.05 (28, d, NH), anthryl: 7.10 (18, d), 7.40 (18, t), 7.85 (18, t), 8.05 (18, d), 8.10 (18, d), 8.15 (18, s), 8.45 (18, d), 8.75 (18, s). 5-{8—[{Bis(2~pyridy1—B-etby1)amine)metby]]-1-antbry1}- 2,8,13,17-tetraetby1-3,7,12,18-tetramethylporpbine 3?: A solution of porphyrin methylsulfonate 39 (300 mg, 0.4 mmol) and-excess amounts of di(2-pyridy1-B-ethy1)amine 3; (227 mg, 1 mmol) in dry dichloromethane (20 mL) was refluxed under nitrogen; the reaction was complete after 48 h as detected by TLC. After that period, more dichloro- methane (30 mL) was added and the solution was washed with 5% hydrochloric acid solution (100 mL) to remove the excess dipyridine ligand, and it was next washed with a saturated solution of Na2C03, and finally washed twice with water -144- (2 X 100 mL). The organic layer was dried over anhydrous sodium sulfate. The solvent was evaporated on a rotary evaporator and the residue was pumped dry overnight. The product was purified by column chromatography (silica gel, 1% MeOH/CHZCIZ) and recrystallized from methanol/methylene chloride to give 170 mg (48% yield). UV-vis, xmaxnm (GmM) 623 (2.5), 570 (6.5), 534 (7.5), 502 (13.3), 404 (132.0). NM”, 8 0.65 (48, t, CH 1.50 (4H, t, CH 1.60 (68, t, 2). 2). Bi), 1.85 (68, t, Et), 2.05 (68, 3, Me), 3.15 (28, 5, C82- N), 3.60 (68, 3, Me), 3.85 (48, q, Et), 4.10 (48, q, Et), pyridyl: 5.70 (28, t), 6.20 (28, t), 7.60 (28, d), 8.45 (28, d), anthryl: 7.05 (18, d), 7.30 (18, t), 7.80 (18, t), 7.90 (18, d), 8.05 (4H, d), 8.90 18, s), meso-H: 9.95 (18, a), 10.10 (28, s), -3.30 (28, broad doublet, NH). 5~{8—[(1,7,10,IS-Tetraoxa—4,13-diazacyclooctadecane)— methy1]—1-antbry1}-2,8,13,17-tetraetby1-3,7,12,18-tetra- methylporpbine 3g: The methyl sulfonate porphyrin 3 (100 mg, 0.13 mmol) 1° and excess amounts of Kryptofix-22, 1w 3 (1,7,10,16-tetraoxa~ 4,l3,diazacyclooctadecane, 68 mg, 0.26 mmol) were refluxed in dry dichloromethane (20 mL) under nitrogen for 10 h. After the reaction was over, more 082012 (30 mL) was added and the solution was washed with 5% hydrochloric acid solu- tion (50 mL), followed by a saturated solution of sodium carbonate (50 mL) and finally with brine (100 mL). The organic layer was dried over anhydrous sodium sulfate and -145- the solvent was removed on a rotavap. The residue was dis- sOIVed in 1 mL of dichloromethane and purified by chromato— graphy on a silica gel column (1 X 5 inch), C82C12 was used first to elute the unreacted methyl sulfonate porphyrin which has a higher Rf-value than the product; chloroform was used next to separate the small amount of porphyrin alcohol which was generated during the reaction possibly by the hydrolysis of the methyl sulfonate, 2% CHaoH/CHZCI2 was used to elute the porphyrin-crown product. Yield was 35 mg (28%); MS, m/e M+ 929; UV-vis, in nm (cm 624 (3.0), 572 ax M). (6.0), 535 (8.0), 504 (15), 405 (148.0). NMR, 8 0.60 (48, t, crown), 1.00 (4H, t, crown), 1.35 (48, t, crown), 1.50 (48, t, crown), 1.60 (68, t, Et), 1.85 (68, t, Et), 2.05 (68, 8, Me), 2.10 (4H, t, crown), 2.50 (4H, t, crown), 3.10 (28, s, CHz-N), 3.80 (48, q, Et), 4.00 (48, q, Et), anthryl: 7.00 (18, d), 7.25 (18, t), 7.85 (18, d), 8.00 (18, d), 8.40 (18, d), 8.50 (2H, m), 8.80 (18, s), 8.95 (18, s), meso: 9.95 (18, 3), 10.10 (18, s), -3.30 (28, broad doublet, NH). 2-[5,5’-Bis(etboxycarbonyJ)—4,4’-dietby1-3,3’-dimethyl- 2,2’-dipyrry1]acenapbthen-I-one 3g: To a solution of acenaphthenequinone 33 (18.20 gm, 0.10 mol) and ethyl 3—ethy1-4-methy1-2-pyrrolecarboxylate 3 (36.20 gm, 0.20 mol) in absolute ethanol (300 mL) was added a concentrated solution of hydrochloric acid (10 mL). The dark red solution was refluxed under argon atmosphere -146- for l h, after which time the solution was chilled in an ice-bath for 2 h. The light brown crystalline product was then taken by filtration and washed with 100 mL of cold 95% ethanol and dried. Recrystallization from aqueous ethanol gave 45 gm (85% yield) of light brown prisms; m.p. 105- 106°C. NMR, 8 1.10 (68, t, Et), 1.30 (68, t, -OEt), 1.65 (68, 5, Me), 2.65 (48, q, Et), 4.30 (48, q, -OEt), 7.4-8.2 (68, m, naphthyl), 8.45 (28, broad, NH). MS, m/e 526 (M+, 100), 480 (53), 437 (50), 391 (39), 346 (51), 240 (44), 217 (58). 