Transition Metal Complexes of Porphyrin Analogs And

TRANSITION METAL COMPLEXES OF PORPHYRIN ANALOGS

AND BORATE-BASED COORDINATION COMPLEXES

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Anıl Çetin

May, 2007

TRANSITION METAL COMPLEXES OF PORPHYRIN ANALOGS

AND BORATE-BASED COORDINATION COMPLEXES

Anıl Çetin

Dissertation

Approved: Accepted:

Advisor Department Chair Christopher J. Ziegler Kim C. Calvo

Committee Member Dean of the College Claire A. Tessier Ronald F. Levant

Committee Member Dean of the Graduate School David A. Modarelli George R. Newkome

Committee Member Date Wiley J. Youngs

Committee Member Rex D. Ramsier

ABSTRACT

The synthesis of low-coordinate metal ions has been a focus of bioinorganic

chemists due to their important roles in active sites in enzymes and protein. Although the

isolation of these types of complexes is challenging, porphyrin analogs with one or two

carbon atoms in the interior position can be good candidates for generating protected low

coordinate metal sites. The metal coordination of one or two carbon substituted

hemiporphyrazines, namely monocarbahemiporphyrazine and dicarbahemiporphyrazine, was investigated. These porphyrin analogs, in which one or two of the central metal binding nitrogen atoms were replaced with C-H groups, were synthesized in the early

1950s by Linstead and co-workers, but their metal binding chemistry remained unexplored. Several low coordinate metal complexes of dicarbahemiporphyrazine, namely silver, copper, manganese, iron and cobalt were synthesized. Three different cobalt complexes of monocarbahemiporphyrazine in +2 and +3 oxidation states were also synthesized.

Porpholactone is another example of a ring modified porphyrin isomer. In this macrocycle one of the four pyrrollic units is oxidized to an oxazolone ring. Metal complexes of porpholactone may be novel catalysts for epoxidation of alkenes. The synthesis and X-ray crystal structure of first manganese complex of the porpholactone

5,10,15,20-tetraphenylporpholactone are reported. The catalytic activity of the complex

iii by using a variety of substrates was explored and it was compared with that of

manganese tetraphenylporphyrin.

Metal complexes of a sulfur containing borate-based chelating ligand,

- tris(imazolyl)borate, HB(mt)3 were examined. Three different modes of interaction were

observed with divalent closed d shell metal metal cations: Ca(II), Ba(II) and Hg(II). This

study exhibits the diversity of binding not typically observed in the scorpionate family.

- With group II and group XII metals, the HB(mt)3 ligand can act as a non-coordinating anion, can engage in B-H agostic bonding, and can form metal cluster compounds.

Molecular organization of guest molecules within nanometer-sized structures is a big challenge in terms of controlling the physical properties and chemical reactivities. In our group, we are investigating borate-based coordination polymer, lead tetrakis(imidazolyl)borate, to organize and sequester anionic guests. A variety of anionic guests were organized within the layers lead(II) borate scaffolds. The possibility of topochemical polymerization of these pre-organized anionic monomers in the crystalline

state was explored.

DEDICATION

To my parents Nurhan and Güney Çetin…

Bu tez annem Nurhan Çetin ve babam Güney Çetin’e adanmıştır. Dünyaya çocuk getirmenin sorumluluğunu en iyi bilen ve şu an bu tezi yaziyor olmamın yegane nedeni olan iki insana…Hayatta başıma ne gelirse gelsin, dünyanın neresinde olursam olayım arkamda o iki insanin olduğunu bilmenin güvenini bana yaşatan anneme ve babama…

Size bütün bunlar için ve bugüne kadar bana gösterdiğiniz sevgi, özveri, çaba ve sabır için teşekkür etmenin ne kadar yetersiz olduğunun farkındayım. Umarim size layık bir evlat olabiliyorumdur.

ACKNOWLEDGEMENTS

The past five years of my life has not been easy; being miles away from my beloved ones, expressing myself in a foreign language, trying to adapt to a new culture and of course being a graduate student in chemistry department. But what I am sure is if I had another chance to return to five years back from now, I would again choose to come here.

And it is because of the things I have learned through my graduate work, the wonderful people I have met and the countless memories I have shared. I am thankful to you all.

First of all, I would like to thank my advisor, Dr. Christopher Ziegler. There is no need to mention the impossibility of writing this thesis without you. Thanks for the support, the guidance and at the same time the freedom you have given me. Thanks for believing in me even at the times that I did not believe in myself.

Many thanks to my parents, Nurhan and Güney Çetin, brother Mert and sister-in-law

Özün and my grandparents for the love, guidance and support. Thanks for being there whenever I need someone, thanks for giving me the comfort of having people who care about me and of course traveling that many miles just to visit me. It is wonderful to have you all.

To my former research lab mates Janet Shaw, John Harvey, Barton Hamilton, and

Tang Ding, thank you for the helpful discussions and the memories we have shared.

Thanks for making me feel like I am home, helping me to find my way in the laboratory

vi and helping me to improve my language skills. Also thanks to the new group members

Saovalak Sripothongnak, Natalie Barone and Roshinee Costa for the friendship and for your friendly support.

Finally, thank you Semih. You are the reason why I am here today; it is your love that brought me here to America. Your love, patience and support helped me go through all the problems I have gone through these past years. And now another chapter of our lives is ending and we are getting ready for a new start. I am not sure where our next stop is going to be or what we are going to be doing next year these times, but what I am sure is that we are going to be together.

vii

TABLE OF CONTENTS

Page

LIST OF TABLES………………………………………..…………………..…….…...... x

LIST OF FIGURES……………………...…………………………………………...….xii

LIST OF SCHEMES…………………...………………………..…………………...….xx

LIST OF ABBREVIATIONS…………...…………………………..……………….…xxi

CHAPTER

I. INTRODUCTION TO ORGANOMETALLIC CHEMISTRY OF PORPHYRINOIDS IN EQUATORIAL PLATFORM…………………..…..………..1

N-confused porphyrins…..…………..……………………………….…...... 7

σ-type M-C bonding...... ………………………10

Divalent metal complexes of NCP with M-C bonds...... 10

Trivalent metal complexes of NCP………....…………...... 23

Side-on M···H-C interaction...... 33

Monomeric metal complexes of NCP...... 34

Dimeric metal complexes of NCP...... 38

Carbaporphyrinoids...... 48

II. METAL MEDIATED C-H BOND ACTIVATION IN A CARBON SUBSTITUTED HEMIPORPHYRAZINE...... ……………...….….....….88

Experimental...... ………………………...90

viii

Results and Discussion ...... ……………………………94

III. COMPLEXES OF LOW-COORDINATE MIDDLE TRANSITION METAL DICARBAHEMIPORPHYRAZINE...... 113

Experimental...... ………………………..114

Results and Discussion ...... …………..………………117

IV. COORDINATIVE Co(II) AND Co(III) COMPLEXES OF MONOCARBAHEMIPORPHYRAZINE...... 127

Experimental...... ………………………..128

Results and Discussion ...... …………..………………131

V. STRUCTURE AND CATALYTIC ACTIVITY OF A MANGANESE(III) TETRAPHENYLPORPHOLACTONE...... 140

Experimental...... ………………………..142

Results and Discussion ...... …………..………………145

VI. COMPLEXES SUMMARY OF THE COORDINATION CHEMISTRY OF PORPHYRIN ANALOGS...... 151

VII. COORDINATIVE FLEXIBILITY IN HYDROTRIS(IMAZOLYL)BORATE DIVALENT METAL COMPOUNDS...... 154

Experimental...... ………………………..156

Results and Discussion ...... …………..………………160

VIII. INVESTIGATING THE TOPOCHEMICAL POLYMERIZATION OF ANILINE DERIVATIVES IN Pb(II) BORATE SCAFFOLDS...... 167

Experimental...... ………………………..170

Results and Discussion ...... …………..………………176

REFERENCES...………………………………...…………………….………………188

LIST OF TABLES

Table Page 2.1 Crystal data and structure refinement for dchp·HCOOH……...... 95

2.2 Crystal data and structure refinement for [dchp]2·H2O.……………...... 96

2.3 Crystal data and structure refinement for dchp·2CH3CN.…………………. 97

2.4 Crystal data and structure refinement for dchp·2py...... 98

2.5 Crystal data and structure refinement for dchp·DMF ……………………... 99

2.6 Crystal data and structure refinement for Ag(dchp)pyNO3·2H2O...... 105

2.7 Crystal data and structure refinement for Cu(dchp-py)py·3py …...... 107

2.8 Crystal data and structure refinement for 2[Cu(dchp-py)]·5H2O ………… 108

2.9 Crystal data and structure refinement for (dchp-py)Cl3…………………… 111

3.1 Crystal data and structure refinement for Mn(dchp)py·3py ……………… 119

3.2 Crystal data and structure refinement for Co(dchp)py·py ………………… 120

3.3 Crystal data and structure refinement for Fe(dchp)py·py ………………… 124

4.1 Crystal data and structure refinement for Co(mchp)py..…………….…….. 132

4.2 Crystal data and structure refinement for Co(mchp)py2..………………….. 135

4.3 Crystal data and structure refinement for Co(mchp-OH)py2……………… 136

5.1 Crystal data and structure refinement for Mn(TPPL)Cl…………………… 148

5.2 Comparison of Mn(TPPL)Cl and Mn(TPP)Cl as catalysts for epoxidation reactions of substrates: 1-hexene, cyclohexene, styrene, 5-hexen-1-ol and 4-penten-1-ol……………………………………………………………….. 151

x 7.1 Crystal data and structure refinement for [Ca(Tm)2]·6H2O ………………. 159

7.2 Crystal data and structure refinement for Ba(Tm)2(H2O)2………………… 161

7.3 Crystal data and structure refinement for Hg4(Tm)4Cl4·13H2O ………….. 163

8.1 Crystal data and structure refinement for Pb[B(Im)4](3-

NH2C6H4COO)·(H2O)…………………………………………………….. 173

8.2 Crystal data and structure refinement for Pb[B(Im)4](4- NH2C6H4COO)·(H2O)………………...... 174

8.3 Crystal data and structure refinement for Pb[B(Im)4](3- NH2C6H4SO3)·(H2O)………………...... 175

LIST OF FIGURES

Figure Page 1.1 Structures of (a) (TPP)Fe(Ph) (b) (TPP)Fe=CCl2 and (c) (TPP)Fe(CO)(py).……………………………………………………...... 2

1.2 Crystal structures of (a) [(η6-cymene)Ru(η5-Ni(OEP))] and (b) {Cp*Ru[η6- Ni(PcOBu)]} ……………...... 3

1.3 Crystal structure of meso- η1-palladioporphyrin…...... 4

1.4 The structure of Ni(II) and Fe(II) complexes of core modified porphyrin with an organometallic bond at the axial position...………………………… 5

1.5 Carbaporphyrin skeletons a. porphyrin b. N-confused porphyrin c. carbaporphyrin d. meta- benziporphyrin e. tropiporphyrin f. azuliporphyrin g. benzocarbaporphyrin……...... …….. 6

1.6 Structures of (a) N-confused hemiporphyrazine (b) Dicarbahemiporphyrazine (c) Monocarbahemiporphyrazine.…...... 7

1.7 Three possible tautomers of N-confused porphyrins.……………………...... 8

1.8 Synthesis of regular porphyrin and N-Confused porphyrin via condensation of pyrrole and aldehydes…...... 9

1.9 Coordination modes for monomeric complexes with M-C bonding...... 9

1.10 Divalent metal complexes of N-confused porphyrins with direct M-C bonds...... 10

1.11 Crystal structures of Ni(II) complexes (a) NCTTPand (b) bromoethyl substituted NCTPP...... 12

1.12 Crystal structures of (a) Pd(II) and (b) Pt(II) complexes of NCP...... 13

1.13 The crystal structure of copper(II) N-confused tetra(pentafluorophenyl) porphyrin, Cu(NCTFPP)...... 15

xii 1.14 Reversible protonation of Ni(II) and Cu(II) NCP...... 16

1.15 Reaction of NiNCP with monohaloalkanes in the presence of a base affording three products...... 17

1.16 Formation of 2,21-linked and 2,2'-linked dimeric NiNCP complexes via the 18 treatment of NiNCP with dihaloalkanes…......

1.17 Reversible deprotonation of the external nitrogen of the NiNCTTP...... 18

1.18 The crystal structures of (a) mono- and (b) dimethylated NiNCTPP…...... 19

1.19 Crystal structures of cyanide substituted NiNCP (a) Ni((C-CN)NCTTP) (b) Ni((C-CN)(C-OMe)NCTTP)…...... 20

II 1.20 Crystal structure of Mo NCTPP(pip)2 with 50% thermal ellipsoids……….. 22

1.21 Crystal structure of CoII(NCTPP)(py)·py with 50% thermal ellipsoids…….. 23

1.22 Crystal structure of NiIII(NCPO) py………………………………………… 24

1.23 Crystal structure of CuIIINCTPFPP…………………………………………. 25

1.24 Crystal structure of Ag(NCTPP)……………………………………………. 26

1.25 Crystal structures of (a) Mn(NCTPP)Br and (b) Mn(NCTPP)py2….………. 28

1.26 Crystal structures of (a) Fe(NCTPP)Br and (b) Fe(ONCTPP)Br…………… 29

III III 1.27 Crystal structures of (a) Co (NCTPP)H2O (b) Co (NCTPP)PPh3 (c) III Co (NCTPP)py2……………………………………………………………. 31

III 1.28 Crystal structure of Co (NCTPP)(1-MeIm)2 with 50% thermal ellipsoids…. 32

V V 1.29 Crystal structure of (a) Sb (NCTPP)(Br)2 and (b)Sb (NCTTP)(OCH3)2…... 33

1.30 Monomeric divalent metal complexes of NCP……………………………… 34

1.31 Crystal structure of (a) MnII(NCTPP)Br (b) MnII(NCTPP)py….………….... 37

1.32 Crystal structure of Yb(NCTPP)(LOMe). …………………………………… 38

1.33 Crystal structure of Zn4 NCTPP dimer……………………………………… 39

1.34 Schematic representations of dimeric NCP complexes…………………….... 40

xiii 1.35 Crystal structure of dimeric Zn(NCDPP)……………………………………. 40

1.36 Crystal structure of [Fe(NCTPP)]2….……………………………………….. 42

1.37 Crystal structure of dimeric Fe(NCTPP) complex bridged via µ-hydroxyl group and a sodium cation…………………………………………………… 42

1.38 Crystal structure of [Mn(NCTPP)2]2...………………………………………. 43

II II 1.39 Crystal structure [Mn (NCTPP)Mn (NCTPPH2)]………………………...... 43

1.40 Crystal structure of [Pd(NCTTP)]2……….………………………………….. 44

1.41 Crystal structure of (a) cis-Pt[NCTBuPP]2Cl (b) trans-Pt[NCTBuPP]2Cl...... 45

1.42 Crystal structure of cis-Pt(Ni(C-benzyl)NCTTP)2Cl.…………..…………… 46

1.43 Crystal structure of Rh2(NCTPP)……………………………………………. 47

1.44 Crystal structure of Re(CO)3(NFTPP)……………………………………….. 48

1.45 Synthesis of β-substituted m-benziporphyrin via 3+1 methodology………… 49

1.46 Tautomerization of m-benziporphyrin………………………………………. 49

1.47 Synthesis of meso-substituted m-benziporphyrin……………………………. 50

1.48 Divalent metal complexes of m-benziporphyrin with direct M-C bond……... 50

1.49 Metallation chemistry of acetoxybenziporphyrin.…...………………………. 51

1.50 Metallation of m-benziporphyrin with divalent metal ions forming organometallic side-on interactions………………………………………….. 52

1.51 Crystal structures of (a) Cd(II) m-benziporphyrin chloride and (b) Ni(II) m- benziporphyrin chloride…………………………………………….……….. 54

1.52 Crystal structure of Fe(II) tetraphenyl-m-benziporphyrin bromide…………. 55

1.53 Crystal structure of tetranuclear Cu-complex of tetraphenyl m- benziporphyrin….……………………………………………………………. 56

1.54 Synthesis of tetraphenyl p-benziporphyrin….……………………………….. 57

1.55 Crystal structure of free base tetraphenyl m-benziporphyrin….…………….. 57

xiv 1.56 Crystal structures of (a) Cd(II) tetraphenyl p-benziporphyrin chloride and (b) Ni(II) tetraphenyl p-benziporphyrin chloride….………………………… 58

1.57 Metallation of vacataporphyrin……………………………………………… 60

1.58 Synthesis of β-substituted 2-oxybenziporphyrin…………………………….. 61

1.59 Keto-enol tautomerization of 2-oxybenziporphyrin…………………………. 61

1.60 2-oxybenziporphyrins with fused ring systems……………………………… 62

1.61 Methoxybenziporphyrins and hydroxybenziporphyrin……………………… 63

1.62 Ni(II) complex of tetraphenyldimethoxybenziporphyrin……………………. 63

1.63 Palladium(II) oxybenziporphyrin……………………………………………. 64

1.64 Reversible protonation of Pd(II) oxybenziporphyrin.…………………...…... 65

1.65 Addition of various electrophiles to the oxygen and electrophilic addition to the internal carbon atom….………………………………………………….. 66

1.66 Reaction of oxybenziporphyrin with silver(I) acetate to generate Ag(III) oxybenziporphyrin…………………………………………………………… 67

1.67 Silver(III) oxynaphthiporphyrin.…...………………………………………... 68

1.68 Core modified oxybenziporphyrin…...……………………………………… 68

1.69 Palladium(II) complex of meso-disubstituted oxybenziporphyrin…....……... 69

1.70 Possible tautomerization of tropiporphyrin………………………………….. 70

1.71 Structure of Silver(III) tropiporphyrin……………………………………….. 71

1.72 Tautomerization of azuliporphyrin…………………………………………... 72

1.73 Crystal structure of Ni(II) azuliporphyrin….………………………………... 73

1.74 Crystal structure of tetrachlorophenyl azuliporphyrin….…………………… 74

1.75 Crystal structure of Cu(II) tetrphenylazuliporphyrin….…………………….. 75

1.76 Crystal structure of free base tetraphenyl benzocarbaporphyrin….………… 76

xv 1.77 Electron delocalization pathway in benzocarbaporphyrin…………………… 76

1.78 Electron delocalization pathway of dimethyl benzocarbaporphyrin….……... 78

1.79 Internal carbon halogenation and subsequent oxidation of benzocarbaporphyrin….……………………………………………………... 78

1.80 Crystal structure of Ag(III) benzocarbaporphyrin….………………………... 79

1.81 Electron delocalization pathway of core modified benzocarbaporphyrin…… 80

1.82 Crystal structure of Pd(II) benzocarbaporhyrin……………………………… 81

1.83 Metallation of S-confused porphyrin with divalent metal ions….…………... 82

1.84 Pyrrole-appended O-confused porphyrin….………………………………… 83

1.85 Crystal structure of Ni(II) complex of pyrrole appended O-confused porhyrin.……………………………………………………………………... 84

1.86 Crystal structure of Ag(III) pyrrole and ethoxy substituted O-confused porphyrin.……………………………………………………………………. 85

1.87 Crystal structure of Ag(III) tetratolylcarbaporpholactone…………………... 86

2.1 The structures of (a) hemiporphyrazine and (b) di-N-deficient analogue dicarbahemiporphyrazine (H2dchp)………………………………………….. 89

2.2 The structure of dication 1 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups and the NH groups…………………………………………………………… 100

2.3 The structure of [1]2·H2O with 50% thermal ellipsoids. Hydrogen atoms have been omitted for clarity………………………………………………… 100

2.4 The structure of 1·CH3CN with 50% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal NH groups... 101

2.5 The structure of 1·pyridine with 50% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal NH groups... 101

2.6 The structure of 1·DMF with 35% thermal ellipsoids………………………. 102

2.7 The structure of 2 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups……………. 106

xvi 2.8 The structure of 3a with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups……… 109

2.9 The structure of second crystal form of 3b with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity……………………………… 109

2.10 The structure of 4 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups and the NH groups……………………………………………………………………….. 112

3.1 The structures of (a) hemiporphyrazine and (b) mono-N deficient analogue monocarbahemiporphyrazine (H3mchp) (c) di-N deficient analogue dicarbahemiporphyrazine (H2dchp)…………………………………………. 114

3.2 Synthesis of compounds 2, 3 and 4 starting from 1…………………………. 118

3.3 The structure of 2 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups……………. 121

3.4 The structure of 3 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups…………….…. 121

3.5 X-band EPR spectra in frozen pyridine/toluene, 1 mM MnII(dchp)py, T = 100 K…………………………………………………………………….. 123

3.6 The structure of 4 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups and the NH groups……………………………………………………………………….. 125

4.1 Structures of (a) Normal porphyrin (b) m-benziporphyrin and (c) monocarbahemiporphyrazine………………………………………………... 128

4.2 The structure of Co(mchp)py with 30% thermal ellipsoids. The hydrogen atoms are omitted for clarity except the ones on the external nitrogen atoms………………………………………………………………………… 133

4.3 The structure of Co(mchp)py2 with 35% thermal ellipsoids. The hydrogen atoms are omitted for clarity except the one on the external nitrogen atom.... 137

4.4 The structure of Co(mchp-OH)py2 with 35% thermal ellipsoids. The hydrogen atoms are omitted for clarity………………………………………. 138

5.1 The structures of (a) Tetraphenylporphyrin (b) Tetraphenylporpholactone and (c) Tetraphenylchlorin………………………………………………….. 141

xvii 5.2 UV-Visible spectra of tetraphenylporphyrin (dotted line) and tetraphenylporpholactone (solid line)……………………………………….. 142

5.3 UV-Visible spectra of Mn(TPP)Cl (dotted line) and Mn(TPPL)Cl (solid line)………………………………………………………………………….. 147

5.4 The structure of Mn(TPPL)Cl with 50% thermal ellipsoids………………… 149

7.1 The structures of (a) Hydrotris(pyrazolyl)borate (b) Hydrotris(imazolyl)borate………………………………………...…………. 154

7.2 Typical geometries for Hydrotris(imazolyl)borate ligand (a) tetrahedral geometry for a 1:1 metal complex with an additional ligand X. (b) Typical octahedral geometry of a 2:1 metal complex ……………………………….. 155

1 7.3 H NMR of Na[HB(mt)3]……………………………………………………. 157

13 7.4 C NMR of Na[HB(mt)3]…………………………………………………… 157

7.5 The structure of Ca[HB(mt)3]2(H2O)6 with 50% thermal ellipsoids. H atoms are omitted for clarity………………………………………………. 162

7.6 The structure of Ba[HB(mt)3]2(H2O)2 with 50% thermal ellipsoids. H atoms are omitted for clarity……………………………………………… 162

7.7 The structure of Hg4[HB(mt)3]4Cl4 with 50% thermal ellipsoids. H atoms are omitted for clarity………………………………………………. 165

8.1 Structure of the borate ligand and the lead complex (a) Structure of - tetrakis(imidazolyl)borate anion, B(Im)4 . (b) The asymmetric unit of Pb[B(Im)4](NO3)(1.35·H2O).(c) The extended structure of Pb[B(Im)4](NO3)(1.35·H2O) along the c axis.………………………………. 169

8.2 The asymmetric unit with 50% thermal ellipsoids of [PbB(Im)4]-(3- NH2C6H4CO2)(H2O (1). The solvent water molecule is not shown…………. 177

8.3 Extended structure of 1 along the a-axis. H-atoms are omitted for clarity….. 178

8.4 The asymmetric unit with 50% thermal ellipsoids of [PbB(Im)4]-(4- NH2C6H4CO2)(H2O) (2). The solvent water molecule is not shown………… 178

8.5 Extended structure of 2 along the a-axis. H-atoms are omitted for clarity….. 179

xviii 8.6 The asymmetric unit with 50% thermal ellipsoids of [PbB(Im)4](3- NH2C6H4SO3)(H2O) (3). The solvent water molecule is not shown………… 179

8.7 Extended structure of 3 along the a-axis. H-atoms are omitted for clarity….. 180

13 8.8 C SSNMR CP/MAS, cp time of 1 ms of (a) Pb[B(Im)4](NO3) (b) 4-amino benzoic acid (c) [PbB(Im)4](4-NH2C6H4CO2)(H2O) (d) [PbB(Im)4](4- NH2C6H4CO2)(H2O) heated for 1 day. (e) [PbB(Im)4](4- NH2C6H4CO2)(H2O) heated for 1 week…………………………………….. 182

13 8.9 C SSNMR CP/MAS, cp time of 2 ms of (a) Pb[B(Im)4](NO3) (b) 3-amino benzoic acid (c) [PbB(Im)4](3-NH2C6H4CO2)(H2O) (d) [PbB(Im)4](3- NH2C6H4CO2)(H2O) heated for 1 day. (e) [PbB(Im)4](3- NH2C6H4CO2)(H2O) heated for 2 days. (f) [PbB(Im)4](3- NH2C6H4CO2)(H2O) heated for 1 week…………………………………….. 183

13 8.10 C SSNMR CP/MAS, cp time of 3 ms of (a) PbBIm4 (b) 3-aniline sulfonic acid (c) [PbB(Im)4](3-NH2C6H4SO3)(H2O) (d) [PbB(Im)4](3- NH2C6H4SO3)(H2O) heated for 1 day. (e) [PbB(Im)4](3- NH2C6H4SO3)(H2O) heated for 2 days. (f) [PbB(Im)4](3- NH2C6H4SO3)(H2O) heated for 1 week……………………………………… 184

xix

LIST OF SCHEMES

Scheme Page 2.1 Synthesis of dicarbahemiporphyrazine (H2dchp),1..………………………… 91

2.2 Synthesis of compounds 2, 3, and 4, starting from 1.Reagents: i) AgNO3, I II pyridine, MeOH, air; ii) Cu or Cu salt, pyridine, MeOH, air; c) CH2Cl2, air.……………...... 103

4.1 Synthesis of monocarbahemiporphyrazine (H3mchp).………..…………...… 130

4.2 Synthesis of Co(mchp)py 8; Co(mchp)py2 9 and Co(mchp-OH)py2 10...... … 134

5.1 Synthesis of free base tetraphenylporpholactone through dihydroxychlorin.……...... ……………………………... 144

5.2 Metallation of free base tetraphenylporpholactone with MnCl2.…...... 145

LIST OF ABBREVIATIONS

H2NCTPP - 5,10,15,20-tetraphenyl-2-aza-21-carba-porphyrin

NCTPP - anion of 5,10,15,20-tetraphenyl-2-aza-21-carba-porphyrin

NCTTP – anion of 5,10,15,20-tetratolyl-2-aza-21-carba-porphyrin

NCTPFPP - anion of 5,10,15,20-tetrapentafluorophenyl-2-aza-21-carba-porphyrin

NCDPP - anion of 5,10-diphenyl-2-aza-21-carba-porphyrin

NCTBuPP – anion of 5,10,15,20-tetrat(4’-tert-butyl)phenyl-2-aza-21-carba-porphyrin

NFTPP - N-fused tetraphenylporphyrin

NCP - N-confused porphyrin

TPP - 5,10,15,20-tetraphenylporphyrin

H2TPPL - 5,10,15,20-tetraphenylporpholactone

TPPL – anion of 5,10,15,20-tetraphenylporpholactone py – pyridine

EPR - Electron Paramagnetic Resonance

HFEPR - High Frequency and Field EPR

EA – Elemental Analysis

DFT – Density functional theory

DDQ - 2,3-Dichloro-5,6-dicyano-1,4-benzoquinone

1-MeIm - 1-methylimidazole

xxi zfs - zero-field splitting

H2dchp – Dicarbahemiporphyrazine

H3mchp – Monocarbahemiporphyrazine

ONSH - Oxidative nucleophilic subsitution of hydrogen

HB(pz)3 – Hydrotris(pyrazolyl)borate anion

HB(mt)3 – Hydrotris(imazolyl)borate anion

B(Im)4 – Tetrakis(imidazolyl)borate

MOF – Metal-organic framework

13C-SSNMR - 13C solid state NMR

CP\MAS – cross-polarization magic angle spinning

HMB - hexamethylbenzene

SQUID – Superconducting Quantum Interference Device

FT-IR – Fourier transform infra-red spectroscopy

xxii

CHAPTER I

INTRODUCTION TO ORGANOMETALLIC CHEMISTRY OF PORPHYRINOIDS IN

THE EQUATORIAL PLATFORM

Porphyrins have long been a major research interest in several areas of chemistry.1

Nature uses porphyrins and metalloporphyrins in many biological systems ranging from oxygen transport to light harvesting and redox chemistry to the production of energy

2 from O2. Porphyrins are also remarkable for their extreme stability and ability to coordinate to almost all of the metal ions found in the periodic table.3 Along with those

attributes, their unusual electronic properties make them candidates for applications such

as photodynamic therapy agents,4 organic catalysts5 and advanced materials.6 As a result

porphyrins are some of the most heavily studied molecules in the chemical literature.1

Although the organometallic chemistry of normal porphyrins has an important place

in metalloporphyrin chemistry, there are not many examples of these types of complexes.

The axial organometallic bonds in metalloporphyrins have been investigated by several

groups due to their biological importance (Figure 1.1a).7 The iron porphyrin complexes

with bound alkyl or aryl groups are relevant to the chemistry of several heme enzymes

such as halocarbon deactivation of cytochrome P-450. One of the several ways of

generating σ-phenyl or σ-alkyliron porphyrins is through electroreductive alkylation or

arylation of iron in porphyrin complexes.8 In this method, the axial organometallic bonds

1 are formed via the reaction of alkyl halides with the electrogenerated Fe(I) complexes.

These types of bonds can also be generated via carbanion transfer from Grignard reagents

to Fe(III) porphyrins.9

II The reaction of Fe TPP with CCl4 in the presence of excess reducing agent generates

10 a (TPP)Fe=CCl2 complex with axial iron carbon double bond (Figure 1.1b). Carbon

monoxide complexes, such as the six-coordinate (TPP)Fe(CO)(py) has been synthesized

and characterized by single crystal X-ray crystallography (Figure 1.1c).11 This complex

is a small-molecule binding model for CO to heme cofactors in proteins. Elucidation of

the crystal structure showed that the carbonyl unit binds to the Fe(II) center in a linear

fashion.

Cl Cl Ph CO Ph C Ph N Ph N Ph N Ph N Fe N Fe N Fe N N N Ph N Ph N Ph N Ph N Ph a. Ph b. c.

Figure 1.1 Structures of (a) (TPP)Fe(Ph)8,9

10 8,11 (b) (TPP)Fe=CCl2 and (c) (TPP)Fe(CO)(py).

There are also a few examples of both σ- and π-type organometallic bonds at the

periphery of the porphyrins. The generation of these complexes is a comperatively new

area of research. A π-type organometallic interaction of porphyrin was discovered in

1996 by Rauchfuss et al.12 These complexes were generated from the reaction of

Ni(OEP) with η6-cymene Ru(II) with π-complexation of the pyrrolic unit with the metal

2 center (Figure 1.2a). Similar chemistry had also been observed with phthalocyanines and

metallophthalocyanines.13 But in the case of the phthalocyanines, the ruthenium(II)

center was coordinated to the benzene ring of the indolene unit (Figure 1.2b). Peripheral

σ-type organometallic bonds were generated via the reaction of bromo-substituted porphyrins with palladium(0) or platinum(0) precursors (Figure 1.3).14 These resulting

complexes, meso-η1-palladio- or platino-porphyrins, are believed to be the intermediate

species in the process of Pd(II) catalyzed carbon-halogen bond formation.

Figure 1.2 Crystal structures of (a) [(η6-cymene)Ru(η5-Ni(OEP))]12

and (b) {Cp*Ru[η6-Ni(PcOBu)]}13

3 In contrast to these normal porphyrin rings, several modifications have been made both in the interior.15,16,17,18,19 and exterior20 positions of the porphyrin macrocycles in order to investigate the potentially interesting electronic and coordination properties of the resulting porphyrinoids. One possible modification is the replacement of one or more internal nitrogen atoms with heteroatom(s) such as oxygen, sulfur, selenium, thus generating core-modified porphyrins.21

Core-modified porphyrins were first prepared in 1971.21 Although their coordination chemistry has not been investigated comprehensively, several studies on the metallation chemistry have been reported. Replacing one of the pyrrolic nitrogens of the porphyrin ring by a heteroatom converts the dianionic ligand to monoanionic which gives the system the ability to stabilize low valent metal ions. Therefore, the resulting coordination complex of this ligand with divalent metal ions requires an axial anionic ligand for charge balance typically a halide. Treatment of these complexes by Grignard reagents or lithium alkyl derivatives results in formation of organometallic bonds, although these complexes

Figure 1.3 Crystal structure of meso- η1-palladioporphyrin.14

4 are not very stable under aerobic conditions. This type of bonding is typified by the nickel(II)22,23 and iron(II) complexes (Figure 1.4).24 The +3 oxidation states of the same metal ions are stabilized by hexacoordination of the metal centers with two anions for charge balance.

