Tuning the Properties of Metal-Ligand Complexes to Modify the Properties

University of Massachusetts Amherst ScholarWorks@UMass Amherst

Open Access Dissertations

5-2012 Tuning the Properties of Metal-Ligand Complexes to Modify the Properties of Supramolecular Materials Ian Henderson University of Massachusetts Amherst, [email protected]

Follow this and additional works at: https://scholarworks.umass.edu/open_access_dissertations Part of the Polymer and Organic Materials Commons

Recommended Citation Henderson, Ian, "Tuning the Properties of Metal-Ligand Complexes to Modify the Properties of Supramolecular Materials" (2012). Open Access Dissertations. 555. https://doi.org/10.7275/w9wa-k694 https://scholarworks.umass.edu/open_access_dissertations/555

This Open Access Dissertation is brought to you for free and open access by ScholarWorks@UMass Amherst. It has been accepted for inclusion in Open Access Dissertations by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].

TUNING THE PROPERTIES OF METAL-LIGAND COMPLEXES

TO MODIFY PROPERTIES OF SUPRAMOLECULAR MATERIALS

A Dissertation Presented

IAN M. HENDERSON

Submitted to the Graduate School of the University of Massachusetts Amherst in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

May 2012

Polymer Science and Engineering

 Copyright Ian M. Henderson 2012

TUNING THE PROPERTIES OF METAL-LIGAND COMPLEXES

TO MODIFY PROPERTIES OF SUPRAMOLECULAR MATERIALS

A Dissertation Presented

IAN M. HENDERSON

Approved as to style and content by:

______Ryan C. Hayward, Chair

______E. Bryan Coughlin, Member

______Dhandapani Venkataraman, Member

______David A. Hoagland, Department Head Polymer Science and Engineering

Dedicated it to whoever makes it past this page

ACKNOWLEDGMENTS

Firstly, I would like to extend my appreciation to my advisor Prof. Ryan C.

Hayward for his guidance and assistance throughout my years at UMass. I am thankful for the knowledgeable guidance he has given, as well as the opportunities he has given me to disseminate my research at conferences and in scientific journals. I would also like to thank the members of my committee Prof. E. Bryan Coughlin and Prof. D.

Venkataraman for their guidance throughout my Ph.D. career. Without their input, it is doubtful that I would have become the scientist that I am today.

I would like to thank my friends from UMass for their friendship and support. I would especially like to thank, in no particular order, Scott and Jess, Sami, Marcos,

Sandy, Erik, Joe, Emily, and C.J. While the list could go on, these individuals have made my time in Massachusetts more enjoyable. I would also like to thank the staff members at PSE, Jack, Vivien, Lisa, and Steve for their help in my graduate research.

I would also like to thank my family members, my mother, father, and my sister, for their love and support. Without you, none of this would have been possible.

ABSTRACT

TUNING THE PROPERTIES OF METAL-LIGAND COMPLEXES TO MODIFY PROPERTIES OF SUPRAMOLECULAR MATERIALS

MAY 2012

IAN M. HENDERSON

B.SC., CASE WESTERN RESERVE UNIVERSITY

M.SC., UNIVERSITY OF MASSACHUSETTS AMHERST

PH.D., UNIVERSITY OF MASSACHUSETTS AMHERST

Directed by: Professor Ryan C. Hayward

Supramolecular chemistry is the study of discreet molecules assembled into more complex structures though non-covalent interactions such as host-guest effects, pi-pi stacking, electrostatic effects, hydrogen bonding, and metal-ligand interactions. Using these interactions, complex hierarchical assembles can be created from relatively simple precursors.

Of the supramolecular interactions listed above, metal-ligand interactions are of particular interest due to the wide possible properties which they present. Factors such as the denticity, polarizability, steric hindrance, ligand structure, and the metal used (among others) contribute to a dramatic range in the physical properties of the metal-ligand complexes. Particularly affected by these factors are the kinetic and thermodynamic properties of the complexes. As a result metal-ligand interactions can vary from inert to extremely transient.

Of the vast number of ligands available for study, this dissertation will center on substituted terpyridine ligands, with a particular focus on terpyridine-functionalized polymers. While polymer-functionalized terpyridine ligands and their complexes with transition metals have been heavily studied, the physical properties, particularly the effects of polymer functionalization on the stability of bis complexes of terpyridines, remain unexplored.

In the course of investigating the kinetic stability of these complexes, polymer functionalization techniques were developed which were found to increase the stability of the metal-ligand interactions compared to conventional techniques. In addition to studying the effect of terpyridine substituents, the effects of solvent on the stability of the complexes was studied as well. As polymer-bound terpyridine complexes are often studied in solvents other than water, knowledge of the stability of the complexes in organic solvents is important to create supramolecular structures with more precisely controlled properties. It was found that, for unsubstituted terpyridyl complexes, the stability of the complexes varied by as many as five orders of magnitude in common solvents. It is believed that this decrease in stability is the result of the ability of the solvent to facilitate the movement of the ligands from the first and second coordination spheres.

Although a large part of this dissertation is dedicated to the study of the kinetic stability of terpyridine complexes, synthetic techniques involving terpyridine and its complexes were investigated as well. It was found that terpyridine functionalized polystyrene could be produced by direction functionalization of terpyridine with polystyryllithium. Additionally heterloleptic terpyridine-based iron complexes were

vii produced with high purity by reduction of the mono terpyridine complex of iron(III) in the presence of a second, functionalized terpyridine ligand. The culmination of these studies was the synthesis of supramolecular organogels, which were crosslinked using metal-terpyridine complexes, yielding dynamic mechanical properties could be broadly tuned by varying the metal used to form the crosslinks.

viii

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ...... v ABSTRACT ...... vi LIST OF TABLES ...... xii LIST OF FIGURES ...... xiv LIST OF SCHEMES...... xvi LIST OF SYMBOLS ...... xvii LIST OF ABBREVIATIONS ...... xviii

CHAPTER

1. INTRODUCTION ...... 1

1.1 Overview ...... 1 1.2 Background ...... 3

1.2.1 A Brief Introduction to Metal-Ligand Chemistry ...... 3 1.2.2 The Chelate Effect ...... 5 1.2.3. Origins of the Terpyridine Ligand ...... 7 1.2.4 Terpyridine Ligands Bearing Functional Groups ...... 10 1.2.5 Terpyridine-Functionalized Polymers ...... 13 1.2.6 Properties of Selected Terpyridine Complexes ...... 16 1.2.7 Terpyridne Kinetics ...... 20

1.3 References ...... 21

2. SYNTHESIS OF END-FUNCTIONALIZED POLYSTYRENE BY DIRECT NUCLEOPHILIC ADDITION OF POLYSTYRYLLITHIUM TO BIPYRIDINE OR TERPYRIDINE ...... 27

2.1 Chapter Overview ...... 27 2.2 Introduction ...... 28 2.3 Experimental ...... 32 2.4 Results and Discussion ...... 38 2.5 Conclusions ...... 48 2.6 Appendix 2A: Side Product Characterization, Detailed Synthetic Procedures, and Equal Reactivity Model ...... 50 2.7 Appendix 2B: Synthesis of Mid-Functionalized Block Copolymers ...... 56 2.8 References ...... 60

3. SUBSTITUENT EFFECTS ON THE STABILITIES OF POLYMERIC AND SMALL MOLECULE BIS-TERPYRIDINE COMPLEXES ...... 62

3.1 Chapter Overview ...... 62 PART 3.1: SUBSITUENT EFFECTS OF ETHER AND AMINO LINKAGES ON KINETIC STABILITY OF BIS TERPYRIDINE COMPLEXES

3.1.1 Introduction ...... 65 3.1.2 Experimental ...... 68 3.1.3 Results and Discussion ...... 77

PART 3.2: EFFECT OF THIOETHER AND SULFONYL LINKAGES ON PEO-BASED MACROMOLECULAR TERPYRIDINE COMPLEXES

3.2.1 Introduction ...... 97 3.2.2 Experimental ...... 100 3.2.3 Results and Discussion ...... 102

3.4 Conclusions ...... 105 3.5 Appendix 3A. Model Compounds in the Investigation of Secondary Peaks ...... 107 3.6Appendix 3B. Cobalt Complex Decay in Mixed Solvent and Nickel Complex Decomposition ...... 109 3.7 References ...... 110

4. KINETIC STABILITY OF BIS COMPLEXES OF TERPYRIDINE WITH IRON(II) AND COBALT(II) IN ORGANIC SOLVENTS ...... 115

4.1 Overview ...... 115 4.2 Introduction ...... 115 4.3 Results and Discussion ...... 117 4.4 Conclusions ...... 123 4.5 Appendix 4A. Preparation of Complexes of Terpyridine with Iron and Cobalt...... 124 4.6 References ...... 125

5. HETEROLEPTIC IRON(II) COMPLEXES OF 4’-SUBSTITUTED TERPYRIDINES: SYNTHESIS, KINETIC STABILITY, AND REDOX PROPERTIES ...... 128

5.1 Chapter Overview ...... 128 5.2 Introduction ...... 128 5.3 Experimental ...... 130 5.4 Results and Discussion ...... 135 5.5 Conclusions ...... 142 5.6 References ...... 143

6. TUNABLE ORGANOGELS UTILIZING TERPYRIDINE-BASED CROSSLINKS ...... 145

6.1 Overview ...... 145 6.2 Introduction ...... 145 6.3 Experimental ...... 147 6.4 Results and Discussion ...... 151 x

6.5 Conclusions ...... 158 6.6 References ...... 158

7. SUMMARY AND OUTLOOK ...... 160

7.1 Overview ...... 160 7.2 Synthesis of Novel Terpyridine-Functionalized Polymers ...... 161 7.3 Kinetics Studies of Terpyridine Complexes ...... 164 7.4 Synthesis of Heteroleptic Terpyridine-Based Complexes of Iron(II) ...... 167 7.5 Organogels with Supramolecular Terpyridne-Based Crosslinks ...... 168 7.6 Conclusions ...... 168 7.7 References ...... 169

BIBLIOGRAPHY ...... 171

LIST OF TABLES

Table Page

1.1. Formation Constants ( β) for mono-, bi-, and tridentate pyridyl ligands ...... 5

1.2. Thermodynamic parameters for the complexation of Cd 2+ with Ethylenediamine (a chelating ligand), and emthyleamine (a monodentate ligand)...... 6

1.3 Kinetic Parameters of Terpyridine Complexation at 25 o C. From Wilkins and coworkers ...... 20

2.1 Properties of Bipyridine and Terpyridine End-Functionalized Polymers...... 39

2.2 End-Functionalities after Purification by Short-Column Chromatography...... 43

2.A.1 Composition of Unpurified Functionalized Sample 4 ...... 52

3.1 Decay Rates of Cobalt bis -Teryridine Complexes ...... 80

3.2 Decay of Iron Bis-Complexes of Terpyridine Ligands ...... 85

3.3 Reduction and oxidation potentials of selected cobalt complexes for the Co(III)/Co(II) redox couple...... 86

3.4 Comparison of the secondary NMR peak intensity with the rate of decay of the cobalt(II) bis complexes of ether-substituted terpyridines...... 90

3.5 Decay Rates of Iron Complexes Based on the Number of Repeat Units ( N) of the PEO Substituent ...... 92

3.6 Decay Rates of Cobalt Complexes Based on the Number of Repeat Units ( N) of the PEO Substituent ...... 94

3.7 Decay Rates of Iron and Cobalt Complexes of Ligands (8) and (9) ...... 103

3.B.1 Decay of Ether-linked Cobalt Complexes in a Water/20% DMSO solution...... 109

3.B.2 Decay Rates of Ether-Linked Nickel Complexes ...... 110

2+ 4.1. Decay of [terpy] 2Fe complexes in various organic solvents...... 117

2+ 4.2. Decomposition of Co[terpy] 2Co in various solvents ...... 122

xii

5.1 Electrochemical Properties of the Terpyridine Complexes Synthesized in Chapter 5...... 140

5.2 Kinetic Stability of the Complexes Synthesized in this Chapter 5...... 140

6.1. β Values of the Gels by Rheology and Stress Relaxation ...... 154

xiii

LIST OF FIGURES

Figure Page

2.1 Ultraviolet-visible spectrophotometry absorption spectra of polystyryllithium, polystyryllithium with terpyridine, and polystyryllithium with bipyridine ...... 38

2.2 1H nuclear magnetic resonance spectrum of functionalized sample 4 in deuterochloroform showing the bipyridyl shifts in detail...... 39

2.3 Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) spectrum of terpyridine-functionalized, purified sample 5 in reflectron mode using dithranol as a matrix and silver(I) trifluoroacetate as a cationization agent...... 42

2.4 Size exclusion chromatographs of end-functionalized sample 3 before and after purification by short column chromatography reveal a decrease in the proportion of double molecular weight chains...... 44

2.5 Ultraviolet-visible spectrophotometry (absorption at 329 nm) of a 0.17 mM solution of terpyridine-functionalized sample 1 titrated with iron (II) chloride hexahydrate...... 46

2.6 1H nuclear magnetic resonance spectra of the pyridyl regions of terpyridine- functionalized sample 5, in deuterochloroform and a 2:1 mixture of functionalized sample 5 with iron (II) chloride in deuterochloroform...... 48

2.A.1 Size exclusion chromatograph of polymer functionalized with an equimolar amount of terpyridine...... 50

2.A.2 SEC trace of the chromatographically separated byproducts of sample 4...... 52

2.B.1. The GPC Chromatographs of Polystyrene-terpy-poly( p-methylstyrene)...... 58

2.B.2. MALDI-TOF spectrum of Polystyrene-terpy-poly( p-methylstyrene)...... 59

2.B.3. GPC Chromatographs of Polystyrene-bipy-polyisoprene...... 59

3.1 Decay of selected bis -terpyridine cobalt complexes, as determined by monitoring the growth in absorbance of the MLCT band formed upon reaction with excess iron ...... 79

3.2 Decay of selected bis -terpyridine iron complexes, as directly determined by monitoring the decrease in MLCT absorbance in the presence of excess nickel ...... 84

3.3 UV-vis spectra reveal that the freshly mixed bis complex of ligand (7) with Co(II) is oxidized to Co(III) upon bubbling of air through the solution...... 89

xiv

3.4 NMR spectra of bis complexes of ligand (7c) with Co(II) in D 2O show a series of secondary peaks that broaden and disappear with increasing temperature...... 90

3.5 Dependence of Iron Complex Decay Rate on Length of Substituent Polymer ...... 93

3.6 The NMR spectra of the ligands used in the sudy Conducted in Chapter 2 Part2. ... 103

3.7 Decay Scans of the Cobalt Complexes of Ligands (8) and (9)...... 104

1 2+ 3+ 3.A.1. H NMR Spectrum of the (7) 2Co and (7) 2Co ...... 108

4.1. Decay of [terpy]2Fe(PF6)2 in selected solvents...... 118

4.2 Solubility of Terpyridine vs. Decay Rate for the Iron Complexes ...... 120

5.1. NMR spectra of the Iron(II) Complexes in CD 3CN...... 137

5.2. UV-Vis Spectrum of Terpyridine Complexs Investigated in Chapter 5 ...... 138

5.3. Kinetic Stability of Terpyridine complexes...... 141

6.1 Storage Moduli (G’) of Iron and Nickel Gels...... 153

6.2. Storage and Loss Moduli of the Copper Gel...... 155

6.3 The Storage and Loss Moduli of Cobalt and Mixed Gels...... 156

LIST OF SCHEMES

Scheme Page

1.1. Numbering Convention Used in Terpyridine Ligands ...... 8

1.2. Synthesis of 2,2':6’,2”-terpyridine from Potts and coworkers...... 9

1.3. Synthesis of 4'-chloro-2,2';6',2"-terpyridine, and 6 ′,6 ″-bis(2-pyridyl)-2,2 ′ : 4,4 ″ : 2″,2 ″′ -quaterpyridine...... 10

1.4. The sythensis of 4’-aryl terpyridine ligands via Kröhnke Condensation ...... 13

1.5. Synthesis of Side Chain Terpyridine-Functionalized Polymers by NMP ...... 15

2.1 End-capping of Polystyryllithium with Terpyridine...... 30

2.2 Apparatus Used for Anionic Polymerization ...... 32

3.1 Ligands used in Decomposition Studies ...... 67

3.2 Reaction of terpyridine ligands with a bivalent metal salt to form a mono and subsequently a bis complex...... 78

3.3 The Macromolecular Ligands used in the study conducted in Chapter 2 Part2...... 99

5.1. General Schematic for the Synthesis of Asymmertic bis -terpyridyl Iron Complexes...... 135

6.1 The Synthesis of Supramolecular Gels Used in this Study...... 151

7.1: Functionalization of Polystyryllithium with Terpyrdine ...... 161

7.2 Decay of Terpyridine Complexes ...... 164

xvi

LIST OF SYMBOLS

β: Bulk relaxation rate

βx: Product of the equilibrium constants of a reaction (x is number of multiplied constants)

ð: Dispersity of a polymer sample ( Mw/Mn)

E1/2 : Difference in half cell potential between to species

G: Standard Gibbs free energy change

H: Standard enthalpy change

H‡: Change in entropy of solvation

S: Standard entropy change

E1/2 : Half-cell poetential

Ea Activation energy of a reaction

G’: Storage modulus

G”: Loss modulus

K: Equilibrium constant kx: Rate constant (x denotes step of reaction)

M: Molarity

Mn: Number average molecular weight

Mw: Weight average molecular weight

N: Number of repeat units of a polymer chain

R: Gas constant

T: Temperature

V: Volts v: volume

xvii

LIST OF ABBREVIATIONS

ACN: Acetonitrile

ATRP: Atom transfer radical polymerization bipy, bipyridine: 2,2’-bipyridine

Cl-terpy,chloro-terpyridine: 4’-chloro-2,2’:6’,2”-terpyridine

CV: Cyclic voltammetry

DGME-terpy: 4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''-terpyridine

DMF: N,N -dimethylformamide

DMSO: Dimethylsulfoxide

EDTA: Ethylene diamine tetraacetic acid

EI-MS: Electron ionization mass spectroscopy

ESI-MS: Electrospray ionization mass spectroscopy

EtOH: Ethanol

FAB-MS: Fast atom bombardment mass spectroscopy

GPC: Gel permeation chromatography

HPLC: High-performance ligiud chromatography

MALDI-TOF: Matrix-assisted laser desorption time-of-flight mass spectroscopy m-CPBA: 3-Chloroperoxybenzoic acid

MeOH: Methanol mesyl: Methanesulfonyl group

MLCT: Metal ligand charge transfer

xviii

NMP: N-methylpyrrolidone, nitroxide-mediated polymerization

NMR: Nuclear magnetic resonance spectroscopy npa-terpy: N-propylamino-2,2’:6’,2”-terpyridine

PEO: Poly(ethylene oxide)

Poly(NIPAAm): Poly( N-isopropylacylamide)

PS: Poly(styrene)

RAFT: Radical atom fragmentation transfer polymerization

RBF: Round-bottom flask

SEC: Size-exclusion chromatography

THF: tetrahydrofuran

TLC: Thin-layer chromatography

Terpy, terpyridine: 2,2':6',2''-terpyridine terpy-OH: 2,6-Bis(2-pyridyl)-4(1H)-pyridone terpy-SH: Terpyridine-4′(1 ′H)-thione

UV-Vis: Ultraviolet/visible spectrophotometry

xix

CHAPTER 1

INTRODUCTION

1.1 Overview

The field of supramolecular chemistry has relates to “chemistry beyond the molecule” as termed by Lehn. 1,2 In particular, it pertains to intermolecular bonds forming associations between discrete molecules whose bonding is non-covalent in, encompassing

hydrogen bonding, π-bonding, hydrophobic effects, host-guest interactions, ionic

interactions, and metal-ligand interactions.

Of these associations, metal-ligand chemistry offers arguably the most varied

array of properties. As compared to covalent bonding, metal ligand bonds can range

from inert to transient in nature. 3 Bond energies of the complexes likewise show a wide

variation in bond strength, from very weak to nearly that of covalent bonds.

Additionally, interesting photophysical and magnetic properties can arise from metal-

ligand bonding. Most importantly, all of these properties can be easily tuned through

variation of the metal and the choice of one of the vast array of ligands available for

complexation.

One particularly interesting subset of metal-ligand chemistry is the study of

tridentate ligands, such as 2,2’:6’,2”-terpyridine with transition metals. Complexes of the

terpyridine ligand with various transition metals have been the focus of intense research

since the ligand was first synthesized in the 1930s. The substituted terpyridines and their

homologues have particularly enjoyed attention from researchers interested in polymer-

based supramolecular structures due to the synthetic ease in which terpyridine

1 functionalities can be incorporated into polymer architectures, as well as the myriad of interesting properties that can be obtained upon the complexation of these ligand with metal ions. While much attention is paid to the structures that can be synthesized with terpyridine, as well as the electronic absorption and emission of the complexes and the electrochemical properties, little is known about how the kinetic stability of terpyridine complexes containing different polymer and small-molecule substituent groups and in different solvents. This information would be invaluable in the synthesis of polymer- based supramolecular structures held together by solvent- and substitutent group-tuned terpyridyl complexes.

The intent of this dissertation is to further expand the knowledge of metal- terpyridine complexes through study of the factors affecting the physical properties of the complexes, as well as to develop novel chemical structures based on the terpyridine ligand, with a particular emphasis on terpyridine-functionalized polymers. This dissertation will serve to outline the effects of polymer substituents and solvents on terpyridine complexes, providing useful and heretofore unknown information on terpyridine kinetics. Additionally, a novel synthetic technique for the synthesis of heteroleptic terpyridine complexes is outlined, as well as the synthesis of a supramolecular gel with terpyridine-based crosslinks which imparted the gel with tunable physical properties.

This dissertation is divided into four sections, each representing specific research initiatives. Chapter 2 involves the functionalization of living polystyryllithium (a living anionic polymerization) with terpyridine and bipyridine, by direct addition of the living

2 anionic polymer to the. The resulting Chichibabin-type reaction results in novel 6- polysrtyryl-2,2’:6’,2”-terpyridines and 6-polystyryl-2,2’-bipyridines.

Chapters 3-4 concern the measurement of the kinetic stabilities of terpyridine complexes of iron(II) and cobalt(II) with various substituent groups and in various solvent environments. Chapter 3 describes the study of the kinetic stability of macromolecular terpyridine complexes utilizing various chemistries to link the ligand to the polymer chain. Kinetic stabilities of both iron(II) and cobalt(II) complexes were tested, as well as the electrochemical properties of complexes of terpyridine-based model compounds with cobalt(II). Chapter 4 records a study of the effect of different organic solvents on the kinetic stabilities of terpyridine complexes of iron(II) and cobalt(II).

Chapter 5 covers the introduction of a new technique for the synthesis of mixed ligand terpyridine complexes of iron(II) by which very pure heteroleptic complexes can be obtained with reasonable yield. This technique is then used to synthesize asymmetric complexes substituted with either electron withdrawing or electron donating groups. The kinetic stability and redox properties of these asymmetrically substituted groups are then measured. The last section of the body of this dissertation is Chapter 6, which covers the synthesis and rheological characterization of supramolecular gels with crosslinks consisting of terpyridine complexes.

1.2 Background

1.2.1 A Brief Introduction to Metal-Ligand Chemistry

Supramolecular chemistry is a field of study concerning the organization of individual molecules into large-scale structures by interactions other than covalent

3 bonds. 4 These supramolecular structures may be formed through a variety of interactions which include hydrogen bonding, electrostatic interactions, π-π interactions, hydrophobic interactions and other host-guest effects, and metal-ligand interactions. Of these interactions, metal-ligand complexes are among the most well-studied, having first been brought to the attention of the broader scientific community with Pedersens’s work on the ability of cyclic polyethers to complex metal salts in the late 1960’s. 5,6

In the simplest terms, metal-ligand coordination revolves around the donation of electon density by the electron-rich ligand to the electron-poor metal center. The electron donation of the ligands may arise through a number of modes, such as a negative charge on the ligand, electron density in the π-electron cloud of sp and sp 2-hybridized bonds, and

the electrons of lone pairs on an organic molecule. Of these metal-ligand interactions, the

electron density donated by the lone pairs of the organic molecule will be the focus of

this writing.

The interaction of electron lone-pairs with metal d-orbitals is observed in σ- donating ligands such as amines and phosphines. One particular example is

7 Co[NH 3]6Cl 3, a compound whose structure was first elucidated by Werner, laying the foundation of the study of coordination chemistry. Upon association of the ligand with the metal, hybridized metal orbitals of the type d xsp y form to accept electron density from the ligand, with the integers x and y depending on the extent of d-ordital filling, ligand properties, and other factors. 8 Often the partial hybridization of the d-subshell of the metal gives rise to a splitting of the orbital energies, resulting in higher energy antibonding hybridized orbitals, and lower energy nonbonding orbitals. This can overcome the pairing energy of the electrons in the orbital, resulting in paired electrons

4 occupying lower energy orbitals. 9 The spectral and magnetic properties of these

compounds could fill several volumes, but the specific implications of ligand field-based

energy splitting will only be dealt with in this work so far as they affect the properties

which are discussed herein.

1.2.2 The Chelate Effect

An important concept in coordination chemistry is the binding of multidentate

ligands to metal centers. These ligands were given the name “chelates” after the greek

name of a lobster’s claw (chela) by Morgan and Drew in 1920, 10 in order to describe

acetylacetonate complexes of tellurium and selenium. The importance of chelate

complexes is due to the increased stability over monodentate ligands. 11 This is demonstrated in Table 1.1, which shows the β (product of the equilibrium constants of all steps involved in complex formation) values for mono-, di-, and tridentate pyridyl ligands. Chelating ligands form more stable complexes for two main reasons.

Table 1.1. Formation Constants ( β) for mono-, bi-, and tridentate pyridyl ligands. 12,13 log β log β log β Cation 6 3 2 (pyridine) (bipyridine) (terpyridine) Fe 2+ - 17.5 20.9 Co 2+ - 16.1 18.3 Ni 2+ 9.8 19.3 21.8 Cu 2+ - 10.2 17.9

The dominate effect can be described from both thermodynamic and kinetic

properties. Kinetically, for a multidentate ligand, if one binding moiety becomes

dissociated from the metal center, the ligand is still bonded to the metal through the

other binding moieties. This means that the dissociated moeity is more likely to re-

5 associate due to its close proximity to the metal center compaired to an equivalent

monodentate ligand. 14

This can also be described in terms of a distinct entropic effect that arises from chelates. During the ligation reaction, coordinated solvent must dissociate from the metal center before complexation can occur. The association of a multidentate ligand to a metal center is more entropically favorable than associations of monodentate solvents due to higher number of total molecules in the system. A ligand with only one binding moiety means that the number of molecules in the system stays constant,meaning there is no net gain in entropy.15-18

This entropic phenomenon becomes apparent when the thermodynamic values for both mono-, and bidentate cadmium complexes displayed in Table 1.2 are taken into consideration. While the enthalpy of complexation HӨ for both mono- and

bidentate species is similar, the entropy of complexation SӨ is much more

favorable for multidentate, providing a larger magnitude of GӨ.

A secondary effect is that the π-electron cloud may delocate though the

metal center. 19,20 This is an advantage of conjugated ligands only, and is

considered to be only a minor contribution to complex stability.

Table 1.2. Thermodynamic parameters for the complexation of Cd2+ with Ethylenediamine (a chelating ligand), and emthyleamine (a monodentate ligand). 21 Complex GӨ(kJ/mole) HӨ(kJ/mole) SӨ(J/mole K -1) log β 2+ Cd(MeNH 2)2 -27.28 -29.37 -6.28 4.81 Cd(en) 2+ -33.30 -29.41 13.05 5.84 2+ Cd(MeNH 2)4 -37.36 -57.32 -66.97 6.55 2+ Cd(en) 2 -60.67 -56.48 13.75 10.62

6 While other considerations, such as the alignment of the binding sites of the

multidentate ligand with the metal d-orbitals, or the pH of the solution, may negate or

lessen the effectiveness of the chelate effect, 22 the complexes formed generally have higher rates of formation, slower rates of dissociation, and higher thermodynamic stability. It is this increased stability, along with magnetic and spectral properties that has helped to intensify research into multidentate ligands such as bipyridine, phenanthroline, and terpyridine.

1.2.3. Origins of the Terpyridine Ligand

The 2,2’:6’,2”-terpyridine ligand (terpyridine, terpy) was first synthesized in 1932 by G.T. Morgan (the same researcher who coined the term “chelate”) and F. H. Burstall by the oxidative coupling of pyridine using anhydrous ferric chloride as a catalyst 23 . The intent of the reaction was to synthesize large amounts of 2,2’-bipyridine (bipyridine, bipy), using a scale-up of a previous synthesis. 24 Due to the large scale of the reaction, side-products (such as terpyridine ) were present in quantities great enough to isolate and characterize. This first paper also contained a characterization the bis complex of ferrous bromide and terpyridine. The complex was considered for dyestuff, a role for which it was poorly suited due to insufficient binding to textiles. For future reference, the numbering convention of terpyridine ligands is shown in Scheme 1.1.

7 4' 3' 5'

3 2' 6' 3" 2 2" 4 N 4" 1' 5 N1 1"N 5" 6" 6

Scheme 1.1. Numbering Convention Used in Terpyridine Ligands

In the beginning, terpyridine research was slow to take off, with only sporadic publication throughout the following five decades, which is most likely due to the difficult synthesis; a problem that wasn’t addressed until the 1980s and 1990s 25,26 . One

of the most commonly employed of these improved synthetic techniques is shown in

Scheme 1.2. The earliest papers published in the 1940s after the initial synthesis of

terpyridine dealt with the detection of iron 27 and cobalt 28 , with some physical measurements (such as acid-dissociation constants and redox potentials) of the iron(II)- terpyridine system 29 being published in the 1950s. Three interesting studies on the kinetics of terpyridine with various metals were conducted in the labs of R. G. Wilkins in the 1960s 30-32 , representing the first and still the most important treatment of terpyridine

kinetics.