8-[(5,5’-Dicarboxy1-4,4’-dietby1-3,3’-dimetnyJ-2,2’- dipyrry1)metby1]napbtbaIene—J-carboxylic acid 32: A suspension of the diester dipyrrylmethylacenaphthe- none 3fi (1.40 gm, 2.66 mmol) in 30% aqueous solution of potassium hydroxide (30 mL) was refluxed for 6 h under argon. The dark red-brown solution was diluted by adding water (70 mL), washed twice with chloroform (2 X 50 mL) and the aqueous layer was cooled in ice before being neutra- lized with glacial acetic acid. The resultant solid was collected by filtration and washed with water (2 X 50 mL); the solid was dissolved in chloroform (100 mL) and dried over anhydrous Na2S04. Evaporation of the organic layer on a rotavap produced the tricarboxylic acid, 1.03 gm (80% yield). No further purification has been done; m.p. 250°C with decomposition; MS, m/e 444 (M+-CO 20), 400 (18), 355 2, (27), 126 (100). NMR, 8 1.05 (68, t, Et), 1.60 (68, s, -147— Me), 2.70 (48, q, Et), 7.30-8.30 (68, m, naphthyl), 8.50 (28, broad singlet, NH). 8[(4,4’-DietbyJ-3,3’-dimethyJ-2,2’-dipyrry1)metby1]- naphthalene-I-carbaxylic acid gg: The decarboxylation of the 5,5’-dicarboxylic groups in 180 Z was achieved by heating a solution of (1.0 gm, 2.04 mmol) of 31 in ethanolamine (20 mL) at reflux under argon for 1/2 h. The hot dark solution was poured onto ice (50 gm), and the yellow precipitation was collected by fil- tration and washed with water. Yield was 700 mg (85%); m.p. loo-110°C (decomposition); MS, m/e 400 (7), 355 (20), 291 (10), 126 (100). NMR, 8 1.05 (68, t, Et), 1.60 (4H, 3, Me), 2.70 (48, q, Et), 6.50 (28, broad, a-H pyrrole), 7.20— 8.30 (68, m, naphthyl), 8.60 (28, broad, NH). .Metbyl 8—5-(2,8,13,17—tetraetbyJ-3,7,12,18—tetrametby1)- porphyrinyljnapbthalene-I-carboxylate jg: (A) From a-free dipyrrylgetggne condenggtion with 5,5’- diforgyl dipyrryl gethgne 3: The 5,5’-unsubstituted dipyrrylmethyl naphthalene 33 (600 mg, 1.50 mmol) and 5,5’-diformyl-3,3’-diethyl-4,4’-di- methyl-2,2’-dipyrrylmethane 3 (429 mg, 1.5 mmol) were dissolved in dry methanol (20 mL). Argon was bubbled through the solution for 15 min before perchloric acid (70%, 0.4 mL) was added. The solution was stirred at room temperature (under argon in the dark) for 15 h, after which -148- time a solution of tetrachloro-o-benzoquinone (200 mg) in methanol (10 mL) was added and stirring was continued for one additional hour. Dichloromethane (50 mL) was added and the solution was washed several times with water (50 mL portions) and dried over anhydrous Na2804. The solvents were removed under vacuum and the residue was dissolved in chloroform (2 mL) and purified by column chromatography (silica gel, 5% methanol/082012). The porphyrin acid 33 was dissolved in dry C8 012 (10 2 mL) and oxalyl chloride (0.5 mL) was added. The solution was refluxed for 2 h under argon. The solvents were then pumped dry, and the residue was heated at reflux in metha- nol (20 mL) for 8 h. The solvent was again removed on a rotavap under vacuum. The residue was purified on a silica C1 gel column (1 X 10 inch) using CH 2 to elute the porphy- 2 rin naphthalene ester, 200 mg (20% yield) was obtained as dark purple plate after crystallization from metha- nol/CHZCIZ. MS, m/e 662 (55), 590 (10), 331 (10), 126 (15), 85 (100); UV-vis, xmaxnm (emM) 625 (3), 573 (16), 536 (7), 504 (14), 404 (180), NMR 5 (B) From a,e:dicarboxyldipyrrylgethgge 3! condenggtion: The 5,5’-dicarboxydipyrrylmethyl naphthalene carboxylic acid 32 (1.0 gm, 2.05 mmol) was dissolved in dry C82012 ethyl-3,3’-dimethy1-2,2’-dipyrry1methane 3 (586 mg, (30 mL) together with 5,5’-diformy1-4,4’-di- 2.05 mmol). To the solution was added p-toluene sulfonic -149- acid (800 mg, 4.1 mmol) and stirred at room temperature for 6 h, a saturated methanolic solution of zinc acetate (10 mL) was then added and stirring continued for 12 more hours. After that a solution of o-chloranil (500 mg) in C8 C12 (5 mL) was added and stirred for an additional hour. 2 Methylene chloride (50 mL) was added and the solution was washed twice with water (2 X 50 mL), evaporation