Alternatively, integrating C-H containing ring systems in place of one or more pyrrolic units generates carbaporphyrinoids.16,17,18,19 One of the most intriguing aspects of these carbon substituted ring systems is their ability to form unusual organometallic bonds both in the axial and the equatorial platforms.25,26 The equatorial organometallic interactions will be the only focus in this chapter. The best studied example of this kind of porphyrinoid is N-confused porphyrin, also called inverted porphyrin.16,25 Since the discovery of N-confused porphyrins in 1994,27,28 more than one hundred reports have been published on this ligand. Because N-confused porphyrins opened up a new field in porphyrin chemistry, several research groups turned their attention to the design of new carbapophyrinoids. True carbaporphyrins, benziporphyrins, tropiporhyrins, azuliporphyrins and benzocarbaporphyrins are a few of the many examples of the carbaporphyrinoids (Figure 1.5).

N M X

X = O, S, Se M = Ni, Fe

Figure 1.4 The structure of Ni(II) and Fe(II) complexes of core modified porphyrin with

an organometallic bond at the axial position. 5 N H N N H N a.

N N N H H N HC NH N N HC H H N N N b. c. d.

N N N H H N NH HC N HC H H N N N

e. f. g.

Figure 1.5 Carbaporphyrin skeletons a. porphyrin b. N-confused porphyrin

c. carbaporphyrin d. meta- benziporphyrin e. tropiporphyrin

f. azuliporphyrin g. benzocarbaporphyrin

Even though the study of carbaporphyrinoids accelerated after the discovery of N-

confused porphyrins, the idea of introducing C-H units into the core of porphyrin-like

molecules was first considered fifty years ago. Linstead and coworkers first synthesized

the N-confused hemiporphyrazines (Figure 1.6a) and carbahemiprophyrazines in the

1950s (Figure 1.6b).29 These molecules are carbon substituted phthalocyanines.30

However, little work has been done with regard to the metal coordination properties of these potentially interesting ligands.

N N N N N N N H NH HN NH HN NH HN

N N N N N N

a. N b. c.

Figure 1.6 Structures of (a) N-confused hemiporphyrazine

(b) Dicarbahemiporphyrazine (c) Monocarbahemiporphyrazine.29

The equatorial organometallic chemistry of carbaporphyrinoids will be detailed in

this chapter and the subject will be summarized under two major divisions: the

coordination chemistry of N-confused porphyrins, and the metal complexes of the

remaining of the carbaporphyrinoids. In each section the character of the equatorial

organometallic bonding, through either direct M-C bonds or agostic type side-on

interactions, will be highlighted.

N-confused porphyrins

As introduced in the earlier section, N-confused porphyrins or so-called inverted

porphyrins are the most extensively studied carbaporphyrinoid systems.16,18,25,31 It is important to note that unlike many carbaporphyrinoids, N-confused porphyrins (NCP) show porphyrin-like aromaticity in their 18-electron π-system. But at the same time, unlike the regular porphyrins, NCPs can act as neutral, anionic, dianionic and also as

7 N N N H

N HC NH HC NH H C N NH 2 N H N N N

Figure 1.7 Three possible tautomers of N-confused porphyrins.

trianionic ligands due to the availability of three different tautomers, which provide the ability to stabilize several different oxidation states (Figure 1.7). Furthermore, the peripheral nitrogen atom can also engage into metal binding. As a result, these rings have the potential of forming interesting monomeric, dimeric and sometimes even tetrameric metal complexes.

NCP was first isolated as a byproduct of regular porphyrin in 1994 by two different groups simultaneously.27,28 Rothemund type condensations of pyrrole and aldehydes

yield meso-tetrasubstituted porphyrin as well as the N-confused porphyrin, with the

inverted pyrrole unit, as a small byproduct. As one might expect, the yield of the

macrocycle was extremely low. However, in 1999, Lindsey and Geier improved the

conditions for NCP synthesis to up to 39% yield, which provides a route for gram-scale

quantities of NCP (Figure1.8).32

Along with the synthesis of meso-substituted NCPs, the β-substituted NCPs were

synthesized because the peripheral modifications could have effect on the stabilization of

organometallic bonds. Dolphin and co-workers employed a “2+2” MacDonald type

condensation of two dipyrrolic units in order to generate the desired product.33 In the

8 Ph Ph Ph Ph O N N N H H 1) CH2Cl2 , MSA + NH NH N HC + N 2) DDQ, Et3N H N N

Ph Ph Ph Ph Figure 1.8 Synthesis of regular porphyrin and N-Confused porphyrin via condensation of

pyrrole and aldehydes.32

following years, an improved yield synthesis using a “3+1” methodology via the

condensation of the tripyrrolic unit with the subsequent dialdehyde was also reported.34

In the following section, the organometallic complexes of the N-confused porphyrin will be summarized under two sections: formation of σ-type bonding as a result of C-H bond activation and side-on interaction of the metal and the C-H unit. Some of the possible monomeric coordination modes of N-confused porphyrin are shown in Figure

1.9.

Ar Ar Ar Ar N N R NH N N M N M

N N

Ar Ar Ar Ar

Ar Ar Ar Ar N N X R N N N M N M

N N

Ar Ar Ar Ar

Figure 1.9 Coordination modes for monomeric complexes with M-C bonding.

9 σ-type M-C Bonding

Insertion of certain metal ions into the coordination core of the N-confused

porphyrins results in the activation of the C-H bond, forming a direct metal-carbon bond.

Because the internal carbon is deprotonated, it is expected the porphyrin will have

carbanionic character. In following years, Ghosh showed that the internal carbon atom in

fact exhibits singlet carbene type properties.35 These complexes often have planar geometries with direct M-C bonds. This type of bonding is observed both with divalent and trivalent metal ions.

Ar Ar N H II N N M

Ar Ar

M = Ni, Pd, Pt, Cu

Figure 1.10 Divalent metal complexes of N-confused porphyrins

with direct M-C bonds.

Divalent metal complexes of NCP with M-C bonds

The first metal complex of N-confused porphyrin, nickel(II) N-confused

tetratolylporphyrin (NiIINCTTP), was reported along with the synthesis of the macrocycle.28 The synthesis involved the activation of the C-H bond. Nickel(II) ion was

10 inserted into the macrocycle under relatively mild conditions and was characterized by spectroscopy as well as crystallography. The macrocycle serves as a dianionic ligand coordinating to the nickel(II) ion through three nitrogen atoms and one internal carbon atom (Figure 1.11a). Deprotonation of the internal carbon atom to form σ-type bonding was initially verified by 1H NMR spectra. The 1H NMR spectra of the metal complex

also showed that the resulting metal complex had a similar pattern to the externally

protonated tautomer of the free base, which indicated the preservation of the proton on

the external nitrogen. The crystal structure of the system verified the planar geometry of

the complex as well as the loss of the internal proton, leaving the internal carbon atom sp2 hybrized. The nickel(II) ion resides in the middle of the macrocyclic cavity with average

Ni-C(N) distance of 1.955(2) Å to 1.963(2) Å. The exact Ni-C bond distance could not be determined because the nitrogen atoms and the carbon atom are undistinguishable from each other due to the disorder. In following years the crystal structure of the Ni(II) bromoethyl substituted N-confused porphyrin was elucidated, which was used to determine the exact metal carbon bond length (Figure 1.11b).36 The corresponding crystal structure has a slightly saddled shape in contrast to the unsubstituted NiNCP but the Ni-N bond lengths observed are similar to that of Ni(OEP) and the Ni-C bond is

1.906(4) Å which is within the limits of previous Ni-C bonds reported.

11 a. b.

Figure 1.11 Crystal structures of Ni(II) complexes (a) NCTTP28 and

(b) bromoethyl substituted NCTPP.36

The group ten metal ions, palladium(II)37 and platinum(II)38 also formed similar structures as the Ni(II) complex under similar conditions. The absorption spectra of the resulting Pd(II) species is very similar to that of the Ni(II) complex with 15 – 20 nm red shifts in all peaks. Although the structures of both complexes were determined by single crystal X-ray crystallography, the exact M-C bond lengths were not determined due to the disorder of the M-C and M-N bonds as in the case of the Ni(II) complex (Figure 1.12).

12 N(C)

N(C)

N(C N(C)

N(C) N(C)

Figure 1.12 Crystal structures of (a) Pd(II)37 and (b) Pt(II)38 complexes of NCP.

Copper metal also coordinated to the macrocycle in the same way as the group 10 metals ions and formed an isostructural complex.39 However this complex has more importance because it is the first example of a stable monomeric organometallic Cu(II) complex. It is also important to note that the metallation reactions were run under anaerobic conditions otherwise decomposition of the porphyrinoid was observed and the degradation product, tripyrrinone copper(II) complex, was isolated. The study on this

13 complex was detailed in a separate report subsequently.40 Although CuII(NCTPP) was not characterized by crystallography, the spectrum of the complex was similar to the externally protonated tautomer of the free base indicating that the outer hydrogen was still present.

Later, a detailed EPR study of CuIINCTPP was reported by Schweiger et al., which shed light on the electronic properties of the N-confused porphyrins as ligands.41 The results of this study were that there were two different kinds of nitrogen atoms in the core unlike the case of CuIITPP in which all four nitrogen atoms were equivalent. Also, the metal ligand bonding has more covalent character than in the regular porphyrin complexes. Subsequently, the crystal structure of copper(II) N-confused tetrapentafluorophenyl porphyrin, Cu(NCTPFP)42 was elucidated along with the Ag(III),

Ni(II) and Pd(II) complexes. (Figure 1.13) The complex had a planar structure and the copper(II) was in a square planar configuration. Again because of the disorder it was not possible to differentiate the coordinated nitrogen atoms from the internal carbon atom but the Cu(II)-N(C) distances were in the range of 1.980(9) to 2.018(9) Å.

Figure 1.13 The crystal structure of copper(II) N-confused

tetra(pentafluorophenyl) porphyrin, Cu(NCTFPP).42

Titration of both Ni(II)43 or the Cu(II)39 complexes with acid resulted in reversible

protonation of the internal carbon atom which is also evidence of the carbanionic

character of the carbon atom (Figure1.14). Titration of the β-substituted Ni(II) complex

with acid was monitored with 1H NMR spectroscopy and it indicated an improvement in

the diamagnetic ring current. The acid titration of the Cu(II) complex was monitored by

EPR spectroscopy and it was affected by the type of acid used. This was accounted for

the protonation of the complex along with the coordination of the conjugate base from the acid.

15 Ar Ar Ar Ar N N H H + H II N +H+ II N N M N M -H+ N N

Ar Ar Ar Ar

M = Ni, Cu M = Ni, Cu

Figure 1.14 Reversible protonation of Ni(II)43 and Cu(II) NCP.39

The carbanionic character of the internal carbon atom in the Ni(II) complex was also demonstrated by the oxidative addition of many different haloalkanes in the presence of proton scavengers to the particular carbon atom.44 Treatment of the corresponding

NiNCPs with several monohaloalkanes or dihaloalkanes afforded a group of monomeric

(Figure 1.15) and dimeric complexes (Figure 1.16), respectively. It is important to note

that the primary target of the nucleophilic addition to the NiNCP is the already activated

internal carbon atom. In the case of free base N-confused porphyrin, the methylation

reaction takes place at the peripheral nitrogen but not at the internal position.45

Treatment of NiNCTPP with monohaloalkanes results in formation of three different

products: an internally methylated form, an internally methylated cationic form with

axially coordinated anion, and also the dimethylated form (Figure 1.15). The

spectroscopic findings showed that, in the second case, the external nitrogen was

protonated to generate a paramagnetic complex. Electrochemical analysis of this

protonated species showed small changes in the potential of Ni(II)/Ni(III) couple

although stabilization of Ni(I) was observed.

16 Ar Ar Ar Ar Ar Ar Ar Ar N N N N X H X R H R R R II N RX / OH- II N II N II N N Ni N Ni N Ni N Ni

N N N N

Ar Ar Ar Ar Ar Ar Ar Ar

Figure 1.15 Reaction of NiNCP with monohaloalkanes in the presence

of a base affording three products.44

Exposure of Ni(II) N-confused porphyrins to dihaloalkanes in the presence of bases results in the formation of covalently linked dimeric units and monomeric derivatives

(Figure 1.16). It is important to note that although the dihaloalkanes are used in excessive amounts, mostly 2,21-linked dimers are observed as well as the 2,2-linked

rings in low yields. It was proposed that the rate determining step is the addition reaction and the dimer formation is the fast step. Electrochemical studies suggest that although the electronic structures are partly similar to the monomeric species, it is apparent that there is interaction between two redox active centers. Furthermore, the Ni(II) centered oxidation potentials are similar to that of the unsubstituted complexes. Formation of monomeric 2- or 21-ethoxy substituted products is observed when sodium ethoxide or ethanol is added to the reaction mixture. Because these reactions only take place in the presence of a base, it can be proposed that the reaction is initiated by deprotonation of the peripheral nitrogen atom. In order to prove this assumption, NiNCTPP was titrated with t-BuOK. Major changes in the electronic structure of NiNCTPP were observed when tracked by UV-vis spectroscopy also the 1H NMR spectrum of the resulted complex indicated the disappearance of the 2-NH proton (Figure 1.17).

Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar N N N N N CH2X2, Base N N N N N N + N Ni Ni N N Ni Ni Ni N

N N N N N Ar Ar Ar Ar Ar Ar Ar Ar Ar Ar

minor Figure 1.16 Formation of 2,21-linked and 2,2'-linked dimeric NiNCP complexes via the

treatment of NiNCP with dihaloalkanes.44

Ar Ar Ar Ar N N H - II N Base II N N Ni N Ni Acid N N

Ar Ar Ar Ar

Figure 1.17 Reversible deprotonation of the external nitrogen of the NiNCTTP.44

Analysis of the crystal structures of mono- and dimethylated species gave further

information on these complexes (Figure 1.18). Unlike the unmethylated form, the

inverted pyrrole unit deviates significantly from the plane with a dihedral angle of -42.2˚

in the monomethylated complex. (Figure 1.18a) Nickel(II) is coordinated to the inverted

pyrrole unit with η1 fashion with a Ni-C bond distance of 2.005(6) Å. This bond length is

longer as compared to bromoethyl substituted NiNCTPP with a Ni-C distance of 1.906(5)

Å36 which can be accounted for the sp3 hybridization of the carbon atom as well as the

II tilted geometry of the pyrrole unit. In the paramagnetic dimethylated form, Ni (2-NCH3-

21-CH3NCTPP)I, the nickel(II) ion is five-coordinate with an iodide ion in the axial

18 position at the opposite side of the internal methyl group (Figure 1.18b). It is also

remarkable that the nickel ion is moved out from the central core towards the iodide ion.

The dihedral angle of the inverted pyrrole is also increased dramatically compared to the

monomethylated species. As one might expect, the Ni(II)-C distance elongated from

2.005(6) to 2.406(9) Å, and the bond lengths around the corresponding carbon atom suggests sp2 hybridization on the carbon atom.

Figure 1.18 The crystal structures of (a) mono- and (b) dimethylated NiNCTPP.44

19 A similar type of methylation reaction was also observed in copper(II) complex of

NCTPP.39 Methylation of the internal carbon atom was achieved by addition of methyl

iodide which proves the carbanionic character of the carbon as in the case of nickel(II)

complexes.

Another example of addition reactions to the internal carbon atom was demonstrated by cyanide addition from DDQ.46 Treatment of NiNCP with sodium ethoxide and DDQ

as the oxidizing agent resulted in the addition of cyanide to the internal carbon atom

along with methoxide addition at the external position. Two products were recovered

from this reaction and both were characterized by X-ray crystallography. Both

complexes deviated from planarity as in the case of methylated species. The comparison of the bond lengths on the inverted pyrrole unit and the Ni-C bond lengths (~2.019 Å and

~2.018 Å) suggested an sp3 hybridization on the carbon atom, as expected.

a. b.

Figure 1.19 Crystal structures of cyanide substituted NiNCP

(a) Ni((C-CN)NCTTP) (b) Ni((C-CN)(C-OMe)NCTTP).46

20 One other remarkable Ni complex of N-confused porphyrins reported is the complex

in which the nickel(II) ion engages in organometallic bonding both equatorially and

axially at the same time in the core of porphyrin.47 The corresponding complexes were

generated by the replacement of the axial chloride anion of the dimethylated species,

II Ni (2-NCH321-CH3NCTPP)Cl by an alkyl or phenyl either by Grignard or lithium

II reagents. The resulting species Ni (2-NCH321-CH3NCTPP)R is a paramagnetic

complex. According to the 1H NMR and EPR spectra and electrochemistry, the macrocycle carries anion radical character with a high-spin nickel(II) center. Addition of

a reducing agent to this complex resulted in one electron reduction and the resulting

complex had a similar molecular geometry as verified by 1H NMR and EPR

spectroscopy.

Mo(NCTPP)(pip)2 is an example to divalent metal complex of NCTPP, which was

reported by our group in 2005 (Figure 1.20).48 This species is unique among N-confused

porphyrin compounds by being isostructural with its regular porphyrin analog; it gives us

the opportunity to make a direct comparison between the electronic properties of the two

compounds. It is also special by being the first demonstration of an early transition metal

ion coordinating to NCP. The porphyrin and N-confused porphyrin complexes are structurally similar in the solid state with identical bond lengths, while the electronic

properties significantly differ from each other. The Mo+2 center is six-coordinate with

the three nitrogen atoms and one carbon atom from the porphyrin core and two nitrogen

atoms from the piperidine molecules. The absorption spectra of the species are different

from each other due to the symmetry differences between the two macrocycles. The

II Figure 1.20 Crystal structure of Mo NCTPP(pip)2.

magnetic measurements showed that the d orbital energy levels of the metal are affected by the switch of the nitrogen atom to a carbon atom in the coordination environment.

One other divalent metal complex of NCP was synthesized in our group. Reaction of the free base NCTPP with cobalt carbonyl under anaerobic conditions and exposure of the resulting complex to pyridine resulted in CoII(NCTPP)py which was elucidated by crystallography (Figure 1.21). The interior pyrrolic C-H bond is broken forming σ type bonding with the metal center. The macrocycle is planar with the three internal nitrogen atoms and one internal carbon coordinating the metal center as in the case of Ni(NCTPP).

One pyridine unit coordinates to the metal center axially whereas one other pyridine molecule remains uncoordinated in the unit cell.

Figure 1.21 Crystal structure of CoII(NCTPP)(py)·py with 50% thermal ellipsoids.

Trivalent metal complexes of NCP

Although the direct metallation of NCP with nickel and copper salts resulted in the

formation of divalent complexes, it was also demonstrated that these complexes could be oxidized to form trivalent metal ion species. Generation of Ni(III) complexes of NCP was accomplished both by electrochemical and by chemical oxidation in situ.49 The spin-

Hamiltonian parameters as well as the 61Ni hyperfine splitting showed that the oxidation

was metal-based but not macrocycle based. It was claimed that the macrocycle acted as a

dianionic ligand with axial ligation of an anion, dependent on the spectroscopic findings.

The oxidation process was reversible and the spectroscopic properties of the oxidized

species were strongly influenced by the axial ligand.

Crystallographic proof for the formation of Ni(III) complex of NCP with a bridging

oxygen atom was reported (Figure 1.22).50 In this study, osmium tetroxide, a strong

23 oxidizing agent, was used to generate a novel Ni(III) complex. The EPR and magnetic

susceptibility measurements suggested that the reaction took place at the metal center but

not on the ring system. The crystal structure showed that an oxygen atom was bridging

the metal center and the carbon atom, whereas a pyridine molecule was axially

coordinated to the metal ion. The macrocycle was distorted from planarity but not as

much as the 21-methylated complex. The internal carbon atom was suggested to be sp3 hybridized and the Ni-C bond length was 2.068(4) Å, longer than the previously reported

Ni(II)-C bond lengths.

One-electron oxidation of the CuIINCTPFPP formed a remarkable Cu(III) complex,

CuIIINCTPFPP, which has square-planar coordination environment around the metal center.51 This species is unique by being the first porphyrin complex in which the metal

center has the same coordination environment in two different oxidation states of the

metal. When CuIIINCTPFPP was treated with 1.5 equiv of DDQ, a metal-based oxidation

Figure 1.22 Crystal structure of NiIII(NCPO) py.50

24 took place generating CuIIINCTPFPP. The resulting complex is a diamagnetic d8 system whereas the Cu(II) complex is paramagnetic. The structure of this complex was elucidated by single crystal X-ray crystallography (Figure 1.23). The crystallographic parameters are similar to that of Cu(II) complex, except the average Cu-N and Cu-C bonds are shorter for CuIIINCTPFPP. In addition, the complex doesn’t have a counter

anion either as an axial ligand or in the unit cell indicates a +3 oxidation state.

Figure 1.23 Crystal structure of CuIIINCTPFPP.51

Reaction of NCP with silver(I) trifluoroacetate resulted in formation of silver(III) complex which was the first example of the direct synthesis of a trivalent metal NCP complex.52 It is a remarkable complex because +3 is not the most common oxidation

state for silver. The oxidation state of the metal ion was verified by magnetic

susceptibility measurements along with electrochemical studies. 1H NMR spectroscopy

showed the loss of both the internal and external protons. The elucidation of the crystal

Figure 1.24 Crystal structure of Ag(NCTPP).52

structure verified the oxidation state because the unit cell lacked any counteranion

(Figure 1.24). The crystal structure showed that the metal ion was coordinated to the ligand through three nitrogen and carbon atoms in a pseudo square planar configuration

with asymmetric bond lengths. The Ag-C bond length is 2.019(2) Å and the Ag-N bond

lengths ranges between ~ 2.050 to 2.065 Å. Silver complexes of tetrapentafluorophenyl42 and diphenyl53 substituted NCPs were also reported in later manuscripts.

Manganese(III)54,55 and iron(III)56 ions can also form σ type bonding with the

internal carbon atom of the inverted pyrrole unit of NCP. Both complexes were

generated by oxidation of the corresponding divalent metal complexes with simultaneous

C-H bond activation. As will be discussed later in this chapter, both Mn(II) and Fe(II) ions form side on interactions with the sp2 hybridized carbon atom. NCP acts either as a

dianionic or a trianionic ligand with the manganese(III) ion forming two different

complexes. In the first case, the macrocycle is not fully deprotonated and bromide ion is 26 coordinated to the metal center for charge balance (Figure 1.25a).54 The Mn(III)-C bond

length is reported as 2.020(6) Å which is shorter than the Mn(II)-C distance of the divalent species. This is expected because the coordination mode is changing from an agostic type interaction to σ-type bonding.

In the latter case, manganese(III) is hexacoordinate with the four equatorial bonds as well as the two axial pyridine molecules (Figure 1.25b).55 In this molecule, NCTPP is

fully deprotonated and the C-H bond is activated. In the following year a high-frequency

and high–field electron paramagnetic resonance (HFEPR) study was carried out on this

complex for better understanding of the effect of the pyrrole inversion on the electronic

properties of the metal ions.57 The complex was measured both in the polycrystalline

state and also in solution and both agreed with the powder pattern simulation. One of the conclusions derived from the spectra is that the ground state of the complex is high spin

(S = 2). The comparison of the spectra with the other Mn(III) porphyrinoids demonstrated the effect of the reduced symmetry having a significant rhombocity of the zfs tensor. All of the known Mn(III) porphyrins have an axial zfs tensor with an E value of zero and Mn(III) corroles have a slight degree of rhombocity (E/D < 0.01) whereas

Mn(NCTPP)(py)2 has a E/D ratio of 0.33. As a result of a comparison of the D values of

the corresponding Mn(NCTPP)(py)2 and Mn(DEHMC)(py)2, it was proposed that the

effect is due to the equatorial ligand field since they both have the same axial ligands.

Similar chemistry was observed with the Fe(II) complex of NCP.56 The metal based

oxidation of the iron(II) species to form FeIII(NCTPP)Br was first observed by 1H NMR

spectroscopy upon exposure of the Fe(II) complex to atmospheric dioxygen. The

27 a.

54 55 Figure 1.25 Crystal structures of (a) Mn(NCTPP)Br and (b) Mn(NCTPP)py2.

magnetic moment measurements indicated a spin state of S = 3/2 for the Fe(III)

center.The formation of the complex was further verified by X-ray crystallography

(Figure 1.26a). The structure highly resembles the Mn(III) complex, the porphyrionid

itself is nearly planar with the iron(III) center sitting above the plane toward the bromide

ion. The Fe-C distance has been found to be 1.981(8) Å. Further addition of O2 resulted in the formation of a high spin Fe(III) center which was then determined to be

[Fe(III)(NCTPPO)Br]-. Protonation of this resulting complex afforded a neutral complex

28 a.

Figure 1.26 Crystal structures of (a) Fe(NCTPP)Br and (b) Fe(ONCTPP)Br.56

which was characterized by X-ray (Figure 1.26b). The molecule was oxygenated and

formed a nonplanar complex which was also observed via 1H NMR spectroscopy. This

complex is similar to the Ni(III) NCP complex which was discovered by Dolphin et al.50

The inverted pyrrole ring deviates from the macrocyclic unit and forms an η2-fashion interaction with the metal center.

Cobalt(III) was also incorporated into NCP and metallation activated the internal carbon atom.58 Cobalt(III) insertion into N-confused porphyrin resulted in formation of 29 three different complexes: one with dianionic NCP character and two with trianionic

characters. The anaerobic reaction of NCTPP with hydrated cobalt nitrate

III (Co(NO3)2·6H2O) yielded Co (NCTPP)(H2O) in which the macrocycle carries a -3

charge with full deprotonation of the porphyrin including the external nitrogen (Figure

1.27a). The complex had a planar geometry which was verified by single crystal X-ray

crystallography. Introduction of triphenyl posphine to this complex induced exchange of the water molecule with PPh3 at the axial position (Figure 1.27b). In this case the

macrocycle was distorted from planarity and had an overall charge of +1 which was

2- compensated by the presence of an unusual anion, [Co2(NO3)4(µ-NO3)2] . This dimeric

anion lies on an inversion center in the unit cell. Each metal center exhibits an octahedral

geometry, with four terminal bidentate nitrates and two bridging nitrates between the

cobalt ions. The external nitrogen atom was protonated which was evidenced by 1H

NMR and IR. The third Co(III) complex, Co(NCTPP)(py)2 was obtained via the

exposure of Co(NCTPP)(H2O) to pyridine (Figure 1.26c). The metal center is six-

coordinate with two axial pyridine molecules like in the case of Mn(NCTPP)py2. The external nitrogen was deprotonated and there was no anion in the unit cell which indicated that the macrocycle had a charge of -3.

Another CoIII complex was also isolated from a cobalt carbonyl reaction. As

mentioned earlier, the reaction of the free base with cobalt carbonyl under anaerobic conditions resulted in the formation of a CoII complex. On the other hand, the exposure of CoII(NCTPP) to 1-MeIm under anaerobic conditions in methanol resulted in the

generation of six-coordinate CoIII complex (Figure 1.28).

30 The stabilization of high valent metal ions by N-confused porphyrin was also demonstrated in antimony(V) complexes. Although antimony(V) is not an example of a trivalent metal ion, it can be classified in this section because it has common structural features with Mn(NCTPP)py2 and Fe(NCTPP)py2. Both the cationic and neutral forms of the Sb(V) complex were isolated.59 In the cationic complex, the macrocycle was dianionic with the external proton present. In order to compensate the remaining +3 charge, two bromide ions were coordinated to the Sb(V) center along with oneuncoordinated bromide anion. Addition of excess pyridine resulted in the

a. b.

III Figure 1.27 Crystal structures of (a) Co (NCTPP)H2O

III III 58 (b) Co (NCTPP)PPh3 (c) Co (NCTPP)py2.

III Figure 1.28 Crystal structure of Co (NCTPP)(1-MeIm)2.

deprotonation of the external nitrogen to give a neutral complex with the two axial

bromide anions (Figure 1.29a). A similar Sb(V) complex of NCTTP this case with

methoxide groups on the axial positions was reported earlier (Figure 1.29b).60 These two

structures were compared with the isostructural [Sb(TPP)Br2]Br3. All three structures have similar parameters such as an elongated octahedral environment around the antimony(V) and almost planar coordination cores. But they differ in their Sb(V)-Br bond distances; [Sb(TPP)Br2]Br3 has the shorter and Sb(NCTPP)Br2 has the longer bond

length. The expected correlation is the higher the charge on the metal, the shorter the bond length. According to this assumption, it can be proposed that the NCP ring is donating positive charge to the metal ion in the case of [Sb(NCTPP)Br2]Br unlike the

regular porphyrin.

32 Side-on M···H-C interaction

As mentioned earlier, side-on interaction of the metal and the internal C-H unit for

N-confused porphyrins is a common type of interaction. Metallation of the macrocycle with certain metals does not result in the activation of the internal C-H bond, but instead results in the formation of a side-on interaction with the metal center. These interactions are often referred as agostic bonds although no study such as coupling constant experiments with NMR spectroscopy was conducted. The metal ions in these monomeric

V V Figure 1.29 Crystal structure of (a) Sb (NCTPP)(Br)2 and (b)Sb (NCTTP)(OCH3)2 33 complexes typically have three coordination sites from the core of the ring. The

availability of a nitrogen atom at the peripheral position gives the macrocycle the

possibility to coordinate through this site. In some cases the metal center coordinates to

the external nitrogen atom of the second complex forming self-coordinating dimeric

species. Both monomeric and dimeric complexes will be discussed in detail.

Monomeric metal complexes of N-confused porphyrin

Ar Ar N L H II N N M

Ar Ar

M = Cu, Mn, Fe, Zn

Figure 1.30 Monomeric divalent metal complexes of NCP.

The first agostic-type interaction was exhibited by copper(II) complexes of NCP in

2000 by Latos-Grażyński and co-workers.39 As mentioned in the previous section this

complex was generated by exposure of CuII(NCTPP) to acid which resulted in the

protonation of the internal carbon atom forming a side-on interaction between the metal

center and the internal arene unit. The resulting complex was characterized by spectroscopic methods but not by X-ray crystallography.

34 The first iron(II) complex of NCP, FeII(NCTPP)Br, also has π-type interaction

between the metal center and the C-H unit of the inverted pyrrole unit. 61 Magnetic

susceptibility measurements showed that the iron(II) has a high spin d6 configuration and

also the absorption spectrum showed a similar pattern with the metal NCP complexes with tilted pyrrole units as seen in Ni(2-NH-21CH3CTPP)Cl. This complex is the first

example of this type of compound to be characterized by crystallography. The bond

lengths around the internal carbon atom indicated that this carbon site has sp2 hybridized

character. The pyrrole ring is tilted from the plane of the macrocycle. The Fe···C

distance (2.361(2) Å) is much longer than the reported Fe-C bond lengths. When the

apical bromide anion is replaced by a thiolate, this causes the iron center to have

intermediate spin and the ligand also affects the overall structure of the molecule. While

the Fe···C distance is only slightly longer, the Fe···H distances differ significantly. It can

be explained by the electronic reasons as well as the steric factors. The electronic

properties of the complex were studied by paramagnetic 1H NMR and 2H NMR spectra in

a later study.62 The presence of the external proton was reflected by the high field

chemical shift at -8.0 ppm while the extremely downfield shifted internal C-H showed up

at 812 ppm. A temperature dependence plot of the internal proton showed a linear

relationship and also titration experiments with 2-methylimidazole and pyridine indicated

the exchange of the bromide anion with the corresponding ligands.

Recently, the iron complexes of the N- or C- methylated NCPs were synthesized and

characterized by 1H NMR.63 The reactions were monitored by paramagnetic 1H NMR spectroscopy. The nature of the inverted pyrrole and methyl groups was monitored for the verification of the electronic state of the iron. The insertion of the iron metal into

35 these macrocycles resulted in the formation of high-spin Fe(II) complexes as in the case

of the unsubstituted macrocycles. Treatment of the N-methylated complex with either O2

or Br2 induced metal centered oxidation along with the activation of the internal C-H

bond generating an intermediate-spin Fe(III) complex. Likewise, the reaction of the C-

methylated and dimethylated compounds with Br2 produced high-spin cationic Fe(III)

complexes with a side-on interaction of the inverted pyrrolic unit with the metal center.