8 1. t-BuOK, THF SMe N 2. CS2 N 3. MeI O O SMe

S 1. N O SMe N N 2.NH4OAc, HOAc N N O SMe

NaBH4

N N NiCl2 N N N N

Scheme 1.2. Synthesis of 2,2':6’,2”-terpyridine from Potts and coworkers. 26 .

Another important stepping stone in terpyridine research was the development of

asymmetrical terpyridine complexes in the 1980s 33-36 . It was found that ruthenium(III) and osmium(III) salts lacked the ability to form a bis complex with terpyridine, forming instead a mono terpyridine complex. These mono complexes were then exposed to a second ligand in the presence of a reducing agent, forming an heteroleptic ruthenium(II) or osmium(II) salt. This represents the first synthesis of a terpyridine-based heteroleptic complex, as rapid ligand exchange and disproportionation makes heterleptic synthesis difficult with first-row transition metals. 30,31

9 1.2.4 Terpyridine Ligands Bearing Functional Groups

On of the strongest features of the terpyridine ligand is the ability to create functionalized terpyridine ligands through a variety of techniques. Often, substitutent groups are placed on the terpyridine ligand in order to modify some of the properties of the complexes made with these ligands. Perhaps the most utilized ligand in this family of substituted terpyridines is 4’-chloro-2,2’;6’,2”-terpyridine (Cl-terpy). The first reported synthesis of this ligand was published by Constable et. al. in 1990, 37 with the aim of homocoupling Cl-terpy, thus forming 6 ′,6 ″-bis(2-pyridyl)-2,2 ′ : 4,4 ″ : 2 ″,2 ″′ - quaterpyridine. The synthesis of Cl-terpy and the homocoupling product is shown in

Scheme 1.3.

Scheme 1.3. Synthesis of 4'-chloro-2,2';6',2"-terpyridine, and 6 ′,6 ″-bis(2-pyridyl)- 2,2 ′ : 4,4 ″ : 2 ″,2 ″′ -quaterpyridine. 37 Reprinted with Permission from Royal Scoiety of Chemistry. Copyright 1990.

Researchers quickly realized that Cl-terpy could easily be functionalized by an

SNAr nucleophilic attack on the 4’-position on the terpyridine molecules. Using the nucleophilic substitution route, terpyridines were produced bearing various ether-linked functional groups. Alcohols bearing a variety of functional groups, such as alkyl chains of various length, 38-40 amines, 40-45 and carboxylic acids 42,46-49 have been employed to give terpyridines bearing these functional groups. The S NAr substitution to the 4’-position has also been show to work with deprotonated thiols. 50,51

The functionalization of Cl-terpy with deprotonated alcohols and thiols is not the limit of the utility of the ligand. For example, amines have been shown to attack the 4’- position of the ligand as well, resulting in functionalized terpyridines with an amino bridge. 52-55 Using an amine as a nucleophile in the functionalization of Cl-terpy has the disadvantage of high reaction temperatures and the addition of metals salts which complex the terpyridines and withdraw electrons from the 4’-postion , thus making the reaction more favorable. These metal salts later have to be stripped from the ligand with

KOH or another strong base, making the workup of the reaction very intolerant to certain functional groups. In addition to ether, thioether, and amino-functionalized terpyridines, nucleophilic addition has been used to create 4 ′-(diphenylphosphino)-2,2 ′:6 ′,2 ″- terpyridine as well. 56 The ligand was synthesized by addition of deprotonated diphenylphospine to Cl-terpy at -95 o C. This ditopic ligand was complexed with a variety of metal salts, the “softer” metals such as platinum and palladium preferentially bound the phosphine moiety, while “harder” metals such as iron and ruthenium preferentially bound the terpyridyl moiety.

11 In addition to nucleophilic aromatic substitution, Cl-terpy has been functionalized

using cross coupling reactions. Buchwald-Hartwig coupling has been used to create

terpyridyl compounds with amino-functionalizations in the 4’ position. 57 Suzuki coupling has been used to create 4’-aryl functionalized terpyridines as well, though the reported yield was poor. 58 In addition to using Cl-terpy, coupling reactions have been undertaken with terpyridines bearing pseudohalide groups as well. The 4’- psuedohalide terpyridines were functionalize by the addition of the psuedohalidochloride to deprotonated 2,6-Bis(2-pyridyl)-4(1H)-pyridone. Stille coupling has been used to synthesize 4’-vinyl-2,2’:6’,2”-terpyridine from these psuedohalidoterpyridines. 59,60

In addition to the chemistry involving 4’-halo and 4’-psuedohalo functionalized terpyridines, several functionalized terpyridines have been synthesized from 2,6-Bis(2- pyridyl)-4(1H)-pyridone (terpy-OH). An intermediate in the synthesis of Cl-terpy, terpy-

OH itself is a useful molecule in the creation of substituted terpyridine compounds.

Terpy-OH can be deprotonated using a mild base, such as potassium carbonate. The

61-64 61,65 resulting anionic species can then perform an S N2 reaction with alkyl, benzyl, and allyl halides 61 ; as well as psuedohalides. 61 Mitsunobu coupling, a technique by which two alcohols are coupled, has been used to react terpy-OH with a variety of alcohols, offering another technique by which ether-linked terpyridines can be functionalized. 66,67

As with tery-OH, the sulfur analogue, 2,2 ′:6 ′,2 ″-Terpyridine-4′(1 ′H)-thione, has been

68,69 used to create thioether-linked compounds via S N2 substitution.

Another commonly utilized synthesis of functionalized terpyridines is the production of 4’-aryl terpyridines via Kröhnke condensation. This synthesis is carried out by reacting two equivalents of the enolate of 2-acetylpyridine with the appropriate

12 arly aldehyde, thus forming the aryl-dipyridyl dione. The central pyridine ring is then closed via condensation of the enone groups with an ammonium species such as ammonium acetate or ammonium hydroxide. The reaction is show in Scheme 1.4.

R R

N O O KOH/NH OH 2 4 EtOh N N N

Scheme 1.4. The sythensis of 4’-aryl terpyridine ligands via Kröhnke Condensation

1.2.5 Terpyridine-Functionalized Polymers

The earliest studies on terpyridine-functionalized polymers involved the synthesis and subsequent polymerization of terpyridine-functionalized vinyl monomers 70-72 . In the studies by Potts, 59,72 a terpyridine ligands with a vinyl group attached to the 4 and 4’ position was synthesized and homopolymerized or copolymerized with styrene. The polymers synthesized in this study were then complexed with cobalt(II) or zinc(II) resulting in precipitation of the polymer. Polymers with pendant metal complexes were also synthesized by polymerizing mixed ligand cobalt(II) complexes of terpyridine/4’- vinyl-2,2’:6’,2”-terpyridine, and crosslinked networks were synthesized using bis-4’- vinyl-2,2’:6’,2”-terpyridine complexes of ruthenium(II).

Another early study conducted by the Hanabusa research group involved copolymerization of 4’-[4-(2-acrylaoxyethoxy)phenyl]-2,2’:6’,2”-terpyridine with styrene and methyl methacrylate. 73 The the polymers were then crosslinked via the addition of

13 iron(II). Another study by the Hanabusa research group involved the copolymerization

of 4’-(p-vinylphenyl)-2,2’:6’,2”-terpyridine with styrene. 74 Pendant iron(II) complexes

were then added to the polymer by first complexing the polymer-bound terpyridine

ligands with iron(III) followed by the reduction of the iron(III) complex with NH 4PF 6 in

the presence of 4’-(p-methylphenyl)-2,2’:6’,2”-terpyridine.

Aside from the studies conducted by Potts and Hanabusa, there have been a

myriad of other studies involving side-chain terpyridine-functionalized polymers. A few

studies have used the approach of adding terpyridine moeities polymers with preexisting

side chain functional groups. One such example by Schubert et al. employed the chlorine

groups of poly(vinyl chloride), displacing the chloro-groups in an S N2 reaction with (2-

mercaptophenyl)-methanol, followed by functionalization with an isocyanyl-

functionalized terpyridine. 75 The terpyridyl functionalized PVC was then mixed with a second terpyridyl ligand complexed with Ru(III), and the metal reduced to Ru(II), to obatian PVC decorated with terpyridyl Ru(II) complexes. Another study by Evrensel et al. involved the direct reaction of Cl-terpy with poly(vinyl alcohol) in the presence of

KOH. 76 , This functionalized polymer was then complexed with zinc acetate and combined with iron particles to create supramolecular magnetorheological gels.

Several side-chain terpyridine-functionalized polymers have been synthesized by

Tew and coworkers, with the aim of synthesizing terpyridine-functionalized polymers with controlled architectures. One such study involved the incorporation of a terpyridine- functionalized styrene molecule into block copolymers via RAFT polymerization with poly(methyl methacrylate) end blocks and a random methyl methacrylate/terpyridyl center block (Scheme 1.5).41 An additional study involved the synthesis of polymers

14 using a macroinitiator consisting of either poly(methyl methacrylate) or poly(styrene) synthesized using ATRP techniques. The bromine-functinalized macroinitiator was then chain extended with N-acryloxysuccinimide, followed by postfunctionalization with 5-

([2,2';6',2'']-Terpyridine-4'-yloxy)-pentylamine in the presence of triethylamine. The polymers were then complexed with various lanthanide metals, and the emission spectra of the complexes were measured. These studies were used as a springboard for further investigations on the emission spectrum of compolymers of styrene and vinyl acetate bearing terpyridine-iridium(III) complexes, 77 white-light-emitting polyermic terpyridyl complexes of lanthanide ions, 78 polymer-based solvent sensors, 79 and polymer-based terpyridyl complexes of lanthanides which could be used to detect nerve agents such as

Sarin and Soman. 80

Scheme 1.5. Synthesis of Side Chain Terpyridine-Functionalized Polymers by NMP. Reprinted from Aamer, K. A.; Tew, G. N. Macromolecules 2004, 37, 1990-1993. 41 Copyright (2004)

More recently, research on end-functionalized polymers has become prevalent,

taking advantage of the addition of hydroxyl-functionalized polymers to Cl-terpy in order

to end-functionalize such polymers as poly(ethylene oxide), polystyrene, poly(ethylene

co-butylene) 81 , polyesters 82 , and poly(vinyl pyridines) 83 with terpyridine moieties.

These end-functionalized polymers have been used to form block copolymers utilizing the aforementioned ruthenium (III/II) redox chemistry 81 . Similar techniques have been used to create block copolymer surfactants 83,84 , and rod-coil-rod assemblies 85 . Other

transition metals such as Ni, Fe, and Co have been used to create supramolecular

condensation polymers using telechelic end-functionalized polymers as macromonomers

along with end functionalized polymers as macromolecular end-cappers 85

1.2.6 Properties of Selected Terpyridine Complexes

In this section the properties of selected metal-terpyridine complexes relevant to this dissertation are discussed, in order to give a basic background for the later chapters.

Cobalt

The ability of terpyridine to complex cobalt was first indicated by the interference of cobalt salts with the colorimetric determination of iron(II) solutions with terpyridine. 86

Soon after, it was determined that terpyridine could be used for the colorimetric

determination of cobalt(II). 87 As cobalt(II) has a d 7 configuration, the bis-terpyridine complex of cobalt(II) is a distinctly Jahn-Teller distorted octahedron in the solid state.88

2 4 The complex is capable of two spin-states, E low-spin manifold and the T1 high- spin manifold, both of which are paramagnetic, though the high spin has a greater magnetic moment. The spin state of the cobalt(II) bis-complex is highly dependent on

16 the counterions and the solvent state. For example, anhydrous [Co(Terpy) 2](ClO 4)2 is has

4 a low spin ground state, with a population of the high spin-state ( T1 manifold) being around 90% at room temperate in the solid state. When the ClO 4 counterions are exchanged with PF 6 counterions, the high-spin state is predominant at lower temperatures, with the low-spin state becoming slightly populated at higher

89 temperatures. Solid [Co(Terpy) 2] complexes crystallized with waters of hydration behaves differently than its anhydrous analogues, even bearing the same counterions. For example, the complex [Co(Terpy) 2](ClO 4)2●0.5H 2O is predominantly high-spin even at low temperatures, unlike its anhydrous counterpart mentioned above, which is predominantly low spin at low temperatures.90 Complexes with simple halide counterions were found to be more independent of hydration, and show a low spin ground state, and were only approximately 30% high-spin at room temperature. In aqueous solution, there is an equilibrium of the two spin states, 91 while in organic solvents, the complex is generally low-spin. 92 An interesting consequence of the differences in the magnetic moments is that the different spin states can have different

NMR spectra.93 The normal low-spin cobalt(II) complexes only reach a maximum downfield shift of approximatly 120 ppm. 95 Adding steric bulk is one way ensure the existence of high-spin cobalt in solution, and can lead to downfield shifts of up to 250 ppm. 92

Along with cobalt(II) complexes, cobalt(III) complexes have been studied as well.

The bis terpyridne complex exists as a low-spin d 6 complex, which is diamagnetic and can be characterized by NMR with “normal” sweep widths (relaxation frequencies commonly employed in the observation of organic molecules), 94 as opposed to cobalt(II)

17 complexes, which are paramagnetic and display Knight-shifted peaks in the spectrum.

The general method for producing the trivalent complex is by oxidation of the divalent complex with an oxidant such as silver nitrate.

Iron

The bis terpyridine complex of iron(II) was the first reported complex of terpyridine, recorded in the Morgan and Burstall paper in which terpyridine was first synthesized. 23 The complex exists as a low-spin d 6 species regardless of the counterion, 96 and the NMR spectrum of the complex is readily obtainable in the “normal” ppm range. 94

Interestingly, the addition of steric bulk to the 6 and/or 6” positions of terpyridine can cause the high-spin state to become the ground state configuration due to weakening of the ligand field strength through distortion of the metal-ligand bonds. 97 An acetonitrile solution of these high-spin complexes could be driven towards the low-spin state under high-pressure, as the low-spin state has a smaller volume that that of the high-spin state.

The kinetics of formation and dissociation of iron(II) terpyridine complexes were measured by Wilkins and coworkers, 12,98 who found the formation rate of the mono complex to be approximately 5.4*10 4 M-1s-1 at 25 oC, similar to that recorded for cobalt(II). This is most likely due to the exchange rate of water being the limiting step in the reaction. 99

Studies have also been conducted on terpyridine complexes of iron(III). When combined with FeCl 3, terpyridine does not form a bis complex, forming instead the mono

100 6 terpyridine iron(III) complex. The complex is a high-spin iron(III) complex with a A1 ground state, and d-d transitions are forbidden, therefore the magnetic moment is temperature independent. 101 In addition to the mono complex, the iron(III) bis-

18 terpyridine complex can be synthesized by oxidizing the iron(II) bis complex with PbO 2 in aqueous sulfuric acid, followed by the addition of sodium perchlorate. 101 Interestingly,

a bulky counterion is required to synthesize the bis complex, as the addition of sodium

chloride precipitate [(terpy)FeCl 3]. The bis complex is a low-spin iron(III) species, with

no apparent Jahn-Teller distortion. 102

Nickel

Aqueous formation rates for the mono and bis nickel complex are 1.3*10 3 M-1s-1

and 2.0*10 5 M-1s-1 at 25 o C, respectively. The formation rates in DMSO have also been

measured at 33 M-1s-1 for the mono comeplex, and 338 M-1s-1 for the bis complex. 103 The

dramatically slower formation rates are attributed to the greater coordinating ability of

DMSO to that of water. In order to determine if this was the case, water exxhange of

the mono complexes was measured by 17 O NMR line broadening. 104 The mono complex

of nickel(II) with terpyridine is ideal for this study due to its stability. It was found that

the water exchange was very similar to that of aquo-nickel(II), suggesting that an S N2

(associative substitution) mechanism, or some form of concerted mechanism may be

responsible for the rate increase. The much faster formation rates for the bis complexes

of all the metals discussed in this section when compared to the formation rates of the

mono complexes had been attributed to labilization of solvent by the initially

coordinating terpyridine, which was disproven, at least in the case of nickel(II), by this

study.

Copper

The bis complex of terpyridine with copper(II) is noteworthy for its rapid

dissociation rate, being too rapid to measure using the techniques employed with other

19 first-row transition metals. 98 As expected for a d 9 metal complex, the copper(II) bis

complex of terpyridine shows a marked Jahn-Teller distortion. 105 Interestingly, the

copper complexes are more thermodynamically stable than those of any of the other first-

row transition metals. 31,106 Another particularly interesting aspect of copper-terpyridine chemistry is the ability to form mixed ligand complexes with bidentate ligands such as bipyridine. When the complexes bis -bipyridine and bis-terpyridine complexes are mixed at a 1:1 ratio in solution, -12.9 kJ/mol of free energy is relesed, and the mixed ligand complex is formed. 107 While the mixed-ligand complexes appear to be a 6-coordinate

species in the solid state, with 5 nitrogens and one counterion making up the coordination

sphere, the coordination number in the solution state remains unknown.

1.2.7 Terpyridne Kinetics

The kinetics of formation and dissociation of terpyridine complexes with various

transition metals have been investigated extensively by Wilkins et. al. 12,98,108 Although

discussion of some kinetic aspects of terpyridine complexes is presented above, the data

from Wilkins et al shown in Table 1.3 presents a more comprehensive treatement of

terpyridine kinetics.

Table 1.3 Kinetic Parameters of Terpyridine Complexation at 25 o C. From Wilkins and coworkers. 12

-1 -1 -1 -1 -1 -1 Ion log k 1 (M s ) log k -1 (s ) log k 2 (M s ) log k -2 (s )

Fe 4.8 -2.2 ~7.0 a -6.8 Co 4.4 -4 ~6.7 a -3.2 Ni 3.1 -7.6 5.3 -5.8 Cu >7.3 a a Recorded at 5 oC

20 In this dissertation, the metals discussed above (Fe, Co, Ni, and Cu) are of the most interest. It can be determined from the table that copper reacts most rapidly

7 -1 -1 with terpyridine, kf ≈ 20*10 M s . The ligand dissociation is also very rapid as

-1 109 compared to the other metals, at kd ≥10 s . For the iron complexes reaction rate is between cobalt and nickel for the mono complex, yet the bis complex is the most rapidly forming after copper. Another important factor regarding iron complexes is the high

-7 -1 kinetic stability as compared to the other bis complexes at k-2 = 1.6*10 s . The stability

-8 -1 of the nickel mono complex is also of note, at k-1 = 2.5*10 s ; though the stability of

-6 -1 the bis complex is an order of magnitude less than iron at k-2 = 1.6*10 s . The high stabiliy of the nickel and iron complexes makes them ideal for forming long-lived supramolecular structures. The stability of the cobalt bis complex is less than that of

-3 -1 iron and nickel, at k-2 = 1.1*10 s . An interesting note on the mono complexes characterized had to be formed by heating the bis complex in vacuuo . This is due to the low stability of the mono complexes in solution, as the complexes disproportionate rapidly, forming the bis complex and free metal ion in solution.

1.3 References

(1) Lehn, J. M. Abstracts of Papers of the American Chemical Society 1981 , 182 , 5. (2) Lehn, J.-M. Supramolecular Chemistry: Concepts and Perspectives ; Wiley-VCH: Weinheim, 1995. (3) Wilkins, R. G. The Study of Kinetics and Mechanism of Reactions of Transition Metal Complexes ; Allyn and Bacon: Boston, 1974. (4) Lehn, J.-M. Science 1993 , 260 , 1762-1762. (5) Pedersen, C. J. Journal of the American Chemical Society 1967 , 89 , 7017- 7036. (6) Pedersen, C. J. Journal of the American Chemical Society 1967 , 89 , 2495- 2496. (7) Werner, A. Z. Anorg. Chem. 1893 , 3, 267.

21 (8) Bailar, J. C. Chemistry of Coordination Compounds ; Reinhold Publishing Company: New York, 1956. (9) Dunn, T. M.; McLure, D. S.; Pearson, R., G. Some Aspects of Crystal Field Theory ; Harper and Row: New York, 1965. (10) Morgan, G. T.; Drew, H. D. K. Journal of the Chemical Society, Transactions 1920 , 117 , 1456-1465. (11) Melson, G. A.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1963 , 2662-2672. (12) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-625. (13) Sillen, S. L.; Martell, A. E.; Bjerrum, J. Stability Constants of Metal-Ion Complexes ; 2 ed.; Chemical Society: London, 1964; Vol. 17. (14) Cotton, F. A.; Wilkinson, G. Advanced Inoeganic Chemistry ; 1 ed.; Wiley Interscience: New York, 1962. (15) Atkinson, G.; Bauman, J. E. Inorganic Chemistry 1962 , 1, 900-904. (16) Atkinson, G.; Bauman, J. E. Inorganic Chemistry 1963 , 2, 64-67. (17) Munro, D. Chemstry in Britain 1977 , 100-105. (18) Spike, C. G.; Parry, R. W. Journal of the American Chemical Society 1953 , 75 , 3770-3772. (19) Hancock, R. D.; McDougall, G. J. Journal of the Chemical Society, Dalton Transactions 1977 , 67-70. (20) Hitoshi, M. Coordination Chemistry Reviews 2001 , 219-221 , 957-992. (21) Spike, C. G.; Parry, R. W. Journal of the American Chemical Society 1953 , 75 , 2726-2729. (22) Frausto da Silva, J. J. R. Journal of Chemical Education 1983 , 60 , 390. (23) Morgan, G.; Burstall, F. Journal of the Chemical Society 1932 , 20-30. (24) Hein, O.; .Eschbaumer, C.; Schubert, U. S. Macrommolecular Rapid Communications 2000 , 21 , 1156-1161. (25) Jameson, D. L.; Guise, L. E. Tetrahedron Letters 1991 , 32 , 1999-2002. (26) Potts, K. T.; Ralli, P.; Theodoridis, G.; Winslow., P.; Chenard, B. L.; Smart., B. E. Organic Synthesis 1986 , 64 , 189. (27) Moss, M. L.; Mellon, G. Industrial and Engineering Chemistry. Analytical Edition 1942 , 14 , 863. (28) Moss, M. L.; Mellon, G. Industrial and Engineering Chemistry. Analytical Edition 1943 , 15 , 74-75. (29) Brandt, W. W.; Wright, J. P. Journal of the American Chemical Society 1954 , 76 , 3082-3083. (30) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society 1962 , 341-351. (31) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wikins, R. G. Inorganic Chemisrty 1966 , 5, 622-626. (32) Farina, R.; Hogg, R.; Wilkins, R. G. Inorganic Chemistry 1968 , 7. (33) Collin, J.-P.; Guillerez, S.; Sauvage, J.-P. Journal of the Chemical Society- Chemical Communications 1989 , 776-778. (34) Sullivan, B. P.; Calvert, J. M.; Meyer, T. J. Inorganic Chemistry 1980 , 19 , 1404-1407.

22 (35) Allen, G. H.; Sullivan, B. P.; Meyer, T. J. Journal of the Chemical Society-Chemical Communications 1981 , 793-795. (36) Collin, J.-P.; Guillerez, S.; Sauvage, J.-P. Inorganic Chemistry 1990 , 29 , 5009-510. (37) Constable, E. C.; Ward, M. D. Journal of the Chemical Society, Dalton Transactions 1990 , 1405-1409. (38) R. Newkome, G.; He, E.; A. Godinez, L.; R. Baker, G. Chemical Communications 1999 , 27-28. (39) Muro, M. L.; Daws, C. A.; Castellano, F. N. Chemical Communications 2008 , 6134-6136. (40) Andres, Philip R.; Lunkwitz, R.; Pabst, Gunther R.; Böhn, K.; Wouters, D.; Schmatloch, S.; Schubert, Ulrich S. European Journal of Organic Chemistry 2003 , 2003 , 3769-3776. (41) Aamer, K. A.; Tew, G. N. Macromolecules 2004 , 37 , 1990-1993. (42) Schubert, U. S.; Eschbaumer, C.; Hien, O.; Andres, P. R. Tetrahedron Letters 2001 , 42 , 4705-4707. (43) Burazerovic, S.; Gradinaru, J.; Pierron, J.; Ward, T. R. Angewandte Chemie International Edition 2007 , 46 , 5510-5514. (44) Zhang, Z.; Mikkola, S.; Lonnberg, H. Tetrahedron 2007 , 63 , 11716- 11723. (45) Cho, E. J.; Jung, S.; Lee, K.; Lee, H. J.; Nam, K. C.; Bae, H.-J. Chemical Communications , 46 , 6557-6559. (46) Newkome, G. R.; He, E. Journal of Materials Chemistry 1997 , 7, 1237- 1244. (47) Newkome, G. R.; Güther, R.; Moorefield, C. N.; Cardullo, F.; Echegoyen, L.; Pérez-Cordero, E.; Luftmann, H. Angewandte Chemie 1995 , 107 , 2159-2162. (48) Füssl, A.; Schleifenbaum, A.; Göritz, M.; Riddell, A.; Schultz, C.; KrÃ¤mer, R. Journal of the American Chemical Society 2006 , 128 , 5986-5987. (49) Göritz, M.; Krämer, R. Journal of the American Chemical Society 2005 , 127 , 18016-18017. (50) Mountford, S. J.; Campi, E. M.; Robinson, A. J.; Hearn, M. T. W. Tetrahedron , 67 , 471-485. (51) Andres, P. R.; Schubert, U. S. Synthesis 2004 , 2004 , 1229,1238. (52) Zhou, Z.; Sarova, G. H.; Zhang, S.; Ou, Z.; Tat, F. T.; Kadish, K. M.; Echegoyen, L.; Guldi, D. M.; Schuster, D. I.; Wilson, S. R. Chemistry – A European Journal 2006 , 12 , 4241-4248. (53) Lowe, G.; Droz, A. S.; Vilaivan, T.; Weaver, G. W.; Tweedale, L.; Pratt, J. M.; Rock, P.; Yardley, V.; Croft, S. L. Journal of Medicinal Chemistry 1999 , 42 , 999- 1006. (54) Duncan, T. V.; Ishizuka, T.; Therien, M. J. Journal of the American Chemical Society 2007 , 129 , 9691-9703. (55) Constable, E. C.; Housecroft, C. E.; Tao, Y. Synthesis 2004 , 2004 , 869,874. (56) C. Constable, E.; E. Housecroft, C.; Neuburger, M.; G. Schneider, A.; Zehnder, M. Journal of the Chemical Society, Dalton Transactions 1997 , 2427-2434. (57) Johansson, O. Synthesis 2006 , 2006 , 2585,2589.

23 (58) Miyachi, M.; Ohta, M.; Nakai, M.; Kubota, Y.; Yamanoi, Y.; Yonezawa, T.; Nishihara, H. Chemistry Letters 2008 , 37 , 404-405. (59) Potts, K. T.; Konwar, D. The Journal of Organic Chemistry 1991 , 56 , 4815-4816. (60) Ji, Z.; Li, Y.; Pritchett, T. M.; Makarov, N. S.; Haley, J. E.; Li, Z.; Drobizhev, M.; Rebane, A.; Sun, W. Chemistry – A European Journal , 17 , 2479-2491. (61) Andres, P. R.; Hofmeier, H.; Lohmeijer, B. G. G.; Schubert, U. S. Synthesis-Stuttgart 2003 , 2865-2871. (62) Huang, W.; Han, C. D. Macromolecules 2006 , 39 , 8207-8209. (63) Ito, M.; Tsukatani, T.; Fujihara, H. Journal of Materials Chemistry 2005 , 15 , 960-964. (64) Majouga, A.; Romashkina, R.; Kashaev, A.; Rahimov, R.; Beloglazkina, E.; Zyk, N. Chemistry of Heterocyclic Compounds , 46 , 1076-1083. (65) Huang, Z.; Chen, X.; Zhang, J.; Yu, X.-Q.; Pu, L. Tetrahedron Letters , 52 , 5497-5500. (66) Jari, H. Tetrahedron Letters 2004 , 45 , 5707-5709. (67) Maanpaa, L.; Luzet, V.; Guillaume, G.; Taherpour, S.; Maki, E.; Mikkola, S. New Journal of Chemistry 2009 , 33 , 1853-1858. (68) Constable, E. C.; Hermann, B. A.; Housecroft, C. E.; Neuburger, M.; Schaffner, S.; Scherer, L. J. New Journal of Chemistry 2005 , 29 , 1475-1481. (69) Wang, Q.; Day, P.; Griffiths, J.-P.; Nie, H.; Wallis, J. D. New Journal of Chemistry 2006 , 30 , 1790-1800. (70) Hanabusa, K.; Nakamura, A.; Koyama, T.; Shirai, H. Die Makromoluculare Chemie 1992 , 193 , 1309-1319. (71) Hanabusa, K.; Nakano, K.; Hoyama, T.; Shirai, H.; Hojo, N.; Kurose, A. Die Makromoluculare Chemie 1990 , 191 , 391-396. (72) Potts, K.; Usifer, D. Macromolecules 1988 , 21 , 1985-1991. (73) Hanabusa, K.; Nakano, K.; Koyama, T.; Shirai, H.; Hojo, N.; Kurose, A. Die Makromolekulare Chemie 1990 , 191 , 391-396. (74) Hanabusa, K.; Nakamura, A.; Koyama, T.; Shirai, H. Die Makromolekulare Chemie 1992 , 193 , 1309-1319. (75) Meier, M. A. R.; Schubert, U. S. Journal of Polymer Science Part A: Polymer Chemistry 2003 , 41 , 2964-2973. (76) Hu, B.; Fuchs., A.; Huseyin, S.; Gordaninejad, F.; Evrensel, C. Journal of Applied Polymer Science 2006 , 100 , 2464-2479. (77) Aamer, K. A.; Tew, G. N. Journal of Polymer Science Part A: Polymer Chemistry 2007 , 45 , 1109-1121. (78) Shunmugam, R.; Tew, G. N. Polymers for Advanced Technologies 2007 , 18 , 940-945. (79) Tew, G. N.; Aamer, K. A.; Shunmugam, R. Polymer 2005 , 46 , 8440-8447. (80) Shunmugam, R.; Tew, G. N. Chemistry – A European Journal 2008 , 14 , 5409-5412. (81) Gohy, J. F.; Lohmeijer, B. G. G.; Schubert, U. S. Macromolecules 2002 , 35 , 4560-4563. (82) Winter, A.; Schubert, U. Macromolecular Chemistry and Physics 2007 , 208 , 1956-1964.