of the organic layer and column chromatography of the residue (silica gel, 1% methanol/CHZCIZ) gave the zinc complex of the porphyrin acid after eluting impurities with high 8f volume by pure 082012. The Zn(II) porphyrin acid solution in C8C1 (50 mL) was shaken with 10% hydrochloric acid 3 (50 mL) in a separatory funnel for ~2 min and the organic layer was washed with water (2 X 50 mL). After evaporation 2C12 (25 mL) together with oxalylchloride (1 mL) and refluxed of the solvents, the residue was dissolved in dry CH under argon for 2 h before the solvents were pumped dry. The porphyrin methyl ester was obtained from the acid chlnride by refluxing in methanol and purifying the product by chromatography as in part (A). 240 mg (17.7% yield) was obtained after recrystallization from methanol/082C12. The product was identical in every respect to the authentic sample prepared by method (A). -150- 5*[8-(EYdroxymetby1)-1—napbtby1]—2,8,I3,17~tetraetby1- 3,7,12,18-tetrametby1porpbyrin 5!: The porphyrin methyl ester 33 (200 mg, 0.30 mmol) was dissolved in dry THF (25 mL). Warming might be necessary in order to assure complete solubility. Lithium aluminium hydride (38 mg, 1 mmol) was added to the solution at room temperature. After stirring for 2 h, the solvent was removed and the corresponding alcohol was separated from the residue by column chromatography (silica gel, 1% metha— nol, C8 C1 2 2)' crystallization from methanol/C8201 155 mg (80.6% yield) was obtained after re- 2; us, m/e 634 (M+, 8), 616 (13), 603 (10), 126 (100). NMR, 8 1.60 (68, t, Et), 1.85 (68, t, Et), 2.00 (68, 3, Me), 3.50 (68, 3, Me), 3.60 (28, 3, C820), 3.85 (48, q, Et), 4.00 (48, q, Et), 9.90 (1“, s, meso—H), 10.20 (28, s, meso-H), naphthyl: 7.3-8.00 624 (4), (68, m), -3.10 (28, d, NH). UV-vis xm nm (em ax M) 572 (14), 535 (10), 503 (12), 402 (177). 5 (B-FormyI-I-napbtby1)-2,8,13,17-tetrametby1-3,7,12,18- tetrametbyl porphine g}: A solution of (8-hydroxymethyl)-l-naphthyl porphyrin g; (170 mg, 0.26 mmol) in pyridine (10 mL) was added at room temperature to chronic trioxide (200 mg) solution in pyridine (20 mL). 'The resultant solution was stirred under argon in the absence of light for 2 h and then poured into cold water (50 mL) and the corresponding porphyrin aldehyde was extracted with chloroform (3 X 25 mL). The combined -lSl- extracts were washed several times with 30 mL portions of water, and the organic layer was evaporated to dryness. The product was separated from the residue by column chromatography using silica gel and CHZCl2 as an eluent, the porphyrin aldehyde has an 8f value of 0.9 compared to 0.6 for the corresponding alcohol. The product was re- crystallized once from methanol/08201 to give 150 mg (88% 2 yield); us, m/e 632 (M+, 7), 604 (26), 603 (30), 475 (15), 126 (100). NMR, 5 1.60 (an, t, Et), 1.80 (68, t, Et), 2.00 (68, 3, Me), 3.50 (68, 5, Me), 3.80 (48, q, Et), 3.95 (48, q, Et), naphthyl: 7.55 (18, t), 7.70 (18, d), 7.80 (18, t), 8.05 (18, d), 8.30 (28, d), 7.90 (18, s, C80) 9.90 (18, s, meso—H), 10.10 (28, s, meso-H), -3.25 (28, d, NH); UV-vis, xmaxnm (cmM) 624 (3), 572 (12), 535 (11), 502 (10) 402 (179). 5{8[(Methanesulfbnate)metby]]—1-napbtby1}-2,8,13,17— tetraetbyl-S,7,12,18—tetramethy1porphine gg: To a solution of the naphthyl porphyrin alcohol 51 (400 mg, 0.63 mmol) in dry methylene chloride (10 mL) was added methane sulfonyl chloride in excess (1 mL) and pyri- dine (0.1 mL). The solution was stirred under argon atmosphere until no more alcohol could be detected by TLC (Rf value of alcohol is 0.6, silica gel/CHZClz), about 8 h. The solvents were then removed on a rotavap and pumped dry under vacuum (~8 h), and the product was purified by chromatography (silica gel plates, CH2C12)° Yield was -152- 229 mg (51%); UV—vis, xmaxnm (emM) 625 (4), 574 (16), 535 (12), 504 (10), 402 (175); MS, m/e, 712 (M+, 5), 649 (14), 617 (23), 603 (18), 478 (55), 126 (100). NMR, 8 -3.05 (28, d, NH), 0.05 (38, 3, -80 -Me), 1.55 (68, t, Et), 1.75 (68, 2 t, Et), 2.00 (68, s, Me), 3.45 (68, 8, Me), 3.55 (28, s, CHz-O), 3.75 (48, q, Et), 4.00 (48, q, Et), 9.85 (18, s, meso—H), 10.05 (28, s, meso-H), naphthyl: 7.2-7.95 (68, m). 