The monomeric manganese(II) complex of NCP with an axial bromide anion

exhibited very similar properties to that of the iron(II) complex (Figure 1.31a).54 As

discussed earlier, oxidation of this complex activated the C-H bond and oxidized the

metal center to the +3 state. Both divalent and trivalent complexes were characterized by

crystallography, which revealed the structural changes that occurred via oxidation. The

distance between the manganese center and the inner carbon atom is 2.437(7) Å which is

longer than a typical Mn(II)-C bonds but shorter than the sum of the van der Waals radii

of the atoms, verifying the agostic-type interaction. Another monomeric manganese(II)

complex of NCTPP with single axial pyridine molecule was reported by our group in

2002 (Figure 1.31b).64 This molecule is different than the Mn(III) complex of NCTPP,

Mn(NCTPP)(py)2, which has a direct Mn-C bond. A noticeable distortion of the pyrrole

ring is observed in this Mn(II) complex as in other complexes with side on interactions.

In addition, switching from the bromide anion to a pyridine unit the metal carbon distance has decreased to 2.357(5) Å from 2.437(7) Å.

Zinc metal ion also coordinates to the NCP core forming an agostic-type

interaction.65 Incorporation of the zinc(II) metal to the inverted porphyrin afforded two

36 a.

Figure 1.31 Crystal structure of (a) MnII(NCTPP)Br54 (b) MnII(NCTPP)py.64

metal complexes: a monomeric species with an axial pyridine and a tetranuclear dimeric complex which will be discussed later in the text. As in the case of the iron(II) and manganese(II) complexes, the inverted pyrrole ring was tilted from the rest of the macrocyclic unit making a side-on interaction with the metal center. The distance between the d10 Zn(II) ion and the sp2 hybridized carbon atom is 2.414(2) Å, which is

again close to the estimated distance of an agostic interaction.

66 Figure 1.32 Crystal structure of Yb(NCTPP)(LOMe).

Two other remarkable complexes which can serve as a new example of an agostic type interaction are the first and only lanthanide complexes of the N-confused porphyrins, Yb(NCTPP)(LOMe) and Er(NCTPP)(LOMe) where LOMe is a tripodal anion

required for stabilization of the complex (Figure 1.32).66 Again the presence of an

agostic type interaction was verified in the solid state by crystallography as well as in the

solution phase by 1H NMR spectroscopy.

Dimeric metal complexes of NCP

Zinc(II) ion coordinates to N-confused porphyrin to generate either a monomeric or a tetranuclear dimeric species depending on the presence of an additional coordinating ligand.65 Whereas these two complexes were characterized by crystallography, the formation of a third complex was reported in this study. The structure of this species was

38 proposed to be a self-coordinating dimer based on the spectroscopic data.

Crystallographic evidence for this complex appeared later in the literature and will be discussed below. As discussed earlier, in the presence of pyridine, a monomeric species with a pentacoordinate Zn(II) center was formed. In the absence of pyridine, a

65 Figure 1.33 Crystal structure of Zn4 NCTPP dimer.

tetranuclear species formed. The coordination geometry of the zinc center is not disturbed significantly but the axial ligand is replaced by an oxygen atom from the

bridging acetate (Figure 1.33). As can be seen in the Figure, the two zinc(II) ions are at

the cores of the porphyrinoids while the other two serve as bridging atoms via

coordination of the acetate ions. The Zn(II) atoms are displaced from the center of the macrocycle cores like in the case of the monomer and each interacts with the C-H unit in a side-on fashion with a distance of 2.489(1) Å.

39 The interaction of other group 12 metals with the diphenyl N-confused porphyrin

was also investigated (Figure 1.34).67 The formation of zinc(II), cadmium(II) and

mercury(II) complexes was verified in solution by 1H NMR spectroscopy but only the

Zn(II) and Cd(II) species were characterized by crystallography (Figure 1.35). Zn(II) and

Cd(II) complexes are isostructural with the pentacoordinate metal centers. As observed

Ar N N M N Ar N

Ar N II N N M M = Zn, Cd, Hg N Ar

Figure 1.34 Schematic representations of dimeric NCP complexes.

Figure 1.35 Crystal structure of dimeric Zn(NCDPP).67

40 in all complexes with side-on interactions, the inverted pyrrole rings for these complexes

are also tilted away from the N3 core and the M-C interactions with bond lengths of

2.534(1) for Zn and 2.555(1) for Cd can be described as agostic-type interactions.

Interestingly, manganese(II)55,64 and iron(II)68 also form very similar types of

dimeric complexes as the group 12 metal ions only with the exception of the orientation of the inverted pyrrole units. In these dimeric complexes the macrocycles are self- coordinating with binding of the external nitrogen atoms to the axial positions. The reaction of Fe(NCTPP)Br with sodium benzeneselenate under anaerobic conditions

results in the formation of the iron dimeric species (Figure 1.36).68 SQUID measurements on this dimer suggest a weak antiferromagnetic coupling between the metal centers. The distance between the metal center and the internal carbon atom is

2.341(5) Å, which is comparable to that of monomeric Fe(NCTPP)Br, which has a bond length of 2.361(1) Å. Also the values of the Fe-N distances on the equatorial plane are similar to the bond lengths reported for the high spin iron(II) complexes.

This dimeric species is air sensitive and exposure of the complex to air resulted in formation of a new complex which was also characterized by spectroscopy as well as crystallography (Figure 1.36). SQUID measurement on the oxidized complex shows a stronger coupling between iron centers. The elucidation of the structure by X-ray crystallography revealed that the internal pyrrolic carbon is oxygenated like in the case of oxidized monomeric iron56 and nickel50 complexes mentioned previously. The two

monomers are connected from the iron centers via a µ-hydroxyl group as well as a

sodium cation through the external nitrogen atoms.

68 Figure 1.36 Crystal structure of [Fe(NCTPP)]2.

Figure 1.37 Crystal structure of dimeric Fe(NCTPP) complex

bridged via µ-hydroxyl group and a sodium cation.68

The manganese dimer complex was reported almost at the same time with the isostructural iron complex.64 Internal hydrogens on both macrocycles are still present with the tilt of the inverted pyrrole rings indicating the presence of an agostic-type interaction (Figure1.38). In another report, it was also observed that an elongated reflux time causes the reduction of one of the porphyrin ring at two opposite meso positions while the second ring remained unreacted (Figure 1.39).55 42

64 Figure 1.38 Crystal structure of [Mn(NCTPP)2]2.

II II 55 Figure 1.39 Crystal structure [Mn (NCTPP)Mn (NCTPPH2)].

A different set of dimeric complexes were synthesized via the complexation of Pd(II) and Pt(II) ions with NCP. The reaction of NCTTP with palladium(II) acetate salt resulted in formation of two types of palladium complexes: an inner-coordinated monomeric species and an outer-coordinated dimeric species (Figure 1.40).69 The nature of this dimeric complex is different than the dimers we have discussed earlier. Each metal site is

43 tetracoordinated but the metals do not reside in the porphyrin core. The coordination

environment of each metal center is composed of two inner nitrogen atoms and outer

nitrogen atom and o-carbon of a meso-tolyl of the adjacent porphyrin. The crystal

structure of the complex is depicted in the Figure.

69 Figure 1.40 Crystal structure of [Pd(NCTTP)]2.

Reaction of platinum(II) with NCP results in different coordination chemistry.70

When tetrakis(4’-tert-butylphenyl)porphyrin (NCTBuPP) was treated with PtCl2, two geometric isomers of a dimeric complex were formed (Figure 1.41). This time the complexes are mononuclear and the internal nitrogen atoms do not participate in coordination. The coordination environment of Pt(II) is made up of two external nitrogen atoms, an ortho-carbon from the tolyl- group of one macrocycle and a chloride. The only difference between two isomers is whether the N-confused porphyrin units are bound cis- or trans- to the metal center. Also, a monomeric platinum complex with Pt binding in the core of NCTBuPP was produced as a minor product.

44 a.

70 Figure 1.41 Crystal structure of (a) cis-Pt[NCTBuPP]2Cl (b) trans-Pt[NCTBuPP]2Cl.

Similar chemistry was applied to the internally coordinated Ni(II) complexes.71 The reaction of Ni(NCTPP) or Ni(NCTTP) with PtCl2 did not affect the internal coordination

and the platinum bound to the external nitrogen atoms of both macrocycles and an ortho

carbon of one of the NCTTP units (Figure 1.42). Only the cis isomer was formed, unlike

in the case of the platinum only complex. 45

71 Figure 1.42 Crystal structure of cis-Pt(Ni(C-benzyl)NCTTP)2Cl.

As mentioned earlier, the nitrogen atom of the inverted pyrrole unit frequently involves in coordination resulting mostly in formation of dimeric species. Furuta and coworkers isolated a monomeric dinuclear Rh(I) complex. Altough the resulting complex of NCP is not a dimeric species, it exhibits both interior and exterior coordination

simultaneously. The reaction of NCTPP with [Rh(CO)2Cl]2 in the presence of sodium

acetate led to the formation of the bis-Rh(I) complex (Figure 1.43).72 As seen in the

Figure, the complex has two rhodium centers one of which is coordinated to the external nitrogen and the second one is coordinated to two nitrogens on the interior of the macrocycle. Both rhodium ions are located above the plane of the macrocycle and exhibit square planar geometries.

72 Figure 1.43 Crystal structure of Rh2(NCTPP).

Some interesting chemistry was observed in the course of metallation of NCP.

Reaction of NCTPP with Re2(CO)10 under reflux conditions resulted in skeletal

modification of NCP ring generating a Re complex of N-fused tetraphenylporphyrin.73

This complex is the first and only metal complex of NFP although the possibility of stabilizing low oxidation state metal complexation was studied computationally.74

Crystal structure of Re (NFTPP)(CO)3 is shown in Figure 1.44. The macrocycle has a bowl like shape and coordinated to the metal center through three internal nitrogen atoms.

The rhenium atom sits above the porphyrin core like in the case of bis-rhenium complex

of regular porphyrin.75

73 Figure 1.44 Crystal structure of Re(CO)3(NFTPP)

Carbaporphyrinoids

Since N-confused porphyrins are capable of forming a wide variety of interesting

coordination complexes, efforts in the field of porphyrin chemistry have been directed to

carbaporphyrinoid synthesis. A carbon atom can be introduced to the core of porphyrins

simply by replacing one or more pyrrolic units with carbon containing ring systems. The

coordination chemistry of these systems has been studied and it was observed that they

can engage in organometallic type of interactions as in the case of NCPs either via direct

M-C bonds or side-on interactions via the equatorial platform.

The first example of carbaporphyrins, the “m-benziporphyrin”, with an incorporated benzene unit was reported in 1994.76 This benzene containing macrocycle was

synthesized via the “3+1” methodology by Berlin and Breitmaier with a 6% yield.76

Later the yield of this particular reaction was enhanced to 28% by changing the acid to

48 trifluoroacetic acid and the oxidizing agent to DDQ (Figure 1.45).77 Analysis of the UV-

vis and 1H NMR spectra of the resulting compound showed no indication of porphyrin- like aromaticity. This could be accounted for the interruption of the 18 π-electron delocalization pathway of the aromatic annulene ring of normal porphyrin. Although there is a possible delocalization pathway for m-benziporphyrin (tautomer b), the benzene unit in the macrocycle has to delocalize its 6 π-electron conjugation which is not favored in this case (Figure 1.46). In addition to its non-aromatic character, the macrocycle is unstable in solution. Degredation of the macrocycle upon metallation makes this β-

substituted m-benziporphyrin unsuitable for coordination studies.

Me CHO Me HO2C Et Et N HN OHC H+ DDQ HO2C N HN NH HN Me Et Me Et Et Et Et Et

Figure 1.45 Synthesis of β-substituted m-benziporphyrin via 3+1 methodology.76

Me Me

Et Et N N

N HN N N Me Et Me Et

Et Et Et Et a b

Figure 1.46 Tautomerization of m-benziporphyrin. 49 A meso-substituted m-benziporphyrin with better stability, tetraphenyl-m-

benziporphyrin, was derived from the reaction of dicarbinol, pyrrole and benzaldeyhde

with a 1:3:2 ratio and a 15% yield (Figure 1.47).78

Ph Ph CHOHPh H N + 32+ PhCHOH Et2O.BF3 N N DDQ H N CHOHPh Ph Ph

Figure 1.47 Synthesis of meso-substituted m-benziporphyrin.78

Successful metallation of the ligand with palladium(II) and platinum(II) metal salts

were also presented in the same study. Both complexes were characterized by

spectroscopic methods. The 1H NMR spectra of the resulting complexes lacked internal

C-H resonances which was an indicative of formation of a direct organometallic bond

(Figure 1.48). Similar chemistry had been observed with the nickel(II) cation, which was

reported in a separate study.81

Ph Ph

N M N

N Ph Ph

M = Pd, Pt, Ni

Figure 1.48 Divalent metal complexes of m-benziporphyrin with direct M-C bond.78, 81

50 The first metallation attempt of tetraphenyl m-benziporphyrin with silver(I) acetate

resulted in formation of acetoxybenziporphyrin. The reaction of the macrocycle with

silver acetate leads to acetoxylation of the internal carbon atom but not the insertion of

the metal ion. The silver chemistry of this macrocycle was further studied in subsequent

years. Reaction of m-benziporphyrin with silver tetrafluoroborate in pyridine under

reflux conditions yielded a pyridine-appended macrocyle.79 It was proposed that

formation of the pyridinated species is initiated by the insertion of Ag(III) ion into core of

the macrocycle. The metal insertion then activates the internal C-H bond resulting in the addition of pyridine unit. These two reactions; acetoxylation and pyridination of the

internal carbon atom, represent carbon atom activation via metal insertion.

The metallation chemistry of the acetoxybenziporphyrin with Pd(II), Ni(II), Zn(II),

Cd(II), and Fe(III) metal salts was also investigated later by the same group.80 In the case

of Ni(II) and Pd(II) complexes, the acetate group was cleaved and a direct M-O bond

formed (Figure 1.49). Similar chemistry had been observed with iron(III) with an axial

chloride anion in order to compensate the charge on the metal. In the case of zinc and

cadmium, the acetate group was preserved and a chloride atom was coordinated to the

metal center as an axial ligand (Figure 1.49).

Ph Ph Ph Ph Ph Ph

O AcO Cl O N M N N M N N M N Cl N N N Ph Ph Ph Ph Ph Ph

M = Ni or Pd M = Zn or Cd M = Fe

Figure 1.49 Metallation chemistry of acetoxybenziporphyrin.80 51 Reaction of the divalent metal ions Zn(II), Cd(II), Hg(II) and Ni(II) with m-

benziporphyrin in refluxing chloroform or chloroform/acetonitrile solvent systems

resulted in the metallation of the macrocycle (Figure 1.50).81 The characterization of the resulting complexes by spectroscopy showed that they do not form direct M-C bonds and instead only have weak metal-arene interactions with an axial chloride ion for charge balance. As mentioned earlier, nickel(II) ion is capable of forming two different types of complexes with this particular macrocycle: one with direct M-C bond and the other with a side-on interaction. The conversion between two species is possible through the activation of the C-H bond.

Ph Ph Ph Ph Cl CHCl3 or CHCl3/CH3CN N N + MCl2 N M N H N N Ph Ph M= Zn, Cd, Hg, Ni Ph Ph

Figure 1.50 Metallation of m-benziporphyrin with divalent metal ions forming

organometallic side-on interactions.81

The 1H NMR spectra of the diamagnetic Zn(II), Cd(II) and Hg(II) complexes showed scalar coupling between the metal nucleus and the 1H and 13C nuclei of the arene moiety.

In addition, it was demonstrated that the magnitude of the coupling is dependent on the

position and the type of the axial ligand. The 1H NMR spectrum of the nickel complex

showed unusual paramagnetic shifts due to the agostic-type interactions between the

Ni(II) center and the hydrogen on the arene unit. The conformational properties of the 52 Ni(II) complexes of m-benziporphyrins, p-benziporphyrins and also m- benziporphodimethene were explored by spectroscopic and theoretical methods in a separate study.82 The conclusion derived from these studies is that these complexes are

dynamic in solution, and that the benzene moiety interconverts between syn- and anti-

conformations as in the case of the free base macrocycles. In addition to the

spectroscopic techniques of characterization, the agostic-type interactions were observed

in the crystal structures of Ni(II) and Cd(II) complexes (Figure 1.51). In the cadmium complex, the Cd(II) ion moved out towards the axial chloride ion from the coordination core of the macrocycle. The Cd···C distance is 2.712(1) Å which is remarkably shorter

than the bond length range observed for direct Cd-C while the Cd···H distance is 2.668(4)

Å.

The crystal structure of the nickel complex also showed the presence of an agostic- type interaction. As in the cadmium complex, the distance reported between the metal center and the internal carbon atom is much shorter than the van der Waals radii of the atoms while they are longer than direct organometallic bonds. The Ni-C bond lengths of

the Ni(II) complexes of NCP and the m-benziporphyrin provide a useful comparison of the difference in the type of the bonding. The nickel atom forms a direct Ni-C bond with the internal carbon atom of the N-confused porphyrin with a bond length of around

1.906(2) Å while the observed distance is 2.549(3) Å in the m-benziporphyrin case.

53 a.

Figure 1.51 Crystal structures of (a) Cd(II) m-benziporphyrin chloride and

(b) Ni(II) m-benziporphyrin chloride.

Another example of an agostic-type interaction was observed in the Fe(II) ion

83 complex of tetraphenyl-m-benziporphyrin. Treatment of the ligand with FeBr2 in THF under refluxing anaerobic conditions led to a high-spin Fe(II) complex. This compound was characterized by spectroscopic methods in solution and the structure was also confirmed by single crystal X-ray diffraction. The Fe(II) ion is coordinated to three

Figure 1.52 Crystal structure of Fe(II) tetraphenyl-m-benziporphyrin bromide.83

pyrrolic nitrogens and an axial bromide ion in a distorted tetrahedral configuration

(Figure 1. 52). The Fe-N distances are in the range of high spin Fe(II)-N distances while the Fe-C distance is much longer than the expected distance with a value of 2.579(2) Å.

This distance is also longer than the Fe(II)···C separation observed in the high spin

Fe(NCTPP)Br complex.

The reaction of tetraphenyl-m-benziporphyrin with CuCl2 in THF at room temperature resulted in a tetranuclear complex.83 The crystal structure of the resulting

84 product revealed that the two monomeric subunits were bridged by a Cu2Cl4 unit forming a tetranuclear dimeric species (Figure 1.53). One interesting aspect about this structure is that the internal C-H bond was activated and the hydrogen is substituted by a chloride ion. While the copper-nitrogen distances are within close proximity to the reported Cu(II)-N distances, elongation of the M···C distance is observed as compared to the iron complex.

Figure 1.53 Crystal structure of tetranuclear Cu-complex of

tetraphenyl m-benziporphyrin.83

In 2002, an aromatic isomer of tetraphenyl-m-benziporphyrin, tetraphenyl-p- benziporphyrin, was synthesized by following a similar methodology.78,.85 Tetraphenyl-

p-benziporphyrin was synthesized from the p- isomer of the precursor in 1% yield. Steric

reasons might have resulted in such low yields of the ring. In addition to being aromatic,

this macrocycle is different by having two carbon atoms instead of one at the core leading

to different electronic and coordination properties. The characterization of the p- benziporphyrin in the solid state by crystallography showed that the macrocycle is close to being planar but the phenylene moiety is significantly tilted from the plane of the tripyrrolic unit (Figure 1.55). The characterization of the resulting macrocycle in the

solution state indicated that the benzene unit is dynamic in solution, exhibiting a seesaw

motion.

56 Ph Ph

H Ph Ph N Et2O.BF3 N N + 3 + 2 PhCHOH DDQ H HO OH N Ph Ph

Figure 1.54 Synthesis of tetraphenyl p-benziporphyrin.85

Figure 1.55 Crystal structure of free base tetraphenyl m-benziporphyrin.85

Demonstration of the first metal complex of this carbaporphyrin was reported in the

85 same study. The reaction of tetraphenyl-p-benziporphyrin with CdCl2 gave the

cadmium(II) complex (Figure 1.56a). Characterization of this species both in solution and solid state verified the formation of chlorocadmium(II) tetraphenyl-p-benziporphyrin

which is similar to the Cd(II) complex of m-benziporphyrin. The 1H NMR spectrum

shows that the metal coordination restricts the flexibility of the p-phenylene moiety. A

closer look at the crystal structure of the complex shows that the cadmium has a trigonal

bipyramidal environment with three nitrogen and two carbon atoms from the ring with an axial bromide anion. The Cd-C(21) and Cd-C(22) distances with values of 2.748(2) Å

57 a.

Figure 1.56 Crystal structures of (a) Cd(II) tetraphenyl p-benziporphyrin chloride85 and

(b) Ni(II) tetraphenyl p-benziporphyrin chloride.82

and 2.762(2) Å, indicate an intermediate-range interaction between the metal center and the internal carbon atoms. The phenylene moiety is tilted from the plane toward the metal center resulting in a boat-like confirmation.

58 Later in a second study, Zn(II)81 and Ni(II)81,82 complexes of the tetraphenyl-p-

benziporphyrin, which are isostructural with the Cd(II) complex, were reported. The

nickel complex was also characterized by X-ray crystallography (Figure 1.56b). The

same type of distortion in the phenylene unit is observed in this compound. The nickel

carbon distances are shorter than that of the cadmium(II) complex due to the better fit of

the nickel(II) ion to the coordination core of the macrocyclic unit. The Ni(II)···C

distances are comparable with the distances for the Ni complex of m-benziporphyrin.

The nickel complex shows paramagnetic behavior and is unusual by having a nickel- arene interaction since these types of bonds are not common for nickel ions.

One other interesting porphyrinoid that also introduces two carbon atoms into the

core is ‘vacataporphyrin’ which is a porphyrinoid missing one of the bridging nitrogen in

the core.86 These macrocycles have the same 18 π-electron delocalization pathway as

their p-benziporphyrin counterparts. The reaction of the macrocycle with the divalent

Zn(II), Cd(II) and Ni(II) ions forms diamagnetic zinc(II) and cadmium(II) and

paramagnetic nickel(II) complexes.87 This core-modified porphyrinoid acts as a

dianionic ligand and coordinates through the three nitrogen atoms with an axial chloride anion. The butadiene fragment of the macrocycle is dynamic in solution and has two different orientations. DFT studies showed that the energy difference between two orientations was small. The axial ligand was also exchanged with imidazole, methanol and acetonitrile. The direct couplings between spin active nucleus of the metal and the internal hydrogen atoms indicated a bonding interaction between these two centers.

59 Ph Ph Ph Ph Ph Ph Cl Cl

MCl2 N N N M N N M N H N N N tol tol tol tol tol tol

Figure 1.57 Metallation of vacataporphyrin.87

Although several metal complexes of benziporphyrins were reported earlier, none of the species exhibited porphyrin type aromaticity. In order to increase the aromaticity of these systems, 18 π-electron delocalization throughout the ring has to be favored over the local aromaticity of the benzene unit. One of the ways of doing this is the incorporation of a hydroxyl group to the benzene unit. This modification would result in a keto-enol tautomerization which then can access to the delocalization pathway (Figure 1.58).88

Using this idea Lash synthesized 2-oxybenziporphyrin in 1995 via the condensation of 5- formyl-salicylaldehyde with tripyrrane in the presence of an acid catalyst with a subsequent oxidation with DDQ.88, 77 This procedure yielded 35-44% of β-substituted 2-

oxybenziporphyrin (Figure 1.58). UV-visible and 1H NMR spectra of the resulting

complex indicate the presence of a strong ring current which agrees with the presence of

the semiquinone moiety (Figure 1.59).

60 Me Me CHO HO2C Et HO Et HO N HN OHC H+

HO2C DDQ N HN NH HN Me Et Me Et Et Et Et Et 2+ Me Me

Et Et O HN HO HN H+

NH N NH HN Me Et Me Et

Et Et Et Et

Figure 1.58 Synthesis of β-substituted 2-oxybenziporphyrin.88

Me Me

Et Et HO N O HN

N HN NH N Me Et Me Et

Et Et Et Et a b

Figure 1.59 Keto-enol tautomerization of 2-oxybenziporphyrin.88

In a detailed study of oxybenziporphyrins, it was demonstrated that these macrocycles could undergo two-step protonation to form an aromatic monocation and a

61 nonaromatic dication.77 In the same study, the possible effects of the fused ring systems on the porphyrinoid were investigated, and benzo-, phenanthro- and acetonaphtho- rings were introduced to the macrocycle (Figure 1.60). It has been observed that the aromaticity of the resulting complexes did not change and they showed the same type of chemistry upon protonation.

Me Me Me

Et Et Et O HN O HN O HN

NH NH N NH N N Me Me Me Et Et Et

Figure 1.60 2-oxybenziporphyrins with fused ring systems.77

Since it was shown that the addition of a hydroxyl group to the benzene ring

significantly alters the electronic properties of the benziporphyrin, Lash et al.

investigated the effects of the presence of an additional hydroxyl group.89 Although the

initial attempts to synthesize dihydroxybenziporphyrin did not work as anticipated, the

desired product was synthesized from dimethoxy- variant of the benziporphyrin (Figure

1.61). A meso-substituted variant of the dimethoxybenziporphyrin was also synthesized

in subsequent years.90

62 H OMe O Me O Me R Me R R

Et Et Et MeO N MeO HN HO HN

N HN NH N NH HN Me Et Me Et Me Et

Et Et Et Et Et Et R = -H or -Me

Figure 1.61 Methoxybenziporphyrins and hydroxybenziporphyrin.90

It was observed that these dimethoxybenziporphyrins showed a slight increase in

aromaticity compared to their benziporphyrin counterparts. In the same report, the Ni(II) complex of dimethoxytetraphenylbenziporphyrin was presented in which the nickel formed a direct M-C organometallic type of bond (Figure 1.62).90 The 1H NMR

spectrum of this complex indicated an increase in the diatropic ring current as a result of

metal insertion. Attempts to metallate the ligand with palladium(II) ion failed to give the desired metal complex.90

OMe Ph R

MeO N

Ph Ni Ph N N

R = H or Me Ph

Figure 1.62 Ni(II) complex of tetraphenyldimethoxybenziporphyrin.90

63 The first metal complex of oxybenziporphyrin was reported in 2001. 91 Treatment of

the free base oxybenziporphyrin with PdCl2 in acetonitrile in the presence of K2CO3 resulted in the formation of the Pd(II) complex (Figure 1.63). Retention of the intense

Soret band and the weak Q bands in the UV-Vis spectrum shows that the resulting metal complex preserves its aromatic character. The 1H NMR spectrum lacks the two NH

protons and the CH proton which shows that the macrocycle acts as a trianionic ligand

forming a σ-type bond between the metal and the internal carbon atom.

Et O N

Pd N N

Me Et

Et Et

Figure 1.63 Palladium(II) oxybenziporphyrin.91

Both UV-Vis and 1H NMR titration of the metal complex with trifluoroacetic acid

resulted in the weakening of the aromatic character indicating that the equilibrium is

pushed toward the enol tautomer (Figure 1.64).

64 Me Me

Et Et O N HO N TFAH Pd Pd K2CO3 N N N N

Me Et Me Et

Et Et Et Et

Figure 1.64 Reversible protonation of Pd(II) oxybenziporphyrin.91

It was also noted that the reactivity of the macrocycle was enhanced toward addition reactions as a result of metallation. The reaction of the metal complex with three different electrophiles resulted in the addition of the corresponding groups to the external oxygen atom. Since the formation of keto tautomer pathway was interrupted, the electronic properties of the resultant species resembled the protonated form. Addition of methyl and butyl groups to the internal carbon atom of the complex indicated an increase

in the reactivity of the particular carbon atom, which was also observed in the case of

nickel(II) N-confused porphyrins.44 In this case, spectroscopic analysis indicated that the

aromaticity of the system was retained (Figure 1.65).

65 Me

Et RO N Ac2O / TsCl / n-BuI Pd Me N N Et Me Et O N Et Et Pd R = Ac, Ts, Bu N N Me

Me Et Et O R N Et Et MeI / n-BuI Pd N N Me Et

Et Et R = Me, Bu

Figure 1.65 Addition of various electrophiles to the oxygen and

electrophilic addition to the internal carbon atom.91

Since the stabilization of the unusual oxidation state of silver metal, Ag(III) ion, by

N-confused porphyrin was observed previously, the reactivity of oxybenziporphyrins

toward silver(I) trifluoroacetate salt was also investigated.92 This macrocycle showed

similar type of chemistry toward silver(III) coordination (Figure 1.66). The 1H NMR spectrum of the resulting metal complex shows that the diatropic ring current is diminished as compared to the free base. The free base serves as a trianoinic ligand resulting a neutral complex, unlike in the case of the Pd(II) complex which carries a negative charge.

66 Me Me

Et Et O HN O N AgOAc Ag

NH N N N

Me Et Me Et

Et Et Et Et

Figure 1.66 Reaction of oxybenziporphyrin with silver(I) acetate

to generate Ag(III) oxybenziporphyrin.92

In the same study, modification of the oxybenzene unit of the porphyrinoid was demonstrated as well as the metallation of this resulting ring with silver(III).

Oxynaphthiporphyrin was synthesized via a ‘3+1’ condensation route and the resultant ring showed relatively higher diatropic ring current. Treatment of this macrocycle with silver acetate at room temperature gave the corresponding metal complex in excellent yields (Figure 1.67). The diatropic ring current for this complex was diminished compared to the free base as in the case of oxybenziporphyrin, but this time the effect was more pronounced.

67 Me

Et O N

Ag N N

Me Et

Et Et

Figure 1.67 Silver(III) oxynaphthiporphyrin.92

Another class of carbaporphyrinoids closely related to oxybenziporphyrin was

developed by two different researchers in 2002. 93,94 These porphyrinoids include two types of modification in the ring: replacement of a pyrrolic nitrogen with a carbon atom and introduction of a heteroatom in the coordination core (Figure1.68).

N X

X = O or S

Figure 1.68 Core modified oxybenziporphyrin.93,94

68 Chandrashekar and coworkers synthesized the meso-disubstituted version of core

modified oxybenziporphyrin in 2002.93 A ‘3+1’ acid catalyzed condensation of the

modified tripyrranes with 5-formylsalisaldehyde afforded core modified

oxybenziporphrins with CNNO and CNNS cores in 28% and 15% yields respectively.

The 1H NMR and UV-vis spectroscopy confirmed that the aromaticity of these new

macrocycles is not disturbed by this modification. The palladium complex of the oxa-

substituted oxybenziporphyrin was also synthesized in good yield and characterized by

spectroscopy (Figure 1.69). The 1H NMR spectrum of the resulting complex lacks the

internal NH and CH signals which are the indicative of bond formations via the interior

nitrogen atoms and the carbon atom. Meso-unsubstituted oxa- and thia-

carbaporphyrionoid systems were also reported, but attempts to metallate these systems

did not lead to the desired metal complexes.94,95

N Pd

N O

Figure 1.69 Palladium(II) complex of meso-disubstituted oxybenziporphyrin.93

Lash and co-workers reported the synthesis of a new aromatic porphyrin isomer, tropiporphyrin, following a ‘3+1’ MacDonald condensation of cycloheptatriene dialdehyde with tripyrrane unit resulting in a 23% yield of the porphyrinoid.96 An 69 improved yield (38%) for this reaction was reported in the following years.97 The same macrocycle was isolated by Braitmaier and coworkers in an independent study.98

Although there is a possibility that the seven membered arene unit might exhibit tropylium character with 6 π-electron aromaticity, the 1H NMR spectrum shows a strong

diatropic ring current throughout the macrocyle (Figure 1.70). This is a clear indication

of the presence of 18 π-electron delocalization. The absorption spectrum of the

tropiporphyrin is broad and ill-defined with a relatively weak Soret band region. These

rings are not very stable in solution and both the electronic and stability characteristics

might be indicative of a nonplanar geometry of the seven-membered ring.

Me Me

Et Et HN HN

NH NH N N Me Et Me Et

Et Et Et Et

Figure 1.70 Possible tautomerization of tropiporphyrin.68,69

The silver(III) complex of tropiporphyrin was synthesized and characterized with a

43-47% yield in 2004.92 The aromaticity of the complex was not disturbed by the

chelation of the metal but it was apparent from the 1H NMR spectrum that the

conformation of the seven-membered ring system changed upon metallation. The

Figure 1.71 Structure of silver(III) tropiporphyrin.97

structure of the metal complex of the diphenyl derivative was verified by single-crystal

X-ray diffraction in a later study which gave a better insight to the geometry of the

resulting complex.97 The silver atom sits at the center of the macrocycle with an

expected square planar geometry with Ag(III)-C distance of 2.126(2) Å. As can be seen from the Figure, the cycloheptatriene moiety is significantly distorted from the plane of the ring (Figure 1.71).