24 (83) Gohy, J.-F.; Lohmeijer, B. G. G.; Schubert, U. S. Chemistry: A European Journal 2003 , 9, 3472-3479. (84) Gohy, J.-F.; Lohmeijer, B. G. G.; Alexeev, A.; Eang, X.-S.; Manners, I.; Winnik, M. A.; Schubert, U. Chemistry: A European Journal 2004 , 10 , 4315-4323. (85) Chiper, M.; Meier, M.; Wouters, D.; Hoeppener, S.; Fustin, C.-A.; Gohy, J.-F.; Schubert, U. Macromolecules 2008 , 41 , 2771-2777. (86) Moss, M. L.; Mellon, M. G. Industrial & Engineering Chemistry Analytical Edition 1942 , 14 , 862-865. (87) Moss, M.; Mellon, M. Industrial & Engineering Chemistry Analytical Edition 1943 , 15 , 74-75. (88) Maslen, E. N.; Raston, C. L.; White, A. H. Journal of the Chemical Society, Dalton Transactions 1974 , 1803-1807. (89) Enachescu, C.; Krivokapic, I.; Zerara, M.; Real, J. A.; Amstutz, N.; Hauser, A. Inorganica Chimica Acta 2007 , 360 , 3945-3950. (90) Kremer, S.; Henke, W.; Reinen, D. Inorganic Chemistry 1982 , 21 , 3013- 3022. (91) Judge, J. S.; Baker Jr, W. A. Inorganica Chimica Acta 1967 , 1, 68-72. (92) Chow, H. S.; Constable, E. C.; Housecroft, C. E.; Kulicke, K. J.; Tao, Y. Dalton Transactions 2005 , 236-237. (93) Constable, E. C.; Housecroft, C. E.; Kulke, T.; Lazzarini, C.; Schofield, E. R.; Zimmermann, Y. Journal of the Chemical Society, Dalton Transactions 2001 , 2864- 2871. (94) Elsbernd, H.; Beattie, J. K. Journal of Inorganic & Nuclear Chemistry 1972 , 34 , 771-774. (95) Chow, H. S., University of Basel, 2005. (96) Constable, E. C.; Ward, M. D.; Corr, S. Inorganica Chimica Acta 1988 , 141 , 201-203. (97) Constable, E. C.; Baum, G.; Bill, E.; Dyson, R.; van Eldik, R.; Fenske, D.; Kaderli, S.; Morris, D.; Neubrand, A.; Neuburger, M.; Smith, D. R.; Wieghardt, K.; Zehnder, M.; Zuberbühler, A. D. Chemistry – A European Journal 1999 , 5, 498-508. (98) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1962 , 341-350. (99) Helm, L.; Merbach, A. E. Chemical Reviews 2005 , 105 , 1923-1960. (100) Cotton, S. A.; Franckevicius, V.; Fawcett, J. Polyhedron 2002 , 21 , 2055- 2061. (101) Reiff, W. M.; Baker, W. A.; Erickson, N. E. Journal of the American Chemical Society 1968 , 90 , 4794-4800. (102) Reiff, W. M. Journal of the American Chemical Society 1974 , 96 , 3829- 3834. (103) Cock, P. A.; Cottrell, C. E.; Boyd, R. K. Canadian Journal of Chemistry 1972 , 50 , 402-411. (104) Rablen, D.; Gordon, G. Inorganic Chemistry 1969 , 8, 395-397. (105) Allmann, R.; Henke, W.; Reinen, D. Inorganic Chemistry 1978 , 17 , 378- 382. (106) Calì, R.; Rizzarelli, E.; Sammartano, S.; Siracusa, G. Transition Metal Chemistry 1979 , 4, 328-332.

25 (107) Chaurin, V.; Constable, E. C.; Housecroft, C. E. New Journal of Chemistry 2006 , 30 , 1740-1744. (108) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1965 , 4, 929-935. (109) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1962 .

26 CHAPTER 2

SYNTHESIS OF END-FUNCTIONALIZED POLYSTYRENE BY

DIRECT NUCLEOPHILIC ADDITION OF POLYSTYRYLLITHIUM

TO BIPYRIDINE OR TERPYRIDINE§

2.1 Chapter Overview

We describe a new approach to synthesize 2,2’-bipyridine- or 2,2':6',2"- terpyridine-terminated polystyrene that relies on anionic polymerization and direct end- capping with the desired pyridyl species. End-functionalization occurs by nucleophilic addition of the living polystyryllithium chain to the 6 position of the pyridine ring, followed by termination and oxidative rearomatization. By using an excess of the pyridyl species to avoid coupling of two living chains through addition to the same bipyridine or terpyridine unit, this technique yielded samples consisting of 77-93% singly end- functionalized chains. The functionality of the polymers was determined by nuclear magnetic resonance spectroscopy and chromatographic separation, while molecular weight and polydipersity were determined by size exclusion chromatography. The crude products were easily purified to near quantitative functionalization by short column chromatography, and the excess pyridyl species could be efficiently recovered and reused. Even though the addition of the polystyrene chain to the 6 position provides some steric hindrance to the ability of pyridyl end-caps to serve as ligands, the

27 terpyridine-functionalized products were found to form bis complexes readily upon

addition of 0.5 equivalents of iron (II) chloride to a solution of the polymers, as

determined by ultraviolet-visible spectrophotometry and nuclear magnetic resonance

spectroscopy.

2.2 Introduction

Polymers bearing ligands with the ability to complex transition metals are a

continued subject of interest due to their ability to form tunable and reversible

supramolecular structures. 1-3 In particular, metal-bipyridyl and terpyridyl complexes

have received much attention since they provide strong non-covalent associations for

which the kinetics and thermodynamics of binding can be tuned through the choice of

metal ion, and because the behavior of these complexes is well characterized for small

molecule systems. 2,4,5 Polymer-bound bipyridyl and terpyridyl complexes of metals such

as Zn, Fe, Ni, Co, Cu and Ru have been used to construct supramolecular step-growth

polymers, 6 block copolymers 7,8 , amphiphilic polymer assemblies, 9 star polymers, 10-14 micelles 15 , tunable self-assembled structures, 16 polymeric chromophores, 17,18 and

polymers for self-healing materials. 19 These constructs make use of two main classes of

polypyridyl-functionalized polymers: side-chain functionalized polymers and end-

functionalized polymers; here, we focus on the latter.

End-functionalization of polymers with terpyridine and bipyridine can be

approached by incorporation of the pyridyl species either pre- or post-polymerization.

Pre-functionalization has been accomplished using a polypyridine-bearing initiator for

nitroxide mediated polymerization (NMP) of polystyrene, polyisoprene, poly( N,N-

28 dimethylacrylamide), poly(butyl acrylate), poly(2-vinylpyridine), and poly(4-

vinylpyridine) with polydispersities in the range of 1.1-1.3. 20 Polypyridyl initiators have

also been used to prepare well-defined, pre-functionalized polystyrene bearing

bipyridine-ruthenium complexes via atom-transfer radial polymerization (ATRP) from a

complexed bipyridyl initiator. 10,14,21 Similar complexes have also been used to initiate

the anionic ring-opening polymerization of 2-ethyl-2-oxazoline, ethylene imine, ethylene

oxide, and lactic acid. 11-13,22 In addition to the aforementioned techniques, reversible

addition fragmentation chain transfer polymerization (RAFT) has been used to

accomplish similar goals with vinyl monomers. 23-26

While effective, pre-functionalization methods suffer from the drawback that an involved synthesis is needed to obtain the appropriately functionalized initiators. Post- functionalization approaches, on the other hand, have most commonly focused on the reaction of hydroxyl-terminated polymers with 4’-chloro-2,2':6’,2”-terpyridine (chloro- terpyridine) under basic conditions, leading to nucleophilic attack of the alkoxide at the

4’ position and displacement of chloride. This approach has been successfully employed for end-functionalization of polystyrene, 27 poly(ethylene oxide), 28 and poly(ethylene-co -

butylene). 8

The application of living anionic polymerization, which remains the method of

choice for obtaining narrow molecular weight distributions, to polypyridyl-functionalized

polymers has so far been constrained by the issue of side reactions with pyridine rings.

The nucleophilic aromatic addition of alkyllithium reagents to pyridine 29-31 and

bipyridine 32 at the 2 or 6 positions to create 1,2- or 1,6-dihydropyridines, followed by

oxidation to form alkyl-substituted pyridines, is well documented. This nucleophilic

29 addition has also been employed in the addition of polystyryllithium and

polyisoprenyllithium to poly(2-vinylpyridine) to form comb-like graft copolymers. 33

However, in most cases this reaction is undesirable; for example, avoiding attack of the living anion on pyridine motivates the use of polar solvents at low temperature as commonly employed for anionic polymerization of 2- and 4-vinyl pyridine. 34 In preparing terpyridine end-functionalized polymers, one route that has been taken to circumvent this reaction is to end-cap the living anionic chains with ethylene oxide, followed by coupling of the hydroxyl-terminated chains with chloro-terpyridine as described above for polystyrene and poly(ethylene-co -butylene). 8 An alternative method is to end-cap or copolymerize polystyrene with 1,1-diphenylethylene. As described by

Schubert and co-workers, the diphenylethylene anion is not sufficiently reactive to add to the 6 positions, thus it reacts selectively with the 4’ position of chloro-terpyridine. 35,36

While both of these approaches provide effective routes to end-functionalized polymers, they require the purification and addition of an intermediate end-capping agent before reaction with chloro-terpyridine, increasing the number of steps in the synthesis.

N N N

Li Li N N

1.H

2.O2/-H2

N N

Scheme 2.1 End-capping of Polystyryllithium with Terpyridine.

Here, rather than seeking to reduce the reactivity of the anionic chain end, we exploit the direct nucleophilic addition of polystyryllithium to bipyridine and terpyridine to provide a simple route to end-functionalized polymers. To avoid coupling by addition of two living chains to the same pyridyl unit, we employ a large excess of purified bipyridine or terpyridine, resulting in predominantly singly-substituted 6-polystyryl polypyridine species upon rearomatization, as shown in Scheme 2.1. We show that this approach allows us to take advantage of the fast reaction times, quantitative conversions, and low-dispersity § products afforded by anionic polymerization to yield polypyridyl end-functionalized polymers without the need for an intermediate end-capping agent.

The primary disadvantage of this approach is the need to use an excess of bipyridine and terpyridine; however, the excess quantities can easily be separated and reused. While the addition of polystyrene to the 6 position might be expected to introduce significant steric hindrance to the ability of the pyridyl end-functionalities to serve as ligands for metal salts, the terpyridine-terminated polymers synthesized here are shown to form bis complexes in 2:1 mixtures with iron(II) chloride, a behavior well-known to occur with terpyridine-functionalized polymers prepared by other techniques. 34

§ In this dissertation, “dipsersity” ( ð) will be used in place of the more often used “polydispersity” as per the recommendations of IUPAC. (37) Stepto, R. F. T. Pure and Applied Chemistry 2009 , 81 , 351-353.

Scheme 2.2: Apparatus Used for Anionic Polymerization b b Solvent and monomer were distilled from flask A, polymerization was carried out in flask B and the living polymers were transferred to flask C for end- functionalization.

2.3 Experimental

• Materials

Styrene (90-100%) was purchased from Mallinckrodt-Baker and stored over calcium hydride (Acros, 92%). 2-2’-Bipyridine (99%) was purchased from Sigma-

Aldrich and sublimated twice at 110 °C before being stored at 50 °C under vacuum until use. 2,2’:6,2”-Terpyridine, obtained partially from Strem Chemicals (98%) and partially from the Tew Research Group at the University of Massachusetts, Amherst, was purified by sublimation at 130 °C followed by recrystallization from methanol. As with bipyridine, terpyridine was stored under vacuum at 50 °C until use (and at least overnight) to maintain dryness. Sec -butyllithium (1.3 M in 92:8 cyclohexane/hexane) and dibutylmagnesium (0.5 M in heptane) were obtained from Acros Organics and used as received. The sec-butyl lithium was titrated periodically to confirm concentration using previously developed techniques. 32 Silica gel (0.032 – 0.063 mm particle size, 60 Å pore size) was purchased from MP Biomedicals and activated by stirring in methanol followed by heating to 115 °C for 12 hours prior to use. Deuterated chloroform (99.8%

32 D) and ferrous chloride tetrahydrate (99%) were purchased from Acros Organics and used without further purification. Ethanol (95%) was purchased from Pharmco-Aaper and used without further purification, tetrahydrofuran (THF) was purchased from Fisher

Scientific (99.9%) and benzene (99.5%) was purchased from TCI America.

• Instrumentation

All 1H NMR spectra were recorded on a Bruker Avance 400 MHz spectrometer in deuterochloroform (CDCl 3) and referenced to tetramethylsilane. UV-Vis spectra were recorded on a Hitachi U-2010 spectrophotometer with optical glass cuvettes obtained from Buck Scientific. Size exclusion chromatography (SEC) was conducted using a

Polymer Laboratories GPC-50 with THF as the solvent and a Polymer Laboratories mixed-C guard column, one Polymer Laboratories mixed D column (5 m pore size, 0.2

– 40 kg/mol range) and two Polymer Laboratories mixed C (5 m pore size, 0.2 – 2000 kg/mol range) columns at 50 °C and 1ml/min flow rate. Signal was generated by an RI detector and molecular weight was determined against polystyrene standards.

• Anionic Polymer Synthesis

Anionic synthesis was carried out using a modified version of a previously described apparatus, 38 as shown in Scheme 2.2. All glassware and stir bars were heated in an oven to 115 °C for at least 2 hours prior to use, and assembled quickly while hot.

Flask C (300 mL RBF) was charged with either 5.0 g of bipyridine or 7.0 g of terpyridine

(25 and 23-fold molar excesses to the initiator, respectively). Vacuum (60-80 mTorr) was pulled inside of the setup while the outside of the glassware was heated via hand torch for further drying. Following the heating of the glassware under vacuum, the setup was

33 evacuated and backfilled three times with dry, deoxygenated nitrogen gas and all glassware joints were fitted with paraffin wrap.

Following the vacuum/backfill steps, the desired amount of styrene was transferred to flask A (250 mL RBF) accompanied by 2 mL of dibutylmagnesium. The mixture was allowed to stir at 70 °C for 30 minutes while the rest of the setup was exposed to vacuum.

At the end of this time, flask B (300 mL RBF) was cooled to -196 °C, and flask A was exposed to vacuum briefly to remove residual atmosphere. The nitrogen/vacuum valve on the Schlenk line was then closed and the vacuum transfer of styrene was completed.

After the vacuum transfer of styrene into flask B, flask A was placed under nitrogen flow and charged with 200 mL of solvent (Benzene or THF) accompanied by 5 mL of styrene. A drying polymerization was initiated by adding 0.5 mL of 1.3 M sec - butyllithium in 92:8 cyclohexane/hexane. During the 10 minutes that the drying reaction was allowed to proceed, the entire setup was subjected to vacuum and flask B was cooled with liquid nitrogen. After 200 mL of solvent was vacuum transferred to flask B, the process was repeated to transfer 50-100 mL of solvent into flask C. Once the vacuum transfers were completed, 3 freeze/pump/thaw cycles were undertaken on flasks B and C simultaneously.

Polymerization of styrene was initiated by addition of 1.0 mL of 1.3 M s ec - butyllithium solution under a positive pressure of nitrogen in flask B. The s ec - butyllithium was added through the septum shown in Scheme 2.2 with a long needle that reached completely through the stem. After the initiator was added, the resulting orange color indicated initiation of the polymerization. The stopcock on flask B was then closed in order to isolate the punctured septum and the reaction was allowed to proceed for 1

34 hour at room temperature in the case of benzene (-78 °C in the case of THF). After the

allotted time was over, the entire setup was tilted so that all but about 10-30 mL of the

reaction mixture was decanted from flask B into flask C, resulting in a dark red color for

bipyridine, or a dark green color for terpyridine. The end-capping reaction was allowed

to proceed for an additional hour at room temperature (- 78 °C in THF) before the

polymer was quickly poured into 400 mL of ethanol to both terminate and precipitate the

polymer. Unfunctionalized polymer remaining in flask B was simultaneously terminated

and precipitated separately by quickly pouring the reaction mixture into methanol.

Following synthesis, the functionalized polymer was filtered, dissolved in acetone

and re-precipitated into ethanol twice more. All ethanol fractions were saved, the solvent

removed in vacuo and the bipyridine and terpyridine recovered and re-purified for use in future polymerizations. Typical recovery of the unconsumed end capping material was greater than 95%.

• Chromatographic Separation

The crude end-functionalized polystyrene (15 to 30 mg) was weighed into a tared vial, and dissolved into 20 mL of chloroform. A fritted glass filter (100 mL) mounted on a vacuum flask with a rubber adaptor was filled to within 2 cm of the top with silica before the polymer solution was poured into the filter. Vacuum was pulled on the flask and 300 mL of chloroform was poured through the filter. Functionalized polymer was retained in the column due to the strong adsorption of terpyridine onto silica gel; this was confirmed by concentrating the eluent in vacuo and characterizing by TLC in chloroform.

In the event that TLC showed material at the baseline (indicating the presence of functionalized polymer), the polymer solution was pulled through (fresh) silica a second

35 time. A control experiment using non-functionalized polystyrene was conducted to verify that the non-functionalized side products did not adhere to the column and could be quantitatively recovered from the eluent.

The resulting solution, which contained only non-functionalized chains, was then evaporated to dryness in a tared vial using a rotary evaporator, and the residual polymer weighed to determine the weight-average percent functionality of the crude polymer sample. Neat triethylamine (300ml) was then introduced into the column to displace the functionalized chains from the silica, the resulting solution was then collected and concentrated in vacuo before being precipitated in methanol and filtered to afford the purified functionalized product.

• Ultraviolet-Visible Spectrophotometry (UV-Vis) Characterization of Chain Ends

An optical glass cuvette fitted with a septum was charged with 2.5 mL of dried

THF and 0.2 mL of styrene. The polymerization was then initiated by 0.1 mL of 1.3 M sec-butyllithium, and the UV-Vis spectrum was immediately measured. After the spectrum was recorded, an excess of dried, sublimated bipyridine was added to the reaction and the UV-Vis spectrum of the resulting dark red solution was recorded. An analogous experiment was conducted with terpyridine.

• MALDI-TOF Characterization of Sample 5

The MALDI-TOF (matrix-assisted laser desorption/ionization time-of-flight spectroscopy) spectrum of sample 5 was recorded on a Bruker Daltronics Reflex III

MALDI-TOF with a 337 nm pulsed laser. The spectrum was recorded in reflectron mode and positive ions were used as the detection species. The MALDI-TOF spectrometer was calibrated using equine cytochrome C, bovine insulin, and oxidized bovine insulin chain

36 B (Sigma Aldrich). To prepare polymer/matrix mixtures, 5 L of a saturated solution of

dithranol (MP Biomedicals, 99%) in THF was mixed with 5 L of a 1 mg/mL polymer

solution in THF. To this solution was added 1 L of a saturated solution of silver (I)

triflouroacetate (Alfa Aesar, 98%) in THF. Approximately 0.5 L of this solution was

dropped onto the target and the solvent allowed to evaporate before the spectrum was

recorded. The number of laser shots was generally 200-300 at 75% intensity.

• UV-Vis Titration of Sample 1

Purified, functionalized sample 1 (38.8 mg) was dissolved in 100 mL of chloroform and allowed to stir continuously. After recording a baseline, a solution of

FeCl 2●4H 2O in methanol (16 mM) was added in 0.1 mL quantities, allowing 5 to 10 minutes between the addition of iron and the removal of 3 mL aliquots for spectral observation. Absorption values for each portion were taken for the peak at 325 nm, after which the aliquot was reintroduced into the reaction vessel.

• NMR Characterization of the Iron Complex of Sample 5

Functionalized sample 5 (50.7 mg) was dissolved in 1 mL of deuterated

chloroform (0.39 mM) containing 0.02 % TMS by volume (Acros Organics). A stock

solution of iron (II) chloride was made by dissolving 79.0 mg of FeCl 2●4H 2O in 2 mL of

methanol for a concentration of 0.195 M. An aliquot (10 µL) of the iron solution was

added to the polystyrene solution to give a molar ratio of 2:1 PS/iron. An NMR spectrum

of this solution was taken and compared with the NMR spectrum of a 0.39 mM solution

of non-complexed functionalized sample 5 in deuterated chloroform.

Figure 2.1 Ultraviolet-visible spectrophotometry absorption spectra of polystyryllithium (solid line), polystyryllithium with terpyridine (grey line), and polystyryllithium with bipyridine (dashed line). No absorption from free bipyridine or terpyridine occurs at these wavelengths, thus the spectral changes result from addition of polystyryllithium to the pyridyl species to form anionic intermediates.

2.4 Results and Discussion

Our approach to prepare end-functionalized polystyrene relies on end-capping of

living anionic chains using an excess of the appropriate pyridyl species. Evidence for

successful nucleophilic addition of the living chains to bipyridine or terpyridine was

provided by a change in color from orange for the polystyryl anion to dark red or dark

green, respectively. To further characterize this process, we performed UV-vis

absorption measurements on each of the anionic chain-end species (Figure 2.1). The

spectrum of polystyryllithium shows no features at wavelengths above 375 nm, in clear

contrast to the spectrum of the PS-bipyridyl anion, which shows an absorption peak

around 425 nm, and that of the PS-terpyridyl anion, which shows a small absorption peak

38 at around 410 nm and an intense broad peak centered at 615 nm, the latter giving rise to the intense dark green color. Since bipyridine and terpyridine on their own do not show any absorption in this region, these spectral changes strongly suggest the formation of anionic pyridyl chain ends. We note that the concentrations of polystyryllithium and pyridyl species used in this experiment were too high for reliable data to be obtained below 375 nm, thus preventing observation of the polystyryllithium peak at 334 nm.

Table 2.1 Properties of Bipyridine and Terpyridine End-Functionalized Polymers.

M in g/mol ( ð) by SEC End-Functionality n Solvent/ Terminating Sample No. Temperature Species Non- Functionalized NMR Chromatography Functionalized 1 2100 (1.11) 2250 (1.13) 88.5 % 85.9 % Benzene/r.t. Terpyridine 2 2560 (1.13) 2790 (1.13) 91.2 % 93.4 % THF/-78°C Bipyridine 3 4400 (1.07) 4500 (1.08) 91.4 % 93.3 % Benzene/r.t. Bipyridine 4 ------a 6300 (1.09) 92.4 % 89.7 % Benzene/r.t. Bipyridine 5 11250 (1.06) 11170 (1.07) 87.3 % 84.2 % Benzene/r.t. Terpyridine 6 13460 (1.09) 12865 (1.13) 82.8 % 77.0 % Benzene/r.t. Bipyridine a Data not collected.

Figure 2.2 1H nuclear magnetic resonance spectrum of functionalized sample 4 in deuterochloroform showing the bipyridyl shifts in detail. The integrated intensity of the peak at 8.62 ppm was compared to the methyl peaks from the initiator (0.88 ppm) to determine the degree of end-functionalization.

39 Having obtained spectroscopic evidence for reaction of anionic polystyrene chains with terpyridine and bipyridine, we used this method to synthesize a range of end- functionalized polymer chains with molecular weights of 2 – 13 kg/mol, as summarized in Table 2.1. In addition to carrying out polymerizations in benzene at room temperature, we prepared sample 2 to verify that the approach also worked for polymerizations in THF at -78 °C. The efficiency of end-functionalization for each sample was calculated in two ways: by NMR spectroscopy and by gravimetric analysis of chromatographically- separated polymer chains. Calculation of efficiency by NMR for bipyridine-terminated samples was performed by comparing the integrated intensity of the singlet at 8.62 ppm from the remaining 6’ hydrogen on bipyridine (see Figure 2.2) to that of the multiplet at

0.88 ppm arising from the two methyl groups on the initiator. In the case of terpyridine, it was not possible to resolve peaks from individual pyridyl protons, thus we used the integrated intensity of the peaks between 7.9 and 8.7 ppm, corresponding to the six terpyridyl protons not obscured by the polystyrene aromatic peaks (the 3, 3’, 4’, 5’, 3” and 6’’ protons). Functionalities determined by NMR ranged from 83 - 93 %, and were generally very close to the values determined by chromatographic separation, which ranged from 77 – 93 %. We note that the dihydropyridyl species formed initially upon termination of the pyridyl anions were never observed by NMR, suggesting that oxidative rearomatization to the substituted bipyridine or terpyridine occurs very rapidly under ambient conditions.

Further evidence for the successful functionalization of polystyrene by this approach is given by comparing the molecular weights of chains end-capped with bipyridine or terpyridine to those determined for the small quantity of polymer from each

40 reaction where polystyryllithium was directly terminated with methanol. As summarized in Table 2.1, most samples showed only a slight change in M n as determined by SEC, consistent with the addition of a single polypyridyl unit to the end of each chain.

(Sample 6 yielded a slightly greater molecular weight for the non-functional polymer compared to its functionalized counterpart, likely indicating that the conversion of styrene had not reached 100 % when the majority of the polymer was transferred to the flask containing terpyridine.) A slight increase in dispersity also occurred between the functionalized and non-functionalized polymers for each sample synthesized in benzene, predominantly reflecting a small quantity of double molecular-weight chains in the functionalized samples, as described below. MALDI-TOF spectroscopy was used to definitively establish terpyridine end-capping of the purified functionalized polymer 5 by examining the specific mass peaks (Figure 2.3). The spacing of peaks in the spectrum corresponds to the molecular weight of styrene (104.15 g/mol), and the central peak shows a molecular weight of 11,955.8 g/mol. This represents 111 styrene repeat units, the sec -butyl group from the initiator (57.1 g/mol), terpyridine minus a hydrogen atom

(232.3 g/mol), and the silver cationation agent (107.9 g/mol), totaling 11,958.0 g/mol.

The small discrepancy of 2.2 g/mol from the measured value is reasonable for MALDI-

TOF spectra of polymers in this weight range. The number average molecular weight determined by MALDI-TOF, Mn = 11.6 kg/mol is also in good agreement with that from

SEC, M n = 11.3 kg/mol.

Figure 2.3 Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) spectrum of terpyridine-functionalized, purified sample 5 in reflectron mode using dithranol as a matrix and silver(I) trifluoroacetate as a cationization agent. The molecular weight of the labeled peak (11955.8 g/mol) corresponds to a polymer with 111 styrene repeat units and a terpyridyl group, confirming successful end- functionalization of this sample. The number average molecular weight calculated from this spectrum was 11.6 kg/mol, in good agreement with the value of 11.3 kg/mol from SEC.

Although the efficiency of end-functionalization was less than 100%, leading to a significant fraction of non-functionalized polymers in the crude reaction products (Table

2.1), samples could easily be purified to higher purity by short column chromatography.

The affinity of terpyridine and bipyridine for the silica gel column packing means that end-functionalized polymers show little movement on the column unless a competing base is introduced into the eluent, while non-functional chains can be washed out of the column. Specifically, we used chloroform as the eluent in all cases, followed by the addition of neat triethylamine to displace the functionalized chains. The purities of the resulting materials, as characterized by NMR, are shown in Table 2.2 for three samples.

In all cases the percentage of functional chains was increased to above 96%, significantly

42 greater than that for the crude reaction products described in Table 2.1. In the most

extreme case, sample 5, the functionality was increased from 87 % to greater than 99 %

by this separation.

Table 2.2 End-Functionalities after Purification by Short-Column Chromatography.

Sample No. End-Functionality (NMR) 1 99.7 % 4 96.5 % 5 99.3 %

2.4.1Coupling of PS chains through addition to polypyridine units

The main shortcoming of our approach, and the reason that large excesses of pyridyl species were used for end capping, is that each bipyridine or terpyridine contains two reactive 6 positions, and thus two living polymer chains may couple during functionalization to yield a double molecular-weight species with a polypyridyl moiety at its midpoint. With a 23 – 25-fold excess of polypyridyl species, the occurrence of double-weight species, as determined by SEC, was fairly low in all samples, ranging from undetectable to 4.8 % of polymer chains in the worst case of sample 3 (Figure 2.4).

In contrast, polymers synthesized with stoichiometric amounts of bipyridine and terpyridine contained ~ 15 % by number of double weight chains (Appendix 2.A). We note that these values may also reflect any polymer chains coupled through the inclusion of oxygen into the sample. 39,40 Thus, to unambiguously establish that the dominant coupling process occurred through addition to polypyridine units, a detailed characterization of the side products for sample 4 was undertaken (see Appendix 2.A for details), which revealed that greater than 75 % of the double-weight species for this sample were bipyridine-centered.

Figure 2.4 Size exclusion chromatographs of end-functionalized sample 3 before (dashed line) and after (solid line) purification by short column chromatography reveal a decrease in the proportion of double molecular weight chains.

Based on a simple model of equal reactivity of both reactive positions on each polypyridine (Appendix 2.A), one would expect that a stoichiometric amount of bifunctional terminating agent would yield 33 % by number of double-weight chains, substantially greater than actually observed. By contrast, a 25-fold excess would be expected to yield 1 % double-weight chains, less than typically observed in our experiments. These observations suggest that the process of coupling through a polypyridine unit is more complicated than independent reactivity of each reactive 6 position; we note that several factors including steric effects, electrostatic interactions, and the tendency of living anionic polymer chains to aggregate in non-polar solvents, 41 may all influence the efficiency of coupling. Further study is required to better understand this process and find suitable conditions to minimize (or enhance) its occurrence. In practice, however, it is straightforward to keep the proportion of mid-

44 functionalized double-weight chains acceptably low by using a sufficiently large excess

of pyridyl species during end-capping. Even in the worst case of sample 3, following purification by column chromatography, double-weight chains made up only 1.6 % of the sample (Figure 2.4).

Finally, we note that the presence of double-weight polymers introduces some

complications into our determination of percent functionality. In terms of

chromatographic separation, TLC analyses suggest that the double-weight chains show a

decreased affinity for the silica gel due to steric hindrance at the polypyridyl center, and

thus they move on the column at a rate intermediate between the end-functionalized

chains and the non-functionalized chains. Since there is significant uncertainty in the

extent to which these species are reflected in the chromatographic yield calculations, the

reported values overestimate slightly the percentage of end-functionalized chains.