5-{8-[fBis(zépyridyI-B~etby1)amine}metbyJ]-1-napbtby1}— 2,8,13,17-tetraetby1-3,7,12,18-tetrametby1porpbine g3: A solution of methanesulfonatenaphthyl porphyrin 33 (450 mg, 0.63 mmol) and excess amounts of bis(2-pyridy1-B- ethyl)amine (300 mg, 1.32 mmol) in dry C8 C1 (20 mg) was 2 2 heated at reflux under nitrogen for ~20 h. The completion of the reaction was tested by TLC. Methylene chloride (20 mL) was added and the solution was washed with dilute hydrochloric acid (2%, 50 mL), then several times with water (3 X 50 mL). The organic layer was dried over anhy- drous sodium sulfate and evaporated. The product was iso- lated from the residue by chromatography (silica gel plate, CHC13) to give 200 mg (37.5% yield), dark purple cubic crystals from methanol/CHZClz. MS, m/e 843 (4), 765 (7), 685 (20), 603 (38), 477 (51), 126 (100); UV-vis, x nm max (émM) 624 (3), 574 (13), 534 (12), 503 (11), 403 (150); NMR, 8 0.30 (4H, t, -CH 0.80 (4H, t, N-CH 1.65 (68, 2), 2), t, Et), 1.85 (68, t, Et), 2.05 (68, 5, Me), 2.75 (28, s, ~153- benzilic-H), 3.50 (68, 5, Me), 3.90 (48, q, Et), 4.00 (48, q, Et), 5.40 (28, d, pyridyl), 6.35 (28, t, pyridyl), 6.60 (28, t, pyridyl), 6.70 (28, d, pyridyl), 7.70-8.40 (68, m, naphthyl), 9.90 (18, s, mesa-8), 10.00 (28, s, meso H), 73.15 (2H, d: NH). 5’(8—Carboxy1-1-antbry1)-2,8,13,17-tetraetbyI-3,7,12,18- tetrametbylporpbine g?: (A) By oxidgtion of the corrggpondigg_glcohol ll: To a cold solution of the 5-(8-hydroxylmethyl-l- anthryl)porphyrin 3! (684 mg, 1 mmol) in acetone (30 mL), (kept at -5 to 0°C), was added 0.3 mL of Jone’s reagent (prepared by dissolving 6.70 gm of CrO3 in 6 mL of concen- trated sulfuric acid and diluting the solution with 50 mL of water). After complete addition of the oxidizing agent, the solution was left to stir at ambient temperature for 10 min, after which time water (25 mL) was added and the porphyrin carboxylic acid was extracted from the aqueous solution with chloroform (3 X 25 mL) and the solvent was removed. The product was purified by column chromatography (silica gel, 5% methanol/CH Cl 8 0.4); 623 mg (89% 2 2’ f yield) was obtained. us, m/e 656 (M+-CO 40), 478 (16), 2’ 178 (55), 44 (100); UV—vis, xmaxnm (emM) 625 (3.0), 570 (5.5), 532 (6.8), 504 (14.7), 404 (122.0). NMR, 8 -3.10 (28, d, NH), 1.70 (68, t, Et), 1.85 (68, t, Et), 2.00 (68, 8, Me), 3.75 (68, 3, Me), 3.90 (48, q, Et), 4.05 (48, q, Et), 9.95 (18, s, meso-H), 10.30 (28, s, meso-H), anthryl: -154- 7.15 (18, d), 7.40 (18, t), 7.85 (28, m), 8.00 (18, d), 8.10 (18, d), 8.40 (18, d), 8.75 (18, s). (B) 31 hydrolysis of porphyrinfigethyl ester 33: To a solution of 5-(8-methoxycarbony1-l-anthryl)~ 2,8,13,l7-tetraethy1-3,7,12,lB—tetramethylporphine 33 (684 mg, 1 mmol) in formic acid (98%, 20 mL) was added con~ centrated 8C1 (5 mL). The solution was heated at reflux under argon for 6 h. The solution was then added to 50 mL of ice-water in a separatory funnel and the porphyrin was extracted with 08013 (3 X 30 mL). The organic layer was washed with water several times (4 X 50 mL) and dried over anhydrous Na2S04 and the solvents were evaporated on a rotavap until dry. Purification of the product was pos- sible by column chromatography (silica gel, 5% metha- nol/CHZCIZ, Hf, 0.4) to give 91 mg (13% yield). The product was identical in every respect with an authentic sample prepared by method (A). 8¢MetboxycarbonyJ-1—antbracenecarboxaldebyde 3!: To a suspension of dimethyl 1,8-anthracene- dicarboxylate 33 (6.00 gm, 20 mmol) in dry diethyl ether (200 mL) was added lithium aluminum hydride (0.76 gm, 0.02 mmol). The mixture was heated at reflux for 3 h, and then cooled in ice before ethyl acetate (4 mL) was added. The mixture was stirred for 5 more minutes and the solvents were removed by evaporation. To the ice-cooled residue, 6N hydrochloric acid (100 mL) was added portion-wise (10 mL -155- each portion); the yellow precipitates which resulted were separated by filtration, washed with dilute HCl and with excess water afterward, and then dried in air to give 5.50 gm. No further purification of the products was done