One other macrocycle that incorporates a cycloheptariene moiety, azuliporphrin, was introduced to the field of carbaporphyrins in 1997.99 This macrocycle was synthesized

with a 28% yield via the same synthetic methodology, a ‘3+1’ MacDonald condensation

of the corresponding cyclic dialdehyde and the tripyrrane unit. The absorption and 1H

NMR spectra of the macrocycle shows that it only has borderline aromaticity, probably

due to the contribution of tautomer a as shown in the Figure (Figure 1.72).

71 Me Me

Et Et N N

N HN N HN Me Et Me Et

Et Et Et Et a b

Figure 1.72 Tautomerization of azuliporphyrin.99

Since azuliporphyrins have only two hydrogen atoms in the core, they are expected to serve as dianionic ligands. Indeed, treatment of the diphenyl substituted azuliporphyrin with nickel(II) acteate under mild conditions gave excellent yields of nickel(II) complexes, the first metal derivative of azuliporphyrins.100 The resulting metal

complex has a slightly higher degree of aromaticity than the free base which is observed

by 1H NMR and absorption spectroscopies. Elucidation of the crystal structure provided

additional confirmation for the formation of the complex (Figure 1.73). The macrocycle

itself deviates from planarity, having a bowl shaped conformation with the nickel(II)

center slightly above the coordination core. While the trans Ni-N bonds are similar in

length, the Ni-C bond has a value of 1.897(3) Å which is slightly shorter than the Ni-N

bond lengths.

Figure 1.73 Crystal structure of Ni(II) azuliporphyrin.100

The palladium(II) complex of the macrocycle was also isolated under similar conditions as the nickel(II) complex from palladium acetate. Although it was not characterized by X-ray crystallography, it was observed that Pd(II) complex had stronger diatropic ring current than the Ni(II) counterpart based on the 1H NMR spectra. It was also reported that the reaction of the ligand with copper(II) acetate afforded a paramagnetic species which was not fully characterized.

The metallation chemistry of the β-substituted azuliporphyrin with the divalent group

10 metal ions, nickel(II), palladium(II) and platinum(II), was presented in a detailed study.101 Although Ni(II) and Pd(II) complexes were reported previously, their electronic properties were discussed in depth in this second report. It was observed that Pt(II) insertion gave lower yields than Ni(II) and Pd(II) counterparts and it was also noted that the aromaticity of the Pt(II) complex was slightly lower than the nickel(II) and the palladium(II) complexes. Attempts to metallate azuliporphyrin with silver(I) acetate led

73 to ring contraction of the seven-membered ring in order to give silver(III)

carbaporphyrin.

One-pot synthesis of meso-substituted azuliporphyrins was also reported by the same group in 2002.102, 103 Since it was discovered that group 10 metals react readily with the

meso-unsubstituted azuliporphyrins, the coordination chemistry of these metal ions was

extended to the meso-tetraarylazuliporphyrins. The structure of the palladium(II)

tetra(chlorophenyl)azuliporphyrin was elucidated by single crystal X-ray diffraction

(Figure 1.74). It was noted that Pd complex is more planar as compared with the Ni

complex, probably due to the better match of the metal size of the metal to the pore. The

palladium(II) ion is only slightly moved out from the center of the ligand. The Pd(II)-C

bond length is 1.980(3) Å, which is comparable with other palladium organometallic

bonds. The electrochemical studies on the free base tetraphenylazuliporphyrin, and

Ni(II) and Pd(II) complexes showed that the coordination of the metal ions improves the

reversibility of the ligand based oxidations.

Figure 1.74 Crystal structure of tetrachlorophenyl azuliporphyrin.101

74 The reaction of tetraphenylazuliporphyrin with copper(II) acetate salt at room

temperature resulted in a paramagnetic copper(II) complex.104 Characterization of this

complex by crystallography showed that the metal complexation took place with simultaneous oxidation of the internal carbon atom (Figure 1.75). It is important to note

that a similar type of complex was observed with the iron and the nickel complexes of N-

confused porphyrin as discussed earlier in the chapter. The crystal structure shows that

the azulene unit deviates from the mean plane of the tripyrrolic unit like having a side-on

interaction with the metal center with a distance of 2.474(2) Å.

Figure 1.75 Crystal structure of Cu(II) tetrphenylazuliporphyrin.104

Another carbaporphyrinoid, benzocarbaporphyrin, was first prepared in 1996 by

Breitmaier and coworkers serendipitously when they attempted to synthesize

azuliporphyrins by using HBr as the acid catalyst.98 As a result of the ring contraction of

the seven-membered ring benzocarbaporphyrin has formed. The free base

benzocarbaporphyrin was characterized by single crystal crystallography (Figure 1.76).

75 Later the same reaction was reproduced by using tert-butyl alcohol in the presence of a base by Lash.105 In later years the yield of the corresponding macrocycle was enhanced up to 43% via the acid-catalyzed condensation of diformylindene with tripyrrane.106 This macrocycle is often referred as a “true carbaporphyrin” since the pyrrole unit of the regular porphyrin is replaced by a five-membered arene unit. The 1H NMR and absorption spectra of the species showed that the ring has a strong diatropic current due to the 18 π-electron delocalization (Figure 1.77).

Fiugure 1.76 Crystal structure of free base tetraphenyl benzocarbaporphyrin.102

Et HN

NH N Me Et

Et Et

Figure 1.77 Electron delocalization pathway in benzocarbaporphyrin. 76 As discussed earlier, N-confused porphyrins are able to serve both as dianionic and

trianionic ligands due to the presence of two tautomeric states. The relocation of the third

proton to the external nitrogen gives this ligand the possibility of acting as a dianionic

ligand. Since this is not applicable to benzocarbaporphyrins, treatment of the macrocycle

with several divalent transition metal ions such as zinc(II), copper(II), nickel(II) and

iron(II) did not lead to formation of any stable metal derivatives. Trivalent metal cations

like chromium(III) and iron(III) did not readily metallate either. But an interesting aspect of benzocarbaporphyrin chemistry was discovered in the course of exploration of the

107 metallation chemistry. The reaction of benzocarbaporphyrin with FeCl3 in CHCl3-

MeOH solvent system resulted in a regioselective oxidation reaction at the inner carbon.

A mechanism was proposed for this regioselective reaction based on the insertion of

iron(III) into the coordination core. This directed the oxidation reaction toward the inner

carbon. All spectroscopic characterization indicated that this dimethyl ketal derivative is

monocationic (Figure 1.78). The reaction also worked in ethanol resulting in the diethyl

derivative. Regioselective oxidation reactions of benzocarbaporphyrins with ferric

chloride were extended to other alcohols such as isopropyl alcohol and ethylene glycol.108

Elongated reaction times resulted in the halogenation of the internal carbon atom and the formation of a nonaromatic dione species (Figure 1.79).

77 Me

Et HN OR OR

NH N Me Et

Et Et

Figure 1.78 Electron delocalization pathway of dimethyl benzocarbaporphyrin.107

Me Me

Et Et HN HN

FeX3 X

H O / CH Cl NH N 2 2 2 NH N M Et Me Et e

Et Et Et Et

Et HN O

NH N Me Et

Et O Et

Figure 1.79 Internal carbon halogenation and subsequent oxidation of

benzocarbaporphyrin.107

78 A successful metallation of the benzocarbaporphyrin system was reported in 2002.109

The reaction of the ligand with silver(I) acetate in methanol-dichloromethane at room temperature gave good yields of the silver(III) complex as in the case of N-confused porphyrins and tropiporphyrins (Figure 1.80). The resulting complex was characterized by various spectroscopic methods as well as the single crystal X-ray diffraction. The oxidation state of the metal ion was verified by electrochemistry. The resulting complex demonstrated a perfect fit of the metal ion into the core of the porphyrinoid, which is very close to being planar with slight tilt of the indene from the tripyrrolic unit. The Ag-C bond distance is 2.015(4) Å indicating the presence of direct Ag-C bonding.

Figure 1.80 Crystal structure of Ag(III) benzocarbaporphyrin.109

The chemistry of the organometallic silver(III) complex of benzocarbaporphyrin was

studied in depth and as well as the isostructural gold(III) complex in a later study.110 In

this report the silver complexes of five different β-substituted benzocarbaporphyrins and

two meso-tetraaryl substituted rings were derived from the related azuliporphyrins via

79 ring contraction.102,103 Also, the formation of gold(III) tetraarylbenzocarbaporphyrin was demonstrated but it was also noted that the metallation products of the meso-unsubsituted

derivatives are not accessible in high yields.

Although benzocarbaporphyrins serve only as trianionic ligands, it was demonstrated

that a similar type of carbaporphyrinoid, core-modified benzocarbaporphyrin, could

coordinate to divalent metal ions (Figure 1.81).94,95 The core modified

benzocarbaporphyrins were synthesized readily from condensation of diformylindene

with the modified tripyrranes in the presence of acid with 31-54% yields.

Et N

NH X Me

Et X = O or S

Figure 1.81 Electron delocalization pathway of

core modified benzocarbaporphyrin.94,95

Since incorporation of the furan unit to the system increased the basicity of the

system, oxa-benzocarbaporphrin was isolated as a hydrochloride salt. Insertion of the

heteroatoms into the core of the porphyrinoids did not affect the aromaticity of the ring.

Reaction of oxa-benzocarbaporphrin with nickel(II) and palladium(II) acetate afforded

the corresponding metal complexes in good yields. The platinum(II) complex was also 80 isolated but only in low yields. In addition to the detailed spectroscopic characterization,

the crystal structure of the palladium(II) complex was also reported (Figure 1.82). The

1H NMR spectrum indicates that the ring current in Pd(II) complex is higher than the

Pt(II) and Ni(II) complexes which can be explained by the better fit of the palladium(II)

cation into the core resulting in a planar configuration. This was also evidenced from the

crystal structure as seen in the Figure. The reaction of the macrocycle with silver salts

did not result any metal insertion.

Figure 1.82 Crystal structure of Pd(II) benzocarbaporhyrin.95

Another set of macrocycles considered as carbaporphyrinoids that resembles N-

confused porphyrins are called as inverted core-modified porphyrins. First example of

this set of porphyrinoids, S-confused tetraphenylthiaporphyrin, was reported in 1999.111

The coordination chemistry of S-confused thiaporphyrin remained unexplored until the synthesis of the cadmium(II) and zinc(II) complexes in 2006 by the same group112

(Figure 1.83). The macrocycle serves as an anionic ligand toward the divalent metal ions

81 and the charge on the complex is balanced by the presence of a chloride anion. These compounds are not examples of direct metal-carbon bonding but of side-on agostic-type

interactions. The 1H NMR spectrum of the resulting species exhibits an effective coupling constant (JCdH ≈ 9.0 Hz) which is a strong indicative of the through space

interaction between the metal center and the internal hydrogen atom. The scalar coupling

values are comparable with those reported before such as cadmium(II) m-benziporphyrin

81 13 (JCdH ≈ 7.4 Hz). The chemical shifts from C NMR also reflect the presence of this

nonbonding interaction. The structure of the cadmium(II) complex was also elucidated

by X-ray crystallography. The crystal structure reveals that the thiophene unit was tilted

out from the tripyrrolic unit and the cadmium(II) ion is displaced from the center of the

macrocycle toward the apical chloride ion. Further investigation of the interaction

between the metal and the thiophene ring was carried out by DFT calculations.

Ph Ph

S S N Cl N CdCl2 / CHCl3 H Ph Ph M or Ph Ph N HN ZnCl2 / THF N N

M = Cd or Zn Ph Ph

Figure 1.83 Metallation of S-confused porphyrin with divalent metal ions.112

82 HN Ar H

O N

Ar Ar N HN

Figure 1.84 Pyrrole-appended O-confused porphyrin.113

A similar porphyrinoid, O-confused oxaporphyrin, was also designed and

synthesized and the metal chemistry of the ligand was investigated by Latos-Grażyński et

al. in 2003.113 They followed a synthetic route in which pyrrole and p-tolylaldehyde was condensed with the appropriate diol which would theoretically produce the desired

product. However, the reaction did not stop after the macrocycle formation step but an

additional equivalent of pyrrole was added to the external carbon of the furan unit (Figure

1.84).

Metallation of the resulting pyrrole-appended O-confused carbaporphyrin with

divalent metals nickel(II) and palladium(II) caused dehydrogenation of the sp3 hybridized

carbon atom forming an aromatic system. The structure of the Ni complex was also

verified by single crystal X-ray crystallography (Figure 1.85). It is apparent from the

crystal structure that the deprotonation of the carbon atom takes place since the carbon

atom has a planar geometry. The macrocycle itself is slightly saddled. The Ni-C bond

Figure 1.85 Crystal structure of Ni(II) complex of

pyrrole appended O-confused porhyrin.113

length 1.892(4) Å is in close proximity to the other Ni(II) porphyrinoid systems with a

CNNN core. The Ni-N bond lengths are also in the range of diamagnetic nickel(II) complexes.

The Ag(III) complex of the macrocycle was also reported. When the parent compound was treated with silver(I) acetate, a silver(III) complex with an unidentified substituent on the C3 position was recovered. This substituent was replaced by an ethoxy group via the addition of ethanol (Figure 1.86). The crystal structure of the resulting complex also clearly indicates that the external carbon atom on the furan ring is sp3 hybridized as seen in the Figure. The Ag(III)-C bond length is in good agreement with the ones observed for other silver(III) carbaporphyrinoid systems which verifies the formation of a direct bond between the metal center and the carbon atom. The titration of

Figure 1.86 Crystal structure of Ag(III) pyrrole and ethoxy

substituted O-confused porphyrin.113

the complex with trifluoroacetic acid resulted in the cleavage of the ethoxy group

forming sp2 hybridized carbon atom. 1H NMR and UV-Visible absorption spectroscopy

showed that the resulting complex is isostructural with the Ni(II) and the Pd(II)

complexes.

When the previous reaction was run in the presence of ethanol, an ethoxy group

replaced the pyrrole group.114 Titration of the resulting macrocycle with acid resulted in

the formation of the cationic form of the “true” O-confused oxaporphyrin and it was claimed that this species is an intermediate in the formation of C3 substituted molecules.

Formation of the dicationic species is reversible, since addition of sodium ethoxide reformed the ethoxide-derivative. The 1H NMR spectrum suggests borderline

aromaticity for this non-appended derivative. It was also metallated with silver and the

resulting silver(III) complex has a strong resemblance to the earlier reported Ag(III)

85 complex of the pyrrole appended carbaporphyrinoid. As with the free base, protonation

of the metal complex resulted in the cleavage of the ethoxide, producing a monocationic

Ag(III) complex. Another type of reactivity observed with this complex is that the

elimination of the silver atom caused furan ring to transform to an oxazolone ring

producing a “carbaporpholactone.” This system is similar to porpholactone which can be

synthesized via the oxidation of regular porphyrins.115, 116, 117 The 1H NMR spectra are indicative of porphyrin-like properties for this macrocycle. Metallation of the resulting macrocycle with silver afforded the Ag(III) metal complex and the structure was elucidated by X-ray crystallography (Figure1.87). The macrocycle slightly deviates from

the plane. The same metal complex was synthesized via the reaction of hydroxyl-

appended oxacarbaporphrin with silver(I) acetate in which the source of the oxygen atom

was H2O.

Figure 1.87 Crystal structure of Ag(III) tetratolylcarbaporpholactone.114

86 Investigation of the coordination chemistry of these macrocycles has been extended to group 12 metal ions such as zinc(II) and cadmium(II).118 One interesting aspect from

these studies is that while nickel(II) ion forms a neutral complex with the pyrrole appended macrocycle, the same macrocycle serves as a monoanionic ligand toward zinc(II) and cadmium(II) cations forming cationic complexes. The resulting complex requires a chloride anion for charge balance. Although the structures were not observed by crystallography, they were supported by spectroscopic data as well as the DFT calculations. DFT calculations suggest that although the distance between the metal ions and the internal carbon is longer than the bonding distance, it is shorter than the sum of the van der Walls radii of the atoms.

Carbaporphyrins can engage into two different types of organometallic interactions:

direct M-C bonds which involve the activation of the internal C-H bond and side-on

interaction of the metal center and the C-H unit. The latter one is referred as agostic-type

of interaction in many reports. But there is no direct evidence for the exact location of

the internal proton that is in contact with the metal centers although some theoretical

studies had been conducted. Although the internal C-H unit is not activated in this π-type

interaction, they often become prone to activation through attack by nucleophiles and

electrophiles. In addition to this interesting organometallic chemistry, these macrocycles

are able to stabilize otherwise unusual oxidation states of metals such as the silver(III)

ion.

CHAPTER II

METAL MEDIATED C-H BOND ACTIVATION IN A CARBON SUBSTITUTED

HEMIPORPHYRAZINE

The activation of C-H bonds in arenes can proceed via a number of routes,119 but

nucleophilic substitution reactions on aromatic rings with hydride as the leaving group

can be difficult to achieve. Nucleophilic substitution of arenes via halide loss has been

known for many decades. However, C-H aromatic substitutions have only been known

since the late 1970s, and have been found to proceed only under certain conditions.120

Even with stable leaving groups, electron withdrawing substituents are required on the aromatic ring. Frequently only strong nucleophiles, such as NH2¯, can readily attack.

Alternative methods can be used to assist replacement of an aromatic hydride. Vicarious

nucleophilic aromatic substitution can be used to activate arene C-H bonds, promoted by

loss of HX.121 In addition, oxidants can also be employed to affect loss of hydride in a

process known as oxidative nucleophilic subsitution of hydrogen (ONSH).122 In this chapter, we present an unusual C-H bond activation via nucleophilic aromatic substitution with the relatively weak nucleophile pyridine in the dicarbahemiporphyrazine macrocycle that may occur through an ONSH mechanism.

Upon metallation with copper, nucleophilic attack of the internal carbon atom with pyridine is promoted by the reduction of the metal from CuII to CuI. This reaction

88 proceeds under very mild conditions in air, and represents a potential new means to activate C-H bonds via transition metals.

The metallation of porphyrin analogues and isomers where an interior nitrogen position has been replaced with a carbon atom has become an area of significant activity.123 The chemistry of the N-confused porphyrins,25,26 benziporphyrins,19 and

azuliporphyrins17,103 all involve macrocycles where an aromatic C-H bond is in close

proximity to the metal binding site. The interior carbon position is not acidic, and often remains intact upon metal binding, as observed in the Mn,55 Fe,61 and Zn65 complexes of

N-confused porphyrin.25, 26 The exact nature of this C-H metal “agostic” interaction is

still under investigation, but the proximity of a metal to this carbon atom activates it to

nucleophilic and electrophilic attack. Indirect evidence of such metal mediated chemistry

has been seen in the copper mediated oxidative degradation of N-confused porphyrin40 and silver catalyzed C-N bond formation in m-benziporphyrin.85

In the early 1950s, Linstead and co-workers reported the synthesis of the hemiporphyrazine macrocycle, (Figure 2.1).124 The hemiporphyrazines bear a superficial

N N N N N

NH HN NH HN

N N N N N

a. b.

Figure 2.1 The structures of (a) hemiporphyrazine and (b) di-N-deficient analogue

dicarbahemiporphyrazine (H2dchp).

89 resemblance to the phthalocyanines—both have an N4 metal-binding core—but the

phthalocyanines are aromatic 18 π-electron systems, while the hemiporphyrazines have a

20 π-electron system.125 Linstead later prepared a series of porphyrazine derivatives in

which one or two of the central metal-binding nitrogen atoms were replaced with C-H

groups. In one case, a 1,3-diaminobenzene replaced each 2,6-diaminopyridine and the macrocycle, 1, is di-N-deficient. This molecule can be considered as a dicarbahemiporphyrazine (H2dchp).

Experimental

General Methods: Unless otherwise noted, all reagents and solvents were purchased

from Sigma, Aldrich, Acros Organics, Strem and used without further purification. Mass

spectra were recorded using an LCT electrospray spectrophotometer at the Mass

Spectrometry and Proteomics Facility of Ohio State University. Elemental analysis was

conducted at the University of Illinois, School of Chemical Sciences Microanalysis

Laboratory.

Single crystal X-ray diffraction data was collected at 100 K (Bruker KRYO-FLEX)

on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped with a

Mo-target X-ray tube (λ = 0.71073 Å) operated at 2000 watts power. The detector was

placed at a distance of 5.009 cm from the crystal. Integration and refinement of crystal

data was done using Bruker SAINT software package and Bruker SHELXTL (version

6.1) software package, respectively. Absorption correction was completed using the

90 NH N N

EtOH NH + NH HN reflux, 24h H2N NH2 NH N N

64%

Scheme 2.1 Synthesis of dicarbahemiporphyrazine (H2dchp), 1.

SADABS program. Crystals were placed in paratone oil upon removal from the mother

liquior and mounted on a plastic loop in the oil.

Preparation of dicarbahemiporphyrazine (H2dchp, 1): The macrocycle was

synthesized as described in the literature (Scheme 2.1).124 Diiminoisoindoline (5.81 g,

0.04 mol) and m-phenylenediamine (4.33 g, 0.04 mol) were heated to reflux in 60 mL of

ethanol for 24 h. Formation of NH3 gas was observed during the reaction. The solvent

was removed under reduced pressure and the resultant precipitate was washed with

ethanol and air dried. The reaction yielded 5.60 g (64.0%) of dchp. Five different crystal

forms of 1 were grown from different solvent systems. Crystals of 1·HCO2H were grown

from formic acid solution diffused with diethyl ether; 1·H2O from hot methanol;

1·CH3CN from hot acetonitrile; 1·py from concentrated pyridine solution and 1·DMF

from hot dimethylformamide solution.

1 (H2dchp, 1): H NMR ([D6]DMSO, 300MHz): δ= 10.39 (s, 2H), 7.96 (m, 4H), 7.74

(m, 4H), 7.34 (t, J = 7.9 Hz, 2H), 6.95 (s, 2H), 6.76 (d, J = 7.9 Hz, 4H); 13C NMR

([D6]DMSO, 300MHz) δ= 156.4, 154.8, 140.0, 137.1, 135.2, 127.3, 123.4, 118.7. UV- visible spectroscopy in DMSO: λmax (nm) = 329. Crystal data and structure refinement 91 parameters for 1·HCO2H, 1·H2O, 1·CH3CN, 1·py and 1·DMF are summarized in Tables

2.1, 2.2, 2.3, 2.4 and 2.5.

Preparation of [Ag(dchp)py]NO3, 2: The silver complex was prepared from 1 (132

mg, 0.30 mmol) dissolved in pyridine (10 mL). AgNO3 (102 mg, 0.60 mmol) was added

to this pyridine solution of the ligand and the mixture was stirred for 30 min. at room

temperature. The resulting solution was filtered and the filtrate was layered by diethyl

ether. Orange-red crystals of 2 were collected after four days.

1 [Ag(dchp)py]NO3, 2: Yield: 0.082 g (40 %) H NMR (300 MHz, [D5]pyridine): δ =

8.32 (q, J = 3.0 Hz, 4H), 8.11 (d, J = 7.2 Hz, 1H), 7.98 (t, J = 3.6 Hz, 1H), 7.91 (m, 2H)

7.36 (s, 5H), 7.12 (dd, J = 2.4, 5.1 Hz, 0.5H), 6.91 (s, 0.5 H), 6.80 (d, J = 6.6 Hz, 0.5H),

6.70 (s, J = 7.5 Hz, 0.5H), 6.62 (m, 0.5H), 6.45 (d, J = 2.1, 7.8 Hz, 0.5H), 109Ag NMR

(14.0 MHz, [D5]pyridine): δ = 377.1 ppm. (Referenced to AgNO3 in pyridine) High res.

ESI MS (positive ion) calculated for Ag(DCHP): 546.4 m/z; found: 547.0 m/z. The

material loses the axial solvent rapidly, with only 0.5 equivalents of pyridine in the bulk

material remaining upon sitting. CHN Analysis Calc. for C30.5H24.5N7.5O5Ag: C, 53.58;

H, 3.58; N, 15.36. Found: C, 53.60; H, 3.07; N, 15.37. Crystal data and structure

refinement parameters are summarized in Table 2.6.

Preparation of Cu(dchp-py)py, 3 Compound 3 was synthesized via three different routes. All reactions were carried out aerobically. Route 1: 1 (439 mg, 1.00 mmol) was dissolved in hot n-butanol (30.0 mL) and a methanol solution (5.00 mL) of

Cu(CH3CO2)2·2H2O (399 mg, 2.00 mmol) was added. The resulting mixture was

refluxed for 12 h. The resulting brown precipitate was collected by filtration and washed

with methanol and diethyl ether. The product was then dissolved in pyridine (10.0 mL)

92 and diffused by diethyl ether. Deep red crystals of 3 were collected after several days, and were suitable for X-ray diffraction (crystal 3a). Route 2: 1 (132 mg, 0.30 mmol) was dissolved in pyridine (10.0 mL) and Cu(CH3CO2)2·2H2O (120 mg, 0.60 mmol) was added

to the solution with stirring. The mixture turned to a deep red color almost immediately

and was heated to reflux for 24 h. The resulting solution was then layered with diethyl

ether after filtering off the precipitates. Deep red single crystals were collected after

several days. Single crystals for X-ray diffraction were grown from toluene/methanol

(crystal 3b) Route 3: 1 (132 mg, 0.30 mmol) and Cu(CH3CN)4BF4 (124 mg, 0.40 mmol)

was mixed in pyridine (10.0 mL). The mixture was stirred at room temperature for 30

min. The resulting pyridine solution was layered with diethyl ether, and the crystals that formed after several days were identical to those from route 1.

Cu(dchp-py)py, 3: Yield: (from route 1) 0.312 g, 48% 1H NMR (300MHz,

[D6]DMSO): δ = 8.34 (s, 4H), 8.18 (t, J = 7.8 Hz, 1H), 7.92 (d, J = 7.3 Hz, 2H), 7.74 (t, J

= 7.4 Hz, 2H), 7.64 (m, 4H), 7.46 (m, 4H), 7.31 (s, 1H), 7.21 (m, 4H), 6.98 (t, J = 7.8 Hz,

1H), 6.87 (t, J = 7.8 hz, 1H), 6.57 (d, J = 7.8 Hz) 2H), 6.42 (d, J = 7.8 Hz, 2H) 13C NMR

(300MHz, [D6]DMSO): 170.2, 164.8, 150.1, 149.9, 147.8, 147.4, 144.8, 140.1, 138.3,

136.1, 133.7, 130.6, 130.2, 126.3, 123.7, 121.5, 120.7, 118.4, 117.2. High res. ESI MS

+ (positive ion): calculated for H2dchp-py : 516.6 m/z; found: 516.2 m/z. CHN Analysis

Calc. for C83H67N16O5Cu2: C, 66.65; H, 4.51; N, 14.98. Found: C, 66.20; H, 4.31; N,

15.96. Crystal data and structure refinement parameters for 3a and 3b are summarized in

Tables 2.7 and 2.8 respectively.

Preparation of (H2dchp-py), 4: Compound 4 was isolated from the reaction of 1 (132

mg, 0.30 mmol) and Cu(CH3CO2)2·2H2O (120 mg, 0.60 mmol) in pyridine (10 mL). The

93 solution turned to a deep red color immediately and was refluxed for 24 h. The resulting

solution was slowly layered with CH2Cl2. The resultant solution turned from a red color

to blue upon oxidation to air. Yellow-orange crystals of 4 were collected after several

days.

1 (H2dchp-py), 4: Yield: 0.016g, 10% H NMR (300MHz, [D6]DMSO) δ = 11.17 (s,

2H), 10.09 (s, 0.5H), 9.62 (d, J = 6.0 Hz, 1H), 9.19 (d, J = 6.0 Hz, 1H), 8.93 (s, 1H), 8.50

(m, 2H), 8.36 (d, J = 7.3 Hz, 2H), 8.14 (t, J = 6.0 Hz, 1H), 8.05 (t, J = 6.0 Hz), 1H), 7.62

(m, 8H), 7.44 (m, 2H), 7.34 (d, J= 8.1 Hz, 2H), 7.07 (d, J = 6.9 Hz, 1H), 6.87 (m, 1H),

13 6.78 (d, J = 7.9 Hz, 2H); C NMR (300MHz, [D6]DMSO) δ = 169.5, 166.8, 148.7,

147.9, 138.8, 37.0, 136.5, 132.9, 131.6, 131.3, 129.2, 128.6, 125.4, 122.1, 122.0, 118.8,

+ 117.2, 116.9. High res. ESI MS (positive ion) calculated for H2dchp-py : 516.6 m/z;

found: 516.2 m/z. CHN Analysis Calc. for C33H32N7OCl3: C, 61.64; H, 4.08; N, 15.25.

Found: C, 61.44; H, 4.10; N, 14.48. Crystal data and structure refinement parameters are

summarized in Table 2.9.

Results and Discussion

Macrocycle H2dchp was synthesized as described in the literature with moderate

124 yields. Crystallization of the H2dchp macrocycle from various solvent systems led to

formation of five different crystal forms. The elucidation of the structures showed the flexibility and tautomerization of the system. Crystallization of the macrocycle from

formic acid resulted in the protonation of the bridging nitrogen atoms and deprotonation

Table 2.1. Crystal data and structure refinement for 1·HCOOH

Identification code dchp·HCOOH·

Empirical formula C32 H26 N6 O8 Formula weight 622.59 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 5.0864(12) Å α = 71.651(4)°. b = 11.867(3) Å β = 89.775(4)°. c = 12.441(3) Å γ = 78.614(4)°. Volume 697.4(3) Å3 Z 1 Density (calculated) 1.482 Mg/m3 Absorption coefficient 0.109 mm-1 F(000) 324 Crystal size 0.15 x 0.07 x 0.03 mm3 Theta range for data collection 1.73 to 28.27°. Index ranges -6<=h<=6, -15<=k<=15, -16<=l<=16 Reflections collected 6172 Independent reflections 3200 [R(int) = 0.0359] Completeness to theta = 28.27° 92.9 % Absorption correction Empirical SADABS Max. and min. transmission 0.9967 and 0.7981 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 3200 / 0 / 260 Goodness-of-fit on F2 1.130 Final R indices [I>2sigma(I)] R1 = 0.0665, wR2 = 0.1298 R indices (all data) R1 = 0.0919, wR2 = 0.1401 Largest diff. peak and hole 0.285 and -0.291 e.Å-3

Table 2.2. Crystal data and structure refinement for 1·H2O.

Identification code [dchp]2·H2O

Empirical formula C62 H62 N12 O7 Formula weight 543.62 Temperature 373(2) K Wavelength 0.71073 Å Crystal system Orthorhombic Space group Fddd Unit cell dimensions a = 7.579(4) Å α = 90°. b = 30.739(18) Å β = 90°. c = 45.94(3) Å γ = 90°. Volume 10702(11) Å3 Z 8 Density (calculated) 1.350 Mg/m3 Absorption coefficient 0.091 mm-1 F(000) 4592 Crystal size 0.30 x 0.10 x 0.01 mm3 Theta range for data collection 1.59 to 25.98°. Index ranges -9<=h<=9, -37<=k<=37, -56<=l<=56 Reflections collected 20009 Independent reflections 2647 [R(int) = 0.0490] Completeness to theta = 25.98° 100.0 % Absorption correction Empirical SADABS Max. and min. transmission 0.9991 and 0.6804 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 2647 / 10 / 250 Goodness-of-fit on F2 1.027 Final R indices [I>2sigma(I)] R1 = 0.0553, wR2 = 0.1560 R indices (all data) R1 = 0.0732, wR2 = 0.1665 Largest diff. peak and hole 0.521 and -0.371 e.Å-3

Table 2.3. Crystal data and structure refinement for 1·CH3CN.

Identification code dchp·2CH3CN

Empirical formula C32 H24 N8 Formula weight 520.59 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Orthorhombic Space group Pnma Unit cell dimensions a = 19.346(3) Å α = 90°. b = 11.1082(19) Å β = 90°. c = 11.979(2) Å γ = 90°. Volume 2574.3(8) Å3 Z 4 Density (calculated) 1.343 Mg/m3 Absorption coefficient 0.084 mm-1 F(000) 1088 Crystal size 0.40 x 0.20 x 0.20 mm3 Theta range for data collection 2.00 to 28.31°. Index ranges -24<=h<=25, -14<=k<=14, -15<=l<=15 Reflections collected 21740 Independent reflections 3320 [R(int) = 0.0913] Completeness to theta = 28.31° 98.8 % Absorption correction SADABS Max. and min. transmission 0.9834 and 0.7377 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 3320 / 0 / 245 Goodness-of-fit on F2 1.048 Final R indices [I>2sigma(I)] R1 = 0.0563, wR2 = 0.1346 R indices (all data) R1 = 0.0783, wR2 = 0.1502 Largest diff. peak and hole 0.295 and -0.276 e.Å-3

Table 2.4. Crystal data and structure refinement for 1·pyridine.