Similarly, our NMR analyses overestimate the percentage of terpyridine end-

functionalized chains, since one double-weight chain is counted as 5/6 th of a functionalized chain due to the remaining contributions of the 3, 3’, 4’, 5’ and 3” positions of mid-functional terpyridine moieties to the NMR spectrum. For bipyridine, the values determined by NMR accurately reflect the fraction of end-functional chains, as the 6’ proton used to identify bipyridine is absent in the case of the double-weight polymers. However, in all cases the errors introduced are several percent or less due to the relatively small proportion of double-weight chains, and thus we have not attempted to correct for these complications.

Figure 2.5 Ultraviolet-visible spectrophotometry (absorption at 329 nm) of a 0.17 mM solution of terpyridine-functionalized sample 1 titrated with iron (II) chloride hexahydrate. The change in slope of the line at 0.5 equivalents of iron indicates nearly quantitative formation of a bis-tepyridyl iron complex, despite the presence of polystyrene chains on the 6 position of each terpyridine ligand.

• Steric effects and formation of supramolecular complexes

An important distinction between the end-functionalized polymers produced here and those prepared via other routes lies in the attachment of polystyrene at the 6 position of the pyridine ring, ortho to the nitrogen atom, compared to the typical 4’ (or 4) position, para to the nitrogen. This positioning of the polystyrene chain may be expected to impose steric limitations on the ability of the polypyridyl species to serve as a ligand for metal ions. Indeed, as described above, we suspect that such steric effects are responsible for the reduced affinity of the mid-functional double molecular weight polymers for silica gel. To establish that the end-functionalized polymers prepared here were nonetheless capable of forming supramolecular complexes, we performed a UV-vis titration of terpyridine-capped polymer 1 with FeCl 2●4H 2O in chloroform (Figure 2.5). The

46 absorption at 325 nm shows a linear increase with the concentration of iron(II) until 0.5

equivalents, where a transition to a smaller slope occurs. This titration curve is a clear

indication of essentially quantitative formation of a bis complex until greater than 0.5

equivalents or iron are added, as typically observed for non-sterically hindered

terpyridine moieties. 35 In addition to the peak at 325 nm from which the titration was

recorded, a metal-to-ligand charge-transfer band was observed at 563 nm; both are

consistent with the values reported by Schubert and co-workers as distinguishing

characteristics of the iron-terpyridine bis complex. 35 Further indication of bis complex formation was provided by an NMR spectrum taken of sample 5 in the presence of 0.5 equivalents of iron(II) (Figure 2.6). While the spectrum of sample 5 in the absence of iron clearly shows terpyridine peaks in the aromatic region, the spectrum of the 2:1 mixture of sample 5 : iron(II) shows no such peaks, which suggests the formation of a paramagnetic high-spin iron(II) complex. This is likely the result of a lengthened Fe-N bond brought about by steric hindrance due to substitution of terpyridine at the 6-position with polystyrene, an effect that has been previously reported for iron(II) complexes of

6,6”-substituted terpyridines. 42 Thus, while the steric hindrance provided by functionalization at the 6 position does influence the complexation behavior, it is not sufficient to prevent formation of a bis complex. It is notable that no Knight shift protons were observed in the NMR spectrum from δ = - 600 to + 600 ppm, as protons with large

upfield or downfield shifts are expected from paramagnetic compounds. 43 The absence

of Knight shift protons in paramagnetic terpyridine-functionalized polymer complexes

7 has been reported previously for aqueous PEO-bound [terpy] 2Co(II) systems, though the

reason for this behavior remains under investigation.

Figure 2.6 1H nuclear magnetic resonance spectra of the pyridyl regions of A) terpyridine-functionalized sample 5, in deuterochloroform and B) a 2:1 mixture of functionalized sample 5 with iron (II) chloride in deuterochloroform. The complete disappearance of the terpyridyl peaks at 0.5 equivalents of iron further supports the near quantitative formation of a bis complex.

2.5 Conclusions

In this work we have shown that bipyridine- and terpyridine-terminated polystyrenes can be effectively synthesized via anionic polymerization and end-capping by direct nucleophilic attack of polystyryllithium on the desired polypyridine species, followed by oxidative rearomatization. Lower molecular weight polymers showed high efficiencies of functionalization (around 90%) by NMR and chromatographic separation, while higher molecular weight samples showed a slightly decreased yield. While the degree of functionalization of the crude products is somewhat lower than for a previously reported approach to terpyridine-functionalized anionically-polymerized polystyrenes that yielded

48 91-100 % functionality when conducted in toluene, 35 the current approach does not require the addition of an intermediate end-capper to the polymer chain. Further, when conducted in THF (sample 2), the present approach yielded 91 % functionalization with bipyridine, which is a considerably better than terpyridine-capped polymers synthesized in THF in the previous study (59-66 %). 35 Additionally, we have shown that it is relatively straightforward to purify samples to obtain highly enriched (> 96 %) end- functionalized polymer. The ability of the purified terpyridine-functionalized polymers to form bis complexes with an iron(II) salt was confirmed by UV-Vis. This process has advantages over existing procedures, namely rapid reactions, low dispersity, and the absence of synthetically demanding initiators or intermediate end-capping agents. The main disadvantage to this process is the large excess of terpyridine and bipyridine needed to minimize the fraction of chains that couple through addition to a single pyridyl unit; however the excess bipyridine and terpyridine can be easily separated from the polymer and purified for later reuse. This technique may also be applicable to the functionalization of dienes and other monomers suitable for anionic polymerization, and further raises the possibility to intentionally prepare homopolymers or block copolymers containing a single terpyridine- or bipyridine mid-chain functionality by allowing the end-functionalized polymers to react with the live anion of a second polymer.

49 2.6 Appendix 2A: Side Product Characterization, Detailed Synthetic Procedures, and Equal Reactivity Model

• Functionalization with Stoichiometric Amounts of Terpyridine/Bipyridine

To demonstrate the necessity of using a large excess of terpyridine or bipyridine to avoid coupling, a sample of polymer was synthesized using a 1:1 molar ratio of terpyridine to butyllithium (1.0 mL of 1.0 M butyllithium/ 0.223 g of terpyridine).

Deconvolution of the chromatogram for the functionalized polymer yielded an estimate of 28 wt % (16 % by number) of double-weight chains (Figure 2.A1). In contrast, the unfunctionalized polymer contained only 0.3 wt % double-weight chains, a typical value for the results of oxidative coupling in our experiments. The experiment was repeated with a 1:1 molar ratio of bipyridine to butyllithium, yielding similar results (26 wt % double-weight chains).

Figure 2.A.1 Size exclusion chromatograph of polymer functionalized with an equimolar amount of terpyridine. Base polymer(solid line): Mn = 10.9 kg/mol, ð = 1.05. Functionalized Polymer (dashed line): Mn = 12.2 kg/mol, ð = 1.18.

• Detailed Characterization of Side Products for Sample 4

50 After purification of sample 4 as described in the main text, the separated side

products were analyzed, with results summarized in Table 2.A.1. After drying in vacuo , side products were dissolved in 0.5 mL of THF. A 5 L aliquot was removed and analyzed by MALDI-TOF; the remainder was characterized by GPC. A second sample of the side products was isolated and characterized by NMR.

The SEC trace of the byproduct (Figure 2.A.2) indicated that 10 ± 1 % (by number) of chains in the byproduct sample were double-weight. To distinguish bipyridine-centered chains from chains coupled due to oxygen inclusion, the doublet centered at 7.65 ppm in the NMR spectrum (corresponding to the 4 and 4’ positions of bipyridine) was integrated and compared to the signal from the polystyrene backbone.

Since no peak was observed at 8.62 ppm (corresponding to remaining 6’ bipyridyl protons on end-functional polymers), all pyridyl signals from the byproduct arose from bipyridine-centered double-weight chains. This analysis revealed that 9 ± 1 % (by number) of the byproduct consisted of bipyridine-centered double weight chains.

Comparison of the results from NMR and SEC indicate that only 10 % (± 14 %) of the double-weight chains can be the result of oxidative coupling; the real number is probably smaller given that a portion of bipyridine-centered chains are not eluted with the rest of the side products during separation. The MALDI spectrum of the separated byproducts showed only PS-H; neither the double-weight polymers nor any other side products were observed. The central peak in the MALDI spectrum was at 6746.6 g/mol, corresponding to 64 styrene repeat units, the butylithium initiator fragment (57.1 g/mol), and a sodium atom (22.9 g/mol), yielding a total of 6745.7 g/mol.

Table 2.A.1 Composition of Unpurified Functionalized Sample 4 Product Percent of Total (number) PS-bpy 90 ± 1 PS-H 9 ± 1 PS-bpy-PS 0.9 ± 0.15 Non-bpy double-weight < 0.3

Figure 2.A.2 SEC trace of the chromatographically separated byproducts of sample 4.

• Equal Reactivity Model for Pyridyl-Centered Coupling

If the molar ratio of polypyridyl end-caps to living chains is m, we take the reactivity of each reactive pyridyl 6 site to be equal and uncorrelated, and all primary chains react with one site, then the probability of any given reactive site being grafted with a primary chain is p = (2 m)-1. The probability that any given primary chain is part of a double-weight species is also p, and therefore the weight fraction of double weight

-1 -1 chains is wd = p, while the number fraction is xd = (2/ wd – 1) = (4 m – 1) , neglecting the weight of the pyridyl species. For the values of m = 25 and 23 employed here for

bipyridine and terpyridine, respectively, this simple treatment predicts 1.0 % and 1.1 %

by number of double weight chains, while for m = 1, it predicts 33 % by number.

• A Detailed Synthesis of Sample 7

To measure synthetic yields, an additional polymer ( 7) was prepared as described

in detail here. Data for this sample are not included in Table 2.1 of the main text.

Terpyridine (7 g, 0.03 moles) was added to a 500 mL RBF (flask C) and dried in vacuo at

40 ºC for 12 h** . The glassware was assembled directly out of the glassware oven (120

ºC) as described in the main text. After evacuating the interior of the glassware setup (60

mTorr) a hand torch and hot-air gun were used to heat the glass, paying special attention

to ground joints to remove water from the silicone joint grease. All surfaces of the glass

were heated except for the flask containing terpyridine, as this would have caused

sublimation of the material into the vacuum transfer manifold.

After heating the glassware, 3 vacuum-backfill cycles were completed with dried,

deoxygenated nitrogen gas †† . Following this step, the left side (flask B and C) of the

setup was evacuated, and closed off from the rest of the setup via the left-side stopcock.

The rest of the setup was filled with nitrogen, and one of the glass plugs closing flask A

(1000 mL RBF) was removed for the injecting of 2 mL of 1.0 M di-n-butylmagnesium

solution in cyclohexane/hexane. After the solution was injected, the flask was closed and

evacuated, removing the solvent and leaving a light grey dibutylmagnesium residue on

the bottom of the flask. Nitrogen was reintroduced to the flask, and the plug was once

again removed to allow the introduction of 22 mL of styrene (0.19 moles) into the flask.

** In previous experiments, terpyridine and bipyridine were further dried by three cycles of evacuation of the flask containing the compound on the synthesis setup, followed by closing the valve to the flask and melting the compound, then resolidifying and pulling vacuum. However, it was found that no significant improvement in yield was obtained, thus this step was dropped from the synthetic procedure. †† The nitrogen was dried trough Drierite™ ( ≈98% calcium sulfate 2% cobalt (II) chloride monohydrate), and deoxygenated with a Restek™ high capacity indicating oxygen trap.

53 The plug was replaced and the flask was evacuated until the styrene had just begun to

boil, at which time the valve was closed, isolating the flask from the rest of the setup.

Flask A was heated to 70 ºC and the styrene/di-butylmagnesium slurry was

stirred at this temperature for 30 min. At end of this time, flask B was placed in a 500

mL dewar flask full of liquid nitrogen. After flask B was cooled to -78 ºC, the valve

connecting the glassware to the gas manifold was closed; the valves on the vacuum

transfer manifold were opened, allowing the vacuum transfer of styrene to proceed. Once

the vacuum transfer of styrene was completed, the right-side valve of the setup was

closed, preserving the vacuum in flasks B and C. Nitrogen was once again introduced

into the rest of the setup, and flask was charged with approximately 400 mL of benzene

and 5 mL of styrene. A drying polymerization was initiated with sec-buyllithium, after

which the flask was evacuated until the benzene had just begun to boil. Next, ~ 400 mL

of benzene was vacuum transferred into flask B in the same manner as styrene.

After 3 freeze-pump-thaw cycles on flask B, ~ 100 mL of the styrene-benzene

solution was transferred into flask C by tipping the setup and allowing the liquid to flow

from flask B to flask C, where the liquid was stirred until the terpyridine was dissolved ‡‡ .

Flasks B and C were then placed under a positive pressure of nitrogen and a 1 mL glass

syringe and 100 mL RBF were removed from the glassware oven; the syringe was set

aside to cool. A septum was fitted to the flask and vacuum/backfilled with nitrogen 3

times ending with a positive pressure of nitrogen. After the syringe was cool enough to

handle, it was fitted with a 24-gauge 12” needle and purged with nitrogen by drawing

from the 100 mL flask several times. The syringe was then used to inject 1 mL of

‡‡ This was an improvement over the technique used for samples 1-6; done so that two flasks would not have to be freeze-pump-thawed at once. This way, the potential “bumping” of the liquids which can cause cross-contamination between the flasks is eliminated and the loss of polymerizable styrene into flask C can be accounted for.

54 nitrogen into the butyllithium container before withdrawing 1 mL of 1.3 M sec-

butyllithium (1.3 mmoles) in heptane into the syringe. §§ The septum on flask B was then

punctured with the needle and the needle inserted into the flask until the tip was 1-2 cm

above the vortex of the vigorously stirred fluid. The butyllithium was rapidly injected

into the solution, and the immediate appearance of orange coloration indicated initiation.

The needle was rapidly withdrawn and the stopcock closed to isolate the punctured

septum.

After stirring the solution for 75 min *** at room temperature the polystyryllithium was end-capped by closing the valve connecting flasks B and C to the rest of the setup and tilting the entire apparatus so that the polystyryllithium solution flowed into flask C.

The resulting dark green color indicated reaction of the terpyridine with the polystyryllithium. Approximately 50 mL of the polystyryllithium solution was allowed to remain in flask B. After 45 min had elapsed, ethanol was injected into flask B to terminate the polystyryllithium ††† , and flask C was precipitated in 1 L of vigorously stirred ethanol. The unfunctionalized polymer was then precipitated into 200 mL of methanol.

The precipitates were filtered and placed in vacuo for 2 h before being weighed.

After this, a portion (3.8 g) the functionalized sample was dissolved in acetone (40 mL) and precipitated into ethanol (150 mL) twice more before further characterization. Since

25 % of the styrene was decanted into flask C prior to polymerization, yields were

§§ We find this technique preferable to overpressuring a bottle of butyllithium prior to withdrawal for safety reasons. *** We have found that this amount of time is insufficient for the molecular weight targeted in this experiment; longer reaction times should be used for polymers of 10 kg/mol with this initiator concentration. ††† Also an improvement over techniques used to synthesize samples 1-6, which can eliminate the oxygen inclusion that can result from simultaneous precipitation and termination.

55 calculated from 16.5 mL (0.14 moles) of styrene. The target molecular weight was 11.5

kg/mol.

The filtered ethanol was concentrated in vacuo to a volume of ~ 150 mL and filtered again to remove any remaining polymer, then rotovapped to dryness in a tared flask and placed under vacuum for 1 h before being weighed. Using this method, 6.6 g of terpyridine was recovered, representing 97% of the unconsumed terpyridine.

Unfunctionalized polymer : M n = 12.0 kg/mol, ð = 1.04, 1.95 g recovered.

Functionalized polymer : M n = 10.8 kg/mol, ð = 1.09, 10.76 g recovered (total recovery

1 12.71g, 85 % yield), functionality (NMR): 83.8 % H NMR (400 MHz, CDCl 3): δ = 8.7-

7.9 (m, 6 H; terpyridine), 7.5-6.2 (m, 518 H; aromatic polystyrene), 2.5-1.0 (m, 311 H;

aliphatic polystyrene backbone), 0.88 (m, 6 H; sec -butyllithium methyl groups).

2.7 Appendix 2B: Synthesis of Mid-Functionalized Block Copolymers

A consequence of end-capping polystyryllithium with bipyridine or terpyridine is

that a second active site is available for nucleophilic substitution. Through nucleophilic

reaction on the 6” position of terpyridine (6’ position of bipyridine) bearing a polystyrene

chain on the 2 postition with a living anionic polymerization, a block copolymer may be

produced with a polypyridine ligand in the center of the polymer. Similar techniques

have been used to successfully produce diblock and multiblock using multifunctional

silanes. 44,45 This approach is convenient due to the fairly inert nature of the linking

molecule, meaning that the first block synthesized can be purified on silica before adding

56 the living polymerization of the second block. This means that a one-pot synthesis is not

necessary, and the potential pitfalls of purifying multiple monomers at once can be

avoided. It also means that multiple block copolymers can be produced from the initial

polymer, which can be stored between reactions. Furthermore a ligand is present in the

center of the copolymer, which has the potential to undergo further reactions.

In this section, a brief treatment of this process is provided. Experimentally, the

process is similar to that described for the end functionalized polymers, with terpyridine

or bipyridine end-functionalized polystyrene being present in vessel C, rather than the

small-molecule ligand. Also unlike the previous procedure, the living polymer was

present in excess in order to ensure full reaction of the functionalized polystyrene. Using

this technique, polystyrene-terpy-polyisoprene and polystyrene-bipy-poly(p-

methylstyrene) were produced. In both cases the copolymers were synthesized in

benzene.

The copolymerization with polyisoprene was accomplished by adding polymer 5

with living polyisoprene. The copolymer was separated from the reaction mixture by

extracting the unfunctionalized polymer with cold acetone, leaving the block copolymer.

The resulting product was characterized by NMR, MALDI-TOF, and GPC. The GPC

chromatograph of the reaction mixture showed a definite shift the higher molecular

weight in the main polymer peak, with the unfunctionalized polyisoprene visible at lower

molecular weights (Figure 2.B.1). Characterization by MALDI showed a peak molecular

weight of 12,583.74 g/mol which corresponds to 2 initiator fragments, 111 styrene repeat

units, 1 terpyridine- 2H, and 10 isoprene repeat units (Figure 2.B.2). The 1H NMR spectrum in CDCl 3 revealed that the material was 96.8% copolymer, and that the isoprene

57 was in predominantly (86%) 1,4 configuration, as is expected for polymerizations of isoprene in non-polar solvents. 46,47

A block copolymer of polystyrene and poly( p-methylstyrene) was produced by applying the polymerization procedure mentioned above using Sample 4 as the beginning polymer. Unlike the PS-PI copolymer, the copolymer could be separated from the reaction mixture using the flash-column chromatography technique described in Section

2.3.1. Figure 2.B.3 shows the GPC chromatographs of the starting polymer, the unpurified reaction mixture, and the purified product. From the chromatograph of the reaction mixture, both the unfunctionalized poly( p-methylstyrene) and the block copolymer can be seen. The block copolymer shows a definite molecular weight increase to M n = 7511 g/mol from the parent polymer, which had a molecular weight of M n =

4400 g/mol. The purified product shows a near total removal of unfunctionalized poly( p- methylstyrene), which was confirmed by 1H NMR. ð of the copolymer was 1.04.

Figure 2.B.1. The GPC Chromatographs of Polystyrene-terpy-poly( p- methylstyrene). The unpurified reaction mixture is shown in dashed lines, note the low molecular weight unfunctionalized polyisoprene shown at higher elution times. The solid line denotes the block copolymer, note the shift to higher molecular weights.

Figure 2.B.2. MALDI-TOF spectrum of Polystyrene-terpy-poly( p-methylstyrene). The peak molecular weight is shown by the arrow, at 12,584.74 g/mol.

Figure 2.B.3. GPC Chromatographs of Polystyrene-bipy-polyisoprene. The starting material is shown at the bottom in dashed lines, the reaction mixture is shown in solid lines, and the purified copolymer is shown at the top in dashed lines. The raw reaction mixture shows a distinct signature of unfunctionalized poly( p- methylstyrene), while the purified mixture shows little to no evidence of unfunctionalized poly( p-methylstyrene). The molecular weight of the purified polymer was found to be Mn = 7511 g/mol, ð = 1.04.

59 2.8 References

(1) Burnworth, M.; Knapton, D.; Rowan, S.; Weder, C. Journal of Inorganic and Organometallic Polymers and Materials 2007 , 17 , 91-103. (2) Schubert, U.; Hofmeier, H.; Newkome, G. Modern Terpyridine Chemisry ; Wiley-VCH: Weinheim, 2006. (3) Friese, V. A.; Kurth, D. G. Current Opinion in Colloid & Interface Science 2009 , 14 , 81-93. (4) Constable, E. C. Chemical Society Reviews 2007 , 36 , 246-253. (5) Thompson, A. Coordination Chemistry Reviews 1997 , 160 , 1-52. (6) Chiper, M.; Meier, M. A. R.; Kranenburg, J. M.; Schubert, U. S. Macromolecular Chemistry and Physics 2007 , 208 , 679-689. (7) Chiper, M.; Meier, M. A. R.; Wouters, D.; Hoeppener, S.; Fustin, C.-A.; Gohy, J.-F.; Schubert, U. S. Macromolecules 2008 , 41 , 2771-2777. (8) Lohmeijer, B. G. G.; Schubert, U. S. Angewandte Chemie International Edition 2002 , 41 , 3825-3829. (9) Guillet, P.; Fustin, C. A.; Wouters, D.; Hoeppener, S.; Schubert, U. S.; Gohy, J. F. Soft Matter 2009 , 5, 1460-1465. (10) Wu, X. F.; Fraser, C. L. Macromolecules 2000 , 33 , 4053-4060. (11) Fiore, G. L.; Klinkenberg, J. L.; Fraser, C. L. Macromolecules 2008 , 41 , 9397-9405. (12) Fiore, G. L.; Edwards, J. M.; Payne, S. J.; Klinkenberg, J. L.; Gioeli, D. G.; Demas, J. N.; Fraser, C. L. Biomacromolecules 2007 , 8, 2829-2835. (13) Lamba, J. J. S.; Fraser, C. L. Journal of the American Chemical Society 1997 , 119 , 1801-1802. (14) Wu, X. F.; Collings, J. E.; McAlvin, J. E.; Cutts, R. W.; Fraser, C. L. Macromolecules 2001 , 34 , 2812-2821. (15) Guillet, P.; Fustin, C. A.; Mugemana, C.; Ott, C.; Schubert, U. S.; Gohy, J. F. Soft Matter 2008 , 4, 2278-2282. (16) Aamer, K. A.; De Jeu, W. H.; Tew, G. N. Macromolecules 2008 , 41 , 2022-2029. (17) Shunmugam, R.; Tew, G. N. Polymers for Advanced Technologies 2007 , 18 , 940-945. (18) Shunmugam, R.; Tew, G. N. Polymers for Advanced Technologies 2008 , 19 , 596-601. (19) El-ghayoury, A.; Hofmeier, H.; de Ruiter, B.; Schubert, U. S. Macromolecules 2003 , 36 , 3955-3959. (20) Lohmeijer, B. G. G.; Schubert, U. S. Journal of Polymer Science Part a- Polymer Chemistry 2005 , 43 , 6331-6344. (21) Fraser, C. L.; Smith, A. P. Journal of Polymer Science Part a-Polymer Chemistry 2000 , 38 , 4704-4716. (22) McAlvin, J. E.; Fraser, C. L. Macromolecules 1999, 32 , 6925-6932. (23) Zhang, L. W.; Zhang, Y. H.; Chen, Y. M. European Polymer Journal 2006 , 42 , 2398-2406. (24) Zhou, G.; Harruna, I. I. Macromolecules 2005 , 38 , 4114-4123. (25) Zhou, G. C.; Harruna, II Macromolecules 2004 , 37 , 7132-7139.

60 (26) Zhou, G. C.; He, J. B.; Harruna, II Journal of Polymer Science Part a- Polymer Chemistry 2007 , 45 , 4204-4210. (27) Lohmeijer, B. G. G.; Schubert, U. S. Polym. Mater. Sci. Eng. 2001 , 85 , 460. (28) Schubert, U. S.; Hien, O.; Eschbaumer, C. Macromolecular Rapid Communications 2000 , 21 , 1156-1161. (29) Barr, D.; Snaith, R.; Mulvey, R. E.; Reed, D. Polyhedron 1988 , 7, 665- 668. (30) Ziegler, K.; Zeiser, H. Berichte Der Deutschen Chemischen Gesellschaft 1930 , 63 , 1847-1851. (31) Ziegler, K.; Zeiser, H. Justus Liebigs Annalen Der Chemie 1931 , 485 , 174-192. (32) Ireland, R. E.; Meissner, R. S. Journal of Organic Chemistry 1991 , 56 , 4566-4568. (33) Gosnell, A. B.; Gervasi, J. A.; Woods, M. D. K.; Stannett, V. Journal of Polymer Science Part C: Polymer Symposia 1969 , 22, 611-620. (34) Quirk, R. P.; Corona-Galvan, S. Macromolecules 2001 , 34 , 1192-1197. (35) Guerrero-Sanchez, C.; Lohmeijer, B. G. G.; Meier, M. A. R.; Schubert, U. S. Macromolecules 2005 , 38 , 10388-10396. (36) Ott, C.; Pavlov, G. M.; Guerrero-Sanchez, C.; Schubert, U. S. Journal of Polymer Science Part a-Polymer Chemistry 2009 , 47 , 3691-3701. (37) Stepto, R. F. T. Pure and Applied Chemistry 2009 , 81 , 351-353. (38) Ndoni, S.; Papadakis, C. M.; Bates, F. S.; Almdal, K. Review of Scientific Instruments 1995, 66 , 1090-1095. (39) Fetters, L. J.; Firer, E. M. Polymer 1977 , 18 , 306-307. (40) Quirk, R. P.; Chen, W. C. Journal of Polymer Science Part a-Polymer Chemistry 1984 , 22 , 2993-3000. (41) Morton, M.; Fetters, L. J. Journal of Polymer Science Part a-General Papers 1964 , 2, 3311-&. (42) Constable, E. C.; Baum, G.; Bill, E.; Dyson, R.; van Eldik, R.; Fenske, D.; Kaderli, S.; Morris, D.; Neubrand, A.; Neuburger, M.; Smith, D. R.; Wieghardt, K.; Zehnder, M.; Zuberbuhler, A. D. Chemistry-a European Journal 1999 , 5, 498-508. (43) Bertini, I.; Luchinat, C.; Aime, S. Coordination Chemistry Reviews 1996 , 150 , R7-+. (44) Fragouli, P. G.; Iatrou, H.; Hadjichristidis, N. Journal of Polymer Science Part A: Polymer Chemistry 2004 , 42 , 514-519. (45) Mavroudis, A.; Avgeropoulos, A.; Hadjichristidis, N.; Thomas, E. L.; Lohse, D. J. Chemistry of Materials 2003 , 15 , 1976-1983. (46) Worsfold, D. J.; Bywater, S. Canadian Journal of Chemistry 1964 , 42 , 2884-2892. (47) Jenner, G. Journal of Macromolecular Science: Part A - Chemistry 1975 , 9, 83-93.

61 CHAPTER 3

SUBSTITUENT EFFECTS ON THE STABILITIES OF POLYMERIC

AND SMALL MOLECULE bis-TERPYRIDINE COMPLEXES∗∗∗

3.1 Chapter Overview

Metallo-supramolecular assemblies of polymeric, particulate, or small-molecule

building blocks based on 2,2':6',2’’-terpyridine most commonly rely on chemical

substitution of this ligand at the 4’ position. In Part 3.1 of this chapter, we investigate the

effects that this modification to the ligand has on the kinetic stabilities of bis transition

metal complexes, and therefore the lifetime of supramolecular bonds that it forms. The

decay of aqueous cobalt(II) and iron(II) complexes in the presence of excesses of

competing metal ions is studied by ultraviolet visible (UV-vis) absorption spectroscopy,

and rate constants for the decay of amino-, ether-, and non-substituted ligands are

determined. Remarkably, for cobalt(II), amino substituents lead to rapid oxidation in air

to kinetically inert cobalt(III) complexes, while ether substituents yield decay constants

three- to six-fold larger than unsubstituted complexes. Ether linkages apparently weaken

cobalt complexes by inductive withdrawal of electrons, while amino substituents donate

electron density, thus lowering the oxidation potential of the complexes. For iron(II),

both ether- and amino-linked complexes decay at least an order of magnitude faster than

unsubstituted complexes. Our results highlight the importance of considering linking

chemistry when designing dynamic supramolecular assemblies based on terpyridine

62 ligands, as it yields significant, and occasionally dramatic, changes in the kinetic stabilities of these complexe

In Part 3.2 of this chapter, the kinetic stability of terpyridine-functionalized poly(ethylene oxide) (Mn = 5000 g/mole) with thioether and sulfonyl linkages is measured. The thioether linked species was produced by the reaction of terpyridine-

4′(1 ′H)-thione with mesyl-PEO under basic conditions. The sulfonyl-linked species was produced by oxidation of the thioether-linked species. It was found that iron complexes of the thioether species decayed at a rate comparable to that of its amino homologues, while the sulfonyl-linked species decay approximately three-fold more rapidly. For complexes of cobalt, both complexes decayed more rapidly than their ether- and amino- linked counterparts, with less difference between the two sulfur species.