at this stage. The dry solid (5.50 gm) was dissolved in dry pyridine (75 mL) and the solution was added to an ice-cooled chromic trioxide, CrO3 (22.0 gm) solution in pyridine (100 mL). The mixture was stirred at 0°C for 15 min. After that the ice bath was removed and the dark brown solution was stirred at room temperature for 4 additional hours. The solution was poured onto water (1000 mL) and the brown precipitate which resulted was obtained by filtration, washed with excess water and dried in air. The products were obtained from the brown solid by extracting with benzene (150 mL) in a Soxhlet extractor. Yellow solid obtained by evaporation of benzene was found to contain three components that were separable by column chroma- tography, on silica gel. The unreduced dimethyl 1,8- authracenedicarboxylate (0.9 gm) was eluted with 1:5 hexane/CHZCl2 (Rf 0.9); the desired product (monomethyl- ester anthracene carboxaldehyde; 2.0 gm, 37% yield) was eluted with CH Cl2 (Rf 0.7). The third component was 1,8- 2 anthracene dicarboxaldehyde (2.1 gm, 45% yield) obtained by elution with CHCl3 (Rf 0.5). Data for 3;: m.p., ISO-152°C; MS, m/e 264 (M+, 12), 236 (16), 221 (28), 206 (60), 192 (65), 178 (94), 83 (100). NMR, 8 4.05 (3H, 3, Me), 7.30- -156- 8.20 (68, m anthryl), 8.45 (18, 3, 10—8 anthryl), 10.50 (18, s, 9-H, anthryl), 10.90 (18, s, 080). 8-{[5,5’-Bis(etboxycarbony1)—4,4’-dietbyI-3,3’-dimetby1~2, 2’~dipyrry11metby1~1 F i p N use gem use messes o.m o.oa ? i p 1‘ j + i L L P > p 4% . «Mum» 3606.3» 23 S mewsmsmsom «N armammaofi as -ae-m~-ssesmssssms-aN.m~.a.m-Nsesmussms-m~.mH.m.m_-eemsn-m.~ kc sassomam ass as use sew New masses gem and- o.o o:~ .xm o.m o.v o.m one fine o.m o.a P — P p p P b P p P F -187- w _ w -188- . «MU 3.398% hf asses as $138 8-368.389-2832 .m-”Nséuzeéissfimssaosess -26 .8 588% as as «a: 8m. 8:. 8.38.8 :Em 0.0 o.H o.N o.m o.¢ o.m o.m 0.5 o.m o.m c.0H » F > L - LP » b p L > LP IIJ j} A l< fijgj -188- . «MU mswxmsons 93 88.33 $12gassesmahfils.28.m-SsésxaéAssfmsssEesS-2-e .8 5:588 as m we: 88 was 8.88.8 N :Em o.o o.H o.N o.m o.v o.m 0.0 0.5 0.x o.m 0.0H > L > L b p > L > _ p L i L > F ll» L > L » IF lq :4 a z fills? o.mn o.NI o.HI I. p » L - L J) -l89- . «NW.» gwemmfiomu m~$m§§umfllmw «NH sum am. 13569331:«MN«m«mnfimfimhseumuSmfimsxmaesobsmumgeoéHumwnw k0 goose $3 m 5% saw m3. 6.83% N as o.o o4 o.~ o.m o4. o.m oé ofi o.m o.m 0.3 rfr Fr FLL rrr rrt+ r r L1? rr. 3 7 A1 Q 711:2: -190- . 22 n . q u u q .. a .. m assumummeesmxmsomNmfims 88 8888358888 8.2..“ m. 35883 2 mm m m S .E . 888-2818 8-388888258 .813 -3883 e888 ,8 5:888 SE mm was 8m 82. 88$ 28 0.0 o.H o.N o.m o.v o.m o.m 0.5 o.m o.m o.oa 1_.L.L1L._»L1bi_1h L_._._ H1 4m? 8.8888 88883-8 N .2 .a .m. -N as 88.8.2 .2 .8 8-: .8584- Sargasiaoms sf 8 888-2818 .m-~s£888-2.2.m .873 833-2882?“ .8 5:888 8% mm um: 88 82. 88$ 28 0.0 o.H o.N o.m o.v o.m 0.0 o.h o.m o.m o.oa 1_+h.L.Li_-L.b-_1LLL._L_ o.¢l o.MI o.NI o.HI -l90- Milne . nv -191- . mm? 838.8 s8 8888.: .2 .8 .m 1N as 88 8 umN.mN.m.mumNms&s:eumu_mesmE nauseamehmumerpmammumcuse Humwnm k6 sxseommm mzé m~ amz.ewm WNQ ossmwm yam o.o o.a o.N o.m o.v o.m o.w o.h o.m o.m c.0H 1 F P P b 1 h 1 b L L 1 P 1 L b 1 j W . % .mWWV mswxasomNmaumEdsemuumm«ma.mqmumeumesumunmm«wwqmamuNmrmxmnw k6 Seapoomm mzé aw mm: qwm ems mssmwk {EN o.o o.H o.N o.m o.v o.m o.© 0.5 o.m o.m 0.0H 1 P 1 L 1 L 1 p 1 L 1 p 1 L 1 L 1 L 1 P 1 L I W o.MI o.NI . 1 L 1 _ 1 O P“? -192- .. . . .88 888.8 138881882 3 a 88888812.2.m818888448818181m .8 2:888 SE 3 as: 88 8:. 8.388 ELL ®.NI 8.9 G.N G4. §.m m.m ®.@.. gm; pppn—ph-ptpbp—bbppwbth—P:bpthp—phbbtnps—>bbh->s-b—h-bshrubs—bhsnhbbhs—shbhtE—smpsbhbbh—Pbbbbmbbb—Dhbbbsth—hbbsnub». —>>b>>>>sn-nhhbsE—bbbbbbbph—nbb-snphi—ss-sbbhbh—bsbbbbs 1 De -l93- -l94- .mwc 83.338: S88888838888812.2.8 813588818128 .8134 .8 5:888 mg as we: 88 Se 882 ELL 8;... fl.NI @.m ®.N 85¢ u.m @.m u.&~ _1 r _ _ _ _ _ _ _ _ r r r F L L 1 1s] j. .1. @© -195- . «Mm.» mrwxmsefi has assess an 18H«MN.8.mnmesmessmsIQN.mm.m.m1-h&axueau~cmwesw~.wumsese ks esseomma axe mm amz,ewm mNm msxmmm ELL gm &.N a4. mm &.m QGL o.m... o.NI o.HI r,_-_1 .Q8 83888 as 8 1888 NN .mN .N 81388881812 .m81388888813382 .m .8 5:888 8.2 mN we: 88 8a. 88.8 _LLL ea e.m 9;. 