Identification code dchp·2py

Empirical formula C38 H28 N8 Formula weight 596.68 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 9.3840(12) Å α = 90°. b = 18.101(2) Å β = 100.076(3)°. c = 17.701(2) Å γ = 90°. Volume 2960.3(7) Å3 Z 4 Density (calculated) 1.339 Mg/m3 Absorption coefficient 0.083 mm-1 F(000) 1248 Crystal size 0.20 x 0.20 x 0.20 mm3 Theta range for data collection 1.62 to 28.31°. Index ranges -12<=h<=12, -23<=k<=24, -23<=l<=22 Reflections collected 25848 Independent reflections 7161 [R(int) = 0.0860] Completeness to theta = 28.31° 97.0 % Absorption correction SADABS Max. and min. transmission 0.9837 and 0.7239 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 7161 / 0 / 527 Goodness-of-fit on F2 1.026 Final R indices [I>2sigma(I)] R1 = 0.0614, wR2 = 0.1346 R indices (all data) R1 = 0.0935, wR2 = 0.1524 Largest diff. peak and hole 0.335 and -0.243 e.Å-3

Table 2.5. Crystal data and structure refinement for 1·DMF.

Identification code dchp·DMF

Empirical formula C31 H25 N7 O Formula weight 511.58 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 9.8240(15) Å α = 90°. b = 15.580(2) Å β = 92.519(3)°. c = 16.240(2) Å γ = 90°. Volume 2483.3(6) Å3 Z 4 Density (calculated) 1.368 Mg/m3 Absorption coefficient 0.087 mm-1 F(000) 1072 Crystal size 0.20 x 0.20 x 0.05 mm3 Theta range for data collection 1.81 to 28.36°. Index ranges -13<=h<=12, -20<=k<=19, -21<=l<=21 Reflections collected 21906 Independent reflections 5844 [R(int) = 0.1254] Completeness to theta = 28.36° 94.1 % Absorption correction ‘Empirical SADABS’ Max. and min. transmission 0.9956 and 0,7020 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 5844 / 0 / 452 Goodness-of-fit on F2 0.860 Final R indices [I>2sigma(I)] R1 = 0.0531, wR2 = 0.0905 R indices (all data) R1 = 0.0943, wR2 = 0.1009 Largest diff. peak and hole 0.267 and -0.237 e.Å-3

Figure 2.2 The structure of dication 1 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity with the exception of the internal CH groups and the NH groups.

Figure 2.3 The structure of [1]2·H2O with 50% thermal ellipsoids.

Hydrogen atoms have been omitted for clarity. 100

Figure 2.4 The structure of 1·CH3CN with 50% thermal ellipsoids. Hydrogen atoms have

been omitted for clarity with the exception of the internal NH groups.

Figure 2.5 The structure of 1·pyridine with 50% thermal ellipsoids. Hydrogen atoms

have been omitted for clarity with the exception of the internal NH groups.

101

Figure 2.6 The structure of 1·DMF with 35% thermal ellipsoids.

of the internal nitrogen atoms (Figure 2.2). The resultant structure is planar due to the deprotonated internal nitrogen atoms. Crystallization of the macrocycle from MeOH gave a dimer of the macrocycle with a water molecule sandwiched between two macrocycles (Figure 2.3). Contrary to the deprotonated form, the macrocyle is not planar in this structure due to the steric hinderance of the internal ptrotons. The internal N-H

units point toward the water molecule forming hydrogen bonding. The asymmetric unit

of 1·CH3CN includes two solvent molecules and one macrocycle (Figure 2.4). One of the

solvent molecules engages in hydrogen bonding with the internal N-H units. Deviation

of the benzene unit in this case is larger than that of 1·H2O with a value of -63.65º,

whereas this value is -40.44 for 1·H2O. In the asymmetric unit of 1·Pyridine there are

two pyridine units per macrocycle (Figure 2.5). Internal nitrogen atoms are protonated 102 and one of the pyridine sits atop of the macrocycle. But, unlike the other solvents, pyridine locates at the other side of the ring with the N-H units pointing to reverse direction. In the case of 1·DMF, there is only one solvent molecule per macrocycle,

(Figure 2.6) and DMF does not engage into any interactions with the macrocycle. The macrocycle deviates from planarity with a dihedral angle of -52.00º.

Although there are only two nitrogen atoms for metal coordination within the core of the ring, macrocycle 1 does bind metal ions (Scheme 2.2). The reaction of 1 with

I Ag(NO3) affords Ag(dchp)py(NO3), a pseudo-three coordinate Ag species where the two interior nitrogen atoms and one axial pyridine is bound to the metal center. The presence

I of a NO3¯ counter ion in the unit cell of 2 indicates that the ligand is binding to the Ag

N N

NH HN

N N iii

1 N N N N N N Cu C H Ag N N H - H NO3 N H N C N N N

N N iii H 2 3 NN

NH HN

N - N N Cl

4 Scheme 2.2 Synthesis of compounds 2, 3, and 4, starting from 1.

I II Reagents: i) AgNO3, pyridine, MeOH, air; ii) Cu or Cu salt, pyridine,

MeOH, air; c) CH2Cl2, air.

103 cation in a neutral tautomeric form. It appears that two of the external nitrogen atoms are

protonated in this complex to achieve charge balance. This is in contrast to 3 (vide infra) where the positive charges of the pyridinium moiety and the CuI ion are balanced by the

charges of the two deprotonated core nitrogen atoms. The metal macrocycle nitrogen

bond lengths are relatively long (~2.36 Å), in large part due to the steric interaction of the

benzene rings. The Ag-C distances are 2.62 and 2.71 Å, too long for a strong silver-arene

interaction,126 but the carbon atoms are within close proximity of the metal ion. The

metal is significantly pushed out of the plane of the macrocycle (as measured by the

plane of the four meso nitrogen atoms) by 1.35 Å, a larger distance than seen in the

metallo-N-confused porphyrins.25,26 This compound is diamagnetic, and the Ag NMR of

this compound is 377 ppm.

II I The reactions of 1 with a variety of Cu and Cu salts such as Cu(CH3COO)2·2H2O

II and Cu(CH3CN)4BF4 in the presence of air and pyridine do not produce stable Cu compounds. Instead, the macrocycle ring reacts with pyridine after metallation to form the species Cu(dchp-py)(py), 3. Species 3 is a CuI compound where an equivalent of pyridine has formed a bond with one of the two internal C-H positions, forming a C-N bond (Scheme 2.2). We isolated two crystal forms of this complex: one with one molecule per asymmetric unit (Figure 2.8) and one with two molecules per asymmetric unit (Figure 2.9). As in the Ag compound, the CuI center is pseudo-three coordinate,

bound to both of the internal nitrogen atoms from the ring and one axial pyridine, with

bond distances ranging between 2.03 and 2.08 Å.

Both the remaining interior C-H bond and the new C-N bond are in close proximity

to the metal center in this new complex, with average bond distances of ~2.41 Å for the

104

Table 2.6. Crystal data and structure refinement for 2.

Identification code Ag(dchp)pyNO3·2H2O·0.5Et2O

Empirical formula C35 H32 Ag N8 O6 Formula weight 768.56 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P1 Unit cell dimensions a = 8.7694(13) Å α = 73.503(2)°. b = 13.3129(19) Å β = 81.501(2)°. c = 14.937(2) Å γ = 73.782(2)°. Volume 1601.1(4) Å3 Z 2 Density (calculated) 1.463 Mg/m3 Absorption coefficient 0.680 mm-1 F(000) 716 Crystal size 0.60 x 0.20 x 0.20 mm3 Theta range for data collection 1.43 to 28.31°. Index ranges -11<=h<=11, -17<=k<=17, -19<=l<=19 Reflections collected 14343 Independent reflections 7400 [R(int) = 0.0381] Completeness to theta = 28.31° 92.7 % Absorption correction Empirical SADABS Max. and min. transmission 0.8760 and 0.7070 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 7400 / 0 / 431 Goodness-of-fit on F2 1.048 Final R indices [I>2sigma(I)] R1 = 0.0472, wR2 = 0.1029 R indices (all data) R1 = 0.0595, wR2 = 0.1075 Largest diff. peak and hole 0.764 and -0.936 e.Å-3

105

Figure 2.7 The structure of 2 with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups.

former and 2.74 Å for the latter. As in species 2, the metal is pushed out of the plane of the macrocycle ring by 0.77 Å, just over half of the distance seen in compound 2. The chemical shift of the proton in the interior C-H bond changes slightly from that seen in the unmetallated ligand 1, moving from 6.77 ppm to 7.31 ppm. We believe that this reaction proceeds via CuII assisted elimination of hydride. The hydride reduces the metal to CuI, which is more stable in the low coordinate environment of the ligand.127 An alternate mechanism to ONSH might involve the initial formation of a highly oxidized

CuIII center with a metal carbon bond (via activation of one of the internal C-H groups). 106

Table 2.7. Crystal data and structure refinement for 3a.

Identification code Cu(dchp-py)py·3py

Empirical formula C48 H35 Cu N10 Formula weight 815.41 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/c Unit cell dimensions a = 14.2673(11) Å α = 90°. b = 13.1552(10) Å β = 126.441(3)°. c = 24.0192(13) Å γ = 90°. Volume 3626.7(4) Å3 Z 4 Density (calculated) 1.349 Mg/m3 Absorption coefficient 0.647 mm-1 F(000) 1520 Crystal size 0.40 x 0.40 x 0.10 mm3 Theta range for data collection 1.77 to 28.32°. Index ranges -18<=h<=19, -17<=k<=17, -31<=l<=31 Reflections collected 31791 Independent reflections 8715 [R(int) = 0.0360] Completeness to theta = 28.32° 96.6 % Absorption correction Empirical SADABS Max. and min. transmission 0.9381 and 0.7818 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8715 / 0 / 482 Goodness-of-fit on F2 1.047 Final R indices [I>2sigma(I)] R1 = 0.0438, wR2 = 0.1013 R indices (all data) R1 = 0.0525, wR2 = 0.1054 Largest diff. peak and hole 0.592 and -0.437 e.Å-3

107

Table 2.8. Crystal data and structure refinement for 3b.

Identification code 2[Cu(dchp-py)]·5H2O·p-xylene

Empirical formula C84 H70 Cu2 N16 O5 Formula weight 1510.64 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 10.2762(11) Å α = 90°. b = 43.463(5) Å β = 101.059(2)°. c = 16.4235(17) Å γ = 90°. Volume 7199.1(13) Å3 Z 4 Density (calculated) 1.394 Mg/m3 Absorption coefficient 0.658 mm-1 F(000) 3136 Crystal size 0.20 x 0.10 x 0.04 mm3 Theta range for data collection 1.35 to 28.29°. Index ranges -13<=h<=13, -57<=k<=57, -21<=l<=21 Reflections collected 63799 Independent reflections 17298 [R(int) = 0.0659] Completeness to theta = 28.29° 96.7 % Absorption correction Empirical SADABS Max. and min. transmission 0.9742 and 0.8796 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 17298 / 0 / 1244 Goodness-of-fit on F2 1.073 Final R indices [I>2sigma(I)] R1 = 0.0568, wR2 = 0.1208 R indices (all data) R1 = 0.0741, wR2 = 0.1282 Largest diff. peak and hole 0.614 and -0.623 e.Å-3

108

Figure 2.8 The structure of 3a with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups.

Figure 2.9 The structure of second crystal form of 3b with 35% thermal ellipsoids.

Hydrogen atoms have been omitted for clarity. 109

Chelation of pyridine, and subsequent reductive elimination would then produce the

pyridine adduct 3.

Recrystallization of 3 in dichloromethane in air results in the loss of pyridine and

demetallation to yield the free ligand. The free base species H2dchp-py, 4, retains the

pendant pyridine at the interior carbon position. The C-N bond in this species has a bond

length of 1.46 Å, which is identical to that seen in both crystal forms of the copper

complex. We believe that dioxygen reacts with 3 and the solvent (dichloromethane),

producing CuCl2 and the protonated free base 4.

In conclusion, metallation of dicarbahemiporphyrazine 1 with silver and copper

metals resulted in formation of metal-complexes with organometallic side-on

interactions. The reaction of 1 with the redox active metal copper activates an interior C-

H bond to nucleophilic substitution. This reaction proceeds through an ONSH mechanism, where the CuII accepts a hydride and is reduced to CuI. Similar internal C-H

activation reactions were observed earlier with the several other carbaporphyrinoids. For

example the reaction of m-benziporphyrin with silver(I) ion resulted in pyridination of the

internal carbon atom.79 N-confused porphyrins also exhibited the same type of reactivity

toward metallation under specific conditions. Exposure of Fe(NCTPP)Br to dioxygen

resulted in the oxygenation of the internal carbon atom56 and reaction of NCTPP with

copper acetate under aerobic conditions resulted in the degredation of the macrocycle via

40 O2 attack of the C-H unit.

110

Table 2.9. Crystal data and structure refinement for 4.

Identification code (dchp-py)Cl3

Empirical formula C33 H24 Cl3 N7 Formula weight 624.94 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 8.5234(11) Å α = 91.287(2)°. b = 13.7264(17) Å β = 97.805(2)°. c = 14.0816(18) Å γ = 96.667(2)°. Volume 1620.0(4) Å3 Z 2 Density (calculated) 1.281 Mg/m3 Absorption coefficient 0.317 mm-1 F(000) 644 Crystal size 0.40 x 0.40 x 0.20 mm3 Theta range for data collection 1.49 to 26.00°. Index ranges -10<=h<=10, -16<=k<=16, -17<=l<=17 Reflections collected 12673 Independent reflections 6313 [R(int) = 0.0353] Completeness to theta = 26.00° 99.2 % Absorption correction Empirical SADABS Max. and min. transmission 0.9394 and 0.7437 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 6313 / 0 / 423 Goodness-of-fit on F2 1.109 Final R indices [I>2sigma(I)] R1 = 0.0814, wR2 = 0.2100 R indices (all data) R1 = 0.0983, wR2 = 0.2196 Largest diff. peak and hole 1.269 and -0.601 e.Å-3

111

Figure 2.10 The structure of 4 with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups and the NH groups.

Nitrogen atoms: blue thermal ellipsoids; carbon atoms: grey thermal ellipsoids

112

CHAPTER III

COMPLEXES OF LOW-COORDINATE MIDDLE TRANSITION METAL

DICARBAHEMIPORPHYRAZINE

The synthesis of low-coordinate metal ions has been a focus of bioinorganic

chemists due to their important roles in the active sites in enzymes and proteins such as

the MoFe cofactor of nitrogenase.128 Recent developments in the synthesis of porphyrin

isomers and core modified analogs have resulted in macrocycles with carbon atoms at

one or more core positions such as N-confused porphyrins,27,28,32 benziporphyrins76-78 and azuliporphyrins.99,105,129 These new macrocycles are able to generate protected low-

coordinate metal geometries.25,26,31 In the course of studying the coordination chemistry

of these porphyrin analogs, a unique metal binding geometry was observed. While in

some cases the internal C-H bond is activated to form a direct metal-carbon bond, in

other cases the C-H bond does not lose its proton and instead participates in the

coordination via forming a side-on agostic-type interaction.25,26 This type of interaction

could help to stabilize low-coordinate metal centers that are relevant to those in

biological systems.

In the early 1950s, Linstead and co-workers synthesized the hemiporphyrazine macrocycle, an expanded phthalocyanine analog where two of the indolene units had been replaced by pyridines.130 Because this modification causes a disturbance in the π-

113 electron delocalization pathway, these systems do not have porphyrin-like aromaticity.

These macrocycles coordinate to metal ions in a square planar geometry, similar to normal porphyrins and phthalocyanines.131 Replacing the pyridines with benzene rings

leads to formation of two carbaporphyrinoids: one is mono-N deficient and the other is

di-N deficient and can be considered as monocarbahemiporphyrazine (H3mchp) and

132 dicarbahemiporphyrazine (H2dchp) respectively. Although these two rings were first

presented over half of a century ago, the coordination chemistry of these free bases

remains largely unexplored. We recently presented a report on the copper(I) and silver(I)

133 complexes of H2dchp, 1. In both cases, the macrocycle retained internal C-H bonds

upon initial metallation.

N N N N N N N N H NH HN NH HN NH HN

N N N N N N N

a. b. c.

Figure 3.1 The structures of (a) hemiporphyrazine and (b) mono-N deficient analogue

monocarbahemiporphyrazine (H3mchp) (c) di-N deficient analogue

dicarbahemiporphyrazine (H2dchp)

Experimental

General Methods: Unless otherwise noted, all reagents and solvents were purchased

from Sigma, Aldrich, Acros Organics, Strem and used without further purification. Mass

114 spectra were recorded using an LCT electrospray spectrophotometer at the Mass

Spectrometry and Proteomics Facility of Ohio State University. Elemental analysis was

conducted at the University of Illinois, School of Chemical Sciences Microanalysis

Laboratory. Magnetic susceptibility measurements were made on a Johnson Matthey

Magnetic Susceptibility Balance using ultra thin bore sample tubes. Infrared

spectroscopy was carried out on a Nicolet Nexus 870 Fourier transform spectrometer.

Single crystal X-ray diffraction data was collected with 100 K (Bruker KRYO-

FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped

with a Mo-target X-ray tube (λ = 0.71073 Å) operated at 2000 watts power. The detector

was placed at a distance of 5.009 cm from the crystal. Integration and refinement of

crystal data was done using Bruker SAINT software package and Bruker SHELXTL

(version 6.1) software package, respectively. Absorption correction was completed using the SADABS program. Crystals were placed in paratone oil before removal from the glove box and mounted on a plastic loop in the oil.

Preparation of Dicarbahemiporphyrazine (H2dchp): This synthesis is based on a procedure by Linstead et al.132 The details of the synthesis can be found in Chapter II.

Preparation of (dicarbahemiporphyrazine) (pyridine) Manganese(II), MnIIdchppy:

H2DCHP was weighed into a vial before transporting it into the glove box. H2dchp (50.0 mg, 0.114 mmol) was dissolved in 2 mL of pyridine and stirred. This solution was yellow in color. Mn2(CO)10 (25.0 mg, 0.06 mmol) was weighed in the glove box and

dissolved in 2 mL of pyridine to produce a light yellow colored solution. The Mn2(CO)10 solution was added dropwise over one minute and the mixture was refluxed. The solution turned to a red color within 10 min. of reflux and the reaction was completed in

115 two hours. The resultant pyridine solution was layered with anhydrous hexanes to afford

transparent yellow crystals.

II Mn (dchp)py: Yield: 30 mg (46%) This complex is paramagnetic (µeff = 6.11, spin

state S = 5/2), High res. ESI MS (positive ion) calculated for K[Mn(dchp)py]: 610.6 m/z; found: 610.2 m/z. CHN Analysis Calc. for: C38H26MnN8: C, 70.26, H, 4.03, N, 17.25.

Found: C, 70.29, H, 4.33, N, 16.85. Crystal data and structure refinement parameters are summarized in Table 3.1.

Preparation of (dicarbahemiporphyrazine) (pyridine) Cobalt(II), CoII(dchp)py:

H2dchp was weighed into a vial before transporting it into the glove box. H2dchp (50.0

mg, 0.114 mmol) was dissolved in 2 mL of pyridine and stirred. This solution was

yellow in color. Co2(CO)8 (22.0 mg, 0.06 mmol) was weighed in the glove box and

dissolved in 2 mL of pyridine to produce a red colored solution. The Co2(CO)10 solution was added dropwise over 1 min. and the reaction was refluxed. The solution turned to a green color within 10 min. of reflux and the reaction was completed in 4 h. The resultant pyridine solution was layered with anhydrous hexanes to afford transparent yellow crystals covered with a green film.

II Co (dchp)py: Yield: 45.0 mg (69%). This complex is paramagnetic (µeff = 4.85, spin state S = 3/2), High res. ESI MS (positive ion) calculated for [Co(dchpH)]+: 498.4 m/z; found: 498.1 m/z. CHN Analysis Calc. for: C33H21CoN7: C, 68.99, H, 3.68, N,

17.07. Found: C, 68.83, H, 3.56, N, 16.73. Crystal data and structure refinement

parameters are summarized in Table 3.2.

II Preparation of (dicarbahemiporphyrazine) (pyridine) Iron(II), Fe (dchp)py: H2dchp

was weighed into a vial before transporting it into the glove box. H2dchp (50.0 mg, 0.110 116 mmol) was dissolved in 2 mL of pyridine and stirred. The solution was yellow in color.

The Fe(CO)5 solution (0.015 mL, 0.110 mmol) was then added drop wise over 1 min. and

the reaction was refluxed. The solution had turned to a green color within 5 min. of

reflux and the reaction was completed in 4 h. The resultant pyridine solution was layered

with anhydrous hexanes to afford transparent orange crystals.

II Fe (dchp)py: Yield: 20.0 mg (31%) This complex is paramagnetic (µeff = 5.15, spin

state S = 2). Under the experimental conditions of mass spectroscopy, the complex

+ demetallates High res. ESI MS (positive ion) calculated for [(H2dchpH)] : 439.5 m/z; found: 441.2 m/z. Anal. CHN Analysis Calc. for: C38H26FeN8: C, 70.16, H, 4.03, N,

17.22. Found: C, 70.66, H, 4.02, N, 16.97. Crystal data and structure refinement

parameters are summarized in Table 3.3. High res. ESI MS (positive ion) calculated for

[Co(dchpH)]+: 498.4

Results and Discussion

As in many other carbaporphyrinoids, reaction of free base H2dchp with simple

metal salts of manganese, iron, and cobalt do not produce the desired metal adducts. As

in the work with N-confused porphyrins, the reaction of the macrocycles with metal

carbonyls under an argon atmosphere resulted in the formation of the desired metallated products. Figure 3.2 shows the reactions of H2dchp free base with dimanganese

dodecacarbonyl, iron pentacarbonyl and dicobalt octacarbonyl. All three reactions

produced metallated products in 46, 31, and 69% yield, respectively.

117 N N N N M N Mn2(CO)10 py H NH HN + N N Co2(CO)8 anaerobic reflux N H N N N

1 M = Mn+2, Co+2 Fe(CO)5 anaerobic py 5, 6 reflux N N

Fe N H N N N H

N H H

Figure 3.2 Synthesis of compounds 5, 6 and 7 starting from 1.

The reaction mixture of H2dchp with either manganese or cobalt carbonyl in pyridine under reflux conditions in argon resulted in color changes to both solutions within minutes. Both reactions produced crystalline products upon diffusion with hexanes, and elucidation by single crystal X-ray methods showed nearly identical structures for both products. Each complex has a formula of M(dchp)py, where a pyridine occupies the axial position in the resultant metalloporphyrinoid (Figure 3.3 and Figure 3.4). The macrocycle ring coordinates to the metal through the two internal nitrogen atoms, for a total of three strongly coordinating nitrogens bound to each metal. The average

118

Table 3.1. Crystal data and structure refinement for 5.

Identification code Mn(dchp)py·3py

Empirical formula C43 H31 Mn N9 Formula weight 728.71 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 16.4993(15) Å α = 90°. b = 12.4912(11) Å β = 110.685(2)°. c = 17.8396(16) Å γ = 90°. Volume 3439.7(5) Å3 Z 4 Density (calculated) 1.407 Mg/m3 Absorption coefficient 0.431 mm-1 F(000) 1508 Crystal size 0.40 x 0.30 x 0.10 mm3 Theta range for data collection 1.45 to 28.30°. Index ranges -21<=h<=21, -16<=k<=16, -23<=l<=23 Reflections collected 30031 Independent reflections 8276 [R(int) = 0.0695] Completeness to theta = 28.30° 96.9 % Absorption correction Empirical SADABS Max. and min. transmission 0.9581 and 0.7023 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8276 / 0 / 579 Goodness-of-fit on F2 1.032 Final R indices [I>2sigma(I)] R1 = 0.0633, wR2 = 0.1665 R indices (all data) R1 = 0.0852, wR2 = 0.1806 Largest diff. peak and hole 0.804 and -0.581 e.Å-3

119

Table 3.2. Crystal data and structure refinement for 6.

Identification code Co(dchp)py·py

Empirical formula C33 H21 Co N7 Formula weight 574.50 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 16.6549(16) Å α= 90°. b = 12.5612(12) Å β= 111.490(2)°. c = 17.5184(17) Å γ = 90°. Volume 3410.2(6) Å3 Z 4 Density (calculated) 1.119 Mg/m3 Absorption coefficient 0.533 mm-1 F(000) 1180 Crystal size 0.50 x 0.40 x 0.20 mm3 Theta range for data collection 2.05 to 28.35°. Index ranges -21<=h<=21, -16<=k<=15, -23<=l<=23 Reflections collected 29223 Independent reflections 8016 [R(int) = 0.0554] Completeness to theta = 28.35° 94.1 % Absorption correction Empirical SADABS Max. and min. transmission 0.9010 and 0.8140 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8016 / 0 / 370 Goodness-of-fit on F2 0.914 Final R indices [I>2sigma(I)] R1 = 0.0521, wR2 = 0.1255 R indices (all data) R1 = 0.0872, wR2 = 0.1377 Largest diff. peak and hole 0.427 and -0.477 e.Å-3

120

Figure 3.3 The structure of 5 with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups.

Figure 3.4 The structure of 6 with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups.

121 equatorial distances are 2.09 Å and 1.98 Å for the manganese and cobalt species,

respectively. The axial pyridine has slightly elongated bonds, with 2.14 Å for

Mn(dchp)py and 2.05 Å for Co(dchp)py. The internal C-H bonds of the benzene rings do

not activate upon metal coordination and are found at a longer distance from the metals in

both compounds. The M-C distances average 2.48 and 2.42 Å for the manganese and

cobalt products, respectively. The geometries of these two compounds can be considered

as low-coordinate, with the two benzene rings of the hemiporphyrazine protecting the

metal site from further coordination. Recently, “agostic” type C-H bonds have been

frequently observed in the metal binding chemistry of the carbaporphyrinoids, including

N-confused porphyrins and benziporphyrins. Interestingly, the H2dchp ligand is the first

example of a porphyrinoid that regularly forms two porphyrinic agostic bonds from

separate subunits upon metal binding.

The reaction of the H2dchp macrocycle with iron pentacarbonyl under same

conditions however, resulted in metalation with a concomitant reduction of the ring. As

in the manganese and cobalt reactions, a low-coordinate geometry is formed with only

three nitrogen atoms strongly coordinating to the metal ion (Figure 3.5). The equatorial

Fe-N distances measure 1.99 and 2.02 Å and the axial pyridine nitrogen is found at 2.07

Å from the metal. The internal carbon sites lie at an average of 2.43 Å from the metal.

The reduction of the ring occurs at one of the indolene Schiff base sites, such that both the α-carbon and meso nitrogen are sp3 hybridized, which is readily observed in the crystal structure.

122

H H d d / / A A d d

Field (Gauss)

Figure 3.5 X-band (9.4 GHz) EPR spectra of 5 in frozen pyridine/toluene,

1 mM MnII(dchp)py, T = 100 K

All three complexes have metals in the M(II) oxidation state. Magnetic susceptibility experiments indicate that all three have high-spin configurations, with spins of 5/2, 2 and 3/2 from µeff values of 6.11, 5.15, and 4.83 for the Mn, Fe, and Co complexes respectively. Preliminary EPR studies on the Mn(dchp)py complex at 100K in frozen pyridine/toluene solution afforded a spectrum that is consistent with either a high-spin manganese(II) ion or a mixed 5/2 and 3/2 spin system.

Generation of low-coordinate transition metal complexes is not favored but is possible via complexation of the metal with bulky ligands. There are limited examples of

123

Table 3.3. Crystal data and structure refinement for 7.

Identification code Fe(dchp)py·py

Empirical formula C33 H23 Fe N7 Formula weight 573.43 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group C2/c Unit cell dimensions a = 31.610(5) Å α = 90°. b = 11.9896(18) Å β = 107.095(3)°. c = 18.876(3) Å γ = 90°. Volume 6838.0(18) Å3 Z 8 Density (calculated) 1.114 Mg/m3 Absorption coefficient 0.470 mm-1 F(000) 2368 Crystal size 0.30 x 0.10 x 0.10 mm3 Theta range for data collection 1.35 to 28.28°. Index ranges -41<=h<=41, -15<=k<=15, -24<=l<=25 Reflections collected 29598 Independent reflections 8215 [R(int) = 0.0647] Completeness to theta = 28.28° 96.8 % Absorption correction Empirical SADABS Max. and min. transmission 0.9545 and 0.5615 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8215 / 0 / 378 Goodness-of-fit on F2 0.966 Final R indices [I>2sigma(I)] R1 = 0.0485, wR2 = 0.1162 R indices (all data) R1 = 0.0753, wR2 = 0.1244 Largest diff. peak and hole 0.573 and -0.503 e.Å-3

124

Figure 3.6 The structure of 7 with 35% thermal ellipsoids. Hydrogen atoms have been

omitted for clarity with the exception of the internal CH groups and the NH groups.

low-coordinate middle-transition metal ions stabilization by sterically hindered ligands.

Complexation of the metals with bulky thiolate,134 alkoxide135 or amide136 based ligands

can result in either monomeric or dimeric complexes. The bis(trimethylsilyl)amide ligand is a well studied ligand used to stabilize low coordination numbers.136 The reaction of this ligand with Mn(II), Fe(II) or Co(II) generated monomeric complexes and the metal centers are complexed by three ligands via the nitrogen atoms. Similar chemistry is also observed with the β-diketimine ligand in the presence of a bis(trimethylsilyl)amide ligand.137 The metal centers in these compounds are again three-

coordinate with the coordination of the bidentate β-diketimide ligand and a

bis(trimethylsilyl)amide unit.

125 This study demonstrates the stabilization of low-coordinate Mn(II), Fe(II) and Co(II)

ions via the coordination of the porphyrinoid system, dicarbahemiporphyrazine. These

systems are also examples of agostic-type interactions which had been observed earlier in other carbaporphyrinoid systems.

126

CHAPTER IV

Co(II) AND Co(III) COMPLEXES OF MONOCARBAHEMIPORPHYRAZINE

The modification of porphyrin by replacing core nitrogen atoms with carbon atoms results in macrocycles with intrinsic electronic properties16,17,18,19 as well as the unusual

coordination modes.25,26,31 This area of research accelerated with the discovery of N- confused porphyrin and its unique coordination chemistry.25,26,31 Incorporation of one or

more carbon atoms into the core of the porphyrins results in compounds with the

possibility of forming organometallic complexes upon metallation. One of the methods

for introducing a carbon atom at the core of the macrocycle is to replace one of the

pyrrole rings with a benzene unit. In normal porphyrin, this leads to the formation of

benziporphyrin, shown in Figure 4.1b. The m-benziporphyrins were first reported in

1994 by Berlin and Breitmaier76 and in later years synthetic routes were improved77 and meso-substituted variants were also reported.78 Although this macrocycle no longer has

porphyrin-like aromaticity, a wide variety of interesting metallation chemistry was

observed, including the metals Pd(II), Pt(II),78 Zn(II), Cd(II), Hg(II), Ni(II)81,82 and

Fe(II).83

The metal chemistry of azaporphyrin analogs of core modified porphyrins are being

studied.133 The azaporphyrin analog of benziporphyrin was first synthesized over half a

127

N N N H N NH HN N N N H H H N N N N N

a. b. c.

Figure 4.1 Structures of (a) Normal porphyrin (b) m-benziporphyrin

and (c) monocarbahemiporphyrazine.

century ago by Linstead and coworkers in their investigations of the synthesis of

phthalocyanine analogs using diiminoisoindolene.29 This family of macrocycles is

known collectively as the hemiporphyrazines,138 and the benziporphyrin analog can be

designated as a monocarbahemiporphyrazine (H3mchp) (Figure 4.1c). This ring closely

resembles m-benziporphyrin, where the meso carbons are replaced with nitrogens, and

the number of internal protons is four rather than two. Although the H3mchp macrocycle

has been known for decades and is readily synthesized, its metallation chemistry has not

been extensively investigated. In this chapter, we are reporting the cobalt coordination

chemistry of monocarbahemiporphyrazine.

Experimental

General Methods: Unless otherwise noted, all reagents and solvents were purchased

from Sigma, Aldrich, Acros Organics, Strem and used without further purification. Mass

spectra were recorded using an LCT electrospray spectrophotometer at the Mass

Spectrometry and Proteomics Facility of Ohio State University. Elemental analysis was

128 conducted at the University of Illinois, School of Chemical Sciences Microanalysis

Laboratory. Magnetic susceptibility measurements were made on a Johnson Matthey

Magnetic Susceptibility Balance using ultra-thin bore sample tubes.