PART 3.1

SUBSTITUENT EFFECTS OF ETHER AND AMINO LINKAGES ON KINETIC STABILITY OF BIS TERPYRIDINE COMPLEXES

3.1.1 Introduction

Metal-ligand complexes provide versatile platforms for the construction of

dynamic supramolecular structures from small molecule, polymer, and particle-based

building blocks. 1-4 It is well established that variations in the metal used to form complexes with these ligands yield changes by many orders of magnitude in both the kinetic rate constants for formation and dissociation of complexes 5,6 and the thermodynamic binding constants. 7 These effects are critical to informing the metal- ligand combination selected, as they determine the properties of the resulting supramolecular structures. 8-10 Another potentially important, although infrequently considered, factor is that the synthetic pathway used to attach the ligands to the building blocks necessarily introduces substituent groups, that may influence ligand behavior. For example, the rate of complex formation is known to be affected by the electron donation/withdrawing character of the substitutent groups on ligands including aliphatic amines 11 and phosphines. 12 Thermodynamic binding constants have been shown to be affected dramatically by electronic donation/withdrawal as well. 13 While several studies on the effects of ligand substituents on complex dissociation have been carried out, 14,15 the electronic effects of the substituents have generally been found to be less important than their steric bulk. The most relevant study on electronic substituent effects for pyridyl ligands 16 showed that electron donation by methyl substituents on bipyridine slowed the decay of nickel(II) complexes, while electron withdrawal by nitro and chloro groups led to more rapid decay.

The 2,2’:6’,2”-terpyridine (terpyridine) ligand, which forms bis complexes with a wide variety of first-row transition metals in the second oxidation state, such as iron(II), 17,18 cobalt(II) 18,19 and nickel(II) 20 has proven to be of particular utility for the preparation of reversibly-associating polymer-based assemblies including block copolymers 21 ,22 crosslinked networks, 23 supramolecular step-growth polymers, 24-27 and surface coatings.28 The rates of reaction and dissociation of several of these metals with terpyridine has been well studied, 6,20,29 and there have been a few studies on the thermodynamics of complex formation. 30,31 However, to the best of our knowledge, there have been no studies on the effect of substituent groups on these parameters.

To attach terpyridine ligands to polymer backbones, two main synthetic strategies have been adopted. In a few cases, the nucleophilic aromatic substitution of living anionic polymers to 4’-chloro-2,2’:6’,2”-terpyridine 32,33 (chloro-terpyridine) or terpyridine 34 has been employed, yielding methylene linkages between the polymer chain and the end-capping terpyridine ligand. More commonly, however, an S NAr reaction of an alkoxide with chloro-terpyridine is employed, yielding either end-functionalized 35-37 or side-chain functionalized 30,38,39 polymers with an ether linkage to terpyridine at the 4’ position. As ether groups can both donate electron density due to resonance effects and withdraw electron density due to inductive effects 40 , this chemical substitution may be expected to significantly perturb the properties of terpyridine as a ligand. There are also several examples of 4’-amino substituted terpyridines in the literature, 41-44 providing an interesting opportunity for a comparison of substituent effects, as amino groups are more electron donating than ether groups. 45 However, while it has been demonstrated that substituents that alter the electron density within the terpyridine pi system affect the

magnetic susceptabilities 46 , redox potentials 46,47 , and photophysical properties 48 of complexes, their effects on complex stability have received little attention.

In the current study, we determined the kinetic stabilities of aqueous bis - complexes for a series of substituted terpyridine ligands, summarized in Scheme 3.1.1, modified at the 4’ position using both ether and amine linking chemistries. For cobalt(II) complexes, ether substituents yielded a significant (three- to six-fold) increase in the dissociation rate constant, compared to terpyridine. Surprisingly, amino-functionalized ligands were found to induce rapid oxidation under aerobic conditions to cobalt(III) complexes, reducing the dissociation rate constants by at least four to five orders of magnitude, to the point that they can no longer be measured using the same methods.

This ligand-facilitated oxidation of cobalt(II) to cobalt(III) may also be of interest in the context of efforts to use cobalt complexes as catalysts for photo-oxidation of water for solar energy storage.49,50 For iron(II) complexes, both ether and amine linking chemistries increased the rate of decay by at least an order of magnitude. Ligands functionalized with poly(ethylene oxide) were compared to small molecule analogs for both cobalt and iron. While the behaviors were found to be qualitatively similar, modest, though easily measurable, changes in dissociation kinetics were found to result from the presence of the polymeric substituents.

(1) R= H O (5) R= N H (2) R = O (3) R= O (6) R = N R O

O N (4a), (4b) R = O O (7) R= N N N Scheme 3.1. Ligands used in Decomposition Studies

3.1.2 Experimental

• Materials

N-methylpropylamine (96%), propylamine (98%), 2-(methylamino)ethanol

(99%), 2-(2-methoxyethoxy)ethanol (99%), iron (II) chloride tetrahydrate (99+%), nickel

(II) chloride hexahydrate (99.9%), calcium hydride (93%), deuterochloroform (99.8% D), deuterium oxide (99.8% D), and D 6-dimethylsulfoxide (99.9% D) were obtained from

Acros Organics. Poly(ethylene oxide) monomethyl ether (M n = 5,000, PDI = 1.04) and

cobalt (II) chloride hexahydrate ( ≥ 98%) were obtained from Fluka Chemical. Water

(HPLC grade), dimethylsulfoxide (BioReagent Grade), chloroform (ACS Grade), sodium

sulfate (ACS Grade), acetone (ACS Grade), sodium hydroxide (ACS Reagent Grade),

sodium carbonate (ACS Reagent Grade), and EDTA disodium salt dehydrate (Certified

ACS grade) were obtained from Fisher Scientific. Ethyl iodide was purchased from

Sigma Aldrich, ethanol was obtained from Alco-AAPER (95%) and 2,2’:6’,2”-

terpyridine (1) (99%) was obtained from Strem Chemical. All chemicals were used

without further purification unless otherwise noted.

• Characterization

All 1H NMR spectra were recorded on a Bruker DPX 300 MHz spectrometer,

except for variable temperature scans, which were measured on a Bruker Avance 400

spectrometer, in deuterochloroform or deuterium oxide, and referenced to

tetramethylsilane.

UV-Vis spectra were recorded on a Hitachi U-3010 spectrophotometer, with

quartz cuvettes (Buck Scientific) for wavelength scans. Kinetic measurements were

performed using HPLC water, with plastic cuvettes (Fisher Scientific) for iron

complexes, or a homemade rapid mixing unit based on a Buck Scientific flow cell for

cobalt complexes. The decays of cobalt complexes were measured following the

procedure of Wilkins et al., 29 while those of iron complexes were characterized using a

modification to this procedure. The cobalt and iron complexes were decayed by adding a

20 – 100-fold excess of iron(II) chloride or nickel(II) chloride, respectively, and tracked

by monitoring the metal-ligand charge transfer (MLCT) band of the iron(II) bis - complexes for respective periods of 3600 or 70000 s. This MLCT band (peak intensities

at 553, 557, and 570 nm for non-substituted, ether-bridged, and amino-bridged

complexes, respectively) was chosen since it falls in a region of the spectrum

unpopulated by competing signals. Experiments were conducted at multiple

concentrations between 0.1 and 0.5 mM to confirm that measured decay rates were first

order in concentration of the bis complexes.

Cyclic voltometry was conducted using a BASi Epsilon electrochemical

workstation with platinum working and auxiliary electrodes and an Ag/AgCl reference.

Measurements were conducted on aqueous solutions containing 100 mM of lithium

bromide as an electrolyte and 10 mM of bis complexes.

• Synthetic Procedures

4’-chloro-2,2’:6’,2”-terpyridine, and 4‘-ethoxy-2,2‘:6‘,2‘‘-terpyridine (2) were

synthesized according to literature procedures 42,51,52 , as was mesylate-functionalized

PEO 53 . The synthesis of all other substituted terpyridines and terpyridine complexes used

in this study are described as follows.

The nitrogen-bridged terpyridine ligands have also been produced using slight

variations on existing procedures. While 4′-(N-2-hydroxyethyl-N-methylamino)-

2,2 ′:6 ′,2 ′′ -terpyridine has been previously synthesized, the literature technique involved producing the iron 42 or zinc 54 complex of 4’-chloro-2,2’:6’,2”-terpyridine. The presence of the metal activated the 4’ positions of the 4’-chloro-2,2’:6’,2”-terpyridine to attack by

N-methylethanolamine, after which the complex was broken with a strong base and then purified. As in some previous studies 41,55-57 , it was found that simply reacting the requisite amines with 4’-chloro-2,2’:6’,2”-terpyridine at high temperature in a pressure vessel followed by precipitation resulted in successful synthesis of the required ligands.

To improve the yield of the end-functionalization of poly(ethylene oxide), 4 ′-(N-

2-hydroxyethyl-N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine was reacted with mesyl-PEO instead of 4’ -(N-2-hydroxyethylamino)-2,2':6',2''-terpyridine. It was found that the addition of the methyl group on the nitrogen atom improved the end-functionality of the product to 95-97% using the same reaction conditions that had yielded only approximately 70% previously. Its is postulated that the similarity of pKa of the alcohol moiety and that of the secondary pyridyl amine 58,59 were responsible for non-selective

deprotonation and the resultant poor yields. As discussed in subsequent sections, the

presence of the methyl group had little effect on the decays of both iron and cobalt

complexes. A diagram of selected ligands used in this study is included in Scheme 3.1.1.

4′-(N-2-Hydroxyethyl-N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine To a 48 mL pressure

vessel were added 0.304 g of 4’-chloro-2,2’:6’,2”-terpyridine, 6 mL of N- methylethanolamine and a stir bar. The dispersion of 4’-chloro-2,2’:6’,2”-terpyridine in

N-methyl ethanolamine was sparged for 30 min with nitrogen before the vessel was

quickly sealed. The vessel was wrapped in aluminum foil to help maintain a uniform

temperature. The vessel was then placed in a sand bath that was held at 250 ºC, as

measured from the bottom of the bath. The reaction was allowed to stir at this

temperature for 48 h. After the vessel was moved from heat and allowed to cool, the

contents were observed to have changed from dispersion to a mauve liquid. This liquid

was added slowly to 125 mL of rapidly stirred water, resulting in the precipitation of the

pure product. The white powder obtained was dried in vacuo overnight to yield 0.286 g

1 (82%) of product. H NMR (CDCl 3): δ = 8.64 (d, 2H), 8.61 (d, 2H), 7.84 (d, 2H), 7.80

(d, 2H), 7.30 (d, 2H), 3.94 (t, 2H), 3.72 (t, 2H), 3.52, (s, broad, hydroxyl group), 3.18 (s,

13 3H). C NMR (CDCl 3): δ = : 156.94, 155.80, 155.58, 148.69, 136.86, 123.50, 121.57,

103.90, 59.72, 53.77, 38.51. ESI-MS m/z = 307.2 (M+H)

4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''-terpyridine (3). An oven-dried 50 mL

round bottom flask was fitted with a septum and purged with nitrogen gas for 15 min.

During the purging process, 0.13 g of powdered KOH was added to the flask. After 15

min, 10 mL of dry DMSO was added to the flask through the septa. The liquid was

sparged with nitrogen for 10 min while the flask was lowered into a 70 °C oil bath and 1

mL of diethylene glycol methyl ether was added via syringe. The mixture was stirred

under heat for 30 min before 0.2175 g of 4’-chloro-2,2’:6’,2”-terpyridine was added. The mixture was stirred at the same temperature for another 10 h before removal from the oil bath. The DMSO was then reduced to a volume of 1-2 mL using short-path distillation

(80 mTorr @ 70° C). The product was then precipitated by addition of the resulting solution to 50 mL of rapidly stirred water in a dropwise manner. The pure product was then filtered and dried, giving a fluffy white powder. Yield: 0.206 g (70.8%). 1H NMR

(CDCl 3): δ = 8.74 (d, 2H), 8.69 (d, 2H), 8.13 (s, 2H), 7.94 (t, 2H), 7.40 (t, 2H), 4.48 (t,

13 2H), 3.95 (t, 2H), 3.76 (m, 2H), 3.59 (m, 2H), 3.39 (s, 3H). C NMR (CDCl 3): δ =

166.99, 157.12, 156.07, 149.06, 136.81, 123.85, 121.35, 107.45, 72.00 70.92, 69.49,

67.82, 59.13 FAB-MS m/z = 352 (M+H)

α-terpyridyl-ω-methoxy-poly(ethylene oxide) (4a). In an oven dried 50 mL round bottom flask 5.51 g of poly(ethylene oxide) methyl ether (MW = 5.0 kg/mol, PDI = 1.03) was placed along with 0.22 g of powdered KOH. The flask was fitted with a septum and purged with nitrogen gas for 15 min, after which 8 mL of anhydrous DMSO was added to the flask via syringe. After adding the DMSO, the mixture was sparged with nitrogen gas for 15 min. The flask was then placed in an oil bath that was heated to 70º C. The mixture was heated for 30 minutes before 1 g of 4’-chloro-2,2’;6’,2”-terpyridine was quickly added. The mixture immediately turned a bright pinkish red, which gradually darkened to a brownish color. After stirring for 48 h, the reaction mixture was transferred into 400 mL of saturated EDTA solution, and the product was extracted into chloroform (3 x 300mL). The organic layers were combined and dried with sodium sulfate before being concentrated down to approximately 10 mL in vacuo . The product was precipitated by dropwise addition into 600 mL of diethyl ether and filtered, followed by drying of the product in vacuo . Yield: 4.67 g (84.9%). The 1H spectra of this species has been published elsewhere. 8 End functionality by NMR: 95-97% An identical procedure was used to synthesize terpyridine functionalized with 1.1 kg/mol PEO (4b) .

4′-(N-propylamino)-2,2 ′:6 ′,2 ′′ -terpyridine (5). To a 48 mL pressure vessel were added

0.300 g of 4’-chloro-2,2’:6’,2”-terpyridine, 4 mL of propylamine and a stir bar. The vessel was wrapped in aluminum foil and placed in a sand bath maintained at 220 ºC, as

measured from the bottom of the bath. The reaction was allowed to stir at this

temperature for 72 h before the mauve reaction mixture added to 75 mL of rapidly

stirring water, thus precipitating the pure product. The product was filtered and dried in

1 vacuo overnight. Yield: 0.2554 g (78%). H NMR (CDCl 3): δ = 8.66 (d, 2H), 8.59(d,

2H), 7.83(d, 2H), 7.66 (d, 2H), 7.30 (d, 2H), 4.38 (s, 1H, N-H), 3.33(t, 2H), 1.72 (t, 2H),

13 1.03 (t, 3H). C NMR (CDCl 3): δ = 156.84, 155.85, 155.47, 148.85, 136.74, 123.48,

121.35, 104.64, 44.75, 22.62, 11.53. ESI-MS m/z = 291.2 (M+H)

4′-(N-n-propyl -N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine (6). To a 48 mL pressure vessel were added 0.289 g of 4’-chloro-2,2’:6’,2”-terpyridine, 2 mL of N- methylpropylamine and a stir bar. The vessel was wrapped in aluminum foil and placed in a sand bath that was maintained at 230 º C, as measured from the bottom of the bath.

The reaction was allowed to stir at this temperature for 48 h. The contents of the vessel were transferred to a 50 mL round bottom flask, where the excess N-methylpropylamine was removed in vacuo , leaving a light brown tar. This material was taken up into 15 mL of chloroform and washed with water (2 x 15 mL). The organic phase was dried over sodium sulfate and the chloroform was removed in vacuo. The resulting light brown tacky solid was determined to be suitably pure by NMR and used without further

1 purification. H NMR (CDCl 3): δ = 8.68 (d, 2H), 8.61 (d, 2H), 7.83(d, 2H), 7.75(d,

2H), 3.53(t, 2H), 3.18 (s, 3H), 1.73 (t, 2H), 0.98 (t, 3H). 13 C: 157.24, 155.71, 155.51,

136.65, 123.32, 121.41, 103.63, 53.35, 38.09, 20.32, 11.48. FAB-MS m/z = 305 (M+H)

Poly(ethylene oxide) functionalized with 4 ′-(N-2-Hydroxyethyl-N-methylamino)-

2,2 ′:6 ′,2 ′′ -terpyridine (7). An oven-dried 50 mL round-bottomed flask was charged with

0.095 g of 4′-(N-2-Hydroxyethyl-N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine, 0.0075 g of

sodium hydride and an egg-shaped stir bar and fitted with a septum. After 10 min of

purging with nitrogen, 5 mL of anhydrous THF was added to the flask. The resultant

milky suspension was stirred while sparging with nitrogen for 10 min. After this time,

0.308 g of mesyl-functionalized PEO was added, and the mixture was heated to 60 ºC for

12 h. The mixture was next added to 75 mL of chloroform, and washed with a saturated

EDTA solution (2 x 50 mL). The organic phase was dried over sodium sulfate and

reduced to approximately 5 mL volume under reduced pressure. The polymer was then

precipitated by adding this solution dropwise to 100 mL of rapidly stirred ethyl ether.

The product was filtered and redissolved in chloroform (5 mL) and precipitated once

more in 100 mL of ethyl ether. The product was then filtered and dried in vacuo at room

1 temperate overnight. Yield: 0.115 g (37.3%) H NMR (CDCl 3): δ = 8.68 (d 2H), 8.63

(d 2H), 7.86 (d, 2H), 7.81(d 2H), 7.32(d, 2H), 3.90-3.40 (m, 454H polymer backbone),

3.25 (s 3H N-Methyl) End functionality by NMR: 93-95%

Iron(II) and Cobalt(II) Complexes of 2,2 ′:6 ′,2 ′′ -terpyridine. Both (1) 2FeCl 2 and

60 (1) 2CoCl 2 were prepared according to previously described methods.

Preparation of Complexes of (2) with Cobalt. Ligand (2) (0.2451 g) was dissolved in 3

mL of a 2:1 ethanol/acetone solution and added to a 1 mL solution containing 0.119 g of

cobalt(II) chloride hexahydrate. The brown solution was stirred in air for 40 h, after

which it was noticed that sedimentation had occurred. Water (2 mL) was added and the

solution was heated until it was no longer turbid, at which point any remaining insoluble

particles were removed via syringe filter. After this, the product was allowed to

recrystallize from solution, giving brown granular crystals. Yield: 0.1811g (57.3%). 1H

NMR (D 2O): δ = 107.24, 66.04, 33.82, 13.13, 8.49, 8.23, 8.12, 8.00, 7.78, 7.47, 7.32,

7.11, 6.21, 1.57. ESI-MS m/z = 648.0 (M-Cl)

Preparation of Complexes of (3) with Cobalt. Ligand (3) (0.0506 g) and 0.0165 g of

cobalt(II) chloride hexahydrate were dissolved in 2.5 mL of ethanol and stirred while

being heated to approximately 60 ° C for 45 min. The solution was allowed to cool to room temperature before being precipitated into rapidly stirring ethyl ether (50 mL). The

1 product was filtered and dried in vacuo . Yield: 0.0233 g (41.8%). H NMR (D 2O): δ =

108.32, 66.35, 65.63, 34.26, 8.49, 7.23, 6.67, 4.61, 4.43, 2.11. m/z = 360.9 (M – Cl -

2+ CH 3-O-(CH 2)2-OH + H)

Preparation of Complexes of (3) with Iron. Ligand (6) (0.044 g) and 0.0116 g of iron(II) chloride tetrahydrate were dissolved in 2.5 mL of ethanol in a small vial. The solution was heated to 60 °C for 45 min. The solution was then allowed to cool to room

temperature before being precipitated into rapidly stirring ethyl ether (50 mL). The

1 product was filtered and dried in vacuo . Yield: 0.0275 g (52.9%). H NMR (D 2O): δ =

8.49 (s, 2H), 8.38 (d, 2H), 7.79 (t, 2H), 7.16 (d, 2H), 6.99 (t, 2H), 4.15 (t, 2H), 3.87 (m,

2H), 3.72 (m, 2H), 3.39 (s, 3H), 2.17 (s, 2H). ESI-MS m/z = 921.0 (M – Cl + FeCl 2),

793.1 (M – Cl), 442.0 (M -(6) –Cl),

Preparation of Complexes of (4a) and (4b) with Iron and Cobalt

An appropriate amount of either iron(II) chloride tetrahydate or cobalt (II) chloride

hexahydrate was weighed into a vial with 2 equivalents of (4) . The appropriate amount

of water (D 2O for NMR studies) was added to achieve the desired concentration, and the purple (iron) or brown (cobalt) mixture was stirred for at least 1 h before further use.

1 (4a) 2FeCl 2 H NMR (D 2O): δ = 8.43 (s, 4H), 8.33(s, 4H), 7.72(s,4H), 7.07 (s, 4H), 6.94

(s,4H), 3.90-3.0 (m, 908H, polymer backbone)

1 (4a) 2CoCl 2 H NMR (D 2O): δ = 107.96, 66.64, 34.29, 13.39, 8.65, 8.20, 7.41, 7.23, 6.75,

5.77, 4.94, 4.35, 4.5-2.5 (m, 908H, polymer backbone)

The complex (4b) 2CoCl 2 was prepared in the same manner as complex (7a) .

Preparation of Complexes of (5) with Cobalt. Ligand (5) (0.0578 g) and 0.0245 g of cobalt(II) chloride hexahydrate were added to 4 mL of water in a small vial. The mixture was stirred for 1 h and filtered into a second vial. Compound (11) was recrystallized from said solution, yielding granular brown crystals. Yield: 0.053 g (74.9%). 1H NMR

(D 2O): δ = 8.32 (d, 2H), 7.96 (d, 2H), 7.31 (d, 2H), 3.53 (t, 2H), 1.76 (m, 2H), 1.01 (s,

3H). ESI-MS m/z = 674.1 (M – Cl – H), 637.2 (M- 2Cl – 2H).

Preparation of Complexes of (5) with Iron. In a 20mL vial, 0.1057 g of (5) and 0.0366 g of iron(II) chloride tetrahydrate were mixed. A 1:1 v/v water/ethanol solution was added (2 mL), and the purple solution was stirred under mild heating for 1 h. During this time sedimentation of the product occurred. After cooling to room temperate, the supernatant was decanted, and the product was dried in vacuo. Yield: 0.083 g (58.3%).

1 H NMR (D 2O): δ = 8.17 (d, 4H), 7.93 (s, 4H), 7.65 (t, 4H), 7.16 (d, 4H), 6.88 (t, 4H),

3.48 (t, 4H), 1.78 (m, 4H), 1.03 (t, 4H). ESI-MS m/z = 671.0 (M – Cl), 603 (M – Cl – 2Pr

+ OH -), 381.0 (M – CL – (2) ), 318 (M – 2Cl) 2+

Preparation of Complexes of (6) with Cobalt. In a 20mL vial, 0.062 g of (6) and

0.0242 g of cobalt(II) chloride hexahydrate were dissolved in 2 mL of 1:1 v/v water/ethanol. The mixture was stirred for 1 h and the product was recrystallized from

1 the reaction mixture, giving square brown crystals. H NMR (D 2O): δ = 8.49 (d, 4H),

8.12 (t, 4H), 8.09 (s, 4H), 7.43 (d, 4H), 7.33 (t, 4H), 3.89 (t, 4H), 3.53 (s, 6H), 1.91 (m,

4H), 1.09 (t, 6H) ESI-MS m/z = 702.1 (M – Cl), 398.0 (M – Cl – (3) ), 334.0 (M – 2Cl) 2+ ,

319.0 (M – 2Cl – 2Me) 2+ Yield: 0.0254 g (30.0%).

Preparation of Complexes of (6) with Iron. In a 20mL vial, 0.052 g of (6) was dissolved in 0.5 mL of acetone, to which 0.019 g of iron(II) chloride tetrahydrate in 0.5 mL water was added. The purple solution was stirred for 30 min, and the product was recrystallized from the reaction mixture, giving granular red crystals. Yield: 0.0233 g

1 (37.0%) H NMR (D 2O): δ = 8.23 (d, 4H), 7.98 (s, 4H), 7.64 (t, 4H), 7.12 (d, 4H), 6.88

(t, 4H), 3.74 (t, 4H), 3.36 (s, 6H), 1.81 (m, 4H), 1.00 (t, 6H). ESI-MS m/z = 699.1 (M –

Cl), 332.1 (M – 2Cl) 2+

Preparation of Complexes of (7) with Iron and Cobalt . Complexes of (7) were prepared using the same technique as those used for (4a) , except that the cobalt complexes were left in air for at least 1 d to ensure complete oxidation of cobalt(II) to cobalt(III).

1 (7) 2FeCl 2 (18) H NMR (D 2O): δ = 8.43 (d, 4H), 8.21 (s, 4H), 7.80 (t, 4H), 7.23 (d,4H),

7.05 (t, 4H), 4.09 (4H), 4.01(s, 6H), 3.90-3.25 (m, 908H, polymer backbone)

1 (7) 2CoCl 2 (7b) H NMR (D 2O): δ = 8.49 (d, 4H), 8.19 (t, 4H), 8.10 (s, 4H), 7.41 (d, 4H),

7.34 (t, 4H), 4.20 – 3.10 (m, 908H, polymer backbone)

3.1.3 Results and Discussion

3.1.3.1 Kinetic stabilities of bis-terpyridine Co(II) complexes

Formation of bis terpyridine complexes of transition metals occurs by a two-step,

reversible addition of ligands, as summarized in Scheme 3.1.2. The forward and

backward rate constants for formation of the mono complex are denoted as k1 and k-1,

respectively, and those for the bis complex are k2 and k-2. In the current study, we are concerned with k-2, which determines the kinetic stability of the supramolecular bond between two ligands.

N N N N N N N k1 k2 N + M(II) N M(II) N M(II) N k-1 k-2 N N N N

Scheme 3.2 Reaction of terpyridine ligands with a bivalent metal salt to form a mono and subsequently a bis complex.

The decay of bis cobalt complexes was measured using the method of Wilkins, et

al. 6 Specifically, a 20-100 fold excess of iron(II) chloride was added to a solution of

cobalt complexes in water, and the growth of the metal ligand charge transfer (MLCT)

band of the resulting iron bis complexes was measured, as shown in Figure 3.1.1 for three

example complexes. To analyze these data, we make two assumptions about the kinetics

in the initial portion of each curve, based on literature values of the rate constants for

terpyridine (1).6,60 First, we assume that all ligands freed by decay of cobalt complexes

rapidly form part of a bis iron complex, and thus the iron MLCT signal is proportional to

the total amount of terpyridine liberated. This assumption is justified for terpyridine,

since the slowest step in formation of iron(II) bis complexes is disproportionation of

- iron(II) mono complexes formed initially; this process is governed by k-1 for iron (6.3*10

3 -1 29 -3 -1 29,60 s ) , which is ~ 4 times larger than k-2 for cobalt (1.1 – 1.4*10 s ) . Second, we

assume that the initial kinetics are dominated by decay of the bis , rather than mono ,

cobalt complexes, since k-1 for cobalt is roughly an order of magnitude smaller than k-2 (k-

-4 -1 -3 -1 29 1= 1*10 s , k-2 = 1.1*10 s ) . As shown in Figure 3.1.1 for the unsubstituted ligand

complex ( 1)2CoCl 2, a single exponential curve could be used to accurately fit the initial

-3 -1 region of the kinetic data (0 - 400 s), thus providing a value of k-2 = 1.1*10 s , which is in excellent agreement with the literature values determined by complex decay and ligand exchange (1.1*10 -3 s-1 and 1.4*10 -3 s-1, respectively). 29,60 The clear deviation from single- exponential behavior at later times in Figure 3.1.1 reflects the contributions from the decay of the cobalt mono complex ( k-1).

Figure 3.1 Decay of selected bis -terpyridine cobalt complexes, as determined by monitoring the growth in absorbance of the MLCT band formed upon reaction with excess iron. The unsubstituted ligand (1) and ether-substituted ligand (4a) yield decay curves that are well described by a single exponential over the first 400 s (solid red line over the fitted range, extended as a dotted red line). At later times, substantial departures from single exponential behavior are apparent. The amine- substituted ligand (7) yielded negligible decay over the same time period. Rate constants extracted from these data are summarized in Table 3.1.1.

Using the same methodology, we determined values of k-2 for the cobalt complexes of each of the different substituted ligands, as summarized in Table 3.1.1.

Notably, the largest values measured approach the literature value of k-1 for iron 79

complexes of terpyridine, calling into question the first assumption described above.

Thus, we provide the caveat that, depending on how k-1 for iron is shifted by substituent

effects, the larger values of k-2 for cobalt reported in Table 3.1.1 may represent underestimates.

Table 3.1 Decay Rates of Cobalt bis -Teryridine Complexes

3 -1 Ligand Complex k-2*10 s

2,2 ′:6 ′,2 ′′ -terpyridine (1) 2Co(II) 1.10 ± 0.05

4‘-Ethoxy-2,2‘:6‘,2‘‘-terpyridine (2) 2Co(II) 5.5 ± 0.1

4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''-terpyridine (3) 2Co(II) 4.77 ± 0.05

5k PEO-terpyridine (4a) 2Co(II) 1.50 ± 0.04

1.1k PEO-terpyridine (4b) 2Co(II) 3.96 ± 0.03 -5 4′-(N-n-propylamino)-2,2 ′:6 ′,2 ′′ -terpyridine (5) 2Co(III) < 7 * 10 -5 4′-(N-n-propyl -N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine (6) 2Co(III) < 1.3 * 10 -5 5k PEO-N(methyl)-terpyridine (7) 2Co(III) < 4.0 * 10

5k PEO-N(methyl)-terpyridine (anaerobic, cobalt (II)) (7) 2Co(II) 1.6 ± 0.5

We first consider cobalt complexes of the ether-substituted ligands (2) – (4) .

Interestingly, the ethoxy-substituted ligand (2) yields a decay constant of 5.5*10 -3 s-1,

five-fold larger than the non-substituted complex (1) . As the ether linkage represents the

most common means for attaching terpyridine to polymer chains, we also considered two

different end-functionalized polyethylene oxide (PEO) ligands (4a) and (4b) with

respective molecular weights of 5 kg/mol and 1.1 kg/mol. Surprisingly, (4a) 2CoCl 2 gave

-3 -1 a value of 1.5*10 s , only about 50% more rapid than terpyridine, while (4b) 2CoCl 2

showed almost four-fold faster decay (4.0*10 -3 s-1). Notably, the diethylene glycol

functionalized ligand (3) gave a value (4.77*10 -3 s-1) similar to the ethoxy-functionalized ligand, but three-times larger than the polymer-bound ligand (4a) . This suggests that the slower decays found for PEO-functionalized ligands with ether linkages, compared to their small molecule analogs, do not reflect a difference in electronic effects, but instead must arise due to the polymeric nature of the substituents. We initially suspected that the

PEO chains slowed decomposition as compared to complexes (2) 2CoCl 2 and (3) 2CoCl 2

either by coordinating to cobalt, thereby enhancing the stability of the bis complexes, or

by increasing the viscosity locally in the vicinity of the complexes. However, when

complex (2) 2CoCl 2 was decomposed in the presence of an equimolar amount of 5k PEO,

no change in k-2 was observed, and even a 5k-PEO concentration of 66 mg/mL (~80 % of the overlap concentration 61,62 ) yielded a minor increase in decay rate to 6.4*10 -3 s-1. We note that excluded volume effects could be important as well, but these are expected to lead to faster decay of polymer-functionalized ligands, i.e. , due to the tension applied to

the chain, whereas in this case we find that the higher molecular weight ligand leads to

significantly slower decay. Further study is required to understand this effect.