95 9m ea: _ L1 r L 1- _ L r F _ _1 1—111L mu, o.m1 o.~1 9N1 LI1111.-_1111111_11 4-2-1 -l97- .Nmmc 83883 sf 883A N .2 .N .N. 1N 88 16.3313 .m N .m «m1SmssNguNuNmsoaseomaofioga«3an 6198.3 .63 Sassoon.» NEE m mm: me N3. 833% N :1... as ®.N as. 9m 9m 98 V r _ _ _ _ _ _ _ _ _ _ o.m1.. o.~1 r _ -l98- . Q8 83.88 ea 88.5 Rd N .NN .N .m 13588818128 .8138.888173885888812«.31281888 .8 8888 see NNN «a: 8“ m8. 88E .mmmv oswxmsomumemseseoaumN«Mquan 18888818128 .mSsésestsfimssaEeseé8.812811.8 .8 8888 $2 m «NE 88 8s. 8:88 -l99- N ELL .N G.¢ ®.m G.m u.®~ F _ _ _ r L r _ _ . NM“? 8:83.83 hf 88.3 3 12.28818881881818.m815888217888128.81281888 .8 58.888 83 anN we: 88 8?. 88.8 . . . it p. a G N 8 ¢ 8. m S. m G. SN b r .7 _ _ _ _ _ _ _ _ _. 1 J \ o.m1 o.NI o.HI -200— .«cmmy mrwxmgommeumEdagmnuAN.mN«mam umeumcLumgnmN.m~«m.mn”menxmurumuoguwrum_mwQIWN«wummo kc Exzuommm mzz mm mm: cum mmw mgxm$m -201- -201- .acmmy mrwxmgommeumsdxumaumm«Mmqmqm -Nmfimcfifi-2 .2.m.m-:mfixugAékfié$3-2$-98 .8 5.298% mg ma 2% 3m a? $ng J W - 4 431.57% a REFERENCES 10. 11. 12. 13. 14. LIST 0? REFERENCES Wasielewski M.R.; Svec, W.A.; CoPe, 0. Chem. Soc. 10 1961 (1978). Netzel, T.L.; Kroger, P.; Chang, C.K.; Fajer J. Chem; Phys. Lett. 61, 223 (1979). Netzel, T.L.; Bergkamp, M.A.; Chang, Acad. Sci. USA 1g, 413 (1982). Chang, C.K.; Kuo, M.-S.; Wang, C.-B. Chem. 14, 943 (1977). Chang, C.K. J. Heterocycl. Chem. J Chang, C.K. J. ChemLiSoc., Chem. Commun., Chang, C.K. ACS Adv. Chem, Ser. 173, B.T. J. Am. Fujita, I.; C.K. Proc. Nat. J. Heterocycl. 14, 1285 (1977). Collman, J.P.; Elliott, C.M.; Halbert, 18 (1977). B.S. Proc. Nat. Acad. Sci. USA 14, Collman, J.P.; Anson, F.C.; Barnes, C.E C.S.; Geiger, T.; Evitt, E.R.; Kreh, R.P.; Pettman, 8.8. J. Ag. Chem. Soc. 105, T.R.; 800 (1977). 162 (1979). Tourog, ., Bencosme, Meier, K.; 2694 (1983). Collman, J.P.; Bencosme, C.S.; Durand, Jr., R.R.; Kreh, R.P.; Anson, F.C. J. Ag; Chem. Soc. 05, 2699 (1983). Collman, J.P.; Bencosme, C.S.; Barnes, C.E.; Miller, 8.0. J. Am. Chem. Soc. 105, 2704 (1983). Durand, R.R.; Bencosme, C.S.; Collman, J.P.; Anson, F.C. J. Am. Chem. Soc. 105, 2710 (1983). Collman, J.P.; Denisevich, P.; Koval, C. J. Electroanal. Chem, Interigcial Electrochgg. 117 (1979). ; Anson, F C. a) Lemberg, M.R. Physiol. Rev. 49, 48 (1969); b) Nicholle, P.; Chance, B. in "Molecular Mechanisms of Oxygen Activation, (O. Hayaishi, Academic PresszNew York (1974). -202- ed.), PP. 479-536, 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. -203- Florkin, M. in "Comprehensive Biochemistry," Vol. 31, (M. Florkin and E.H. Stolz, eds.), ElsevierzAmsterdam, pp. 187-235 (1962). a) Caughey, W.S.; Wallace, W.T.; Volpe, T.A.; Yoshikawa, S. in "The Enzymes," (P.D. Boyer, ed.), Vol. 13, Academic PresszNew York, pp. 299-337 (1976); b) Hikstram, M.; Krab, R.; Saraste, M. in "Cytochrome Oxidase,” Academic PresszNew York, p. 86 (1981). Challoner, D.R. Nature (London) 217, 78 (1968). Tucker, A. Science 154, 150 (1966). Malmstram, B.G. Quart. Rev. Biophvg. 6, 389 (1973). Keilin, 0.; Bartree, E.F. Proc. Roy. Soc. London B127, 167 (1939). Babcock, G.T.; Vickery, L.E.; Palmer, G. J. Biol. Chem. 251, 7907 (1976). Palmer, 6.; Antolis, T.; Babcock, G.T.; Gracia- Iniguez, L.; Tweedle, M.; Wilson, L.J.; Vickery, L.E. in "Mechanisms of Oxidizing Enzymes," (T.P. Singer and R.N. 0ndarza, eds.), Elsevieerew York, p. 221 (1978). Tsao, M.S.; Wilmarth, W.K. Adv. Chem. Ser. _fi, 113 (1962). Sehested, K.; Rasmussen, 0.L.; Fricke, H. J. Phxs. Chem. 12, 626 (1968). a) George, P. in "Oxidazes and Related Redox Systems," (T.E. King, B.S. Mason and M. Morrison, eds.), WileyzNew York, p. 3 (1965); b) Wikster, M.; Krab, R.; Saraste, M. in "Cytochrome Oxidase," Academic PresszNew York, p. 12 (1981). Wilkins, R.G. Adv. Chem. Ser. 100, 111 (1971). Taube, H. J. Gen. Physiol. 49, 29 (1965). Methews, G.T.; Robins, R.G. Australas. Inst. Mining Met., Proc. [31,242, 47 (1972). Simplicio, J.; Wilkins, R.G. J. Am. Chem. Soc. _g, 6092 (1967). Stadtherr, L.G.; Prados, R.; Martin, R.B. Inorg. Chem. 12, 1814 (1973). 