Single crystal X-ray diffraction data was collected with 100 K (Bruker KRYO-

FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped with a Mo-target X-ray tube (λ = 0.71073 Å) operated at 2000 watts power. The detector was placed at a distance of 5.009 cm from the crystal. Integration and refinement of crystal data was done using Bruker SAINT software package and Bruker SHELXTL

Preparation of Monocarbahemiporphyrazine (H3mchp): This two-step synthesis is based on the procedure by Linstead et al.29 (Scheme 4.1). Diiminoisoindoline (14.5 g;

0.10 mol) was dissolved in 100 mL of hot ethanol. The solution was cooled down to 0 ºC and 0.5 equivalents of 1,4-diaminobenzene (5.40 g, 0.05 mol) were added to the solution dropwise. The resulting mixture was stirred at 0ºC for 24 h. and a yellow-white precipitate was formed. The resulting precipitate was washed with cold ethanol and diethyl ether. Yield 12.5 g, 63%.

The intermediate compound (12.0 g; 0.03 mol) and diaminobenzene (3.30 g, 0.03 mol) was dissolved in 100 mL n-butanol and heated to reflux for 24 h. The resulting deep purple-red precipitate of H3mchp was washed with ethanol and diethyl ether. Yield

6.70 g, 51%.

129 NH

NH NH N N HN NH H N NH H2N NH2 2 2 NH NH NH HN o ethanol, 0 C NH HN n-butanol, 100 oC H N NH N N NH HN

29 Scheme 4.1 Synthesis of monocarbahemiporphyrazine (H3mchp).

Preparation of (monocarbahemiporphyrazine) (pyridine) Cobalt(II), CoII(mchp)py:

Free base H3mchp was weighed into the reaction flask before transporting into glove box.

H3mchp (125 mg, 0.26 mmol) was dissolved in 10 mL of pyridine. The resultant solution

was reddish brown in color. Co2(CO)8 (50.0 mg, 0.130 mmol) was weighed in the glove

box and dissolved in 2 mL of pyridine producing a red colored solution. The Co2(CO)8 solution was added dropwise over 1 min. and the reaction was refluxed for 3 h. The color changed to a darker brown color. The solution was filtered through a fine frit and the solution was layered with anhydrous hexanes to afford red crystalline blocks within three days.

Co(mchp)py: Yield: 71 mg (45%) This complex is paramagnetic. (µeff = 1.80, spin

state S = ½) High resolution ESI MS (positive ion) calculated for Co(mchp): 532.4 m/z, found: 532.0 m/z. CHN analysis calc. for C35H22N8Co: C, 69.46, H, 3.79, N, 18.23.

Found C, 70.26, H, 4.50, N, 18.87. Crystal data and structure refinement parameters are summarized in Table 4.1.

Preparation of (monocarbahemiporphyrazine) (dipyridine) Cobalt(III),

III III Co (mchp)py2: Air exposure of a pyridine solution of Co (mchp)py results change in

130 the color of the solution from red to green. Crystallization of this product from

pyridine/hexanes resulted in the formation of green block crystals.

Co(mchp)py2: High res. ESI MS (positive ion) calculated for Co(mchp): 532.4 m/z,

found: 532.0 m/z. Analysis calc. for C30H17CoN7: C, 67.42, H, 3.21, N, 18.34. Found: C,

67.37, H, 3.36, N, 16.35. Crystal data and structure refinement parameters are

summarized in Table 4.3.

Preparation of (monocarbahemiporphyrazinehydroxide) (dipyridine) Cobalt(III),

III Co (mchp-OH)py2: Co(OAc)2·4H2O (100 mg, 0.400 mmol) and mchp (190 mg, 0.390

mmol) was dissolved in 10 mL of DMF. The resulting mixture was refluxed for 24 h.

and added to about 30 mL of H2O. The deep brown precipitate was collected, washed

with ethanol and air dried. The product was dissolved in 1:1 pyridine/p-xylene solvent mixture and the solvent was slowly evaporated. Deep red crystals are formed within one week.

Co(mchp-OH)py2: Yield 56.4 mg (20%) High resolution ESI MS (positive ion)

550.0 m/z for [Co(MCHPOH)] CHN analysis calculated for C30H17N7CoO1: C, 65.44, H,

3.11, N, 17.82. Found C, 64.19, H, 3.52, N, 16.71. Crystal data and structure refinement

parameters are summarized in Table 4.2.

Results and Discussion

The H3mchp macrocycle can be synthesized as described in the literature in

moderate yields.29 Cobalt can be cleanly inserted into the ring by using cobalt carbonyl

131

Table 4.1. Crystal data and structure refinement for 9.

Identification code Co(mchp)py

Empirical formula C45 H32 Co N10 Formula weight 771.74 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 15.4472(15) Å α = 90°. b = 18.5201(18) Å β = 116.181(2)°. c = 16.7680(17) Å γ = 90°. Volume 4304.9(7) Å3 Z 4 Density (calculated) 1.324 Mg/m3 Absorption coefficient 0.449 mm-1 F(000) 1768 Crystal size 0.30 x 0.20 x 0.05 mm3 Theta range for data collection 1.49 to 28.26°. Index ranges -20<=h<=20, -23<=k<=24, -21<=l<=22 Reflections collected 36370 Independent reflections 9933 [R(int) = 0.0375] Completeness to theta = 28.26° 93.1 % Absorption correction Empirical SADABS Max. and min. transmission 0.9783 and 0.8420 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 9933 / 0 / 513 Goodness-of-fit on F2 1.036 Final R indices [I>2sigma(I)] R1 = 0.0401, wR2 = 0.1010 R indices (all data) R1 = 0.0502, wR2 = 0.1054 Largest diff. peak and hole 0.697 and -0.447 e.Å-3

132

Figure 4.2 The structure of Co(mchp)py with 35% thermal ellipsoids. The hydrogen

atoms are omitted for clarity except the ones on the external nitrogen atoms.

in a reaction similar to that used for metal insertions into N-confused porphyrin (Scheme

4.2). The reaction of H3mchp with Co2(CO)8 followed by crystallization under anaerobic

conditions resulted in the formation of the CoII(mchp)py complex, 8. The activation of

the internal C-H bond to form a direct Co-C bond is observed with this complex. The

Co(II) ion fits well into the coordination core of the macrocycle, forming a slightly

distorted square pyramidal complex. The equatorial Co-N bond lengths vary between

1.975(1) Å to 2.016(1) Å while the axial Co-N distance is 2.178(1) Å which is shorter

than the value of 2.44 Å of Co(II) TPP piperidine complex.139 The complex is

133 N Co2(CO)8 H mchp N 3 pyridine N N N Co N HN NH Co(OAc)2 8 DMF exposed to air pyridine N OH N N N N N N N N N Co Co N N N N N NH N N

10 9

Scheme 4.2 Synthesis of Co(mchp)py 8; Co(mchp)py2 9

and Co(mchp-OH)py2 10.

paramagnetic, and magnetic susceptibilty is µeff =1.80 which correlates to a low spin (S=

½) electron configuration. The ring is tetraanionic, requiring the protonation of two meso

nitrogen positions for charge balance.

This compound can then be oxidized by air in the presence of pyridine to produce the

III corresponding Co(III) complex, Co (mchp)py2, 9. (Scheme 4.2) The complex is six

coordinate, with an additional pyridine occupying an axial position. (Figure 4.3) The Co-

C bond remains intact, but the distances about the metal reflect the increase in oxidation

state, with 1.97 and 2.00 Å Co-N and Co-C bond lengths, respectively. The average axial

Co-N bond lengths of 1.98 Å is comparable with the values observed for the

134

Table 4.2. Crystal data and structure refinement for 9.

Identification code Co(mchp)py2

Empirical formula C55 H41 Co N12 Formula weight 928.93 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/c Unit cell dimensions a = 9.9002(16) Å α = 90°. b = 10.0469(16) Å β = 93.700(3)°. c = 44.850(7) Å γ = 90°. Volume 4451.8(12) Å3 Z 4 Density (calculated) 1.396 Mg/m3 Absorption coefficient 0.441 mm-1 F(000) 1932 Crystal size 0.18 x 0.05 x 0.01 mm3 Theta range for data collection 1.82 to 28.36°. Index ranges -12<=h<=12, -13<=k<=13, -59<=l<=59 Reflections collected 37698 Independent reflections 10627 [R(int) = 0.0990] Completeness to theta = 28.36° 95.4 % Absorption correction Empirical SADABS Max. and min. transmission 0.9957 and 0.6374 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 10627 / 0 / 563 Goodness-of-fit on F2 0.978 Final R indices [I>2sigma(I)] R1 = 0.0760, wR2 = 0.1627 R indices (all data) R1 = 0.1369, wR2 = 0.1835 Largest diff. peak and hole 0.574 and -1.054 e.Å-3

135

Table 4.3. Crystal data and structure refinement for 10.

Identification code Co(mchp-OH)py2·3p-xylene

Empirical formula C56 H46 Co N9 O Formula weight 919.95 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 12.5789(11) Å α = 75.7250(10)°. b = 13.3090(12) Å β = 76.5220(10)°. c = 15.3136(13) Å γ = 69.3530(10)°. Volume 2294.6(3) Å3 Z 2 Density (calculated) 1.331 Mg/m3 Absorption coefficient 0.426 mm-1 F(000) 960 Crystal size 0.28 x 0.18 x 0.08 mm3 Theta range for data collection 1.66 to 25.00°. Index ranges -14<=h<=14, -15<=k<=15, -18<=l<=18 Reflections collected 16526 Independent reflections 8035 [R(int) = 0.0254] Completeness to theta = 25.00° 99.5 % Absorption correction Empirical SADABS Max. and min. transmission 0.9667 and 0.7065 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8035 / 0 / 609 Goodness-of-fit on F2 1.120 Final R indices [I>2sigma(I)] R1 = 0.0906, wR2 = 0.2454 R indices (all data) R1 = 0.1095, wR2 = 0.2593 Largest diff. peak and hole 4.080 and -0.493 e.Å-3

136

Figure 4.3 The structure of Co(mchp)py2 with 35% thermal ellipsoids. The hydrogen

atoms are omitted for clarity except the one on the external nitrogen atom.

Co(NCTPP)py2 complex and also with the distance (2.06 Å) observed for the cationic

Co(III) TPP piperidine complex.140 The metal is a diamagnetic Co(III) center, as expected for the geometry and oxidation state. Once again, the macrocycle is tetraanionic, requiring a single proton on a meso position for charge balance.

III The Co (mchp)(py)2 complex is unstable in solution exposed to air, often demetallating to form organic fragments and paramagnetic cobalt oxides. This decomposition process may occur through oxidation of the ligand. Torres and Hannack noted that the lack of aromaticity and iminic nature of these macrocycles can result in ligand decomposition.138 Partial oxidation of the ligand upon metallation of the free base 137 with Co(OAc)2·4H2O in DMF under aerobic conditions was observed. The resultant

III product, Co (mchp-OH)py2, is precipitated from DMF with the addition of water and recrystallized from pyridine/p-xylene. Single crystal X-ray structure elucidation of the

Figure 4.4 The structure of Co(mchp-OH)py2 with 35% thermal ellipsoids.

The hydrogen atoms are omitted for clarity except the one on hydroxyl group.

molecule shows that a hydoxide at a β-carbon position. As a result, the macrocycle

deviates from planarity due to this new sp3 hybridized carbon in the ring. The metal is a

low spin +3 metal ion, and the macrocycle is trianionic and no additional protons are

observed on the external nitrogen positions. The Co-C distance (1.96 Å) is similar in

length to the Co-C distance of 9. The equatorial Co-N distances vary between 1.90 and

1.96 Å while the axial Co-N distances are 1.98 and 1.99 Å, the same as seen in compound 9. 138 Interestingly, the metallation of the closely related macrocycle m-benziporphyrin

with divalent metal ions such as Cd(II), Ni(II) and Fe(II) did not result in the activation of

the internal C-H bond as seen with the H3mchp macrocycle. In these compounds the

macrocycle serves as a monoanionic ligand and an anion is coordinated axially for charge

balance. The benzene unit is tilted from the porphyrin plane forming an agostic-type interaction.

In conclusion, three different cobalt complexes of monocarbahemiporphyrazine are reported in two oxidation states. The metal can be inserted using either cobalt carbonyl

or acetate, however in the latter case oxidation of one of the Schiff base C=N bond

occurs.

139

CHAPTER V

STRUCTURE AND CATALYTIC ACTIVITY OF A MANGANESE(III)

TETRAPHENYLPORPHOLACTONE

Ring modified porphyrins and their metal complexes have been receiving much

attention due to their altered electronic and metal coordination properties. 15,,141

Porpholactone is an example of a porphyrin analog in which one of the pyrrole rings is converted to an oxazolone ring. (Figure 5.1b) With the introduction of an ester functionality to the macrocycle, the conjugation around the ring is restricted but not completely blocked, as in the case of chlorin (Figure 5.1c.).142 As a consequence,

porpholactone shares electronic properties both with chlorin and porphyrin (Figure

5.2).142 The absorption spectra of porphyrin and porpholactone are comparable since the

•-electron pathway of porpholactone is not affected to great extend by the modification

because of the keto-enol tautomerization of oxazolone ring. Although the spectra of

these species are similar, the Soret/Q Band intensity ratio of porpholactone is smaller

than that of porphyrin due to the lowered symmetry.

140 Ph Ph Ph Ph Ph Ph N N N H H O H N N N N N N O H H H N N N Ph Ph Ph Ph Ph Ph

Figure 5.1 The structures of (a) Tetraphenylporphyrin (b) Tetraphenylporpholactone and

Although porpholactone was first synthsized in 1984 by Crossley et al.,143 the metal

coordination properties of this macrocycle have not been extensively probed to date, with

the exception of a study by Khalil and Gouterman of the Ni, Zn, Pd and Pt complexes of

tetra(pentafluorophenyl)-porpholactone.144 In this chapter, the synthesis and characterization of a Mn(III) tetraphenylporpholactone is presented. This compound was examined as a catalyst for the epoxidation of olefins. As in normal manganese

porphyrins, manganese porpholactones catalytically epoxidize a variety of olefins.145

Whereas the same products are produced as observed in tetraphenylporphyrinatomanganese(III) chloride, Mn(TPP)Cl, the presence of the oxazolone ring does effect the turnover numbers of the catalysis, but does not provide a site for hydrogen bonding for substrate. This work represents one of the few examples of a catalysis study using a porphyrin analog or isomer.146

141 4

2 Absorbance

1 x5

0 350 400 450 500 550 600 650 700 Wavelength (nm)

Figure 5.2 UV-Visible spectra of tetraphenylporphyrin (dotted line)

and tetraphenylporpholactone (solid line).

Experimental

General Methods: Unless otherwise noted, all reagents and solvents were purchased from Sigma, Aldrich, Acros Organics, Strem or Sorbent Technologies and used without further purification. Mass spectra were recorded using an ES MS Bruker Esquire-LC ion-trap mass spectrophotometer. Elemental analysis was conducted at the University of

Illinois, School of Chemical Sciences Microanalysis Laboratory. UV-visible spectroscopy was performed using a U-3010 Hitachi spectrophotometer. GC analysis

142 was performed using Shimadzu GC-17A Gas Chromatograph equipped with a FID

detector and Rtx-50 column. Argon was used as the carrier gas.

Single crystal X-ray diffraction data was collected with 100 K (Bruker KRYO-

FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped

with a Mo-target X-ray tube (λ = 0.71073 Å) operated at 2000 watts power. The detector

was placed at a distance of 5.009 cm from the crystal. Integration and refinement of

crystal data was done using Bruker SAINT software package and Bruker SHELXTL

(version 6.1) software package, respectively. Absorption correction was completed using the SADABS program. Crystals were placed in paratone oil upon removal from the mother liquor and mounted on a plastic loop in the oil.

Preparation of tetraphenylporpholactone (H2TPPL): This synthesis is a two-step

oxidation of free base tetraphenylporphyrin (Scheme 5.1).147 The first step is based on the procedure by Dolphin et al. and the second step on Brückner’s report.147 Free base

tetraphenylporphyrin (0.110 g, 0.180 mmol) was dissolved in 40 mL of CHCl3 (10%

pyridine). OsO4 (52.0 mg, 0.200 mmol) was added to the stirring solution and the

reaction was mixed at ambient temperature for 4 d. protected from light. An excess

amount of sulfur (~100 mg) was added to the solution and the mixture is filtered through

a frit. The solvent was removed under reduced pressure. The mixture was placed on a

silica gel column and eluted with CH2Cl2. The unreacted H2TPP was completely removed by CH2Cl2 and H2TPPL was eluted with MeOH/CH2Cl2 (1.5%). Evaporation of

the solvent yielded 400 mg of tetraphenyl dihydroxychlorin. Overall yield of the reaction was 42%, which is comparable to 49% from the literature.

143

Ph Ph Ph Ph Ph Ph N N N H H H O OsO4 OH KMnO4 N N N N N N 18-crown-6 O H CHCl3/pyr. H OH H N 4 days N THF N 12-hours Ph Ph Ph Ph Ph Ph

Scheme 5.1 Synthesis of free base tetraphenylporpholactone through dihydroxychlorin.

The second step is the oxidation of dihydroxychlorin to porpholactone. Tetraphenyl

dihydroxychlorin (400 mg, 0.640 mmol) from the first step was dissolved in 80 mL of

THF and 18-crown-6 (56.0 mg, 0.200 mmol) was added to the stirring solution. KMnO4

(504 mg, 3.20 mmol, ~5 equiv.) was slowly added to the stirring solution. The reaction mixture was stirred for 12 h. at ambient temperature. The resulting solution was passed through a short plug of silica and washed with CH2Cl2 until the filtrate was colorless.

The resulting solution was dried in vacuo. The product was isolated by chromatography

eluted with CHCl3 and was recrystallized from EtOH. 400 mg of product was isolated.

The purity of the sample was checked by TLC plate and UV-visible spectroscopy in

CHCl3: λmax (nm) 408.0, 518.0, 544.0, 592.0, 644.0 nm.

Preparation of tetraphenyl(porpholactone) (chloride) Manganese (Mn(TPPL)Cl):

The free base H2TPPL (100 mg, 0.160 mmol) was dissolved in 20.0 mL of DMF and

MnCl2·4H2O (35.0 mg, 0.180 mmol) was added to the solution. The reaction mixture

was refluxed for 2 h. DMF was removed under reduced pressure and the resulting

precipitate was crystallized from CH2Cl2-hexanes mixture.

144 Mn(TPPL)Cl: Yield: 100 mg (73%), UV-visible (benzene); λmax (nm) 637.0, 486.5,

380.5, CHN Anal. Found (Calculated) C: 68.85 (71.14), H: 4.52 (3.70), N: 6.52 (7.90),

ESI-MS Found (Calculated): 685.3 M/z (685.1 M/z) Crystal data and structure refinement parameters are summarized in Table 6.1.

Catalysis Reactions: In a typical experiment, 5 µmol of the catalyst and 5 mmol of olefin substrate were dissolved in 5 mL of anhydrous benzene and stirred well under air.

Iodosylbenzene (0.5 mmol) was added to the reaction mixture and stirred for 3 h.

Products are analyzed by GC relative to the internal standard nitrobenzene.

Results and Discussion

A manganese porpholactone was generated using 5,10,15,20-

tetraphenylporpholactone (H2TPPL) freebase as the macrocycle. The porphyrinoid

freebase can be readily generated using the procedure developed by Brückner and

coworkers, which involves two sequential oxidations of normal tetraphenyl porphryin

using osmium tetroxide followed by permanganate anion (Scheme 5.1).147 The manganese complex of this macrocycle can the be produced by metallating with

MnCl2•4H2O in DMF (Scheme 5.2). Crystallization of the product from

chloroform/hexanes afforded pure single crystals of Mn(TPPL)Cl.

Cl Ph Ph Ph Ph O N N Mn(Cl)2.4H2O O H N O N Mn N N H O N Ph DMF N Ph Ph Ph

Scheme 5.2 Metallation of free base tetraphenylporpholactone with MnCl2.

145 The UV-visible spectrum of Mn(TPPL)Cl compared to Mn(TPP)Cl shows some

significant differences. While both spectra exhibit hyper type features with split Soret

bands typical for Mn(III) porphyrins (Figure 5.3), the extinction coefficent of the 486 nm

Soret band is much lower than that observed for the analagous band in Mn(TPP)Cl. In

addition, the Q bands of Mn(TPPL)Cl are ~20 nm lower in energy than those of

Mn(TPP)Cl. Both observations are consistent with the break in symmetry and reduction

in aromaticity arising from the presence of the oxazolone ring.

Crystals of Mn(TPPL)Cl suitable for single crystal X-ray diffraction were isolated by

recrystalization from a CHCl3/hexane diffusion. Elucidation of the structure showed that

the macrocycle contains a five coordinate Mn(III) porphyrinoid with an axial chloride

(Figure 5.4). The structure of Mn(TPPL)Cl exhibits a coordination geometry very similar

to that observed in Mn(TPP)Cl.148 The metal nitrogen bond lengths range between

1.993(4) and 2.022(4) Å, which is the same range as observed in normal manganese(III) porphyrins. The axial Mn-Cl bond also shows an identical length, 2.362(2) Å, as seen in

Mn(TPP)Cl structures. Overall, the manganese porholactone ring has a slight saddle shape, similar to that observed in some the crystal forms of porphyrinatomanganese(III) chloride. The metal center is pulled away from the plane of the four interior nitrogen by

0.257 Å. This geometry and out of plane deviation is also in good agreement with the structure of the non-planar variant of Mn(TPP)Cl.

146 1.5

1 Absorbance 0.5

0 300 400 500 600 700 Wavelength (nm)

Figure 5.3 UV-Visible spectra of Mn(TPP)Cl (dotted line) and Mn(TPPL)Cl (solid line)

147

Table 5.1. Crystal data and structure refinement for Mn(TPPL)Cl.

Identification code Mn(TPPL)Cl⋅CHCl3

Empirical formula C43 H26 Cl Mn N4 O2⋅CHCl3 Formula weight 840.45 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P2(1)/n Unit cell dimensions a = 12.031(9) Å α = 90°. b = 21.864(16) Å β = 102.581(13)°. c = 14.551(11) Å γ = 90°. Volume 3736(5) Å3 Z 4 Density (calculated) 1.282 Mg/m3 Absorption coefficient 0.466 mm-1 F(000) 1480 Crystal size 0.50 x 0.20 x 0.02 mm3 Theta range for data collection 1.71 to 25.00°. Index ranges -14<=h<=14, -26<=k<=25, -17<=l<=17 Reflections collected 24320 Independent reflections 6568 [R(int) = 0.1081] Completeness to theta = 25.00° 99.8 % Absorption correction Empirical SADABS Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 6568 / 0 / 453 Goodness-of-fit on F2 0.911 Final R indices [I>2sigma(I)] R1 = 0.0752, wR2 = 0.1811 R indices (all data) R1 = 0.1203, wR2 = 0.2009 Largest diff. peak and hole 1.320 and -0.363 e.Å-3

148

Figure 5.4 The structure of Mn(TPPL)Cl with 50% thermal ellipsoids. Hydrogen atoms

are omitted for clarity.

The catalytic efficiencies of Mn(TPPL)Cl and Mn(TPP)Cl for the epoxidation of olefins using iodosylbenzene as the oxidant were compared. In all cases, the products of

the epoxidation using Mn(TPPL)Cl were identical and produced in the same ratios to

those observed for Mn(TPP)Cl (Scheme 5.3). This led the conclusion that Mn(TPPL)Cl

catalyzes the epoxidation of olefins via the same mechanism as in Mn(TPP)Cl.149 It is

believed that Mn(TPP)Cl catalyzes oxidations have two reactive intermediate species:

MnV-oxoporphyrin (MnV(TPPO)) and MnIV-oxoporphyrin (MnIV(TPPO)). For this reason, other oxidation products are observed in Mn(III) porphyrin catalyzed alkene oxidation reactions. Electrophilic addition of the substrate to MnV(TPPO) generates a

carbocation which then undergoes to ring closure generating an epoxide and carboxylic

149 products. The reaction of the other reactive intermediate, MnIV(TPPO) with the alkene

substrate results in the formation of a radical species which also forms an epoxide via

ring closure. Alternatively, the resulting radical species can react with atmospheric O2 and generate the peroxo radical which can result in formation of other products. Some differences were observed between the two macrocycles in the epoxidation of alkenes; the turnover numbers from reactions with 1-hexene, cyclohexene, styrene, 5-hexen-1-ol and 4-penten-1-ol are listed in Table 5.2. The results demonstrate that Mn(TPPL)Cl is only slighly better in catalytic efficiency compared to Mn(TPP)Cl for all of the transformations with the exception of 1-hexene. Interestingly, Mn(TPPL)Cl is significantly more efficient than normal Mn(TPP)Cl for the epoxidation of 1-hexene, exhibiting nearly twice the efficiency for this electron poor substrate. Because the largest efficiency difference between these two catalysts was observed with electron poor substrate, 1-hexene, the difference appears to arise from electronic reasons. Both catalysts were examined using 4-penten-1-ol and 5-hexen-1-ol as substrates. It was desirable to determine whether hydrogen bonding of the hydroxide group to the oxazolone oxygens might increase the efficency of epoxidation of this substrate with

Mn(TPPL)Cl. For this part of the investigation, two different substrates with different chain lengths were chosen. While the turnover numbers are increased for the porpholactone versus normal porphyrin of the alkene-alcohols, the small increase indicates that electronic factors, rather than hydrogen bonding, governs the efficiency of the catalysis.

150 Table 5.2 Comparison of Mn(TPPL)Cl and Mn(TPP)Cl as catalysts for epoxidation reactions of substrates: 1-hexene, cyclohexene, styrene, 5-hexen-1-ol and 4-penten-1-ol.

Turnover no.(h-1)a

Catalyst Hexene Cyclohexene Styrene Pentenolb Hexenolc

Mn(TPPL)Cl 13.0 63.4 59.4 27.4 38.5

Mn(TPP)Cl 6.7 60.5 58.3 21.2 34.6 a Turnover numbers are determined by GC analysis using a FID detector with nitrobenzene as internal standard. The reactions were run at least twice. b 4-penten-1-ol. c 5-hexen-1-ol.

In conclusion, Mn(TPPL)Cl exhibits a similar structure to Mn(TPP)Cl and catalyzes the epoxidation of olefins with similar efficency. Mn(TPPL)Cl is a slightly better catalyst than Mn(TPP)Cl for the most alkene substrates. However, for the electron poor substrate 1-hexene, the turnover number is appreciably larger than in normal manganese porphyrin. Hydrogen bonding does not seem to play a significant role in catalysis with manganese porpholactone; the turnover numbers are only slightly larger than for

Mn(TPP)Cl.

151

CHAPTER VI

SUMMARY OF THE COORDINATION CHEMISTRY OF PORPHYRIN ANALOGS

Core modified hemiporphyrazines, monocarbahemiporphyrazine and

dicarbahemiporphyrazine, have been known for more than five decades, but the metal

chemistry of these macrocycles remained unexplored. Metallation chemistry of these two

macrocycles was studied. It has been shown that dicarbahemiporphyrazines bind to

metal ions, affording complexes “agostic” type interactions whereas

monocarbahemiporphyrazines form direct metal-carbon bonds.

Reaction of dicarbahemiporphyrazine with silver nitrate affords a cationic three-

coordinate silver(I) complex with two internal C-H bond interactions. The macrocycle

serves as a neutral ligand with two protons at the external nitrogen atoms. Copper(II)

binding leads to substitution at an internal C-H position, possibly through an oxidative

nucleophilic substitution of hydrogen resulting a Cu(I) complex. Recrystallization of the

Cu(I) complex results in the loss of axial pyridine and demetallation to yield the free

ligand.

Coordination chemistry of dicarbahemiporphyrazine with middle transition metal ions is also investigated. Reaction of dicarbahemiporphyrazine with iron and cobalt carbonyl under anaerobic conditions results in the formation of the corresponding metal

152 complexes. The resultant complexes have three coordinate high-spin metal centers with

two side-on interactions with the internal C-H units. Metallation of the ligand with iron

pentacarbonyl under same conditions results in a slightly different chemistry. Whereas

the metallation chemistry takes place, partial reduction of the macrocycle occurs. This

iron(II) complex is similar to the cobalt(II) and manganese(II) complexes although

protonation occurs at the two external positions.

Metallation chemistry of monocarbahemiporphyrazines are also reported. Cobalt

coordination of this ligand is studied extensively. Reaction of the ligand with cobalt

carbonyl under anaerobic conditions generates a cobalt(II) complex. Unlike in the case

of dicarbahemiporphyrazine, activation of the internal C-H bond is observed and a direct

metal-carbon bond is formed. The metal center has a five-coordinate environment with

one carbon and three nitrogen atoms from the macrocycle and one nitrogen from the axial

pyridine. Exposure of this complex to air results in a metal-based oxidation, affording a

cobalt(III) complex with a six-coordinate metal center. Reaction of the ligand with cobalt

acetate under aerobic conditions results in partial oxidation of the ring along with the

metallation affording a Co(III) complex.

Metallation of a porphyrin analog, porpholactone, with manganese is also reported

along with the catalytic activity of the resultant complex. It is observed that manganese porpholactone has similar structural features with the manganese complex of regular porphyrin. Comparison of the catalytic activities of these two complexes shows that they

catalyze epoxidation reactions via the same mechanism. The catalytic activities of both

species are almost the same with a slight increase in effectiveness of manganese(III) porpholactone over Mn(III) porphyrin with electron-rich substrates.

153

CHAPTER VII

COORDINATIVE FLEXIBILITY IN HYDROTRIS(IMAZOLYL)BORATE

DIVALENT METAL COMPOUNDS

Since its development in the late 1960s, scorpionate metal chemistry has continued to be an active area of inquiry, especially in borate-based chelating ligands.150 Over the past few years, research in this area has grown to include oxygen and sulfur containing

- 151 chelate systems in addition to the pyrazolylborates, HB(pz)3 (Figure 7.1a). Recently, a

- tri-sulfur ligand, tris(imazolyl)borate, HB(mt)3 (1, Figure 7.1b) was developed as a soft- tridentate ligand by Reglinski et al.152, 153 The metal coordination properties of the ligand have been investigated and several coordination modes were observed for various main

- - CH S 3 N H N H N B B S N N S N N N N CH a. b. H C 3 3

Fig.7.1 Structures of (a) Hydrotris(pyrazolyl)borate (b) Hydrotris(imazolyl)borate.

154

group and transition metals, typically forming either 1:1 complexes with additional

ligands154, 155, 156, 157 (Figure 6.2a) or 2:1 complexes with divalent metals153, 156, 158 (Figure

7.2b) similar to that observed for tris(pyrazolyl)borate metal complexes. In this report, three different modes of interaction with divalent metal cations which exhibit a diversity of binding not typically observed in the scorpionate family were reported. With group II

- and group XII metals, the HB(mt)3 ligand can act as a non-coordinating anion, can

engage in B-H agostic bonding, and can form metal cluster compounds, all of which are

previously unknown modes of metal binding for this ligand with d0 or d10 divalent metals.

H B N N N N N N N N S S S S S HB N M M N X S S S N N N S N N N N a. N b. B H

Fig.7.2 Typical geometries for Hydrotris(imazolyl)borate ligand (a) tetrahedral geometry

for a 1:1 metal complex with an additional ligand X. (b) Typical octahedral geometry of a

2:1 metal complex.

155

Experimental

All reagents and solvents were purchased from either Aldrich or Acros and were

used as received. Water was purified by using a Milli-Q reagent water system.

Elemental analysis was carried out at the School of Chemical Sciences, Microanalytical

Laboratory at the University of Illinois at Urbana-Champaign. Nitrogen analyses tend to be lower than calculated presumably due to the formation of refractory boron nitrides.

X-Ray Crystallography: The X-ray intensity data for compounds 1-3 were measured at 100K (Bruker KRYO-FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped with a Mo-target X-ray tube (λ = 0.71073 Å) operated at

2000 W power. The crystals were mounted on a cryoloop using paratone N-Exxon oil and placed under a stream of nitrogen at 100 K. The detector was placed at a distance of

5.009 cm from the crystals. Frames collected with a scan width of 0.3° in ω. The frames for each data set were integrated with the Bruker SAINT software package using a narrow frame integration algorithm. The data were corrected for absorption with the

SADABS program. The structure was solved and refined using the Bruker SHELXTL

(version 6.1) software package until the final anisotropic full-matrix least-squares refinement on F2 converged.