Perhaps the most interesting aspect of the decay of cobalt complexes is that the

amino-substituted terpyridine complexes are almost completely kinetically inert . Unlike the non-functionalized and ether-bridged cobalt complexes, there was no visual indication of decomposition over the span of one week. While small changes in the UV- vis spectrum were observed, the absorbance at 570 nm actually decreased slightly over time, as can be seen in Figure 3.1.1 for a cobalt complex of ligand ( 7), most likely due to formation of iron oxide from iron(II) chloride. In Table 3.1.1, we list upper bounds of k-2 for bis cobalt complexes of ligands (5) - (7) , corresponding to the fastest possible decays

that could be obscured by this decrease in absorbance. These values are of order 10 -8 s-1,

corresponding to at least a four to five order of magnitude reduction in decay rate

compared to the ether- and non-functionalized terpyridine ligands.

The remarkable stability of the amino-linked compounds reflects a ligand-

facilitated oxidation of the cobalt center to Co(III) by atmospheric oxygen. This behavior

is common for cobalt complexes of ammonia and aliphatic amines, 63 but the only

example of which we are aware for terpyridine ligands involves 4,4’,4”-

tris(diethylamino)terpyridine. 47 However, there are also previous examples in which

cobalt(II) complexes of terpyridine ligands were intentionally converted to cobalt(III) by

chemical oxidation to increase complex stability. 22,64

To prove our supposition that oxidation to cobalt(III) is responsible for the

enhanced stability of amine-linked complexes, we prepared an anaerobic solution of

complex (7) 2CoCl 2 by sparging the water, and purging the mixing apparatus and all

containers, with nitrogen, and then measuring decomposition within 10 min of complex

-3 preparation. Under these conditions, the bis cobalt complex of (7) yielded k-2 = 1.6 *10 s-1, a value very similar to that of its ether-linked counterpart (4a) , and just within the uncertainty (95% confidence interval) of that for terpyridine. These results indicate that the excess electron density provided by the 4’ amino substituent inherently has little effect on the bis complex stability relative to non-substituted terpyridine, with perhaps a

slight destabilization in similar fashion to the ether substituents, though the large

uncertainty in the anaerobic measurement precludes making this conclusion with a high

level of confidence. The main effect of the nitrogen-based substituents is that they

82 greatly facilitate oxidation of the complex to Co(III), thereby providing an apparently dramatic increase in stability.

The effects of cobalt oxidation on stability of the complexes are twofold. Apart from the increase in electrostatic attraction between the cobalt center and the ligand due to the change of oxidation state from (II) to (III), there is an additional gain in stability due to the change from a distorted d 7 to a truly octahedral d 6 complex. The Jahn-Teller distortion of cobalt(II) complexes of terpyridine is most likely the reason for their much decreased stability as compared to d6 terpyridine complexes, such as iron(II) and the kinetically inert Ru(II) complexes of terpyridine. 65,66 Further characterization of the resulting Co(III) complexes are provided in a subsequent section.

3.1.4.2 Kinetic stabilities of bis-terpyridine Fe(II) complexes

The importance of substituent chemistry on the lifetimes of cobalt complexes raises the question of whether similar effects apply to complexes of different metal ions.

Toward this end, we have also measured the values of k-2 for bis -complexes of iron(II) for most of the same ligands, with nickel(II) as the decomposing ion. Since k-1 is more than

29,60 four orders of magnitude larger than k-2 for iron(II) terpyridine complexes, decay of the bis complex is the rate limiting step and k-2 can be directly determined from a single exponential fit to the decrease of the MLCT band of the bis-terpyridine iron(II) complex.

While we are unaware of previous reports of using nickel(II) to decompose iron(II) terpyridine complexes, we found the process to be suitable due to the relative thermodynamic stability of the nickel mono complex. Specifically, a comparison

29 20.9 between the value of β12 = K1*K2 for the bis-terpyridine iron(II) complex (10 ) and

29 2 21.4 the square of K1 for the nickel(II) mono complex (K 1 = 10 ), reveals that in the

83 presence of excess nickel, the equilibrium state will be predominantly the nickel mono complex. As shown in Figure 3.1.2, a single exponential fit to the decay of the MLCT

-7 -1 signal yields a value of k-2 for (1) 2FeCl 2 of 1.5 * 10 s (at 21.5 ºC) which agrees well with the published value (1.66 * 10 -7 s-1 at 25 ºC) obtained by ligand exchange, 60 confirming the validity of our procedure. For the substituted complexes, this method is also suitable provided that the substituent groups do not dramatically stabilize the iron mono or bis complexes. Our studies show directly that the bis iron(II) complexes of substituted ligands are decomposed by excess nickel(II), and even were the iron mono complex to be stabilized, the smaller optical density of its MLCT band (less than one tenth that of the bis complex 60 ) means that the observed signal would still be dominated by bis complex decay.

Figure 3.2 Decay of selected bis -terpyridine iron complexes, as directly determined by monitoring the decrease in MLCT absorbance in the presence of excess nickel. For the unsubstituted, ether-funtionalized, and amine-functionalized ligands (1), (4a), and (7), respectively, data could be well described by a single exponential fit (solid red line). The resulting rate constants, summarized in Table 3.1.2, varied by as much as a factor of roughly 30.

The decay constants for iron(II) complexes of terpyridine ligands also show a

significant dependence on 4’- substitution, as summarized in Table 3.1.2. Interestingly,

ether linked ligands (3) and (4a) yield rate constants of 2.8 * 10 -6 s-1, and 5.0 * 10 -6 s-1, or

19 and 33 times faster than the unmodified terpyridine complex, respectively. For the amine-linked ligands (5) - (7) , the values range from 1.6 * 10 -6 s-1 to 2.2 * 10 -6 s-1, each at least an order of magnitude faster than for terpyridine. Thus, the iron(II) complexes show even more pronounced decreases in kinetic stability than cobalt(II) due to substitution at the 4’ position with either amine or ether functionalities. The values in Table 3.1.2 further demonstrate the complexity in the behavior of the polymer-bound terpyridine ligands. While for ether-linked species, the 5 kg/mol PEO sample (4a) showed an ~ 80% larger decay constant than the corresponding small molecule compound (3) , for the amine-linked species the 5 kg/mol PEO ligand (7) showed a roughly 25% smaller value than the corresponding ligands (5) and (6) . Though this effect needs further study, these

data suggest that the electronic effects due to 4’ substitution and those arising from the

presence of a polymeric ligand are not additive in a simple way.

Table 3.2 Decay of Iron Bis-Complexes of Terpyridine Ligands

6 -1 Ligand Complex k-2*10 s

2,2 ′:6 ′,2 ′′ -terpyridine (1) 2Fe(II) 0.150 ± 0.002 4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''-terpyridine (3) 2Fe(II) 2.79 ± 0.01 5k PEO-terpyridine (4a) 2Fe(II) 5.00 ± 0.03

4′-(N-n-propylamino)-2,2 ′:6 ′,2 ′′ -terpyridine (5) 2Fe(II) 2.10 ± 0.01 4′-(N-n-propyl -N-methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine 2.23 ± 0.01 (6) 2Fe(II)

5k PEO-N(methy)-terpyridine (7) 2Fe(II) 1.64 ± 0.01

3.1.3.3 Cyclic Voltammetry of Cobalt Complexes

To directly test the electron donating or withdrawing characteristics of the 4’

amine and ether substituents, we conducted cyclic voltammetry scans on bis cobalt complexes of ligands (1) , (3) , and (6) . Electron donating substituents should facilitate

oxidation to Co(III), thus leading to a more negative oxidation potential, and vice versa

for electron withdrawing substituents. As shown in Table 3.1.3, the nitrogen-linked

ligand (5) yields the most negative oxidation potential of -0.17 V for the cobalt(II)/cobalt(III) transition, with the corresponding reductive process at -0.29 V, consistent with our observations that amine substitution at the 4’ position leads to oxidation in air to cobalt(III) complexes. The ether-linked ligand (3) , in contrast, shows

oxidation at 1.10 V and reduction at 0.74 V. The bis-terpyridine complex shows a

cobalt(II)/cobalt(III) transition at 0.08 V with the corresponding reduction at -0.02 V, in

good agreement with previously published data. 67 Interestingly, a further reduction from

cobalt(II) to cobalt(I), previously reported for bis-terpyridine cobalt(II) complexes, 68,69

was found for terpyridine at -0.92 V (with the corresponding oxidation at -0.79 V), but

was not observed in either of the substituted complexes.

Table 3.3 Reduction and oxidation potentials of selected cobalt complexes for the Co(III)/Co(II) redox couple. Complex Oxidation Potential Reduction Potential

(3) 2Co(II) 1.10 V 0.74 V (1) 2Co(II) 0.08 V -0.02 V (5) 2Co(III) -0.17 V -0.29 V

For all three complexes, the difference in oxidation and reduction potentials Ep

was larger than the value of 59 mV expected for a reversible one-electron process.70

The oxygen-substituted complex had the largest Ep, of around 360 mV, which is a greater than for any other reported terpyridine complex of which we are aware. The current measured during oxidation was also greater than during reduction, denoting some degree of irreversibility. The amino complex showed a Ep of around 120 mV, while the unsubstituted complex showed a Ep of around 100 mV, both values being closer to those previously reported for bis -tridentate ligand complexes of Co(II). 71,72 For the amino- and unsubstituted complexes, the magnitude of the oxidation and reduction currents were approximately equal.

Previous studies of d 6 systems have shown that properties such as the redox

potentials depend on the Hammet constant of the substituent group, which places the

electron donating ability of ether groups below that of amino groups, 46-48 but above that

of unsubstituted compounds. However, the highly positive oxidation potential of the bis

complex of ligand (3) indicates that the ether-linkage leads to a net withdrawal of

electron density from the ligand, whereas the opposite is true for the amino-linked ligand

(6) . This suggests that, for the ligands considered here, the inductive electron withdrawal

of the oxygen atom dominates over resonance stabilization.

Considered in light of these CV data, the rate constants reported in Table 3.1.1

point to an unexpected result. While ether and amine substituents at the 4’ position are

found to be electron withdrawing and donating, the unmodified ligand shows a smaller

dissociation rate constant k-2 for cobalt(II) complexes than either (or possibly equal in the

case of amine-substituents). Thus, the unmodified ligand appears to be nearly optimal in

electron density, as far as kinetic stability of cobalt(II) complexes is concerned.

3.1.3.4 Spectroscopic Characterization of the Oxidation of Amine- Substituted Cobalt Complexes

Further information on the facilitation of the oxidation from Co(II) to Co(III) by

amine-linked ligands was provided by NMR spectroscopy and UV-Vis

spectrophotometry. For NMR measurements, a solution containing a molar ratio of 2:1

(7) /cobalt(II) chloride was prepared in D 2O and measured within 5 min of mixing.

Knight-shifted proton peaks between 20 and 100 ppm revealed the presence of a

paramagnetic cobalt(II) complex. The solution was not kept oxygen free, however, and

within 1 h the formation of (7) 2Co(III) was noticeable from the disappearance of the

Knight-shifted protons. After 48 h, (7) 2Co(III) was the sole detectable species in solution

(spectra are shown in Appendix 3A, Figure 3.A.1).

For characterization by UV-Vis, a solution containing 0.1 mM (6) and 0.05 mM

cobalt(II) chloride was prepared, and a spectrum recorded within 10 min of mixing. Air

was then bubbled into the solution over a period of 30 min, resulting in oxidation of the

intermediate complex into (6) 2Co(III), after which a second spectrum was recorded. As

can be determined from Figure 3.1.3, the absorption band corresponding to either the d-d

transitions or MLCT band 73-75 centered at approximately 376 nm is reduced dramatically upon oxidation from cobalt(II) to cobalt(III), similar to previously reported observations. 76

Figure 3.3 UV-vis spectra reveal that the freshly mixed bis complex of ligand (7) with Co(II) (gray line) is oxidized to Co(III) (black line) upon bubbling of air through the solution.

3.1.3.5 NMR Spectra of Cobalt Complexes

While the presence of Knight shifted peaks in NMR spectra for Co(II) species are well documented, 77-79 all of the substituted terpyridine-cobalt(II) complexes measured here (including the unoxidized amino complexes), displayed a series of additional smaller peaks. Impurities, mono -complexes, and Co(III) species were ruled out as the source of these peaks by studies on model compounds (see Appendix 2A), and the observed shifts did not match those of uncomplexed ligands. Interestingly, these peaks were present in

6 3 D2O, as well as d -DMSO and d -acetonitrile, suggesting that the peaks are not the result of a spin equilibrium, as the presence of a spin crossover for cobalt-terpyridine complexes is known to depend on the solvent. 79,80 Together, these observations suggest that the secondary peaks arise from the influence of the paramagnetic metal center on the chemical shifts of protons on the 4’-subsituent groups. Notably, with increasing temperature, these peaks gradually broadened and eventually disappeared, as shown in

Figure 3.1.4 for (7) 2CoCl 2. matching the typical behavior for paramagnetically perturbed 89

spins.81-83 Interestingly, as shown in Table 3.1.4 for the ether-substituted species, the intensities of these peaks was found to be positively correlated with the kinetic stability of the bis complexes. The reasons for this behavior remain under investigation.

Figure 3.4 NMR spectra of bis complexes of ligand (7c) with Co(II) in D 2O show a series of secondary peaks that broaden and disappear with increasing temperature.

Table 3.4 Comparison of the secondary NMR peak intensity with the rate of decay of the cobalt(II) bis complexes of ether-substituted terpyridines.

Relative 3 -1 Ligand Complex Secondary Peak k-2*10 s Intensity 4‘-Ethoxy-2,2‘:6‘,2‘‘- (2) Co(II) 5.5 terpyridine 2 0.10 4'-[2-(1-methoxyethoxy)ethoxy]- (3) Co(II) 0.13 4.77 2,2':6',2''-terpyridine 2

1.1k PEO-terpyridine (4b) 2Co(II) 0.29 3.96

5k PEO-terpyridine (4a) 2Co(II) 0.31 1.5

3.1.3.6 Dependence of Polymer Molecular Weight on Kinetics

In the previous sections, the terpy-PEO samples use were all of molecular weight

Mn =5,000 g/mol, save for sample 4b (Mn = 1.1k g/mol), which was only characterized

as a cobalt complex. Interestingly, the ether-linked cobalt species had a wide range of

-3 -1 -3 -1 decay rates, from 1.70*10 s to 5.5*10 s . For the iron samples too, (3) 2Fe(II)

(diethylene glycol methyl ether substitutent) decayed at approximately half the rate of

(4a) 2Fe(II) (5k PEO substitutuent). Control experiments conducted on the cobalt

complexes (see section 3.1.4.1 ) eliminated the possibility of the PEO chain acting as a

ligand, as well as local viscosity effects, leaving the reason for this phenomenon still

poorly understood.

To gain a more complete picture of the effect of molecular weight on the stability

of iron and cobalt complexes, terpyridine-functionalized PEO samples of 12.3k g/mol

and 41k g/mol molecular weights were utilized, and a decay experiment for iron with the

1.1k PEO ligand (4b) was conducted. These new ligands were designated (4c) (12.3k

PEO) and (4d) (41k PEO), and were synthesized using the same methods as for (4b) and

(4a) , from PEO-methyl ether polymers purchased from Polymer Source. All these additional ligands had a >95% end capping with terpyridine. A 93k g/mol PEO-based ligand was also considered, but was too large to be accurately studied by NMR and had too low of an overlap concentration for practical kinetics studies. For kinetics studies involving ligands (4c) and (4d) , concentration of the complexes was lowered to keep the solution well below the overlap regime.

The decay constants cobalt complexes of the new ligands are shown in Table

3.1.5, with the previously recorded data for ligands (3) and (4a) . It can be deduced from

the data that there is a definite increase in decay rate for the iron complexes as the

-6 -1 molecular weight increases from the 5k ligand ( k-2 = 5.00*10 s ), with the complexes

-6 -1 -6 -1 (4c)2Fe(II) and (4d)2Fe(II) decaying at k-2 = 57.*10 s and k-2 = 6.61*10 s , respectively. Additionally, the complex of the 1.1k ligand (4b) 2Fe(II) decayed at a rate

-6 -1 of k-2 = 3.80*10 s , and the N = 2 PEO ligand (diethylene glycol methyl ether

-6 -1 substitutent) (3) 2Fe(II) decayed at a rate of 2.79*10 s , giving a slight, but unmistakable

dependence of the decay rate on the molecular weight.

Table 3.5 Decay Rates of Iron Complexes Based on the Number of Repeat Units ( N) of the PEO Substituent 6 -1 Ligand Complex N k-2*10 s 4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''- (3) Fe(II) 2 2.79 ± 0.01 terpyridine 2

1.1kk PEO-terpyridine (4b) 2Fe(II) 25 3.80 ± 0.05

5k PEO-terpyridine (4a) 2Fe(II) 114 5.00 ± 0.03

12.3k PEO-terpyridine (4c)2Fe(II) 280 5.7 ± 0.1

41k PEO-terpyridine (4d)2Fe(II) 932 6.61 ± 0.02

One possible explanation for these results is that the excluded volume effect

factors into the decay rate of these complexes. Tension applied to the polymer chain via

the excluded volume effect could lower the activation energy of dissociation for the

complex, resulting in faster decays as the complex increases in molecular weight. A plot

of the decay rate versus chain length for the iron complexes is shown in Figure 3.1.5,

along with a power law fit of the data. The power law fit reveals a relationship of k-2 ~

N0.14 for the iron complexes. Incidentally, the exponent is very close to the inverse of the relationship predicted by Khokhlov for two equal-length chains participating in an end-

-0.16 84 to-end chemically controlled association reaction ( ka ~ N ), and agrees reasonably

well with the dependence of polymer bond strength on excuded volume by Khalatur. 85

While this is not hard evidence of a pronounced excluded volume effect, it certainly

underlines the need for more study of these complexes.

7E-6

6E-6

5E-6 ) -1

(s 4E-6 -2 k

3E-6

1 10 100 1000 Repeat Units ( N)

Figure 3.5 Dependence of Iron Complex Decay Rate on Length of Substituent Polymer. The data points from Table 3.5 are denoted by black dots, while the power law fit is shown as a red line.

The results of the kinetic decay studies of cobalt complexes of the new ligands are

shown in Table 3.1.5, alongside the previously recorded values. It can clearly be

determined from Table 3.1.6 that ligand (4a) (5k PEO) is the most slowly decaying

species, which is surprising given the results for the iron complexes discussed above.

The most rapidly decaying cobalt species was found to be N = 1 species (2) 2Co(II) at k-2

-3 -1 -3 -1 = 5.5*10 s , closely followed by the N = 2 species, (3) 2Co(II) at k-2 = 4.77*10 s .

The inverse relationship between chain lengths and decay rates is continued by the 1.1k

-3 -1 -3 -1 and 5k PEO ligands, with decay rates of k-2 = 3.96*10 s and k-2 = 1.5*10 s , respectively. This trend is reversed at higher molecular weights, however, with the 12.3k

-3 -1 -3 -1 species decaying at k-2 = 2.3*10 s , and the 41k species decaying at k-2 = 5.09*10 s .

Table 3.6 Decay Rates of Cobalt Complexes Based on the Number of Repeat Units (N) of the PEO Substituent 3 Ligand Complex N k-2*10

4‘-Ethoxy-2,2‘:6‘,2‘‘-terpyridine (2) 2Co(II) 1 5.5±0.1 4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''- (3) Co(II) 2 4.77±0.05 terpyridine 2 1.1k PEO-terpyridine (4b) 2Co(II) 25 3.96±0.03 5k PEO-terpyridine (4a) 2Co(II) 114 1.5±0.04 12.3K PEO-terpy (4c)2Co(II) 280 2.3±0.04 41K PEO-terpy (4d)2Co(II) 932 5.09±0.09

The dependence of the molecular weight on the decay of these cobalt complexes

is puzzling indeed; it appears that ligand of weight over 5,000 g/mol behave similar to the

iron complexes, while those below this molecular weight decay more rapidly as the

polymer chains become smaller. It is interesting to note that the increase in decay rate

with increasing molecular weight does not begin until the 5k ligand, which when

associated (final weight ~ 10k g/mol) is the first polymer in the series above the rod-coil

threshold of ~ 6k g/mol, 86, which would suggest that the polymer chain architecture in

solution may have an effect on the decay rate.

What is most puzzling is why the iron complexes do not exhibit a similar

behavior, as the structures of the iron and cobalt complexes on the macromolecular scale

should be similar. The largest difference in the complexes is the Jahn-Teller distortion of

the cobalt complex due to the unpaired E g d-electron. This leads one to wonder if the

dynamics of the polymer coils provides some stabilizing effect to the distorted complex

94 which the lower molecular weight polymer rods to not. Regardless of the origin of the phenomenon, more research is required before specific theories can be tested.

PART 3.2

EFFECT OF THIOETHER AND SULFONYL LINKAGES ON PEO-

BASED MACROMOLECULAR TERPYRIDINE COMPLEXES

3.2.1 Introduction

In the previous section, the effects of amino and ether substitutions on the 4’

position of terpyridine molecules on the kinetic stability of the bis complexes of these

ligands with iron(II) and cobalt(II) were investigated. This study was performed

primarily with determining how the chemistry commonly used to link terpyridine to

polymers affected kinetic stability. To more expand this study, sulfur linking chemistries

were added to the ligands employed, using the techniques previously defined to

investigate the kinetic stability of bis complexes of thioether and sulfone-linked

terpyridine-fiunctionalized poly(ethylene oxide).

Previously, Constable et al. outlined a convenient synthesis of 2,2 ′:6 ′,2 ″-

Terpyridine-4′(1 ′H)-thione from 4’-chloro-2,2’:6’,2”-terpyridine. 87 The compound was

found to be able to self-react by forming disulfide linkers in the presence of a base and

iodine, thus creating a difunctional ligand. 88 It was also shown to be capable of forming a

monolayer on a gold surface, creating redox-active surfaces. 89 Most important to this study, however, is the ability of the thione-terpyridine analogue to displace mesyl 87 and

90 tosyl via S N2 reaction after being deprotonated with a base. This reaction is similar to that employed with the oxygen analogue, 2,6-Bis(2-pyridyl)-4(1H)-pyridone.91 The

reaction is subtable for use with mesyl-functionalized PEO, in order to create thioether-

linked terpyridine-functionalized polymers.

Another factor influencing the choice of sulfur as the atom linking the terpyridine

to the polymer is the ability of the thioether to be conveniently oxidized by m- chloroperbenzioc acid, thus creating a sulfoxide. This reaction has been used to create

4’-methansulfonyl-2,2’:6’,2”-terpyridine from 4’-methylthio-2,’2:6’,2”-terpyridine, 92 an intermediate in the synthesis of 2,2’:6’,2”-terpyridine. 93 This reaction gives another method by which the electron density of the ligand can be modified. The Hammet constants as measured by McDaniel and Brown 94 for the p-thioether and p- methylsulfonyl groups are 0.034 and 0.71 in 1:1 water/ethanol. While this is a considerable difference, it was found in our previous study that the Hammet constants of the substitutent do not necessarily determine how the kinetic stability of the bis complexes will be affected.

In this study, terpyridine-functionalized poly(ethylene oxide) is synthesized from mesyl-functionalized PEO (Mn=5000 g/mol), creating thioether-linked PEO-terpy. The macromolecular ligand is further modified via oxidation of the sulfur linkage to produce

PEO-terpy with a sulfonyl linkage. The ligands used in this study are shown in Scheme

3.2.1. The kinetic stabilities of these complexes of these ligands in aqueous solution were measured by metal exchange in the manner presented previously described.

O (8) O n S N

N O O (9) O n S N O N

Scheme 3.3 The Macromolecular Ligands used in the study conducted in Chapter 2 Part2.

The ligands used in this study are found in Scheme 3.2.1. We find that, for the iron complexes, the stability of thioether-linked complex is approximately an order of magnitude less stable than the complex with unsubstituted terpyridine. The iron complex is, however approximately as stable as the amino-linked complex described in the last chapter. The sulfonyl-linked iron complex decayed at a rate nearly three times more rapid than the thioether, suggesting that the electron-withdrawing nature of the sulfonyl group destabilizes the complex. For the cobalt complexes, the thioether-linked complex decays approximately twice as fast as its ether- and amino-linked homologues, and the sulfonyl complex decays slightly more rapidly than the thioether-linked complex. These results suggest that the effect on the cobalt complex is more complicated than simple electron donation/withdrawl.

3.2.2 Experimental

• Equipment

All NMR spectra were recoded on a Bruker Avance 300MHz spectrometer. All

decay experiments were recorded on a Hitachi U-3010 spectrophotometer. The

concentration of the complex for the decay studies was 0.5mM, while the concentration

of the decomposition ion was held at 10mM. The decay was measured at 569nm for the

thioether-linked complexes and at 570nm for the sulfonyl-linked complexs, the peak of

the MLCT band for the iron complex of each ligand. The decay of the iron complex was

measured directly, while the formation of the resultant iron complex was measured in the

decay of the cobalt complex.

• Materials

Monomethoxy PEO (Mn = 5000 g/mole) and cobalt(II) chloride hexahydrate

(≥98%) were purchased from Fluka chemicals; sodium hydrosulfide hydrate (100%), triethylamine (reagent grade), and m-chloroperbenzoic acid (70-75%) were purchased from Fisher Scientific. Iron(II) chloride tetrahydrate (99+%), and 4’-chloro-2,2’:6’,2”- terpyridine (99%) were purchased from Acros Organics. The mesylate of PEO was prepared using previously published techniques, 53 as was terpyridine-4′(1 ′H)-thione.87

• Synthetic Procedures

Synthesis of PEO-S-Terpy (8). To a 250 mL round bottom flask, 2.19 g of mesyl-PEO,

0.173 g of terpyridine-4′(1 ′H)-thione, and 0.38 g of potassium carbonate were added.

The flask was fitted with a condenser and purged with N 2 for 15 min. The flask was then charged with 150 mL of acetone and sparged with N 2 for an additional 15 min. The mixture was then heated to reflux under nitrogen for 18 hours. After this time, the

100

acetone was removed under vacuum, and the product was dissolved in 200 mL

chloroform, and washed with 2x200 mL sat. sodium bicarbonate. The organic phase was

then dried over sodium sulfate and reduced in volume to approximately 5ml, with product

was then precipitated in diethyl ether (250 mL) and filtered. The product was washed

with another 250 mL of ether and allowed to air dry. 1.27g (58%) recovered. 1H NMR

(CDCl 3): δ 8.70 (d, 2H J = 4.52 Hz; H6,H6”), 8.61 (d, 2H, J = 7.52Hz; H3,H3”), 8.37(s,

2H, H3’,H5’), 7.86(t, 2H, J = 7.16Hz; H4,H4”), 7.35(t, 2H, J= 5.65 Hz), 5.31(d, 2H,

Terpy-S-CH2), 3.88 (t, 2H, J = 4.57Hz Terpy-S-CH 2CH2), 3.75-3.55(m, 455H, polymer

backbone), 3.41 (t, 3H, J = 4.52Hz, PEO-O-CH3).

Synthesis of PEO-SO 2-Terpy (9). To a 50ml round bottom flask, (1) (0.441 g, 0.088

mmoles) was added along with 30 mL methylene chloride. The polymer was stirred until

dissolution and cooled in an ice/water bath for 15 min. After this 0.039 g m-CPBA

(assuming 72% purity) was added to the mixture, which was stirred in the ice bath before

removal after 5 min. The solution was then allowed to stir at room temperature for 18

hours, after which the solution was washed with 2x50 mL of saturated sodium

bicarbonate solution, and the organic phase dried over sodium sulfate. The volume of the

organic phase was then reduced and the product precipitated into 100ml diethyl ether,

after which the product was filtered and washed with a second 100 mL of diethyl ether

1 before being allowed to dry in air. Yield: 0.252 g (57%) H NMR (CDCl 3): δ 8.75(s, 2H,

H3’,H5’), 8.73 (d, 2H J = 3.96 Hz; H6,H6”), 8.64 (d, 2H, J = 78.64Hz; H3,H3”), 7.90(t,

2H, J = 7.54Hz; H4,H4”), 7.39(t, 2H, J= 5.09 Hz), 3.75-3.55(m, 455H, polymer

backbone), 3.41 (t, 3H, J = 4.52Hz, PEO-O-CH3).

101

Complexes of the ligands were produced by addition of the ligand and appropriate metal salt (2:1 ratio) in the proper concentration for kinetics testing in water. The mixture was stirred for approximately 30 min before kinetic experiments commenced.

3.2.3 Results and Discussion

The synthesis of thioether-linked terpyridine-functionalized PEO (8) was carried out by mixing mesyl-PEO with terpyridine-4′(1 ′H)-thione and sodium bicarbonate in acetone under reflux. The 1H NMR spectrum of the product revealed that the polymer was >95% functionalized. The NMR spectrum of (8) was sufficiently similar to that of its oxygen-linked analogue 35 (save for the difference in the shift of the 3’/5’ protons) to allow for convenient assignment. The sulfonyl-linked species (9) was produced by reacting (8) with 2 equivalents of m-CPBA in methylene chloride at 0 o C, follwed by stirring at room temperate, in a manner similar to that described by Potts. 92 In order to perform this reaction successfully, care must be taken to add only 2 equivalents of m-

CPBA, as the peroxyacid is capable of oxidizing the nitrogen atoms of terpyridine under the same conditions. 95 In the product that was obtained in the synthesis of (9) , no evidence of terpyridine N-oxides were detected in the NMR spectrum.