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. -204— Cohen, I.A.; Caughey, W.S. in "Remes and Hemo- proteins," (B. Chance, R.W. Bstabrook and T. Yonetani, eds.), Academic Press:New York, p. 577 (1966). Caughey, W.S.; Davies, T.L.; Fuchsman, R.B.; McCoy, S. in "Structure and Function of Cytochromes," (K. Okunuki, M.D. Kamen and Sekuzu, eds.), University of Tokyo PresszTokyo, p. 20 (1968). George, P. J. Chem. Soc., 4349 (1954). Caughey, W.S. Ann. N.Y. Acad. Sci. 174, 148 (1970). Bockris, J.O’M.; Srinivasan, S. in "Fuel Cells: Their Electrochemistry,” McGraw-HillzNew York (1969). Vielstich, W. in "Fuel Cells," Wiley-IntersciencezNew York (1970). Gould, R.F. in "Fuel Cell Systems, II," in ACS Adv. Chem. Ser., Vol. 90 (1969). Cohen, R. Proc. Ann. Power Sources Conf. 29¢ 21 (1966). Maget, H.J.R. in "Handbook of Fuel Cell Technology," (C. Berger, ed.), Chapter 4, Prentice Ha11:Eng1ewood Cliffs, NJ (1968). Schanz, J.L.; Bullock, E.: Paper presented in American Rocket Society, September 25-28, Santa Monica, CA (1962) Dolpin, D.; Riom, J.; Pain III, J.B. Heterocycleg 16, 417 (1981). Abdalmuhdi, I.; Chang, C.K. J. Org. Chem. 48, 5388 (1983). a) Abdalmuhdi, I.; Chang, C.K. Angew. Chem. 96, 154 (1984); b) Abdalmuhdi, I.; Chang, C.K. Angew. Chem. Int. Ed. Engl. gg, 164 (1984). Abdalmuhdi, I.; Chang, C. J. Org; Chem. (1985). 0, 411 Abdalmuhdi, I.; Chang, C.K.; Liu, H.Y. J. Am. Chem; Soc. 106, 2725 (1984). Abdalmuhdi, I.; Chang, C.K.; Anson, F.C.; Liu, H.Y. J. Phys. Chem. 89, 665 (1985). 47. 48. 49. 50. 51. 52. 53. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. -205- Abdalmuhdi, I.; Fillers, J.P.; Ravichandran, K.G.; Chang, C.K.; Tulinsky, A. J. Am. Chem. Soc., in press. Abdalmuhdi, I.; Chang, C.K.; Kondylis, M. J. Am. Chem. Soc., in press. Johnson, A.W.; Kay, I.T. J. Chem: Soc., 1620 (1965). Ogoshi, R.; Sugimoto, R.; Nishiguchi, T.; Watanabe, T.; Matsuda, Y.; Yoshida, 2. Chem: Lett., 29 (1978). Akiyama, S.; Misumi, S.; Nakagawa, M. Bull. Chem. Soc. Jpn. 35, 1829 (1962). Akiyama, S.; Misumi, S.; Nakagawa, M. Bull. Chemg Soc. Jpn. 33, 1293 (1960). Stock, L.M.; Golden, R. J. Am;,Chem. Soc. 94, 3080 (1972). Abchibald, J.L.; Walker, D.M.; Shaw, K.B.; Markovac, A.; McDonald, S.F. an. J. Chem. ii: 345 (1966). Fischer, R.; Berg, M. and Schormuller, S. Ann. 80, 109 (1930). Shneerson, L. and Leibush, M. J. Applied Chem; USSR 19. 869 (1946). Chong, R.; Clezy, P.S.; Liepa, A.J.; Nichol, A.W. Amst. J. Chem. 22, 229 (1969). Arsenault, G.P.; Bullock, E.; MacDonald, S.F. J. Am. Chem. Soc. gg, 4384 (1960). Fischer, R.; Kurzinger, A.Z. Physiol. Chem. 196, 213 (1931). Thonson, A.W.; Kay, I.T. J. Chem, Soc., 2418 (1961). Badger, G.M.; Harris, R.L.N.; Jones, R.A. Aust. J. Chem. 17, 1013 (1961). Fischer, 8.; Stangler, G. Justus Leibigg Ann. Chem. 459, 84 (1927). Wang, C.—B.; Chang, C.K. S nthesis, 548 (1979). Wilcox, Jr., C.F.; Uetrecht, J.P.; Grohmann, L. gm Am. Chem. Soc. 94, 2532 (1972). 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. -206- Campbell, C.D.; Rees, C.W. J. ChemmiSoc. C, 742 (1969). Logullo, F.M.; Seitz, A.H.; Friedmann, L. Org. Sgnth. 19, 12 (1968). Wilcox, Jr., C.F.; Granthan, G.D. Tetrahedron 9;, 2889 (1975). Gunter, M.J.; Mander, L.N. J. Org. Chem. 49, 4792 (1981). Young, R.; Chang, C.K. J. Am; Chem. Soc. 1 7, 898 (1985). Yoshida, M.; Kadokura, A.; Minabe, M.; Suzuki, K. Tetrahedron ii. 2237 (1979). Stille, J.K.; Foster, R.T. J. Org. Chem. 99, 2703 (1963). Nelson, S.M.; Rodgers, J. Inorg, Chem. 9, 1390 (1967). Kondylis, M.; Chang, C.K. J. Am. Chem. Soc., in press. Jahnke, R.; Schonborn, M.; Zimmerman, G. Top. Cmrr. Chem. fil: 133 (1976). Zagal, J.B.; Sen, R.K.; Yeager, E.J. Electroanal. Chem. 83, 207 (1977). Behret, R.; Clauberg, W.; Sandstede, G. Ber. Bunsenges. Phyg. Chem. 9;, 54 (1977). Albery, W.J.; Hitchman, M.L. in "Ring-Disk Electrodes," Clarenden Press:0xford (1971). Thirsk, H.R.; Harrison, J.A. "A Guide to the Study of Electrode Kinetics," Academic Press:New York (1972). Brown, A.; Koval, C.; Anson, F.C. J. Electroanal. Chem. 19, 379 (1976). Brown, A.P.; Anson, F.C. J. filectroangl. Chem. 19, 203 (1977). Damjanovic, A.; Genshaw, M.A.; Bockris, J.O’M. ;; Chem. Phys. 39, 4057 (1966). Evans, J.F.; Kuwana, T. Anal. Chem. 19, 1632 (1977) and references cited therein. 