Preparation of sodium tris(methimazolyl)borate, Na[HB(mt)3]: The ligand was

synthesized according to a procedure from the literature.152 The starting material 2-

mercapto-1-methylimidazole (5.00g, 0.044 mol) was mixed with sodium borohydride

156

1 Figure 7.3 H NMR of Na[HB(mt)3].

13 Figure 7.4 C NMR of Na[HB(mt)3] 157

(0.55 g, 0.014 mmol) in a 50 mL round bottom flask. The reagents were heated up to

180ºC without addition of solvent. H2 gas evolution was monitored with a bubbler and gas evolution ceased within three hours. After cooling the mixture down to room

temperature, the resulting solid was washed with hexane and the product extracted with

CHCl3. The resulting white powder was filtered and dried. The purity of the product was

checked by 1H NMR and 13C NMR which are shown in Figure 7.3 and Figure 7.4,

respectively.

Preparation of Ca[HB(mt)3]2(H2O)6: Sodium tris(methimazolyl)borate (12.0 mg,

0.03 mmol) was dissolved in 2 mL of deionized water and a separate solution of

Ca(NO3)2.4H2O (5.0 mg, 0.02 mmol) was prepared in 3 mL of deionized water. The

metal solution was added to the ligand slowly. The slow evaporation of the water led to

formation of a crystalline product. Yield: 12.0 mg (87%) Anal. Calcd for

C24H32B2BaN12S6 (%): C, 35.38; H, 4.95; N, 20.63 Found: C, 36.34; H, 4.46; N, 20.45

Crystal data and structure refinement parameters are summarized in Table 7.1.

Preparation of Ba[HB(mt)3]2(H2O)2: Sodium tris(methimazolyl)borate (12.0 mg,

0.03 mmol) was dissolved in 2 mL of deionized water and a separate solution of

Ba(NO3)2 (5.20 mg, 0.02 mmol) was prepared in 3 mL of deionized water. The metal

solution was added to the ligand slowly. The slow evaporation of the water led to formation of crystalline product. Yield: 10 mg (76%) Anal. Calcd for C24H32B2BaN12S6

(%): C, 34.32; H, 2.57; N, 20.01 Found: C, 35.43; H, 2.57; N, 20.01 Crystal data and structure refinement parameters are summarized in Table 7.2.

158

Table 7.1. Crystal data and structure refinement for [Ca(Tm)2]·6H2O.

Identification code [Ca(Tm)2]·6H2O

Empirical formula C24 H52 B2 Ca N12 O10 S6 Formula weight 924.84 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 8.8138(6) Å α = 91.5160(10)°. b = 9.4094(7) Å β = 93.0220(10)°. c = 13.1360(9) Å γ = 90.5400(10)°. Volume 1087.44(13) Å3 Z 2 Density (calculated) 1.409 Mg/m3 Absorption coefficient 0.493 mm-1 F(000) 486 Crystal size 0.48 x 0.26 x 0.17 mm3 Theta range for data collection 2.17 to 28.28°. Index ranges -11<=h<=11, -12<=k<=12, -17<=l<=17 Reflections collected 8988 Independent reflections 4814 [R(int) = 0.0314] Completeness to theta = 28.28° 89.0 % Absorption correction Empirical SADABS Max. and min. transmission 0.9208 and 0.4220 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 4814 / 0 / 297 Goodness-of-fit on F2 1.100 Final R indices [I>2sigma(I)] R1 = 0.0413, wR2 = 0.1241 R indices (all data) R1 = 0.0436, wR2 = 0.1261 Largest diff. peak and hole 0.632 and -0.450 e.Å-3

159

Preparation of Hg4[HB(mt)3]4Cl4·13H2O: Sodium tris(methimazolyl)borate (12.0 mg, 0.03 mmol) was dissolved in 2 mL of deionized water and a separate solution of

HgCl2 (5.40 mg, 0.02 mmol) was prepared in 3 mL of water/acetone. The metal solution

was added to the ligand slowly. The slow evaporation of the water/acetone lead to

formation of crystalline product. Yield: 10.0 mg (76%) Crystal data and structure

refinement parameters are summarized in Table 7.3.

Results and Discussion

- We examined the reaction of the HB(mt)3 ligand (produced via the reaction of 2-

mercapto-1-methylimidazole with borohydride)152 with three metals across the families;

calcium, barium and mercury. All reactions were carried out in aqueous or

aqueous/acetone solution, and formed crystalline products. When sodium

tris(imazolyl)borate is reacted with calcium nitrate, the resultant compound

Ca[HB(mt)3]2(H2O)6, 2, has a metal-ligand stoichiometry similar to that of other divalent

- 158 2+ metal HB(mt)3 compounds, such as observed in Fe[HB(mt)3]2. However, with Ca , the borate acts as a non-coordinating species. As can be seen from the single X-ray structure elucidation in Figure 7.2, the coordination sphere of the metal ion is occupied by solvent water molecules, and the borate does not bind to the metal center. The only interaction between metal and ligand occurs through hydrogen bonding interactions between the ligand sulfur atoms and the metal bound solvent waters (~3.2 -3.3 Å).

Interestingly, the structure of the ligand shows that the sulfur groups are pointing in the

160

Table 7.2. Crystal data and structure refinement for Ba(Tm)2(H2O)2.

Identification code Ba(Tm)2(H2O)2

Empirical formula C24 H36 B2 Ba N12 O2 S6 Formula weight 875.96 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 7.5055(6) Å α = 81.1580(10)°. b = 10.5872(8) Å β = 80.5120(10)°. c = 12.5053(9) Å γ = 69.6220(10)°. Volume 913.77(12) Å3 Z 2 Density (calculated) 1.592 Mg/m3 Absorption coefficient 1.472 mm-1 F(000) 442 Crystal size 0.20 x 0.10 x 0.02 mm3 Theta range for data collection 1.66 to 28.28°. Index ranges -9<=h<=9, -14<=k<=13, -16<=l<=16 Reflections collected 8131 Independent reflections 4225 [R(int) = 0.0193] Completeness to theta = 28.28° 93.5 % Absorption correction Empirical SADABS Max. and min. transmission 0.9711 and 0.8074 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 4225 / 0 / 229 Goodness-of-fit on F2 1.098 Final R indices [I>2sigma(I)] R1 = 0.0304, wR2 = 0.0714 R indices (all data) R1 = 0.0338, wR2 = 0.0729 Largest diff. peak and hole 1.049 and -0.427 e.Å-3

161

Fig.7.5 The structure of Ca[HB(mt)3]2(H2O)6 with 50% thermal ellipsoids.

H atoms are omitted for clarity.

same direction as the B-H bond, a conformation that can lead to B-H agostic bonding as will be shown below.

- When reacted with an element lower in group II, HB(mt)3 also forms a species with a 2:1 stoichiometry with the metal. The reaction of NaHB(mt)3 with Ba(NO3)2 affords

Fig.7.6. The structure of Ba[HB(mt)3]2(H2O)2 with 50% thermal ellipsoids.

H atoms are omitted for clarity.

162

Table 7.3. Crystal data and structure refinement for Hg4(Tm)4Cl4.

Identification code Hg4(Tm)4Cl4·13H2O

Empirical formula C48 H94 B4Cl4 Hg4 N24 S12 O13 Formula weight 2588.36 Temperature 100(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 12.923(3) Å α = 81.587(4)°. b = 18.019(4) Å β = 76.885(4)°. c = 20.599(5) Å γ = 70.578(4)°. Volume 4392.5(18) Å3 Z 2 Density (calculated) 1.934 Mg/m3 Absorption coefficient 7.440 mm-1 F(000) 2448 Crystal size 0.18 x 0.10 x 0.02 mm3 Theta range for data collection 1.20 to 28.33°. Index ranges -17<=h<=17, -22<=k<=23, -26<=l<=26 Reflections collected 33523 Independent reflections 19743 [R(int) = 0.1200] Completeness to theta = 28.33° 90.2 % Absorption correction Empirical SADABS Max. and min. transmission 0.8658 and 0.1880 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 19743 / 48 / 886 Goodness-of-fit on F2 0.753 Final R indices [I>2sigma(I)] R1 = 0.0663, wR2 = 0.1435 R indices (all data) R1 = 0.1447, wR2 = 0.1554 Largest diff. peak and hole 3.388 and -2.577 e.Å-3

163

crystals of the species Ba[HB(mt)3]2(H2O)2, 3, from aqueous solution. Elucidation of the

structure via single crystal X-ray methods reveals that the metal-borate interaction is

much different from that in calcium. In the case of Ba+2, two equivalents of the ligand do

coordinate to the metal, but do so in a different fashion from the commonly observed tris-

sulfur mode. The Ba(II) metal is in a square prismatic coordination environment with

two imazolyl ligands and two water molecules chelated to the metal center. Each

hydrotris(imazolyl)borate ligand is coordinated to the metal center in a κ3-H,S,S′ coordination mode with a three centered, two-electron Ba···H-B bond. The distance between the metal center and the H atoms is 2.625Å and the Ba-S distances are 3.1784(7)

Å and 3.2401(6) Å. The coordination environment of Ba(II) is completed by two water molecules.

The first example of this type of agostic interaction was proposed as an intermediate by Hill et al., but it wasn’t fully characterized,159 although a few examples of this type of

coordination mode were subsequently structurally elucidated in Ru and Ag

complexes.156,160 Very recently, Goh and Webster carried out some electrochemical

studies on a RuII/RuIII hydotris(imazolyl)borate system in order to analyze the dynamic

coordination exchange reactions in solution from κ3-S,S′,S′′coordination mode to κ3-

H,S,S′ mode about the Ru center.161 Compound 3 is the first example of this coordination

type with a d0 metal.

164

Fig.7.7 The structure of Hg4[HB(mt)3]4Cl4 with 50% thermal ellipsoids.

H atoms are omitted for clarity.

Moving to group XII, the reaction of NaHB(mt)3 with HgCl2 in water/acetone

produces a complex with a strikingly different structure of the formula Hg4[HB(mt)3]4Cl4,

3. Single crystals of this complex were grown from water/acetone and the structure was elucidated by X-ray crystallography. The structure of the cluster is shown in figure 6.2 c.

In this case, the borate forms a metal cluster complex with four Hg(II) ions in a tetrahedral arrangement around a central chloride anion. The chloride ion exhibits a tetrahedral geometry as well, with a Hg-Cl-Hg bond angle of 102.34(12)° to 114.85(12)°.

The ratio of metal to ligand is 1:1, and each borate coordinates to three different Hg(II) ions through Hg-S bonds with a varying distance from 2.453(3)Å to 2.516(4)Å. Each metal ion has a tetrahedral coordination geometry. Interestingly, the borate anions are inverted toward the center of the cluster, in the direction of the central chloride atom.

The boron to chloride bond separation is too long to be considered a hydrogen bond,

165

- however, with a distance of > 4Å. Previously, only one Hg(II) complex with HB(mt)3

154 has been elucidated: a 1:1 species of the formula Hg[HB(mt)3]Br.

In conclusion, three distinct modes of coordination for

hydrotris(methimazolyl)borate ligand were observed for three different closed d shell

divalent metals. The ligand acts as a non-coordinating anion for the group II metal ion

Ca+2, and only engages in H-bonding with the metal bound water molecules. In the case of Ba2+, the ligand binds via two sulfur atoms as well as a Ba···H-B agostic interaction

in a 2:1 ligand to metal ratio. Upon binding to Hg(II), the ligand forms a metal cluster

compound, with an interstitial chloride in the center of a tetrahedral arrangement of metal

- ions. This diversity of metal binding reflects the soft nature of the HB(mt)3 ligand in contrast to tris(pyrazolyl)borate. Similar diversity may be observed in other soft tripodal ligands, such as in the systems developed by Riordan using tetrakis(thioether)borates162 and Peters using tris(phosphine)borates.163

166

CHAPTER VIII

INVESTIGATING THE TOPOCHEMICAL POLYMERIZATION OF ANILINE

DERIVATIVES IN Pb(II) BORATE SCAFFOLDS

The design of metal-organic frameworks continues to be an active area of research in

part due to the promise of advanced applications.164 In general, one of the most difficult

challenges in crystal design is to develop functional materials based on coordination

polymers. Functional network solids can be broadly classified into two categories: materials that exhibit specific physical properties, like magnetism or non-linear optical behavior,165 and compounds that incorporate chemical reagents into the solid for reaction,

such as seen in zeolite catalysis.166 While there are many examples of naturally occurring

network solids (i.e. minerals) that exhibit these two types of functionality,167 synthetic network solids, such as metal-organic frameworks (MOFs), still frequently display neither topology dependent physical properties nor chemical reactivities.

An important class of potentially functional MOFs is layered coordination polymers.168 They are structurally reminiscent of naturally occurring compounds such as clays and metal oxides. Layered minerals can be used in many applications such as molecular intercalation and storage,169 ion exchange,170 and catalysis.171 The interlayer

spacing in these compounds can provide a micro-environment for many chemical

reactions such as polymerization reactions.172 Using the confined space of minerals as

167

reaction media can be advantageous in several ways, such as isolating only certain

desired products, allowing the reaction to occur in milder conditions, or even forcing

otherwise unreactive species to couple. Layered coordination polymers have the

potential to provide a similar micro-environment for chemical reactions which can also

be known as a “ship in a bottle” effect.173

We are currently investigating the coordinating anion tetrakis(imidazolyl)borate as a

component of coordination solids (Figure 8.1a). This structurally rigid multidentate

ligand has a tetrahedral environment and can form an extended network solid. In an

earlier study the Ziegler group reported that the reaction of Pb(NO3)2 salt with the ligand

leads to the formation of a layered coordination polymer with the formula of

Pb[B(Im)4](NO3)(1.35·H2O) (Figure 8.1b.). Previously, it was shown that this compound

can incorporate anionic organic guest molecules into the interlayer spacing of the

material. 174,175 When lead-borate crystals are grown in the presence of buffered benzoic acid solution, benzoates replace the nitrate anions and self-assemble into the interlayer spacing of the material without the disruption of the lead-borate network structure.175

Upon the observation of the nanoscale organization of benzoates in this lead-borate material, we realized other organic anions could be similarly enclathrated. The Ziegler group has incorporated several anionic organic guests into Pb[B(Im)4]X based materials

(where X is the anionic guest) and have recently reported the nanoscale organization of carboxylate modified TEMPO and PROXYL.176 This coordination polymer could also be used as a host material for generating conducting polymers where the polymer is

grown in the interlayer spacing.

168

N N -

N N B N N

N N a. b.

c. Figure 8.1. Structure of the borate ligand and the lead complex (a)

- 177 tetrakis(imidazolyl)borate anion, B(Im)4 . (b) The asymmetric unit of

174 Pb[B(Im)4](NO3)(1.35·H2O). (c) The extended structure of

174 Pb[B(Im)4](NO3)(1.35·H2O) along the c axis.

There are some examples of in situ polymerizations in between the layers of

inorganic materials and several different methods for polymerization within the confined

spaces of a variety of inorganic materials have been reported in the literature. For redox

active materials, no external oxidizing agents are required since the host itself initiates

polymerization of monomeric species such as thiophene,178 aniline,179 and pyrrole.180

Several different oxidizing agents have been utilized for the polymerization of various monomers within the matrix of redox inactive inorganic host materials.181 Kanatzidis et

al. in 1993 used ambient oxygen as the oxidizing agent to polymerize aniline within the

169

layers of uranyl phosphate.182 Later work confirmed this method for generating polymers

using atmospheric dioxygen. 183, 184

In this chapter, we present the enclathration of three different aniline derivatives between the layers of this lead-borate material and an investigation into the polymerization of these pre-organized monomers. Both ambient oxygen and solution phase oxidizing agents were used to facilitate the polymerization and we monitored the reactions by 13C-solid state NMR (13C-SSNMR). We observed that all three monomers

show different reactivities toward polymerization based on their orientation within the

solid and on the nature of the anionic derivative.

Experimental

All reagents and solvents were purchased from either Aldrich or Acros and were used as received. Water was purified by using a Milli-Q reagent water system.

Elemental analysis was carried out at the School of Chemical Sciences, Microanalytical

Laboratory at the University of Illinois at Urbana-Champaign. Nitrogen analyses tend to be lower than calculated due to the formation of refractory boron nitrides. 13C solid state

NMR spectra were obtained at 50.8 MHz on a Varian Unityplus-200 (4.7 T) using a Doty

Scientific Magic Angle Spinning (MAS) probe. The samples were packed into 7 mm

silicon nitride rotors with Kel-F caps. All 13C spectra were collected using cross

polarization and magic angle spinning (CP/MAS) with a sample spinning speed of 5

KHz. Chemical shifts were referenced to solid hexamethylbenzene (HMB) (δ(CH3)= 17.3

170 ppm). Infrared spectra were recorded on a Nicolet Nexus 870 Fourier transform spectrometer.

Intercalation of aniline derivatives into Pb[B(Im)4]X: A solution was made of a 3- fold excess of the aniline derivative (2.00 mmol) and 1 equivalent of sodium tetrakis(imidazolyl)borate (202 mg, 0.670 mmol) in 4 mL of deionized water (1,2) or 1:1 water/EtOH (3). The pH of the solution was adjusted to pH ~ 8.0 with sodium hydroxide and placed at the bottom of a vial. Pb(NO3)2 (222 mg, 0.670 mmol) was dissolved in 4 mL of deionized water. The lead(II) solutions were carefully layered on top of the aniline/borate solutions and immediate formation of white cloudy precipitate was observed. The vials were kept at room temperature for 3 days, after which white plate- like crystals of products were collected by filtration and washed with deionized water.

Pb[B(Im)4](3-NH2C6H4COO)·(H2O), 11: Yield: 285 mg (66%) The material was dried before the CHN analysis. Anal. Calcd for C19H19BN9O2Pb (%): C, 36.60; H, 3.07;

N, 20.22. Found: C, 34.88; H, 2.73; N, 17.34. FT-IR: 3367, 3204, 1602, 1509, 1489,

1387, 1310, 1177, 866, 852, 824, 781, 704, 669, 648, 639, 629, 620 cm-1. 13C SSNMR:

176, 155, 153, 151, 149, 142, 132, 127, 126, 122, 115 ppm.

Pb[B(Im)4](4-NH2C6H4COO)·(H2O), 12: Yield: 268 mg (62%) The material was dried before the CHN analysis. Anal. Calcd for C19H19BN9O2Pb (%): C, 36.60; H, 3.07;

N, 20.22. Found: C, 35.43; H, 2.57; N, 17.75. FT-IR: 3250, 1627, 1592, 1550, 1515,

1483, 1451, 1389, 1354, 1311, 1262, 1234, 1211, 1075, 1019, 996, 925, 815, 793, 754,

673, 658, 649, 639, 628, 618 cm-1. 13C SSNMR: 175, 147, 146, 140, 139, 136, 129, 126,

121, 119 ppm.

171

Pb[B(Im)4](3-NH2C6H4SO3)·(H2O), 13: Yield: 140 mg (31%) The material was

dried before the CHN analysis. Anal. Calcd for C18H19BN9O3PbS (%): C, 32.80; H, 2.90;

N, 19.10. Found: C, 31.69; H, 2.72; N, 18.33. IR: 1483, 1299, 1247, 1213, 1174, 1108,

1082, 1029, 991, 927, 815, 765, 692, 660 cm-1. 13C SSNMR: 148, 143, 130, 123, 115,

113 ppm.

+ Thermal polymerization of aniline derivatives between PbB(Im)4 layers: Fine crystals were placed in an open porcelain crucible and heated at 473 K for 24 h., 48 h. and 1 w. under atmospheric oxygen.

+ Chemical polymerization of aniline derivatives between PbB(Im)4 layers: Fine

crystals were suspended in H2O at 273 K and treated with aqueous solution of K2S2O8 at ratios ranging between 2:1 and 1:2 aniline:oxidant. The reaction mixtures were stirred at

273 K for 1 h. and then allowed to warm up to room temperature and react for 12 h. at room temperature.

X-Ray Crystallography: The X-ray intensity data for compounds 11-13 were measured at 100K (Bruker KRYO-FLEX) on a Bruker SMART APEX CCD-based X-ray diffractometer system equipped with a Mo-target X-ray tube (λ = 0.71073 Å) operated at

2000 W power. The crystals were mounted on a cryoloop using paratone N-Exxon oil and placed under a stream of nitrogen at 100 K. The detector was placed at a distance of

172

Table 8.1. Crystal data and structure refinement for 11.

Identification code Pb[B(Im)4](3-NH2C6H4COO)·(H2O)

Empirical formula C19 H19 B N9 O3 Pb Formula weight 639.43 Temperature 373(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 8.8415(11) Å α = 81.486(2)°. b = 8.9151(11) Å β = 80.621(2)°. c = 14.2785(18) Å γ = 79.454(2)°. Volume 1083.5(2) Å3 Z 2 Density (calculated) 1.960 Mg/m3 Absorption coefficient 7.828 mm-1 F(000) 614 Crystal size 0.30 x 0.10 x 0.05 mm3 Theta range for data collection 1.46 to 28.35°. Index ranges -11<=h<=11, -11<=k<=11, -18<=l<=18 Reflections collected 9380 Independent reflections 4933 [R(int) = 0.0314] Completeness to theta = 28.35° 91.4 % Absorption correction Empirical SADABS Max. and min. transmission 0.6956 and 0.4686 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 4933 / 198 / 299 Goodness-of-fit on F2 1.150 Final R indices [I>2sigma(I)] R1 = 0.0498, wR2 = 0.1307 R indices (all data) R1 = 0.0531, wR2 = 0.1325 Largest diff. peak and hole 6.020 and -2.926 e.Å-3

173

Table 8.2. Crystal data and structure refinement for 12.

Identification code Pb[B(Im)4](4-NH2C6H4COO)·(H2O)

Empirical formula C19 H21 B N9 O3 Pb Formula weight 641.45 Temperature 373(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 8.7632(5) Å α = 84.1630(10)°. b = 9.0574(5) Å β = 82.5060(10)°. c = 14.0903(8) Å γ = 77.1090(10)°. Volume 1077.82(11) Å3 Z 2 Density (calculated) 1.976 Mg/m3 Absorption coefficient 7.870 mm-1 F(000) 618 Crystal size 0.22 x 0.10 x 0.05 mm3 Theta range for data collection 2.31 to 28.28°. Index ranges -11<=h<=11, -11<=k<=11, -18<=l<=18 Reflections collected 9614 Independent reflections 5014 [R(int) = 0.0198] Completeness to theta = 28.28° 93.6 % Absorption correction Empirical SADABS Max. and min. transmission 0.6943 and 0.3931 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 5014 / 0 / 307 Goodness-of-fit on F2 1.047 Final R indices [I>2sigma(I)] R1 = 0.0168, wR2 = 0.0415 R indices (all data) R1 = 0.0172, wR2 = 0.0416 Largest diff. peak and hole 1.072 and -1.158 e.Å-3

174

Table 8.3. Crystal data and structure refinement for 13.

Identification code Pb[B(Im)4](3-NH2C6H4SO3)·(H2O)

Empirical formula C18 H21 B N9 O4 Pb S Formula weight 677.52 Temperature 373(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 8.3409(11) Å α = 89.295(2)°. b = 11.1636(15) Å β = 71.341(2)°. c = 12.3672(16) Å γ = 82.638(2)°. Volume 1081.5(2) Å3 Z 2 Density (calculated) 2.080 Mg/m3 Absorption coefficient 7.945 mm-1 F(000) 654 Crystal size 0.30 x 0.10 x 0.04 mm3 Theta range for data collection 1.84 to 28.31°. Index ranges -10<=h<=11, -14<=k<=14, -16<=l<=15 Reflections collected 9595 Independent reflections 5016 [R(int) = 0.0349] Completeness to theta = 28.31° 93.0 % Absorption correction Empirical SADABS Max. and min. transmission 0.7417 and 0.5337 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 5016 / 0 / 308 Goodness-of-fit on F2 1.084 Final R indices [I>2sigma(I)] R1 = 0.0334, wR2 = 0.0751 R indices (all data) R1 = 0.0370, wR2 = 0.0767 Largest diff. peak and hole 2.490 and -1.367 e.Å-3

175

SADABS program. The structure was solved and refined using the Bruker SHELXTL

(version 6.1) software package until the final anisotropic full-matrix least-squares

refinement on F2 converged. Experimental details for the three structures are shown in

Table 8.1-8.3.

Results and Discussion

We examined the enclathration of three modified anilines and followed the same

route as used in the preparation of the benzoate material Pb[B(Im)4](C6H5COO)·0.5H2O

175 for the incorporation of the three aniline derivatives. The reaction of Na[B(Im)4] with

Pb(NO3)2 in the presence of three-fold excesses of the aniline derivatives in pH = 8 water

or water/ethanol solutions give the desired crystalline products of Pb[B(Im)4](3-

NH2C6H4COO)·(H2O) 11, Pb[B(Im)4](4-NH2C6H4COO)·(H2O) 12, and Pb[B(Im)4](3-

NH2C6H4SO3)·(H2O), 13 in good yields. The three aniline derivatives namely m- aminobenzoic acid, p-aminobenzoic acid, and m-amino benzenesulfonic acid, vary in the location of substitution or in the identity of the anionic functional group. Both aspects affect the orientation and binding properties of the monomer in the interlayer spacing which also affects the degree of polymerization as discussed below.

All three syntheses afforded crystalline products, and the structures of the resultant compounds were elucidated by single-crystal X-ray analysis. Figures 8.2, 8.3, and 8.4 show the asymmetric units and extended network structures of compounds 11, 12, and 13 respectively. The two amino benzoic acid materials 11 and 12 exhibit very similar

176

binding properties and similarly sized asymmetric units. In both cases, the layered lead-

borate network is retained (relative to the Pb[(Im)4](NO3) parent compound) and the

carboxylates are bound to the Pb(II) metal ion in a symmetric bidentate fashion. In addition, the layered organization of the anions in the two materials are similar to each

175 other and the benzoate substituted material Pb[B(Im)4](C6H5COO)·0.5H2O. All three

materials have an alternating monolayer of arene rings aligned in a parallel fashion in the

interlayer spacing. In compounds 11 and 12, the monomers are aligned at 48.6° and

34.3° angles respectively to the plane of the lead atoms.

The organization of the aniline sulfonic acid units (deprotonated metanilic acids) in

13 is slightly different than that in compounds 11 and 12. Although the sulfonates are

located near to the lead center, no covalent bonding seems to occur between the metal

and the sulfonate group. This lack of binding appears to induce a change in the reactivity

of the monomer towards polymerization (vida infra). The extended structure of the

material shows that in compound 13, the aniline derivatives lined up parallel to the lead-

borate layer.

Figure 8.2. The asymmetric unit with 50% thermal ellipsoids of [PbB(Im)4]-(3-

NH2C6H4CO2)(H2O) (1). The solvent water molecule is not shown. 177

Figure 8.3 Extended structure of 11 along the a-axis. H-atoms are omitted for clarity.

Figure 8.4 The asymmetric unit with 50% thermal ellipsoids of

[PbB(Im)4]-(4-NH2C6H4CO2)(H2O) (2). The solvent water molecule is not shown.

178

Figure 8.5 Extended structure of 12 along the a-axis. H-atoms are omitted for clarity.

Figure 8.6 The asymmetric unit with 50% thermal ellipsoids of

[PbB(Im)4](3-NH2C6H4SO3)(H2O) (13). The solvent water molecule is not shown.

179

Figure 8.7 Extended structure of 13 along the a-axis. H-atoms are omitted for clarity.

In order to investigate the possibility of ambient oxygen induced polymerization of

compounds 11-13 we simply heated the crystalline materials at various temperatures

under aerobic conditions and monitored the material via solid state NMR and infrared

spectroscopy. Figure 8.5 summarizes the 13C solid state NMR spectra of heated samples

of 11 along with the starting materials 4-aminobenzoic acid and Pb[B(Im)4](NO3) for comparison. After heating a crystalline sample of 11 to 473 K in an open porcelain crucible for 24 h., the color of the sample remains unchanged and the NMR spectrum shows no significant change. Even upon heating the sample for one week at 473 K, no noticeable change is observed either in the color of the species or in the 13C NMR

spectrum shifts. Similarly, FT-IR experiments on the heated samples showed no

noticeable change after extended heating. As a result we have observed that 4-amino

benzoic acid is not reactive under these conditions when confined in the lead-borate

material. Two factors appear to contribute to this lack of reactivity. 180

First, the acid units are bound to Pb sites at the layer interface, which limits their

mobility in the layer. As a result, even though the monomers are closely packed, they

cannot move close enough to react in the solid. In addition, the orientation of the amine

group is such that it is not close to another arene C-H bond. The closest NH2-CH distance is 6.14 Å, which is too far for the restricted 4-aminobenzoic acid to move in compound 11.

In compound 12, the orientation of amino group on the benzoic acid is different than that in compound 11, and there is a clear difference in reactivity upon heating (Figure

7.6). The reactivity of m-aminobenzoic acid had been shown in earlier studies.183 When a

crystalline sample of 12 was heated up to 473 K under ambient oxygen, there was an

obvious color change from white to light purple which can be indicative of

polymerization of m-aminobenzoic acid monomer. The solid state NMR spectrum of the

heated sample changes as compared to the starting material as can be seen in Figure 8.6.

The major change in the spectrum is observed in the 110-150 ppm region where the

peaks are observed to broaden slightly. This result is consistent with some

polymerization of the m-aminobenzoic acid as the linewidths in the 13C spectrum for the

polymer are larger than the monomer. Upon extended heating for up to one week, there

was not much additional change in the spectrum. Although some reactivity with m- aminobenzoic acid is observed, it seems like the polymerization reaction only proceeds to

a small extent. This is in stark contrast to the monomer alone, which polymerizes readily

at 473 K. As in compound 11, this may result from the fact that the monomers in the host

181

180 160 140 120 100 80 13C ppm

13 Figure 8.8. C SSNMR CP/MAS, cp time of 1 ms of (a) Pb[B(Im)4](NO3)

(b) 4-amino benzoic acid (c) [PbB(Im)4](4-NH2C6H4CO2)(H2O)

(d) [PbB(Im)4](4-NH2C6H4CO2)(H2O) heated for 1 day.

(e) [PbB(Im)4](4-NH2C6H4CO2)(H2O) heated for 1 week. 182

200 180 160 140 120 100 80 13C ppm

13 Figure 8.9. C SSNMR CP/MAS, cp time of 2 ms of (a) Pb[B(Im)4](NO3)

(b) 3-amino benzoic acid (c) [PbB(Im)4](3-NH2C6H4CO2)(H2O) (d) [PbB(Im)4](3-

NH2C6H4CO2)(H2O) heated for 1 day. (e) [PbB(Im)4](3-NH2C6H4CO2)(H2O) heated for 2

days. (f) [PbB(Im)4](3-NH2C6H4CO2)(H2O) heated for 1 week.

183

200 180 160 140 120 100 80 60 13C ppm

13 Figure 8.10. C SSNMR CP/MAS, cp time of 3 ms of (a) PbBIm4 (b) 3-aniline sulfonic acid (c) [PbB(Im)4](3-NH2C6H4SO3)(H2O) (d) [PbB(Im)4](3-NH2C6H4SO3)(H2O) heated

for 1 day. (e) [PbB(Im)4](3-NH2C6H4SO3)(H2O) heated for 2 days. (f) [PbB(Im)4](3-

NH2C6H4SO3)(H2O) heated for 1 week. 184

material do not have an optimal orientation for polymerization, and also are not free to

move because they are anchored to the Pb(II) centers.

The reactivity of aniline derivatives 11 and 12 using potassium persulfate, K2S2O8, as an oxidant to carry out the polymerization reaction was investigated. Materials 11 and

12 were suspended in H2O and treated with a variety of concentrations of the oxidant at

room temperature with gentle mixing for 12 h. With a 1:2 ratio of the sample to the

oxidant a noticeable color change to purple was observed in the samples. Whereas the

solid state 13C NMR of the resultant compounds is reminiscent of that of polyaniline, loss of the Pb-borate material peaks was observed, most likely due to oxidative degradation of the host material. To avoid oxidizing the lead-borate layer, we exposed the Pb-borate parent compound Pb[B(Im)4](NO3) to only one equivalent of oxidant. This concentration

of K2S2O8 also causes decomposition of the material. Using lower than one equivalent of

13 K2S2O8 with materials 11 and 12 does not produce any observable change in the C

NMR spectrum.

As mentioned above, the different orientation and lack of metal-host bonding makes

compound 13 potentially improved candidate for polymerization. Heating crystalline

compound 3 under the same conditions as 11 and 12 results in a color change from a

white color to a light purple color. The 13C solid-state NMR spectra of the resulting

products show major changes as compared with the starting material. Although the peaks

corresponding to the lead-borate material itself are preserved, the 110 ppm-150 ppm

region of the 13C NMR spectrum shows significant broadening. With additional heating

of the sample for one week, the monomeric substituted aniline peaks almost completely

185

disappear as shown in Figure 8.10. Heating the same material to 473 K in the absence of

oxygen did not lead to any observable change in color or the 13C SSNMR spectrum. This

observation supports the theory that the polymerization proceeds via the oxidation of the

aniline by the ambient oxygen as previously reported.