The success of the oxidation reaction was measured by comparing the NMR peaks arising from the 3’/5’ protons, due to their adjacency to the sulfonyl group. In the spectrum of (8) in CDCl 3 these proton shifts occur at 8.37ppm (as compared to 8.04ppm in the oxygen analogue), 35 however in the spectrum of (9) the peaks appear to shifted to

8.75ppm, overlapping with the 6/6” proton peaks. Recording the spectrum in acetone-d6 more fully separates the peaks, giving a better view of the 3’/5’ and 6/6” peaks. The

102 spectra of both (8) and (9) in CDCl 3 and the spectrum of (9) in acetone are displayed in

Figure 3.2.1.

Figure 3.6 The NMR spectra of the ligands used in the sudy Conducted in Chapter 2 Part2. The spectrum of ligand (8) in CDCl 3 is shown in (A). The spectrum of ligand (9) is shown in CDCl 3 in (B) and in acetone-d6 in (C).

Table 3.7 Decay Rates of Iron and Cobalt Complexes of Ligands (8) and (9)

-1 Metal Ligand k-2 in s Fe(II) (8) 2.06*10 -6 (9) 5.88*10 -6 Co(II) (8) 3.69*10 -3 (9) 4.76*10 -3

The kinetic data for the complexes studied in this writing can be found in Table

3.2.1. The kinetics measurements of stability of the complex of (8) with iron(II) returned a decay rate of 2.06*10 -6 s-1, which is comparable to the values previously measured for amino-linked terpy-PEO at 1.64*10 -6 s-1. This particularly interesting because in terms of

103

the Hammett constants, the amino group would be much more electron donating,

presumably giving a much slower decay rate. The thioether-linked species decays more

rapidly than its oxygen analogue, as we have established previously that oxygen may

affect complex stability primarily though inductive withdrawal, this is sensible due to the

higher electronegativity of oxygen. The sulfonyl-linked species (9) shows nearly 3fold more rapid decay rate than (8) , decaying at 5.88*10 -6 s-1. This is expected due to the more electron-withdrawing nature of the sulfonyl group, as the withdrawl of electron density from the π-cloud of the ligand by the substitutent group decreases that with is

available for bonding with the metal center.

Figure 3.7 Decay Scans of the Cobalt Complexes of Ligands (8) and (9).

The decay scans for the cobalt complexes are shown in Figure 3.2.2. The

cobalt(II) complex of (8) decayed at a rate of 3.69*10 -3 s-1, more that twice as fast as the decay of its ether- and amino-linked homologues (1.70*10 -3 s-1 and 1.6*10 -3 s-1 , respectively). This unexpected result suggest that the action on the substitutent group on the stability of the complex is more complex that simple electron withdrawal or donation. 104

Other factors, such as the effect of the substitutent group on the polarizability of the

ligand, or the interaction of the substitutent group with solvent may be at play. The decay

-3 -1 scan of (9) showed a small increase in k-2 from that of (8) ; measured at 4.76*10 s . This is once again an expected result due to the electron withdrawal of the sulfonyl group, but the relative difference between the two ligands is much less than that for iron, further underlining the notion that more complicated factors than electron withdrawal determine the stability of these ligands.

3.4 Conclusions

In Part 3.1 of this chapter, the effects of both amine and ether substituents at the

4’-position of terpyridine on the kinetic stabilities of bis -terpyridine complexes of iron(II) and cobalt(II) we studied. For cobalt complexes, nitrogen substitution causes rapid oxidation by air, resulting in kinetically inert cobalt(III) complexes, while ether-linkages yield modestly more rapid decay (three to six-fold faster) than unsubstituted terpyridine complexes. Cyclic voltammetry measurements indicated that ether substituents withdraw electrons from the ligands, thus rendering the complexes more labile, while nitrogen substituents donate electron density, facilitating oxidation of the cobalt center.

Iron complexes are also sensitive to 4’ substitution, with both ether and amino groups leading to faster decay by at least an order of magnitude compared to the unsubstituted ligand. In contrast to the d 7 cobalt complex, which has a distorted octahedral architecture, the octahedral d 6 iron complex is already very stable, and the nitrogen-substituted ligands do not yield the same oxidation-induced stability increase.

A brief study on the effect of polymer chain length on the stability revealed an decrease in complex stability with increasing chain lengths for iron. The results suggest a

105

0.14 power-law dependence of the decay rate to the chain length of k-2 ~ N . The cobalt complexes demonstrated a much different behavior, with decay rates initially falling with increasing molecular weight, then increasing. These are interesting preliminary results which encourage further study on the topic.

Comparisons between poly(ethylene oxide) chains end-functionalized with terpyridine through ether and amine linkages pointed to a complex interplay between linking chemistry and the effects of polymeric substituents. For both ions, however, the amine-linked species were found to provide enhanced stability over the more conventional ether-linked materials, with dramatic results for the case of cobalt due to ligand facilitated oxidation to Co(III). The deeper understanding gained of the effects of ether and amino substituent groups on terpyridine complexes will hopefully lead to new channels of investigation into how properties of dynamic supramolecular assemblies can be tuned by substituents.

In Part 3.2 of this chapter, we have demonstrated that thioether linkages in terpyridine-functionalized PEO are capable of forming an iron complex of approximately the same stability of its amino-linked homolougue. When the thioether is oxidized to a sulfonyl, the rate of decay increased nearly threefold, due to electron withdrawl. For cobalt complexes of the same ligands, the decays are faster than either their ether-or amino-linked homologues, suggesting the action on the stability of the ligand is not merely electron withdrawal. Also, the difference between the decay rates of the two cobalt complexes is less, with the sulfonyl species being only slightly less stable than the thioether.

106

3.5 Appendix 3A. Model Compounds in the Investigation of Secondary Peaks

In order to determine if the secondary peaks in the NMR spectra of cobalt

3+ complexes were an oxidized species (4b )2Co was synthesized using a procedure similar

to existing methods. 95 Also, the mono complex of terpyridine with cobalt was

synthesized and investigated via NMR. Both species were not a match for the secondary

peaks. Impurities giving rise to the secondary peaks was also ruled out by repeated

recrystallization and NMR investigation of the small-molecule compounds.

3+ Synthesis of (4b) 2Co . A 20 mL vial was charged with 0.1 g of (4b) , 0.0024 g

CoCl 2*6H 2O and 0.0034 g AgNO 2. Dueterium oxide (1 mL) was then added, and the

mixture stirred for 1 hour. After this time, all particulates where removed via syringe

1 filter and the solution was transferred to an NMR tube. H NMR (D 2O): δ = 8.56 (s, 2H),

8.52(d, 2H), 7.34 (2, 4H), 4.2-3.3 (m, 100H, polymer backbone)

NMR Shifts of (1)CoCl 2 in D 2O. The complex was prepared using an existing

60 1 procedure. H NMR (D2O): δ = 108.06 (s, 2H), 84.77 (s, 2H), 51.71 (s, 2H), 34.83 (s,

1H) 15.96 (s,2H). Note : Only five proton shifts are observed in the window examined.

Since the sixth shift has recently been found to be in the range 130-170 ppm in organic solvents, 1 it is conceivable that the sixth peak populates this region in aqueous solution as well.

107

2+ • NMR Spectra of (7) 2Co Before and After Oxidation.

1 2+ Figure 3.A.1. H NMR Spectrum of the paramagnetic (7) 2Co species can be viewed in the top spectrum, note the presence of Knight-Shifted peaks between 90 and 20 3+ ppm. The sample was gradually oxidized in air to diamagnetic (7) 2Co , as shown in the lower spectrum. Note the lack of Knight-shifted peaks in the lower spectrum.

• NMR Measurement of the Decay of (1) 2Co(II)

The decay of the unsubstituted complex (1) 2Co(II) was characterized by measuring the Knight-shifted peaks of a solution of the complex along with FeCl 2 in

-4 -1 D2O. An exponential fit of the data returns k-2 = 6.12*10 s , a decay of a little more than half of that measured by stop flow.

108

3.6 Appendix 3B. Cobalt Complex Decay in Mixed Solvent and Nickel Complex Decomposition

The decay over several of the cobalt complexes in a mixture of water in 20%

DMSO, the results of the study is shown in Table 3.B.1. Little change observed from the decay rates in aqueous solution, except for complex (4a) 2Co(II), which showed a four- fold increase in decay rate.

Table 3.B.1 Decay of Ether-linked Cobalt Complexes in a Water/20% DMSO solution.

3 Ligand Complex k-2*10

4'-[2-(1-methoxyethoxy)ethoxy]-2,2':6',2''-terpyridine (3) 2Co(II) 5.1±0.02 1.1k PEO-terpyridine (4b) 2Co(II) 3.86±0.06 5k PEO-terpyridine (4a) 2Co(II) 4.45±0.03 12.3K PEO-terpy (4c)2Co(II) 2.54±0.04

• Decay of Ether-Linked Nickel(II) Complexes

Ether-linked Nickel(II) complexes was recorded, using FeCl 2 as the decaying species, the data was not included in the main study due to the fact that the unsubstituted complex could not be decayed with iron, and therefore could not be checked against previously published data. The results are presented in Table 3.B.2, as a precursor to potential future studies. Al the complexes show decay rates on the order of 10 -5 s-1 which would be expected for Nickel complexes based on the study by Hogg and Wilkins. 60

109

Table 3.B.2 Decay Rates of Ether-Linked Nickel Complexes

3 Ligand Complex k-2*10

1.1k PEO-terpyridine (4b) 2Ni(II) 0.90±0.02 5k PEO-terpyridine (4a) 2Ni(II) 1.4±0.1 12.3K PEO-terpy (4c)2Ni(II) 1.6±0.1

3.7 References

(1) Constable, E. C.; Emeléus, H. J. In Advances in Inorganic Chemistry ; Academic Press: 1986 ; Vol. Volume 30, p 69-121. (2) Guillet, P.; Fustin, C.-A.; Mugemana, C.; Ott, C.; Schubert, U. S.; Gohy, J.-F. Soft Matter 2008 , 4, 2278-2282. (3) Schubert, U. S.; Hofmeier, H.; Newkome, G. R. Modern Terpyridine Chemistry ; Wiley-VCH: Weinheim, 2006 . (4) Ruben, M.; Lehn, J.-M.; Muller, P. Chemical Society Reviews 2006 , 35 , 1056-1067. (5) Yount, W. C.; Loveless, D. M.; Craig, S. L. Angewandte Chemie International Edition 2005 , 44 , 2746-2748. (6) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1965 , 4, 929-935. (7) Danil de Namor, A. F.; Zapata-Ormachea, M. L.; Jafou, O.; Rawi, N. A. The Journal of Physical Chemistry B 1997 , 101 , 6772-6779. (8) Chiper, M.; Meier, M. A. R.; Kranenburg, J. M.; Schubert, U. S. Macromolecular Chemistry and Physics 2007 , 208 , 679-689. (9) Meier, M. A. R.; Lohmeijer, B. G. G.; Schubert, U. S. Journal of Mass Spectrometry 2003 , 38 , 510-516. (10) Schubert, U. S.; Eschbaumer, C.; Andres, P.; Hofmeier, H.; Weidl, C. H.; Herdtweck, E.; Dulkeith, E.; Morteani, A.; Hecker, N. E.; Feldmann, J. Synthetic Metals 2001 , 121 , 1249-1252. (11) Yoshida, H.; Ozawa, T.; Jitsukawa, K.; Einaga, H. Polyhedron 1993 , 12 , 1319-1328. (12) Romeo, R.; Arena, G.; Monsu Scolaro, L. Inorganic Chemistry 1992 , 31 , 4879-4884. (13) Ostermeier, M.; Berlin, M.-A.; Meudtner, R. M.; Demeshko, S.; Meyer, F.; Limberg, C.; Hecht, S. Chemistry – A European Journal , 2010 , 16 , 10202-10213. (14) Elias, H.; Hasserodt-Taliaferro, C.; Hellriegel, L.; Schoenherr, W.; Wannowius, K. J. Inorganic Chemistry 1985 , 24 , 3192-3198. (15) Cusumano, M.; Giannetto, A.; Imbalzano, A. Polyhedron 1998 , 17 , 125- 129. (16) Ellis, P.; Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1959 , 3308-3313.

110

(17) Brandt, W. W.; Wright, J. P. Journal of the American Chemical Society 1954 , 76 , 3082-3083. (18) Farina, R. D.; Hogg, R.; Wilkins, R. G. Inorganic Chemistry 1968 , 7, 170- 172. (19) Miller, R. R.; Brandt, W. W. Analytical Chemistry 1954 , 26 , 1968-1969. (20) Morgan, G.; Burstall, F. H. Journal of the Chemical Society (Resumed) 1937 , 1649-1655. (21) Gohy, J.-F.; Lohmeijer, B. G. G.; Schubert, U. S. Macromolecular Rapid Communications 2002 , 23 , 555-560. (22) Mugemana, C.; Guillet, P.; Hoeppener, S.; Schubert, U. S.; Fustin, C.-A.; Gohy, J.-F. Chemical Communications 2010 , 46 , 1296-1298. (23) Hofmeier, H.; Schubert, U. S. Macromolecular Chemistry and Physics 2003 , 204 , 1391-1397. (24) Hofmeier, H.; Schmatloch, S.; Wouters, D.; Schubert, U. S. Macromolecular Chemistry and Physics 2003 , 204 , 2197-2203. (25) Lohmeijer, B. G. G.; Schubert, U. S. Macromolecular Chemistry and Physics 2003 , 204 , 1072-1078. (26) Chiper, M.; Fournier, D.; Hoogenboom, R.; Schubert, U. S. Macromolecular Rapid Communications 2008 , 29 , 1640-1647. (27) Schmatloch, S.; González, M. F.; Schubert, U. S. Macromolecular Rapid Communications 2002 , 23 , 957-961. (28) Haensch, C.; Chiper, M.; Ulbricht, C.; Winter, A.; Hoeppener, S.; Schubert, U. S. Langmuir 2008 , 24 , 12981-12985. (29) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-625. (30) Shunmugam, R.; Gabriel, G. J.; Aamer, K. A.; Tew, G. N. Macromolecular Rapid Communications , 2010 , 31 , 784-793. (31) Dobrawa, R.; Würthner, F. Journal of Polymer Science Part A: Polymer Chemistry 2005 , 43 , 4981-4995. (32) Guerrero-Sanchez, C.; Lohmeijer, B. G. G.; Meier, M. A. R.; Schubert, U. S. Macromolecules 2005 , 38 , 10388-10396. (33) Ott, C.; Pavlov, G. M.; Guerrero-Sanchez, C.; Schubert, U. S. Journal of Polymer Science Part A: Polymer Chemistry 2009 , 47 , 3691-3701. (34) Henderson, I. M.; Hayward, R. C. Macromolecules , 2010 , 43 , 3249-3255. (35) Lohmeijer, B. G. G.; Schubert, U. S. Angewandte Chemie International Edition 2002 , 41 , 3825-3829. (36) Schubert, U. S.; Hien, O.; Eschbaumer, C. Macromolecular Rapid Communications 2000 , 21 , 1156-1161. (37) Lohmeijer, B. G. G.; Schlaad, H.; Schubert, U. S. Macromolecular Symposia 2003 , 196 , 125-135. (38) Shunmugam, R.; Tew, G. N. Chemistry – A European Journal 2008 , 14 , 5409-5412. (39) Shunmugam, R.; Tew, G. N. Polymers for Advanced Technologies 2007 , 18 , 940-945. (40) Wade, L. G. Organic Chemistry ; 5 ed.; Pearson Education, inc.: Upper Saddle River, NJ, 2003.

111

(41) Zhou, Z.; Sarova, G. H.; Zhang, S.; Ou, Z.; Tat, F. T.; Kadish, K. M.; Echegoyen, L.; Guldi, D. M.; Schuster, D. I.; Wilson, S. R. Chemistry – A European Journal 2006 , 12 , 4241-4248. (42) Lowe, G.; Droz, A. S.; Vilaivan, T.; Weaver, G. W.; Tweedale, L.; Pratt, J. M.; Rock, P.; Yardley, V.; Croft, S. L. Journal of Medicinal Chemistry 1999 , 42 , 999- 1006. (43) Johansson, O. Synthesis 2006 , 2006 , 2585,2589. (44) Winter, A.; Wild, A.; Hoogenboom, R.; Fijten, M. W. M.; Hager, M. D.; Fallahpour, R.-A.; Schubert, U. S. Synthesis 2009 , 2009 , 1506,1512. (45) McMurry, J. Organic Chemistry ; 6 ed.; Brooks/Cole-Thomas Learning: Belmont, CA, 2004 . (46) Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.; Slattery, S. J. Inorganica Chimica Acta 2006 , 359 , 2400-2406. (47) Hathcock, D. J.; Stone, K.; Madden, J.; Slattery, S. J. Inorganica Chimica Acta 1998 , 282 , 131-135. (48) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; Thompson, A. M. W. C. Inorganic Chemistry 1995 , 34 , 2759-2767. (49) Kanan, M. W.; Surendranath, Y.; Nocera, D. G. Chemical Society Reviews 2009 , 38 , 109-114. (50) Kanan, M. W.; Nocera, D. G. Science 2008 , 321 , 1072-1075. (51) Constable, E. C.; Ward, M. D. Journal of the Chemical Society-Dalton Transactions 1990 , 1405-1409. (52) Huang, W.; Han, C. D. Macromolecules 2006 , 39 , 8207-8209. (53) Park, M. K.; Kim, H. S.; An, J. H.; Kim, J. Journal of Industrial and Engineering Chemistry 2005 , 11 , 222-227. (54) End, N.; Schoening, K.-U.; Froehling, B. Euro. Pat. Appl. WO2006EP61990 20060503, 2006 . (55) Whittle, B.; Batten, S. R.; Jeffery, J. C.; Rees, L. H.; Ward, M. D. Journal of the Chemical Society, Dalton Transactions 1996 , 4249-4255. (56) Ruloff, R.; Koten, G. v.; Merbach, A. E. Chemical Communications 2004 , 842-843. (57) Lowe, G.; Droz, A.-S.; Park, J. J.; Weaver, G. W. Bioorganic Chemistry 1999 , 27 , 477-486. (58) Harris, M. G.; Stewart, R. Canadian Journal of Chemistry 1977 , 55 , 3800- 3806. (59) Olmstead, W. N.; Margolin, Z.; Bordwell, F. G. The Journal of Organic Chemistry 1980 , 45 , 3295-3299. (60) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1962, 341-350. (61) Graessley, W. W. Polymer 1980 , 21 , 258-262. (62) Khan, M. S. Journal of Applied Polymer Science 2006 , 102 , 2578-2583. (63) Bjerrum, J.; McReynolds, J. P.; Oppegard, A. L.; Parry, R. W. Hexamminecobalt(III) Salts ; John Wiley & Sons, Inc., 2007 . (64) Constable, E. C.; Thompson, A. M. W. C. Journal of the Chemical Society, Dalton Transactions 1995 , 1615-1627. (65) Pearson, R. G. The Journal of Physical Chemistry 1959 , 63 , 321-326.

112

(66) Powell, D. H.; Lothar, H.; Andre, E. M. The Journal of Chemical Physics 1991 , 95 , 9258-9265. (67) Kjaer, A. M.; Ulstrup, J. Inorganic Chemistry 1986 , 25 , 644-651. (68) Musumeci, S.; Rizzarelli, E.; Sammartano, S.; Bonomo, R. P. Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 1973 , 46 , 109-117. (69) Rao, J. M.; Hughes, M. C.; Macero, D. J. Inorganica Chimica Acta 1976 , 16 , 231-236. (70) Nicholson, R. S.; Shain, I. Analytical Chemistry 1964 , 36 , 706-723. (71) Ayers, T.; Scott, S.; Goins, J.; Caylor, N.; Hathcock, D.; Slattery, S. J.; Jameson, D. L. Inorganica Chimica Acta 2000 , 307 , 7-12. (72) Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.; Slattery, S. J. Inorganica Chimica Acta 2006, 359 , 2400-2406. (73) Enachescu, C.; Krivokapic, I.; Zerara, M.; Real, J. A.; Amstutz, N.; Hauser, A. Inorganica Chimica Acta 2007 , 360 , 3945-3950. (74) Kremer, S.; Henke, W.; Reinen, D. Inorganic Chemistry 1982 , 21 , 3013- 3022. (75) Judge, J. S.; Baker, W. A. Inorganica Chimica Acta 1967 , 1, 68-72. (76) Vitello, L.; Erman, J. Journal of Biological Inorganic Chemistry 2001 , 6, 578-589. (77) C. Constable, E.; E. Housecroft, C.; Neuburger, M.; G. Schneider, A.; Zehnder, M. Journal of the Chemical Society, Dalton Transactions 1997 , 2427-2434. (78) Epperson, J. D.; Ming, L.-J.; Baker, G. R.; Newkome, G. R. Journal of the American Chemical Society 2001 , 123 , 8583-8592. (79) Chow, H. S.; Constable, E. C.; Housecroft, C. E.; Kulicke, K. J.; Tao, Y. Dalton Transactions 2005 , 236-237. (80) Beattie, J. K.; Binstead, R. A.; Kelso, M. T.; Del Favero, P.; Dewey, T. G.; Turner, D. H. Inorganica Chimica Acta 1995 , 235 , 245-251. (81) Ivano Bertini; Claudio Luchinat; Parigi, G. Solution NMR of Paramagnetic Molecules: Applications to Metallobiomolecules and Models ; Elsevier, 2001 . (82) Brennan, L.; Turner, D. L. Biochimica et Biophysica Acta (BBA) - Protein Structure and Molecular Enzymology 1997 , 1342 , 1-12. (83) Webb, G. A. Nuclear Magnetic Resonance ; Royal Society of Chemistry, 2004 ; Vol. 33. (84) Khokhlov, A. R. Die Makromolekulare Chemie, Rapid Communications 1981 , 2, 633-636. (85) Khalatur, P. G., A. S. Pavlov, et al. Doklady Akademii Nauk Sssr 1984, 276, 6,: 1431-1434. (86) Kawaguchi, S.; Imai, G.; Suzuki, J.; Miyahara, A.; Kitano, T.; Ito, K. Polymer 1997 , 38 , 2885-2891. (87) Constable, E. C.; Hermann, B. A.; Housecroft, C. E.; Neuburger, M.; Schaffner, S.; Scherer, L. J. New Journal of Chemistry 2005 , 29 , 1475-1481. (88) Constable, E. C.; Housecroft, C. E.; Neuburger, M.; Price, J. R.; Schaffner, S. Dalton Transactions 2008 , 3795-3797.

113

(89) Park, J.; Pasupathy, A. N.; Goldsmith, J. I.; Chang, C.; Yaish, Y.; Petta, J. R.; Rinkoski, M.; Sethna, J. P.; Abruna, H. D.; McEuen, P. L.; Ralph, D. C. Nature 2002 , 417 , 722-725. (90) Wang, Q.; Day, P.; Griffiths, J.-P.; Nie, H.; Wallis, J. D. New Journal of Chemistry 2006 , 30 , 1790-1800. (91) Andres, P. R.; Hofmeier, H.; Lohmeijer, B. G. G.; Schubert, U. S. Synthesis-Stuttgart 2003 , 2865-2871. (92) Potts, K. T.; Cipullo, M. J.; Ralli, P.; Theodoridis, G. The Journal of Organic Chemistry 1982 , 47 , 3027-3038. (93) Potts, K. T.; Cipullo, M. J.; Ralli, P.; Theodoridis, G. Journal of the American Chemical Society 1981 , 103 , 3585-3586. (94) McDaniel, D. H.; Brown, H. C. The Journal of Organic Chemistry 1958 , 23 , 420-427. (95) Thummel, R. P.; Jahng, Y. The Journal of Organic Chemistry 1985 , 50 , 3635-3636. (96) Mugemana, C.; Guillet, P.; Hoeppener, S.; Schubert, U. S.; Fustin, C.-A.; Gohy, J.-F. Chemical Communications 2010 , 46 , 1296-1298.

114

CHAPTER 4

KINETIC STABILITY OF BIS COMPLEXES OF TERPYRIDINE

WITH IRON(II) AND COBALT(II) IN ORGANIC SOLVENTS

4.1 Overview

We have studied the effect of solvent environments on the kinetic stability of bis

complexes of 2,2':6',2’’-terpyridine with iron(II) and cobalt(II), and found variations in

rate of greater than five orders of magnitude for iron and one order of magnitude for

cobalt in common solvents. While the effects governing the decay of cobalt complexes

remain unclear, a correlation is found between the solubility of terpyridine in the selected

solvents and the decay rates of the iron complexes. These results not only underscore the

importance of solvent selection to the kinetic stability of terpyridine complexes, but also

provide an opportunity to tune the kinetic properties of these complexes.

4.2 Introduction

The 2,2':6',2’’-terpyridine ligand and its family of analogues have been studied as

complexes of numerous of transition metals such as Fe, Ni, Co, Cu, Ru, and Pt, and with

a wide variety of substituent groups such as ethers 1,2 , amines 3-5, phenyl rings 6, and halides. The synthetic versatility of chloro-2,2’:6’,2”-terpyridine as a precursor for syntheses including polymer functionalization 7-10 has made the terpyridine ligand a

common motif in supramolecular chemistry.

Wilkins, et. al. performed studies on the rate constants of formation and

dissociation of 2,2':6',2’’-terpyridine complexes for several first-row transition metals in

115

the 1960s.11,12 More recently we have shown that substituent groups in the 4’ position of terpyridine affect kinetic stabilities of iron and cobalt complexes.13 Notably, all of these studies were completed in aqueous environments, while most recent studies on poymer assemblies use organic solvents such as acetonitrile, 14,15 N,N-dimethylformamide 16 , acetone 14,17 , and various alcohols 14,18 . With the wide variety of studies that have been conducted on terpyridine-based materials, relatively few studies have been completed on the kinetic stability of terpyridine complexes in different solvent environments. The classical kinetic studies may provide a qualitative guide to whether a particular complex is “inert” or “labile”, but not an accurate picture of the specific behavior of complexes in organic solvents. To better understand the stability of terpyridine complexes in organic solvents, we undertook the current study on effects of solvent on the decay of bis- complexes of 2,2':6',2’’-terpyridine with iron (II) and cobalt (II).

To measure the kinetic stabilies of the terpyridine bis complexes (the rate constant for dissociation, k-2), the complex was allowed to decay in the presence of a more strongly binding metal ion. Cobalt complexes were decayed by iron (II) chloride tetrahydrate, while iron complexes were decayed by nickel (II) chloride hexahydrate; both schemes have previously been shown to yield accurate rate constants.9,12,19 The solvents used in this study were chosen based on their ability to dissolve both the complex and the decomposing metal salts. Dimethylsulfoxide (DMSO), N,N- dimthylformamide (DMF), N-methylpyrrolidine (NMP), and ethanol were initially chosen based on their decomposing metal/complex compatibility. Acetone and acetonitrile were also chosen as solvents of interest, but as the decomposing metal salts were not soluble in either, these solvents were studied as mixtures with ethanol.

116

4.3 Results and Discussion

Complexes were synthesized by mixing two equivalents of 2,2’:6’,2”-terpyridine with one equivalent of either iron(II) chloride tetrahydrate or cobalt(II) chloride hexahydrate in water. After stirring until no solid ligand remained, an excess of ammonium hexaflourophosphate was added, causing precipitation of [terpy]2Fe(PF 6)2 or

[terpy]2Co(PF 6)2. After the complexes were separated by filtration and washed with water, they were allowed to dry in air before being washed with ethyl ether to remove any excess ligand. Due to the poor solubility of [terpy]2Fe(PF 6)2 in ethanol, [terpy]2FeCl 2 was also synthesized according to existing methods.12

2+ Table 4.1 Decay of [terpy] 2Fe complexes in various organic solvents. aChloride counterions. bData from Chapter 3. cReference 43

3 -1 Solubility of Solvent k *10 (s ) -2 Terpyridine (g/L) DMF 21.3 625 DMSO 2.60 227 NMP 1.50 312 Acetone/Ethanol 1.1*10 -1 278 Ethanol 2.4*10 -2 a 91 Acetonitrile/Ethanol (1:1) 2.2*10 -2 227 Acetonitrile/Ethanol (95:5) 9.7*10 -2 - Water 1.5*10 -4 a,b 0.049 c

117

Figure 4.1 Decay of [terpy]2Fe(PF6)2 in selected solvents.

Decay scans of the iron complexes were recorded by monitoring the absorbance of the MLCT band of the complex, which occurs between 550 and 555 nm, depending on solvent. The decay of the MLCT signal was measured for 70,000 s at 21.5 oC, and the data were fitted with a single exponential that directly yielded k-2, as shown for a few examples in Figure 4.1. The concentration of the iron complex was 0.25 mM, while concentration of nickel(II) chloride hexahydrate (the decomposing metal) was 10 mM.

Iron complex stability scans in DMF where repeated with 4-100 fold excesses of nickel(II) to confirm zero-order dependence on the concentration of the nickel. As can be

-5 seen in Table 4.1, the decay of the iron complex ranged from k-2 = 2.26*10 for 1:1

-2 ACN/Ethanol to k-2 = 2.13*10 in DMF, a three order of magnitude difference. When

-7 -1 9 the value previously recorded for water is taken into account ( k-2 = 1.5*10 s ), the range of decay rates becomes more than five orders of magnitude.

118

There are a few different ways by which solvent can affect metal-ligand reactions,

and these effects can even influence which reaction pathways are taken. 22 For example, a

solvent that stabilizes the transition state of an associative interchange reaction may cause

the complex in question to exchange ligands by that pathway rather than the dissociative

mechanism to which it would adhere in other solvent systems. The affinity of the solvent

for the coordination sites on the metal may also be important in determining mechanism,

as solvents that have a high affinity for a metal might conduct an S N2-type attack on an already bound ligand.