83. 84. 85. 86. 87. 88. 89. 90. 91. 93. 94. 95. 96. 97. 98. -207- Liu, R.Y.; Weaver, M.J.; Wang, C.B.; Chang, C.K. 99 Electromnal. Chem: Interfacial Electrochgm. 145, 439 (1983). Levich, V.G. "Physicochemical Hydrodynamics" Prentice-Hallenglewood Cliffs, NJ (1962). Koutecky, J.; Levich, V.G. Zh. Fig. Khim. 2, 1565 (1956) and reference 84, pp. 345-357. Durand, Jr., R.B.; Anson, F.C. J. Electroanal. Chemm Interfacigl Electrochem. 134, 273 (1982). a) Fujihara, M.; Sunakawa, K.; Osu, T.; Kuana, T. 99 Electroanal. Chem: Interfacial Electrochem. 99, 299 (1978).; b) Bettelheim, A.; Kuwana, T. Anal. Chem. 9;, 2257 (1979); c) Forshey, P.; Kuwana, T. Inorg. ghgm. 99, 693 (1981) 99, 699 (1983); d) Zagal, J.; Bindra, P.; Yeager, E. J. Electroanal. Chem. lnterfgcial Electrocggm. 991, 1506 (1980). Davies, C.; Sutin, N.; Watkins, K.O. J. Am. Chemm Soc. 224 1892 (1970). Ni, C.—L.; Anson, F.C., private communication. Abdalmuhdi, I.; Chang, C.K.; Kondylis, M. J. Am. Chem. Soc., in press. Collman, J.P.; Denisevich, P.; Konai, Y.; Morrocco, M.; Koval, C.; Anson, F.C. J. Am. Chem; Soc. 102, 607 (1980). Gouterman, M. J. Mol. Spectrosc. 9, 138 (1961). Ichimura, K. ChemmiLett., 641 (1977). Pain III, J.B.; Dolphin, D.; Gouterman, M. Can. J. Chem. 99, 1712 (1978). Kagan, N.E.; Manzerall, D.; Merrifield, R.B. J. Am. Chem. Soc. 99, 5484 (1977). Wasielewski, M.R.; Svec, W.A.; Cope, B.T. J. Am. Chem. Soc. 100, 1961 (1977). Chang, C.K. in "ACS Advances in Chemistry Series," (R. Bruce King, ed.), no. 173 (1979). El-Bayoumi, M.A.; Rawls, H.R.; Kasha, M. Emre Appl. Chem. ll, 371 (1965). 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. -208- Gouterman, M.; Holten, D.; Lieberman, E. Chem. Phys. 29. 139 (1977). Boxer, S.G.; Gloss, G.L. J. Am. Chem. Soc. 99, 5406 (1976). Wasielewski, M.R.; Smith, U.H.; Cope, B.T. J. Am. Chem. Soc. 99, 4172 (1977). Abraham, R.J.; Burbidge, P.A.; Jackson, A.H.; MacDonald, D. . J. Chemm_Soc. 9, 620 (1966). a) Abraham, R.J.; Eivazi, F.; Pearson, H.; Smith, K.M. J. Chem. Soc., Chgm. Commun., 698, 699 (1976); b) Abraham, R.J.; Barnet, G.H.; Bretschneider, B.S.; Smith, K.M. Tetrgmgdron 99, 553 (1973); c) Abraham, R.J.; Eirazi, F.; Pearson, R.; Smith, K.M. Tetrmhedron 99, 2277 (1977). Hudson, M.F.; Smith, K.M. Tetrghegron 99, 3077 (1975). Subramanian, J. in "Porphyrins and Metalloporphyrins", (K.M. Smith, ed.), Elsevieerew York, pp. 560-571 (1975). Kokoszka, C.; Duerst, R.W. Coord. Chemvaev. 9, 209 (1970). Sykes, A.G. Frog. Inorg. Chem. 13, 1 (1970). Weil, J.A.; Kinnraid, J.K. J. Phyg. Chem. 9;, 3341 (1967). Chasteen, N.D.; Belford, R.L. Inorgm Chgm. 9, 169 (1970). Belford, R.L.; Chasteen, N.D.; So, H.; Tapscott, R.B. J. Am. Chem. Soc. 9;, 4675 (1969). Boas, J.F.; Pilbrow, J.B.; Smith, T.D. J. Chem. Soc. A, 723 (1969). Kokoszka, J.F.; Linzer, M.; Gordon, G. Inorg. Chem. 1, 1730 (1968). Wasson, J.B.; Shyr, C.-I.; Trapp, C. Inorg, Chem. 1, 469 (1968). Blumberg, W.E.; Peisach, J. J. Biol. Chem. 240, 870 (1965). Boas, J.F.; Pilbrow, J.B.; Smith, T.D. J. Chem. Soc. A, 721 (1969). -209- 116. Boyed, P.D.; Smith, T.D.; Price, J.K.; Pilbrow, J.R. J. Chem; Phyg. 99, 1253 (1972). 117. Chang, C.K.; Ward, B.; Wang, C.B. J.Amm, Chgm. Soc. 103, 5236 (1981). 118. Chang, C.K. in "Biochemical and Chemical Aspects of Oxygen," (W.S. Caughey, ed.), Academic PresszNew York, pp. 437-454 (1979). 119. Eaton, 8.8.; More, K.M.; Sawant, B.M.; Eaton, G.R. J2 Am. Chem. Soc. 105, 6560 (1983). 120. Chang, C.K.; Eaton, S.S.; Eaton, G.R. J. AmmmChemL Soc. 07, 3177 (1985). 121. Smith, T.D.; Pilbrow, J.R. Coord. Chem. Rev. 99, 173 (1974). 122. Batad Soc. a, M.R.; Tulinsky, A.; Chang, C.K. J. Am. Chem. 103, 5623 (1981). 123. Collman, J.P.; Chong, A.D.; Jameson, C.B.; Oakley, R.T.; Rose, 8.; Schmittou, B.R.; Ibers, J.A. J. Am. Chem. Soc. 516 (1981). 124. Chang, Soc. C.K.; Tulinsky, A.; Hatada, M.H. J. Am. Chemm 02, 7115 (1980). 125. Slip angle = sin‘1 [(magnitude of slip)/(meta1-meta1 distance)]. 126. Jones, T.A. in "Computational Crystallography," (D. Sayer, ed.), Clarendon Press:0xford, pp. 303—317 (1982).