Chemical oxidant induced polymerization of aniline-sulfonic acid in our host

material was also explored. K2S2O8 was used in aqueous conditions in a 2:1 molar ratio of the aniline to the oxidant. The first physical observation was a change in the color of the solution. After filtering off the solid we analyzed the product by solid state NMR.

The spectrum showed that the solid material was the inorganic host material without the aniline inside. At this point there are two possibilities; either polymerization takes place within the host material and the polymer is released to the solution or alternatively the polymerization reaction takes place in solution upon monomer anion exchange with persulfate. A solution of aniline sulfonic acid was generated with the persulfate oxidant at the same molar ratio under the same conditions as above. Polymerization of the aniline sulfonic acid monomers was observed under these conditions. It can not be ruled out that polymerization is taking place in solution rather than within the confines of the solid when oxidized with K2S2O8.

Three different aniline derivatives have been enchlathrated in a layered Pb-borate

network for the purpose of polymerization within the confines of the solid. Although all

three monomers line up in the interlayer spacing, they orient differently which affects

their reactivity. The aniline sulfonic acid derivative shows the highest reactivity towards

thermal polymerization at 473K which appears to result from the mobility of the

186 monomer within the layer. The other two compounds have the monomer units bound to metal sites at the layer interface, and this decrease in mobility inhibits the polymerization reaction.

187

REFERENCES

1 Kadish, K. M.; Smith, K. M.; Guilard, R, (Eds.), The Porphyrin Handbook, Vol. 1-10, Academic Press, New York, 2000.

2 Smith, K. M. (ed) Porphyrins and Metalloporphyrins Elsevier, Amsterdam 1975.

3 Buchler, J. W. in Dolphin, D. (Ed.) The porphyrins, vol. 1, Academic Press, New York, 1979, p.389.

4 Bonnett, R. Chem. Soc. Rev. 1995, 24, 19.

5 Poulos, T. L., in Kadish, K. M.; Smith, K. M.; Guilard, R. (Eds), The Porphyrin Handbook, Vol. 4, Academic Press, New York, 2000, p. 189.

6 Kadish, K. M.; Smith, K. M.; Guilard, R, (Eds.), The Porphyrin Handbook, Vol. 6, Academic Press, New York, 2000.

7 Mansuy, D.; Lange, M.; Chottard, J. C. J. Am. Soc. 1979, 100, 3213; Collman, J. P.; Brauman, J. I.; Doxsee, K. M.; Sessler, J. L.; Morris, R. M.; Gibson, Q. H. Inorg. Chem. 1983, 22, 1427; Guilard, R.; Lagrange, G.; Tabard, A.; Lançon, D.; Kadish, K. M. Inorg. Chem. 1985, 24, 3649.

8 Lexa, D.; Mispelter, J.; Savéant, J.-M. J. Am. Chem. Soc. 1981, 103, 6806.

9 (a) Clarke, D. A.; Dolphin, D.; Johnson, A. W.; Pinnock, H. A. J. Chem. Soc.( C) 1968,881. (b) Clarke, D. A.; Grigg, R.; Johnson, A. W. Chem. Commun. 1966, 208.

10 Mansuy, D.; Lange, M.; Chottard, J. C.; Guerin, P.; Morlidre, P.; Brault, D.; Rougee, M. J. Chem. Soc., Chem. Commun., 1977, 648.

11 Peng, S.-M.; Ibers, J. A. J. Am. Chem. Soc. 1976, 98, 8032.

12 Dailey, K. K.; Yap, G. P. A; Rheingold, A. L.; Rauchfuss, T. B. Ang. Chem. Int. Ed. Engl. 1996, 35, 1833.

13 Contakes, S. M.; Beatty, S. T.; Dailey, K. K.; Rauchfuss, T. B.; Fenske, D. Organometallics 2000; 19(23); 4767.

14 Arnold, D. P.; Worthington, E. I; Sakata, Y.; Sugiura, K.; Chem. Commun. 1998, 2331.

188

15 Latos-Grażyński, L. in Kadish, K. M.; Smith, K. M.; Guilard, R, (Eds.), The Porphyrin Handbook, Vol. 2, Academic Press, New York, 2000, p. 361.

16 Furuta, H.; Maede, H.; Osuka, A. Chem. Commun. 2002, 1795.

17 Lash, T. D. Synlett 1999, 3, 279.

18 Srinivasan, A.; Furuta, H. Acc. Chem. Res. 2005, 38, 10.

19 Stępień, M.; Latos-Grażyński, L Acc. Chem. Res. 2005, 38, 88.

20 Senge, M. O. in Kadish, K. M.; Smith, K. M.; Guilard, R, (Eds.), The Porphyrin Handbook, Vol. 1, Academic Press, New York, 2000, p. 239.

22 Chmielewski P.J., Latos-Grażyński, L. Inorg. Chem. 1992, 31, 5231.

23 Chmielewski P.J., Latos-Grażyński, L. Inorg. Chem. 1998, 37, 4179.

24 Chmielewski P.J., Latos-Grażyński, L. Inorg. Chem. 2004, 43, 5564.

25 Harvey, J. D.; Ziegler, C. J. Coord. Chem. Rev. 2003, 247, 1.

26 Chmielewski P.J., Latos-Grażyński, L. Coord. Chem. Rev. 2005, 249, 2510.

27 Furuta H., Asano T., Ogawa T., J. Am. Chem. Soc. 1994,116, 767.

28 Chmielewski P.J., Latos-Grażyński, L., Rachlewicz K., Glowiak T., Angew. Chem. Int. Ed. 1994, 33, 779.

29 (a) Clark, P. F.; Elvidge, J. A.; Linstead, R. P. J. Chem. Soc. 1954, 2490. (b) Elvidge, J. A.; Golden, J. H. J. Chem. Soc. 1957, 700.

30 Phthalocyanines. Properties and Applications; Leznoff, C. C., Lever, A. B. P., Eds.; VCH publishers, Inc.: Weinheim, 1989, 1993, 1996; Vols. 1-4.

31 Harvey, J. D.; Ziegler, C. J. J. Inorg. Biochem. 2006, 100, 869.

32 Geier III, G.R., Haynes D.M., Lindsey J.S., Org. Lett. 1999, 1, 1455.

33 Liu, B. Y.; Brückner, C.; Dolphin, D. Chem. Commun. 1996, 2141.

34 Lash, T.D.; Richter, D.T.; Shiner, C.M. J. Org. Chem. 1999, 64, 7973. 189

35 Ghosh, A. Angew. Chem. Int. Ed. Engl. 1995, 34, 1028.

36 Schmidt, I.; Chmielewski, P. J.; Ciunik, Z. J. Org. Chem. 2002, 67, 8917.

37 Furuta H.; Kubo N.; Maeda H.; Ishizuka T.; Osuka A.; Nanami H.; Ogawa, T. Inorg. Chem. 2000, 39, 5424.

38 Furuta H.; Youfu, K.; Maeda H.; Ishizuka T.; Osuka A. Angew. Chem. Int. Ed. 2003, 42, 2186.

39 Chmielewski, P. J.; Latos-Grażyński, L.; Schmidt, I. Inorg. Chem. 2000, 39, 5475.

40 Furuta, H.; Maeda, H.; Osuka, A. Org. Lett. 2002, 4(2), 181.

41 Mitrikas, G.; Calle, C.; Schweiger, A. Angew. Chem. Int. Ed. 2005, 44, 3301.

42 Maeda, H.; Osuka, A.; Ishikawa, Y.; Aritome, I.; Hisaeda, Y.; Furuta, H. Org. Lett. 2003, 5(8), 1293.

43 Lash, T.D.; Richter, D.T.; Shiner, C. M. J. Org. Chem. 1999, 64, 7973.

44 Chmielewski, P. J.; Latos-Grażyński, L.; Głowiak, T. J. Am. Chem. Soc. 1996, 118, 5690.

45 Chmielewski, P. J.; Latos-Grażyński, L. J. Chem. Soc., Perkin Trans.2 1995, 503.

46 Xiao, Z.; Patrick, B. O.; Dolphin, D. Chem. Commun. 2003, 1062.

47 Chmielewski, P. J.; Latos-Grażyński, L. Inorg. Chem. 2000, 39, 5639.

48 Harvey, J. D.; Shaw, J. L.; Herrick, R. S.; Ziegler, C. J. Chem. Commun. 2005, 4663.

49 Chmielewski, P. J.; Latos-Grażyński, L. Inorg. Chem. 1997, 36, 840.

50 Xiao, Z.; Patrick, B. O.; Dolphin, D. Inorg. Chem. 2003, 42, 8125.

51 Maeda, H.; Ishikawa, Y.; Matsuda, T.; Osuka, A.; Furuta, H. J. Am. Chem. Soc. 2003, 125, 11822.

52 Furuta, H.; Ogawa, T.; Uwatoko, Y.; Araki, K. Inorg. Chem. 1999, 38, 2676.

53 Furuta, H.; Morimoto, T.; Osuka, A. Org. Lett. 2003, 5(9), 1427.

54 Bohle, D. S.; Chen, W.-C.; Hung, C.-H. Inorg. Chem. 2002, 41, 3334.

55 Harvey, J. D.; Ziegler, C. J. Chem. Commun. 2003, 2890. 190

56 Rachlewicz, K.; Wang, S.-L.; Ko, J.-L.; Hung, C.-H.; Latos- Grażyński, L. J. Am. Chem. Soc. 2004, 4420.

57 Harvey, J. D.; Ziegler, C. J.; Telser, J.; Ozarowski, A.; Krzystek Inorg. Chem. 2005, 44, 4451.

58 Harvey, J. D.; Ziegler, C. J. Chem. Commun. 2004, 1666.

59 Liu, J.-C.; Ishizuka, T.; Osuka, A.; Furuta, H. Chem. Commun. 2004, 1908.

60 Ogawa, T.; Furuta, H.; Takahashi, M.; Morino, A.; Uno, H. J. Organomet. Chem. 2000, 611, 551.

61 Chen, W.-C.; Hung, C.-H. Inorg. Chem. 2001, 40, 5070.

62 Rachlewicz, K.; Wang, S.-L.; Peng, C.-H.; Hung, C.-H.; Latos- Grażyński, L. Inorg. Chem. 2003, 42, 7348.

63 Rachlewicz, K.; Gorzelańczyk, D.; Latos- Grażyński, L. Inorg. Chem. 2006, 45, 9742.

64 Harvey, J. D.; Ziegler, C. J. Chem. Commun. 2002, 1942.

65 Furuta, H.; Ishizuka, T.; Osuka, A. J. Am. Chem. Soc. 2002, 124, 5622.

66 Zhu, X.; Wong, W.-K.; Lo, W.-K.; Wong, W.-Y. Chem. Commun. 2005, 1022.

67 Furuta, H.; Morimoto, T.; Osuka, A. Inorg. Chem. 2004, 43, 1618.

68 Hung, C.-H; Chen, W.-C.; Lee, G.-H.; Peng, S.-M. Chem. Comun. 2002, 1516.

69 Furuta, H.; Kibo, N.; Maeda, H.; Ishizuka, T.; Osuka, A.; Nanami, H.; Ogawa, T. Inorg. Chem. 2000, 39, 5424.

70 Furuta, H.; Youfu, K.; Maeda, H.; Osuka, A. Angew. Chem. Int. Ed. 2003, 42, 2186.

71 Chmielewski, P. J.; Schmidt, I. Inorg. Chem. 2004, 43, 1885.

72 Srinivasan, A.; Furuta, H.; Osuka, A. Chem. Commun. 2001, 1666.

73 Toganoh, M.; Ishizuka, T.; Furuta, H. Chem. Commun. 2004, 2464.

74 Kiran, B.; Nguyen, M. T. J. Organomet. Chem. 2002, 643-644, 265.

75 Tsutsui, M.; Hrung, C. P.; Ostfeld, D.; Srivastava, T. S.; Cullen, D. L.; Meyer, E. F. J. Am. Chem. Soc. 1975, 97, 3952.

191

76 Berlin, K.; Breitmaier, E. Angew. Chem. Int. Ed. Engl. 1994, 33, 1246. See also: Berlin, K.; Breitmaier, E. Angew. Chem. Int. Ed. Engl. 1994, 33, 219.

77 Lash, T.D.; Chaney, S. T.; Richter, D.T. J. Org. Chem. 1998, 63, 9076.

78 Stępień, M.; Latos-Grażyński, L. Chem. Eur. J. 2001, 7(23), 5113.

79 Stępień, M.; Latos-Grażyński, L. Org. Lett. 2003, 5(19), 3379.

80 Stępień, M.; Latos-Grażyński, L. Inorg. Chem. 2003, 42, 6183.

81 Stępień, M.; Latos-Grażyński, L.; Szterenberg L.; Panek, J.; Latajka Z. J. Am. Chem. Soc. 2004, 126, 4566.

82 Stępień, M.; Latos-Grażyński, L.; Szterenberg L. Inorg. Chem. 2004, 43, 6654.

83 Hung, C.-H.; Chang, F.-C.; Lin, C.-Y.; Rachlewicz, K.; Stępień, M.; Latos-Grażyński, L.; Lee, G.-H.; Peng, S.-M. Inorg. Chem. 2004, 43, 4118.

84 A similar bridging unit was reported before: Rybak-Akimova, E. V.; Busch, D. H.; Kahol, P. K.; Pinto, N.; Alcock, N. W.; Clase, H. J. Inorg. Chem. 1997, 36, 510.

85 Stępień, M.; Latos-Grażyński, L. J.Am. Chem. Soc. 2002, 124, 3838.

86 Pacholska, E.; Latos-Grażyński, L.; Ciunik, Z. Chem.-Eur. J. 2002, 8, 5403.

87 Pacholska-Dudziak, E.; Skonieczny, J.; Pawlicki, M.; Szterenberg L.; Stępień, M.; Latos-Grażyński, L. Inorg. Chem. 2005, 44, 8794.

88 Lash, T.D. Angew. Chem. Int. Ed. Engl. 1995, 34, 2533.

89 Richter D. T.; Lash T. D. Tetrahedron, 2001, 57, 3657.

90 Szymanski, J. T.; Lash, T. D. Tetrahedron. Lett. 2003, 44, 8613.

91 Stępień, M.; Latos-Grażyński, L.; Lash, T. D.; Szterenberg, L. Inorg. Chem. 2001, 40, 6892.

92 Lash, T. D.; Rasmussen, J. M.; Bergman, K. M.; Colby, D. A. Org. Lett. 2004, 6(4), 459.

93 Venkatraman, S.; Anannd, V. G.; Pushpan, S. K.; Sankar, J.; Chandrashekar, T. K. Chem. Commun. 2002, 462.

94 Liu, D.; Lash, T. D. Chem. Commun. 2002, 2426.

192

95 Liu, D.; Ferrence G. M.; Lash, T. D. J. Org. Chem. 2004, 69, 6079.

96 Lash, T.D.; Chaney, S.T. Tetrahedron Lett. 1996, 37, 8825.

97 Bergman, K. M.; Ferrence G. M.; Lash, T. D. J. Org. Chem. 2004, 69, 7888.

98 Berlin, K; Steinbeck, C.; Breitmaier, E. Synthesis 1996, 336.

99 Lash, T.D.; Chaney, S.T. Angew. Chem. Int. Ed. Engl. 1997, 36, 839.

100 Graham, S. R.; Ferrence, G. M.; Lash, T. D. Chem. Commun. 2002, 894.

101 Lash, T. D.; Colby, D. A.; Graham, S. R.; Ferrence, G. M.; Szczepura, L. F. Inorg. Chem. 2003, 42, 7326.

102 Colby, D. A.; Lash, T. D. Chem. Eur. J. 2002, 8, 5397.

103 Lash, T. D.; Colby, D. A.; Ferrence, G. M. Eur. J. Org. Chem. 2003, 42, 4533.

104 Colby, D. A.; Ferrence G. M.; Lash, T. D. Angew. Chem. Int. Ed. 2004, 43, 1346.

105 Lash, T. D. Chem. Commun. 1998, 1683.

106 Lash, T. D., Hayes, M. J. Angew. Chem. Int. Ed. 1997, 36, 840.

107 Hayes, M. J.; Spence, J. D.; Lash, T. D. Chem. Commun. 1998, 36, 2409.

108 Lash, T. D.; Muckey, M. A.; Hayes, M. J.; Liu, D.; Spence, J. D.; Ferrence, G. M. J. Org. Chem. 2003, 68, 8558.

109 Muckey, M. A.; Szczepura, L. F.; Ferrence, G. M.; Lash, T. D. Inorg. Chem. 2002, 41, 4840.

110 Lash, T. D.; Colby, D. A.; Szczepura, L. F. Inorg. Chem. 2004, 43, 5258.

111 Sprutta, N.; Latos-Grażyński, L. Tetrahedron Lett. 1999, 40, 8457.

112 Chmielewski, M. J.; Pawlicki, M.; Sprutta, N.; Szterenberg L.; Latos-Grażyński, L. Inorg. Chem. 2006, 45, 8664.

113 Pawlicki, M.; Latos-Grażyński, L. Chem.-Eur. J. 2003, 9, 4650.

114 Pawlicki, M.; Latos-Grażyński, L. J. Org. Chem., 2005, 70, 9123.

115 Crossley, M. J.; King, L. G. J. Chem. Soc., Chem. Commun. 1984, 920.

193

116 Gouterman, M.; Hall, R.; Khalil, G. E.; Martin, P. C.; Shankland, E. G.; Cerny, R. L. J. Am. Chem. Soc. 1989, 111, 3702.

117 McCarthy, J. R.; Jenkins, H. A.; Brückner, C. Org. Lett. 2003, 5, 19.

118 Pawlicki, M.; Latos-Grażyński, L.; Szterenberg L. Inorg. Chem. 2005, 44, 9779.

119 Arndtsen B. A.; Bergman R. G.; Mobley T. A.; Peterson T. H. Acc. Chem. Res. 1995, 28, 154; Jia C. G.; Kitamura T.; Fujiwara Y. Acc. Chem. Res. 2001, 34, 633; Jones W. D. Acc. Chem. Res. 2003, 36, 140.

120 Mąkosza, M.; Winiarski, J. Acc. Chem. Res. 1987, 20, 282.

121 Mąkosza, M.; Kwast, A. Eur. J. Org. Chem. 2004, 10, 2125; Florio, S.; Lorusso, P.; Luisi, R.; Granito, C.; Ronzini, L.; Troisi, L. Eur. J. Org. Chem. 2004, 10, 2118; Drew, M. D.; Jackson, D. A.; Lawrence, N. J.; Liddle, J.; Pritchard, R. G. Chem. Commun. 1997, 2, 189.

122 Mąkosza, M.; Wojciechowski, K. Chem. Rev. 2004, 104, 2631.

123 The Porphyrin Handbook, Vol. II (Eds.: K. M. Kadish, K. M. Smith, R. Guilard), Academic Press, New York, 2000.

124 Elvidge, J. A.; Linstead, J. P. J. Chem. Soc. 1952, 5008.

125 Fernández-Lázaro, F.; Torres, T.; Hauschel, B; Hanack, M. Chem. Rev. 1998, 98, 563.

126 Shelly, K.; Finster, D. C.; Lee, Y. J.; Scheidt, W. R.; Reed, C. A. J. Am. Chem. Soc. 1985, 107, 5955; Taylor, I. F., Jr.; Hall, E. A.; Amma, E. L. J. Am. Chem. Soc. 1969, 91, 5745.

127 Kadish K. M.; Araullo C.; Maiya G. B.; Sazou D.; Barbe J.-M.; Guilard R. Inorg. Chem. 1989, 28, 2528.

128 Einsle, O.; Tezcan, F. A.; Andrade, S. L. A.; Schmid, B.; Yoshida, M.; Howard, J. B.; Rees, D. C. Science 2002, 297(5587), 1696.

129 See also the reviews: Lash, T. D. Synlett 1999, 3, 279; Furuta, H.; Maeda, H.; Osuka, A. Chem. Commun. 2002, 1795; Stępień, M.; Latos-Grażyński, L. Acc. Chem. Res. 2005, 38, 88; Srinivasan, A.; Furuta, H. Acc. Chem. Res. 2005, 38, 10.

130 Elvidge, J. A.; Linstead, R. P. J. Chem. Soc. 1952, 5008.

194

131 Agostinelli, E.; Attanasio, D.; Collamati, I.; Fares, V. Inorg. Chem. 1984, 23, 1162; Dirk, C. W.; Marks, T. J. Inorg. Chem. 1984, 23, 4325; Fernández-Lázaro, F.; Torres, T.; Hauschel, B.; Hanack, M. Chem. Rev. 1998, 98, 563.

132 Clark, P. F.; Elvidge, J. A.; Linstead, R. P. J. Chem. Soc. 1954, 2490; Elvidge, J. A.; Golden, J. H. J. Chem. Soc. 1957, 700.

133 Wu, R.; Çetin, A.; Durfee, W. S.; Ziegler, C. J. Ang. Chem. Int. Ed. 2006, 45, 5670.

134 Power, P. P.; Shoner, S. C. Angew. Chem. Int. Ed. Engl. 1991, 30, 330; Nguyen, T.; Panda, A.; Olmstead, M. M.; Richards, A. F.; Stender, M.; Brynda, M.; Power, P. P. J. Am. Chem. Soc. 2005, 127, 8545.

135 Murray, B. D.; Power, P. P. J. Am. Chem. Soc. 1984, 106, 7011.

136 Ellison, J. J.; Power, P. P.; Shoner, S. C. J. Am. Chem. Soc. 1989, 111, 8044.

137 Panda, A.; Stender, M.; Wright, R. J.; Olmstead, M. M.; Klavins, P.; Power, P. P. Inorg. Chem. 2002, 41, 3909.

138 Fernández-Lázaro, F.; Torres, T.; Hauschel, B.; Hanack, M. Chem. Rev. 1998, 98, 563.

139 Scheidt, W. R. J. Am. Chem. Soc., 1974, 96, 84.

140 Scheidt, W. R.; Cunningham, J. A.; Hoard, J. L. J. Am. Chem. Soc. 1973, 95, 8289.

141 (a) Sessler, J. L.; Seidel, D. Angew. Chem. Int. Ed. 2003, 42, 5134. (b) Sessler, J. L.; Gebauer, A.; Vogel, E. Porphyrin Handbook 2000, 2, 1.

142 Jayaraj, K.; Gold, A.; Austin, R. N.; Ball, L. M. Inorg. Chem. 1997, 36, 4555.

143 Crossley, M. J.; King, L. G. J. Chem. Soc., Chem. Commun. 1984, 920.

144 Khalil, G.; Gouterman, M.; Ching, S.; Costin, C.; Coyle, L.; Gouin, S.; Green, E.; Sadilek, M.; Wan, R.; Yearyean, J.; Zelelow, B. J. Porphyrins Phthalocyanines 2002, 6, 135.

145 (a) Hill, C. L.; Schardt, B. C. J. Am. Chem. Soc. 1980, 102, 6374. (b) Groves, J. T.; Kruper, W. J.; Haushalter, R. C. J. Am. Chem. Soc. 1980, 102, 6375. (c) Meunier, B. Bull. Soc. Chim. Fr. 1986, 4, 578. (c) Castellino, A. J.; Bruice, T. C. J. Am. Chem Soc. 1988, 110, 158. (c) Traylor, T. G.; Miksztal, A. R. J. Am. Chem. Soc. 1989, 111, 7443.

146 Gross, Z.; Simkhovich, L.; Galili, N. Chem. Commun. 1999, 7, 599.

195

147 (a) Brückner, C.; Rettig, S. J.; Dolphin, D. J. Org. Chem. 1998, 63, 2094. (b) McCarthy, J. R.; Jenkins, H. A.; Brückner, C. Org. Lett. 2003, 5, 19.

148 Armstrong, R. S.; Foran, G. J.; Hambley, T. W. Acta. Cryst. C. 1993, 49, 236.

149 Ostovic, D.; Bruice, T. C. Acc. Chem. Res. 1992, 25, 314.

150 (a) Trofimenko, S. (Ed.), Scorpionates: The Coordination Chemistry of the Polypyrazolylborate Ligands, Imperial College Press, London, 1999. (b) Edelmann, F. D. Angew. Chem. Int. Ed. 2001, 40(9), 1656. (c) Marques N.; Sella A.; Takats, J. Chem. Rev. 2002, 102(6), 2137.

151 Hammes, B. S.; Kieber-Emmons, M. T.; Letizia, J. A.; Shirin, Z.; Carrano, C. J.; Zakharov, L. N.; Rheingold, A. L. Inorg. Chim. Acta 2003, 346, 227; Smith, J. N.; Shirin, Z.; Carrano, C. J. J. Am. Chem. Soc. 2003, 125, 868.

152 Garner, M.; Reglinski, J.; Cassidy, I.; Spicer, M. D.; Kennedy, A. R. Chem. Commun. 1996, 1975; Reglinski, J.; Garner, M.; Cassidy, I. D.; Slavin, P. A.; Spicer, M. D.; Armstrong, D. R. J. Chem. Soc., Dalton Trans. 1999, 2119.

153 Reglinski, J.; Spencer, M. D.; Garner, M.; Kennedy, A. R. J. Am. Chem. Soc. 1999, 121, 2317.

154 Cassidy, I.; Garner, M.; Kennedy, A. R.; Potts, G. B. S.; Reglinski, J.; Slavin P.A.; Spicer, M. D. Eur. J. Inorg. Chem. 2002, 1235.

155 Dodds, C. A.; Lehmann, M.-A.; Ojo, J. F.; Reglinski, J.; Spicer, M. D. Inorg. Chem 2004, 43, 4927.

156 Foreman, M. R. St.-J.; Hill, A. F.; Owen, G. R.; White, A. J. P.; Williams, D. J. Organometallics 2003, 22, 4446.

157 Graham, L. A.; Fout, A. R.; Kuehne, K. R.; White, J. L.; Mookherji, B.; Marks, F. M.; Yap, G. P. A.; Zakharov, L. N.; Rheingold, A. L.; Rabinovich, D. Dalton Trans. 2005, 171.

158 (a) Slavin, P. A.; Reglinski, J.; Spicer, M. D.; Kennedy, A. R. J. Chem. Soc., Dalton Trans. 2000, 239. (b) Garner, M.; Lewinski, K.; Pattek-Janczyk, A.; Reglinski, J.; Sieklucka, B.; Spicer, M. D.; Szaleniec, M. Dalton Trans. 2003, 1181. (c) Dodds, A.; Jagoda, M.; Reglinski, J.; Spicer, M. D. Polyhedron 2004, 23, 445.

159 Hill, A. F.; Owen, G. R.; White, A. J. P.; Williams, D. J. Angew. Chem. Int. Ed. 1999, 18(38), 2579.

196

160 Santini, C.; Pettinari, C.; Lobbia, G. G. ; Spagna, R.; Pellei, M., Vallorani, F. Inor. Chim. Acta 1999, 285, 81.

161 Kuan, S. L.; Leong, W. K.; Goh, L. Y.; Webster, R. D. Organometallics 2005, 24, 4639.

162 Ge, P.; Haggerty, B. S.; Rheingold, A. L.; Riordan, C. G. J. Am. Chem. Soc. 1994, 116, 8406.

163 Thomas, J. C.; Peters, J. Inorg. Chem. 2003, 42, 5055.

164 (a) James, S. L. Chem. Soc. Rev. 2003, 32(5), 276. (b) Rosi, N. L.; Eddaoudi, M.; Kim, J.; O'Keeffe, M.; Yaghi, O. M. Cryst. Eng. Comm. 2002, 4, 401.

165 (a) Miller, J. S.; Manson, J. L. Acc. Chem. Res. 2001, 34(7), 563. (b) Evans, O. R.; Lin, W. Acc.Chem. Res. 2002, 35(7), 511. (c) Lin, W.; Ngo, H. L. Chemistry of Nanostructured Materials 2003, 261. (d) Noveron, J. C.; Lah, M. S.; Del Sesto, R. E.; Arif, A. M.; Miller, J. S.; Stang, P. J. J. Am. Chem. Soc. 2002, 124(23), 6613.

166 (a) Yoshizawa M.; Takeyama Y.; Okano T.; Fujita M. J. Am. Chem. Soc. 2003, 125(11), 3243-3247. (b) Ohmori O.; Fujita M. Chem. Commun. 2004, 14, 1586. (c) Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312(5771), 251.

167 a) Greenwell, H. C.; Jones, W.; Coveney, P. V.; Stackhouse, S. J. Mater. Chem. 2006, 16(8), 708. (b) Carrado, K. A. Handbook of Layered Materials 2004, 1. (c) Ogawa, M.; Kuroda, K. Chem. Rev. 1995, 95(2), 399. (d) Pinnavaia, T. J. Science 1983, 220(4595), 365.

168 Biradha, K.; Fujita, M. Perspectives in Supramolecular Chemistry 2003, 7, 211.

169 Rowsell, J. L. C.; Yaghi, O. M. Ang. Chem., Int. Ed. 2005, 44(30), 4670.

170 (a) Thomas, S.; Bertrand, J. A.; Occelli, M. L. Chem. Mater. 1999, 11(1), 184. (b) Hutson, N. D.; Gualdoni, D. J.; Yang, R. T. Chem. Mater. 1998, 10(11), 3707.

171 (a) Moyano, E. L.; del Arco, M.; Rives, V.; Yranzo, G. I. J. Org. Chem. 2002, 67(23), 8147. (b) Zhou, Y.; Hu, N.; Zeng, Y.; Rusling, J. F. Langmuir 2002, 18(1), 211. (c) Zhu, H. Y.; Zhao, J.-C.; Liu, J. W.; Yang, X. Z.; Shen, Y. N. Chem. Mater. 2006, 18(17), 3993.

172 (a) Stackhouse, S.; Coveney, P. V.; Sandre, E. J. Am. Chem. Soc. 2001, 123(47), 11764. (b) Itoh, T.; Shichi, T.; Yui, T.; Takagi, K. Langmuir 2005, 21(8), 3217. (c) Zeng, C.; Lee, L. J. Macromolecules 2001, 34(12), 4098.

197

173 (a) Kusukawa, T.; Fujita, M. J. Am. Chem. Soc. 1999, 121(6), 1397. (b) Yoshizawa, M.; Kusukawa, T.; Fujita, M.; Yamaguchi, K. J. Am. Chem. Soc 2000, 122(26), 6311. (c) Dahne, L.; Leporatti, S.; Donath, E.; Mohwald, H. J. Am. Chem. Soc. 2001, 123(23), 5431.

174 Hamilton, B. H.; Kelly K. A.; Wagler, T. A.; Espe, M. P.; Ziegler, C. J. Inorg. Chem. 2002, 41, 4984.

175 Hamilton, B. H.; Kelly K. A.; Wagler, T. A.; Espe, M. P.; Ziegler, C. J. Inorg. Chem. 2004, 43, 50.

176 Hamilton, B. H.; Cardon, T. B.; Lorigan, G. A.; Ziegler, C. J. Dalton Trans. 2005, 17, 2941.

177 (a) Trofimenko, S. J. Am. Chem. Soc. 1967, 89, 3170. (b) Trofimenko, S. J. Coord. Chem. 1972, 2, 75.

178 Kanatzidis, M. G.; Marcy, H. O.; McCarthy, W. J.; Kannewurf, C. R.; Marks, T. J., Solid State Ionics 1989, 32/33, 594.

179 (a) Kanatzidis, M. G.; Wu, C.-G. J. Am. Chem. Soc. 1989, 111, 4139. (b) Kanatzidis, M. G.; Marcy, H. O.; Wu, C. G.; DeGroot, D. C.; Kannewurf, C. R. Adv. Mater. 1990, 2, 364. (c) Mehrotra, V.; Giannelis, E. P. Solid State Commun. 1991, 77, 155.

180 Kanatzidis, M. G.; Tonge, L. M.; Marks, T. J. J. Am. Chem. Soc. 1987, 109, 3797.

181 (a) Enzel, P.; Bein, T., J. Phys. Chem. 1989, 93, 6270. (b) Wu, C.-G.; Bein, T. Chem. Mater. 1994, 6, 1109.

182 Liu, Y.-J.; Kanatzidis, M. G. Inorg. Chem. 1993, 32, 2989.

183 Isupov, V. P.; Chupakhina, L. E.; Ozerova, M. A.; Kostrovsky, V. G.; Poluboyarov, V. A. Solid State Ionics 2001, 231, 141.

184 Moujahid, E. L.; Dubois, M.; Besse, J.-P.; Leroux, F. Chem. Mater. 2002, 14, 3799.

198