Though the affinity of the solvent for metal coordination sites often decides the rate-limiting step of metal-ligand formation reactions, 11,21,23,24 stabilization of the metal center after dissociation of the ligand may also be important, as it makes the dissociation reaction more thermodynamically favourable. 25,26 Although it was difficult to reach a conclusion due to the wide variation of reported values, a survey of the solvent exchange rates and coordination enthalpies 27-35 showed little or no correlation with the teryridine-

complex decay rates which were measured for both cobalt and iron. In spite of the lack

of reliable data, it seems unlikely that any of the solvents investigated could complete

with the terpyridine ligand in terms of binding strength due to the rapid exchange of the

solvents with the metal center. Additionally, since the solvent is the only competing

ligand in this case, an interchange reaction seems unlikely, so the stabilization of the

transition state of an interchange reaction is probably not the way solvent governs the

reaction rate.

119

A far more likely scenario by which solvent affects the decay of these complexes

is the interaction of solvent with the transition state of the dissociation reaction, rather

than the metal center. As a result, the rate of dissociation is accelerated in solvents which

lower the energy of the transition state of the dissociation reaction. Solvent interaction

with bound ligands has previously been shown to affect the decay rate of calixspherand,

cryptate, and crown ether complexes, 36-40 as well as the rate of racemazation and dissociation of bipyridine and phenanthroline complexes.25,26,41 Most relevant to this study, however, is the work done by Burgess and coworkers on solvent effects on racemazation and dissociation of tris -phenathroline iron(II) complexes. 42 It was found that the rate of dissociation of the phenanthroline complex was dependant on the ability of the solvent to lower the energy of the transition metal complex by interaction of the organic solvent with the hydrophobic ligands. As a result the rates of dissociation increased as the organic fraction of solvent/water mixtures increased.

Figure 4.2 Solubility of Terpyridine vs. Decay Rate for the Iron Complexes

120

For the complexes used in this study, it was decided that the solubility of

terpyridine in the organic solvents would be a good indicator of the ability of the solvent

to interact with the hydrophobic terpyridine ligands, and therefore lower the energy of the

transition state . The solubility of terpyiridine in the solvents investigated is plotted

versus the decay rate for the iron complex in Figure 4.2 and the numerical values are

displayed along with the kinetic data for both iron and cobalt. The plot in Figure 4.2

clearly shows a correlation between the solubility of terpyridine and the decay rate of the

iron complex, with the solvents in which terpyridine is most soluble also displaying the

most rapid decay rates.

For the pure solvent systems, DMF showed the most rapid decay rate ( k-2 =

2.13*10 -2 s-1) as well as the highest solubility of terpyridine (625 g/L). DMSO and NMP

showed decay rates approximately one order of magnitude lower that DMF, as well as

decreased terpyridine solubility (312 g/L and 227 g/L, respectively), while the complex in

ethanol decayed three orders of magnitude less rapidly than in DMF as well as displaying

a much lower solubility of terpyridine than DMF, DMSO, and NMP (91 g/L). The

-7 -1 slowest decay of the complex was observed in water ( k-2 = 1.5*10 s ), which was also

43 the solvent in which terpyridine was least soluble (0.049 g/L).

For the mixed solvents, the dependence of terpyridine solubility on decay rate was

less sraightfoward than that of pure solvents. While both solvent mixtures displayed

similar terpyridine solubility (278 g/L for 1:1 acetone/ethanol and 227 g/L for 1:1

acetonitrile/ethanol), the mixture of acetone/ethanol showed an order of magnitude more

-4 -5 rapid decay than acetonitile/ethanol ( k-2 = 1.1*10 versus k-2 = 2.2*10 , respectively).

Additionally, decay rate in both mixed solvents was less rapid than would be expected

121

from pure solvents with similar terpyridine solubility. This could possibly be attributed

to a higher concentration of alcohol in the immediate vicinity of the metal, thus making

the observed decay rate more closely resemble that in ethanol.

The cobalt complexes were decomposed with iron(II) chloride tetrahydrate, and the

growth of the MLCT band from the iron complex signal was monitored and used to trace

the decay. An Advanced Photophysics RX.2000 rapid mixing unit was used for the

decay scan; the concentration of complex being 0.25 mM and the concentration of

iron(II) being 10 mM. Single exponential decays were fit to the linear region of the plot,

11 as has previously been done in water. The k-2 data from the cobalt complex scans are shown in Table 4.2.

2+ Table 4.2 Decomposition of Co[terpy] 2Co in various solvents aChloride counterions, bData from Chapter 3. c Data from Reference 43

Solubility of 3 -1 Solvent k-2 *10 (s ) Terpyridine (g/L) DMSO 35.0 227 DMF 32.0 625 Acetone/Ethanol 16.0 278 NMP 3.8 312 Acetonitrile/Ethanol 1.9 227 Water 1.1 a,b 0.049 c

In contrast to the iron decay scans, variations in the cobalt decay rate constants

were limited to a range of only one order of magnitude. The least rapid decay rate

measured in organic solvent was that of the cobalt complex in 1:1 acetonitrile/ethanol, at

-3 -1 k-2 = 1.90*10 s . This is only slightly more rapid that the decay measured for water ( k-2

= 1.1*10 -3 s-1). Since no scans of the complex could be completed in ethanol due to precipitation(even with Cl - counterions) of what is assumed to be the cobalt(II) mono

complex, the decay rate in ethanol could not be measured. Surprisingly, the second

122

-3 -1 slowest decay occurred in NMP (k-2 = 3.8*10 s ), further underlining the differences between the iron and cobalt complexes. The solvents in which the cobalt complexes decayed fastest in were 1:1 acetone/ethanol, DMF, and DMSO. The acetone/ethanol

-2 -1 mixture was the slowest of these, decaying at k-2 = 1.6*10 s . The decay rates of the cobalt bis complex in DMF and DMSO were similar, at 3.2*10 -2 and 3.8*10 -2, respectively.

In addition to displaying a weaker dependence on solvent, the decay of the cobalt complexes did not show a clear correlation to the solubility of terpyridine in the solvent used. It is currently unknown why the cobalt complexes show a much weaker solvent effect than iron, although many other factors such as the conformational and spin-state changes in a particular solvent could be particularly important for a distorted d 7 complex such as that of terpyridine with cobalt. Additionally, it may be the case that a different mechanism may be responsible for the cobalt complex decay. It is possible that an interchange reaction with a different mode of solvent interaction may be responsible for the decay of cobalt complexes.

4.4 Conclusions

In conclusion, we have documented the kinetic stability of the bis complexes of iron(II) and cobalt(II) with terpyridine in a variety of organic solvents. When have found that the iron(II) complexes experience a large increase in decay rate from that in water for all of the solvents tested. Furthermore, the range of decay rates of the iron complex in

-7 -1 different solvents spanned five orders of magnitude, from k-2 = 1.5*10 s for water to

-2 -1 k-2 = 2.13*10 s in DMF. The cobalt complexes did not show as much variation in the

123

-3 -2 -1 decay rate, from k-2 = 1.1*10 in water, to k-2 = 3.5*10 s for DMSO. A correlation between the solubility of terpyridine and decay of the iron complexes was found, with the more strongly solvating solvents displaying the most rapid decay rates. Such a correlation could not be established with the cobalt complexes, denoting the possibility of more complicated solvent interactions being at play in the decay of the distorted d 7 complex.

4.5 Appendix 4A. Preparation of Complexes of Terpyridine with Iron and Cobalt

[Terpy 2Fe](PF 6)2 A vial was charged with 0.150 g of 2,2’:6’,2”-terpyridine, to which was added to 0.064 g FeCl 2●4H 2O in 5 mL of water. The mixture was diluted to 100 mL, and sonicated for 10 min, before being run through a 0.45 m syringe filter in order to remove any free ligand. This solution was stirred rapidly, and 0.5 g of NH 4PF 6 was added, causing a purple precipitate to form. After 20 min of stirring, the precipitate was filtered and allowed to air dry for 18 hours before being washed with 300 ml of diethyl ether. The product was once again allowed to air dry for 18 h. Yield: 0.198 g (75.8%)

1 3 3 H NMR (300 MHz, acetonitrile-d6): δ 8.91 (d, 4H, J = 7.91 Hz, H3’ ,H 5’ ), 8.68 (t, 2H, J

3 = 8.10 Hz, H4’ ), 8.47 (d, 4H, J = 7.91 Hz, H3,H 3” ), 7.88(m, 4H, H4,H 4” ), 7.07 (m, 8H, H-

5,H 5” ,H 6,H 6” )

[Terpy 2Co](PF 6)2 In 100ml of water were combined 0.176 g of 2,2’:6’,2”-terpyridine and 0.090 g of CoCl 2●6H 2O. The mixture was vortexed, then sonicated for 40 minutes before being allowed to sit overnight. The solution was filtered through a 0.45 m syringe filter before 0.7 g NH 4PF 6 was added, giving a tan precipitate. after 20 min of 124

stirring, the precipitate was filtered and dried in air for 18 h. The tan precipitate was then

washed with 300 mLl of diethyl ether and dried in air for an additional 18 h. Yield:

1 0.204g (66.7%) H NMR (300 MHz, acetonitrile-d6): 99.46 (broad 4H, H6,H 6” ), 57.15(s,

4H, H3,H 3” ), 47.29(s, 4H, H3’ ,H 5’ ), 34.62(s, 4H, H5,H 5” ) 22.26(s, 2H, H4’ ), 9.14 (s, 4H,

H4,H 4” )

Note: No oxidation of the cobalt complex was observed on the timescale that the complex was tested in the kinetic scans.

4.6 References

(1) Hathcock, D. J.; Stone, K.; Madden, J.; Slattery, S. J. Inorganica Chimica Acta 1998, 282 , 131-135. (2) Andres, P. R.; Hofmeier, H.; Lohmeijer, B. G. G.; Schubert, U. S. Synthesis 2003, 2003 , 2865,2871. (3) Zhou, Z.; Sarova, G. H.; Zhang, S.; Ou, Z.; Tat, F. T.; Kadish, K. M.; Echegoyen, L.; Guldi, D. M.; Schuster, D. I.; Wilson, S. R. Chemistry – A European Journal 2006, 12 , 4241-4248. (4) Lowe, G.; Droz, A.-S.; Park, J. J.; Weaver, G. W. Bioorganic Chemistry 1999, 27 , 477-486. (5) Johansson, O. Synthesis 2006, 2006 , 2585,2589. (6) Spahni, W.; Calzaferri, G. Helvetica Chimica Acta 1984, 67 , 450-454. (7) Aamer, K. A.; de Jeu, W. H.; Tew, G. N. Macromolecules 2008, 41 , 2022- 2029. (8) Guerrero-Sanchez, C.; Lohmeijer, B. G. G.; Meier, M. A. R.; Schubert, U. S. Macromolecules 2005, 38 , 10388-10396. (9) Henderson, I. M.; Hayward, R. C. Macromolecules , 43 , 3249-3255. (10) Shunmugam, R.; Gabriel, G. J.; Aamer, K. A.; Tew, G. N. Macromolecular Rapid Communications , 31 , 784-793. (11) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966, 5, 622-625. (12) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1962, 341-350. (13) Henderson, I. M.; Hayward, R. C. Polymer. (14) Constable, E. C.; Baum, G.; Bill, E.; Dyson, R.; van Eldik, R.; Fenske, D.; Kaderli, S.; Morris, D.; Neubrand, A.; Neuburger, M.; Smith, D. R.; Wieghardt, K.; Zehnder, M.; Zuberbuhler, A. D. Chemistry-a European Journal 1999, 5, 498-508. 125

(15) Aihara, M.; Kishita, H.; Misumi, S. Bulletin of the Chemical Society of Japan 1975, 48 , 680-683. (16) Mugemana, C.; Guillet, P.; Hoeppener, S.; Schubert, U. S.; Fustin, C.-A.; Gohy, J.-F. Chemical Communications , 46 , 1296-1298. (17) Constable*, E. C.; Housecroft*, C. E.; Neuburger, M.; Schneider, A. G.; Springler, B.; Zehnder, M. Inorganica Chimica Acta 2000, 300-302 , 49-55. (18) Chan, Y.-T.; Li, X.; Yu, J.; Carri, G. A.; Moorefield, C. N.; Newkome, G. R.; Wesdemiotis, C. Journal of the American Chemical Society , 133 , 11967-11976. (19) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1965, 4, 929-&. (20) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1965, 4, 929-935. (21) Mechanisms of Inorganic Reactions ; Amercian Chemical Society, 1965. (22) Dimitris, K.; Gilbert, G. Mechanisms of Inorganic Reactions ; John Wiley and Sons: New York, 1987; Vol. 1. (23) Chattopadhyay, P. K.; Coetzee, J. F. Inorganic Chemistry 1973, 12 , 113- 120. (24) Wilkins, R. G. Accounts of Chemical Research 1970, 3, 408-416. (25) Seiden, L.; Basolo, F.; Neumann, H. M. Journal of the American Chemical Society 1959, 81 , 3809-3813. (26) Van Meter, F. M.; Neumann, H. M. Journal of the American Chemical Society 1976, 98 , 1388-1394. (27) Cossy, C.; Helm, L.; Merbach, A. E. Helvetica Chimica Acta 1987, 70 , 1516-1525. (28) Funahashi, S.; Jordon, R. B. Inorganic Chemistry 1977, 16 , 1301-1306. (29) Orland, W. K. Transactions of the Kansas Academy of Science (1903-) 1978, 81 , 303-309. (30) Vigee, G. S.; Ng, P. Journal of Inorganic & Nuclear Chemistry 1971, 33 , 2477-&. (31) Ducommun, Y.; Newman, K. E.; Merbach, A. E. Inorganic Chemistry 1980, 19 , 3696-3703. (32) Sisley, M. J.; Yano, Y.; Swaddle, T. W. Inorganic Chemistry 1982, 21 , 1141-1145. (33) Swift, T. J.; Connick, R. E. Journal of Chemical Physics 1962, 37 , 307-&. (34) Matwiyoff, N. A. Inorganic Chemistry 1966, 5, 788-795. (35) McAteer, C. H.; Moore, P. Journal of the Chemical Society, Dalton Transactions 1983, 353-357. (36) Bakker, W. I. I.; Haas, M.; Denhertog, H. J.; Verboom, W.; Dezeeuw, D.; Reinhoudt, D. N. Journal of the Chemical Society-Perkin Transactions 2 1994, 11-14. (37) Cox, B. G.; Garciarosas, J.; Schneider, H. Journal of the American Chemical Society 1981, 103 , 1384-1389. (38) Lewandowski, A.; Malinska, J. New Journal of Chemistry 1996, 20 , 653- 657. (39) Rounaghi, G. H.; Mofazzeli, F. Journal of Inclusion Phenomena and Macrocyclic Chemistry 2005, 51 , 205-210.

126

(40) Thaler, A.; Cox, B. G.; Schneider, H. Inorganica Chimica Acta 2003, 351 , 123-132. (41) Schweitzer, G. K.; Lee, J. M. The Journal of Physical Chemistry 1952, 56 , 195-198. (42) Blandamer, M. J.; Burgess, J.; Cope, S. D.; Digman, T. Transition Metal Chemistry 1984 , 9, 347-350. (43) Mytides, P.; Rozou, S.; Benaki, D.; Antoniadou-Vyza, E. Analytica Chimica Acta 2006, 566 , 122-129.

127

CHAPTER 5

HETEROLEPTIC IRON(II) COMPLEXES OF 4’-SUBSTITUTED

TERPYRIDINES: SYNTHESIS, KINETIC STABILITY, AND REDOX

PROPERTIES

5.1 Chapter Overview

In this study we have developed a convenient synthetic method of producing a heteroleptic bis-terpyridine(terpy) complexes of iron(II).. Complexes were able to be synthesized with >95% asymmetry, and the kinetic stability and redox properties of the complexes were compared with symmetric complexes bearing the same substiutent groups. Kinetic studies demonstrated a strong dependence of kinetic stability on the electron withdrawing chloro substitutent, with both the symmetric complexes decaying more rapidly than the unsubstituted complex, the symmetric chloro-substituted complex decaying the most rapidly. The n-propylamino (npa) substitutents were found to impart only minor effects on the kinetic stability. Redox potentials were found to be nearly linearly dependant on both the chloro and npa groups, with the Fe(II/III) couple becoming more positive with each chloro group, and more negative with each amino group, as compared to the unsubstituted complex.

5.2 Introduction

Terpyridine ligands and their metal complexes have gained a fair amount of notoriety since 2,2’:6’,2”-terpyridine was first synthesized by Morgan and Burstall in

1932. 1 As a tridentate ligand, terpyridine is predisposed toward forming complexes of

128

high thermodynamic and kinetic stability with a variety of transition metals. 2-4 In

addition, the straightforward synthesis of 4’-chloro-2,2’:6’,2”-terpyridine and its S NAr

functionalization with various nucleophiles have resulting in the synthesis of a wide

variety of substituted terpyridines, thus popularizing the use of the ligand in polymer

functionalization and the production of self-assembled supramolecuar structures. 5-8

One particularly interesting facet of terpyridine research is the formation of

heteroleptic or asymmetric complexes with certain transition metals. Early studies were

conducted by synthesizing a terpyridyl mono -complex with osmium(III) or

ruthenium(III) followed by reduction in the presence of a phosphine or substituted

terpyridine ligand, resulting in the mixed-ligand ruthenium(II) or osmium(II) complex. 9-12

While the original asymmetric ruthenium complexes were investigated mainly for

their photoelectric properties, the same synthetic techniques have been applied to the

creation of block copolymers, 13-15 as well as polymers with side-chain heteroleptic

complexes. 16,17 The ease of synthesis of these compounds and the stability of the

resultant Ru(II) complex have made this technique a practical way of producing

controlled structures with supramolecular linkages. Recently, Schubert et. al expanded

this concept to the include block copolymer synthesized using terpyridine-functionalized

polymers complexed with cobalt(III) and nickel(II). 18

With the interest in asymmetric terpyridine complexes, particularly those

formed with d 6 metals such as ruthenium(II), osmium(II), and cobalt(III), it is surprising that more work has not been done on the synthesis of asymmetric iron(II) terpyridine complexes. We are aware of only a few studies to date on asymmetrical iron-based terpyridine complexs. 19-21 In all the cases there is a high incidence of symmetric complex

129

present as byproducts in the products of these reactions, thus purification via column was

necessary to obtain the desired product, and isolated yields were not given. A simple

and high-yield technique for the synthesis of asymmetric iron-terpyridine complexes

would be beneficial as a new way to facilitate directed formation of supramolecular

structures, and will provide further insight into how substituent groups affect the

properties of terpyridine complexes.

In this study, we have demonstrated a simple and effective way of synthesizing

asymmetric terpyridine-iron(II) complexes, using the technique to create complexes with

either one electron donating or electron withdrawing group. We have used 4’-chloro-

2,2’;6’,2”-terpyridine (cl-terpy) for the ligand with the electron withdrawing group, and

4’-N-(n-propylamino)-2,2’;6’,2”-terpyridine (npa-terpy) as the ligand with an electron donating group. Additionally, we have studied the kinetic stability, redox properties, and spectrophotometric properties of these heteroleptic complexes as compared to their symmetric counterparts.

5.3 Experimental

• Materials

Iron (II) chloride tetrahydrate (99+%), nickel (II) chloride hexahydrate (99.9%),

4’-chloro-2,2’;6’2”-terpyridine (99%), tetrabutylammonium hexaflourophosphate, (98%) and d 3-acetonitrile (99% D, 1% v/v TMS) were obtained from Acros Organics.

Acetonitrile (optima grade) and acetone (ACS grade) were obtained from Fisher

Scientific. 2,2’;6’,2”-terpyridine (99%) was obtained from Strem Chemical and ethanol

(200 proof) was obtained from Alco-AAPER. All chemicals were used as recieved

without further purification.

130

• Instrumentation

COSY and 1H NMR spectra were recorded on a Bruker DPX 300 MHz spectrometer, and all UV-Vis experiments were conducted on a Hitachi model U-3010 spectrophotometer. Decay scans were conducted at a 70,000 s scan time in 1:1 v/v acetonitrile/ethanol, the wavelength being set at the peak of MLCT band of the complex being tested. Wavelength scans were conducted in acetonitrile in a glass cuvette with a sealed top. A solution of the complex (0.5 mM, 1.5 mL) in acetonitrile was mixed with a solution of nickel (II) chloride hexahydrate (20 mM, 1.5 mL) and quickly transferred to the spectrophotometer, where the decay of the complex was measured at the peak of its

MLCT band. The decay was measured over a 70,000 s scan time. Measurement of the absorption spectrum was done on the same spectrophotometer in acetonitrile using quartz cuvettes at concentration of 0.025 mM.

Cyclic voltometry was conducted using a BASi Epsilon electrochemical workstation with platinum working and auxiliary electrodes and an Ag/AgCl reference.

Measurements were conducted on acetonitrile containing 100 mM of tetrabutylammonium hexaflourophosphate as an electrolyte and 10 mM of the terpyridine complex.

• Synthetic Procedures

4’-N-(n-propylamino)-2,2’;6’,2”-terpyridine (npa-terpy) was prepared by a

22 23 previously published procedure, as was Fe(terpy)Cl 3. Other compounds were synthesized as follows. The NMR spectra of the complexes were assigned using COSY

131

1H NMR and the data determined from previous studies. 24 For asymmetric compounds, the proton shifts of the substituted ligand are recorded in bold type.

Preparation of Fe(Cl-terpy) 2[PF 6]2. (1) Cl-terpy (0.192 g) was dissolved in 10 mL of hot acetone, to which a solution of 0.073 g FeCl 2x 4H 2O in 3 mL of water was added.

The resultant reddish-purple solution was added to 100 mL of water, which was stirred rapidly as 1.0 g of ammonium hexaflourophosphate was added, resulting in the precipitation of the compound. The mixture was stirred for 10 min before filtering the precipitate. The filtrate was allowed to dry in air overnight before the red-purple powder was washed with diethyl ether (200 mL). 0.217g recovered (68.7%). 1H NMR (300

3 MHz, acetonitrile-d6): δ 9.00 (s, 4H, H3’ ,H5’ ), 8.45 (d, 4H, J = 7.91 Hz, H3,H3” ), 7.90 (t,

3 -1 -1 4H, J = 6.41, H4,H4” ), 7.12 (m, 8H, H5,H5” ,H6,H6” ). UV-Vis λmax /nm ( ε, M cm )

274(47,090), 281(42,981), 317(52,109), 361(6,145), 512(8,472), 557(15,054)

Preparation of Fe(terpy)(Cl-terpy)[PF 6]2. (2) Fe(terpy)Cl 3 (0.131 g) was dissolved in a mixture of acetone and water (4 mL acetone/6 mL water) in a 20 mL vial, and stirred until dissolution of the solid, resulting in a dark brown solution. The vial was then fitted with a septum and cooled in an ice/water bath for 15 min. Cl-terpy (0.088 g) and ascorbyl palmitate (0.21 g) were dissolved in 10 mL of acetone and cooled over the course of 10 minutes in an ice/acetone bath, as the solution was stirred. The cooling caused some clouding of the solution due to the precipitation of the ascorbyl palmitate.

After cooling, the solution of Fe(terpy)Cl 3 was loaded into a dropping funnel which was fitted with an ice/water bath. The solution of Fe(terpy)Cl 3 was then added dropwise to the rapidly stirring solution of Cl-terpy and ascorbyl palmitate. At the end of this

132 addition, the resultant reddish-purple solution was added to 150 mL of rapidly stirring water. Ammonium hexaflourophosphate (2.0 g) was added, resulting in the precipitation of the complex. The mixture was allowed to stir for 10 min before filtration of the solid.

The reddish-purple solid was then dried in air overnight, after which it was washed with

200 mL of diethyl ether. Residual ascorbyl palmitate was then removed from the product by sonication of the complex in 20 mL diethyl ether. 0.112 g recovered (41%). 1H

3 NMR (300 MHz, acetonitrile-d6): 9.02 (s, 2H, H3’ ,H5’ ), 8.92 (d, 2H, J = 8.10 Hz,

3 3 H3’ ,H 5’ ), 8.71 (t, 1H, J = 7.91 Hz, H4’ ), 8.48 (d, 4H, J = 7.54 Hz, H3,H 3” ,H3,H3” ), 7.91

3 (t, 4H, J = 7.35 Hz, H4,H 4” ,H4,H 4” ), 7.10 (m, 8H, overlapped signals, H5,H 5’ , H-

+ + + 6,H 6’ ,H5,H 5” ,H 6,H 6” ) FAB/MS m/z 701.1 ( (2) - PF 6), 591.5 ( (2) - 2PF 6 + Cl), 575.0 ( (2)

+ + - 2PF 6 + F), 324.0 ( (2) - Cl-Terpy – 2PF 6 + Cl), 309.0 ( (2) - Cl-Terpy – 2PF 6 + F),

2+ -1 -1 278.0 ( (2) - 2PF 6) UV-Vis λmax /nm ( ε, M cm ) 273(43,345), 280(38,509),

318(55,454), 370(5,127), 512(7,054), 555(13,527)

Preparation of Fe(terpy) 2[PF 6]2. (3) The method for synthesizing this compound was

22 1 3 recorded in a previous study. H NMR (300 MHz, acetonitrile-d6): δ 8.91 (d, 4H, J =

3 3 7.91 Hz, H3’ ,H 5’ ), 8.68 (t, 2H, J = 8.10 Hz, H4’ ), 8.47 (d, 4H, J = 7.91 Hz, H3,H 3” ), 7.88

(m, 4H, Hz, H4,H 4”), 7.07 (m, 8H, overlapped signals, H5,H 5’ , H 6,H 6’ ). UV-Vis λmax /nm

(ε, M -1cm -1) 272(25,963), 280(22,727), 319(34,372), 374(2,945), 552(7,200)

Preparation of Fe(terpy)(npa-terpy)[PF 6]2. (4) Fe(terpy)Cl 3 (0.131 g) was dissolved in a mixture of acetone and water (4 mL acetone/6 mL water) in a 20 mL vial, and stirred until dissolution of the solid, resulting in a dark brown solution. The vial was then fitted with a septum and cooled in an ice/water bath for 15 min. 4′-(N-n-propylamino)-

2,2 ′:6 ′,2 ′′ -terpyridine (0.084 g) and ascorbyl palmitate (0.20 g) were dissolved in 10 mL

133

of acetone and cooled over the course of 10 min in an ice/acetone bath, as the solution

was stirred. Once again, precipitation of the ascorbyl palmitate occurred. The

Fe(terpy)Cl 3 was then loaded into the cooled dropping funnel. The solution of

Fe(terpy)Cl 3 was then added dropwise to the rapidly stirring solution of Cl-terpy and

ascorbyl palmitate. At the end of this addition, the resultant dark-purple solution was

added to 150 mL of rapidly stirring water. After stirring 10 min, the solution was run

though a vacuum filter, in order to remove precipitated ascorbyl palmitate. Ammonium

hexaflourophosphate (2.0 g) was then added, resulting in the precipitation of the

complex. The dark-purple solid was then dried in air overnight, after which it was

washed with 200 mL of diethyl ether. 0.182 g recovered (68.0%). 1H NMR (300 MHz,

3 3 acetonitrile-d6): δ 8.87 (d, 2H, J = 8.10 Hz, H3’ ,H 5’ ), 8.60 (t, 1H, J = 7.91 Hz, H4’ ), 8.46

3 3 (d, 2H, J = 7.91, H3,H 3” ), 8.35 (d, 2H, J = 7.91 Hz, H3,H 3” ), 8.06 (s, 2H, H3’ ,H5’ ), 7.86

3 (m, 4H, overlapping peaks, H4,H 4” ,H4,H 4” ), 7.28 (d, 2H, J = 4.90, ,H6,H 6” ), 7.15 (t, 2H,

3 3 J = 6.40 Hz, ,H5,H 5” ), 7.03 (m, 4H, H5,H 5” ,H 6,H 6” ), 6.56 (s, 1H, N-H ), 3.65 (q, 2H, J =

3 6.40 Hz, NH-CH 2-CH 2- CH 3 ), 1.29 (m, 2H, NH-CH 2-CH 2- CH 3 ), 1.21 (t, 3H, J = 7.16

-1 -1 Hz NH-CH 2-CH 2- CH 3) UV-Vis λmax /nm ( ε, M cm ) 274 (64,618), 281(63,890),

306(44,181), 319(47,454), 375(7,309), 502(10,981), 565(15,927)

Preparation of Fe(npa-terpy) 2[PF 6]2. (5) 4′-(N-n-propylamino)-2,2 ′:6 ′,2 ′′ -terpyridine

(0.0476 g) was dissolved in 2 mL of hot ethanol. To this solution was added 0.0157 g of

FeCl 2x4H 2O in 6mL of water, resulting in the formation of the dark purple complex. The

solution was stirred for ten minutes before 0.4 g of NH 4PF 6 were added, resulting in

precipitation of the dark-purple complex. The product was then filtered and allowed to dry in air

before being washed with 150 mL of diethyl ether. 0.471 g recovered (75.5%). 1H NMR (300

3 MHz, acetonitrile-d6): δ 8.33 (d, 2H, J = 7.91 Hz, H3,H 3” ), 8.03 (s, 2H, H3’ ,H 5’ ), 7.83 (t, 134

3 3 3 2H, J = 7.54 Hz, H4,H 4” ), 7.27 (d, 2H, J = 5.09 Hz, H6,H 6” ), 7.09 (t, 2H, J = 6.22 Hz, H-

3 5,H 5’ ), 6.43 (s, 1H, N-H), 3.63 (q, 2H, J = 6.03 Hz, NH-CH 2-CH 2- CH 3), 2.18 (m, 2H,

3 NH-CH 2-CH 2- CH 3), 1.20 (t, 3H, J = 7.35 Hz, NH-CH 2-CH 2- CH 3) UV-Vis λmax /nm ( ε,

M-1cm -1) 273(49,417) 281(61,745), 292(49,236) 303(39,636), 375(6,145), 517(9,272),

570(12,945)

5.4 Results and Discussion

• Compound Synthesis

Synthesis of the asymmetric complexes was successfully carried out by reduction of the mono -terpyridine iron(III) complex with ascorbyl palmitate. A solution of the iron(III) complex in 2:3 acetone/water was added dropwise to a stirring solution of of the second ligand in acetone. Initially, the synthesis was conducted at room temperature, however this resulted in a high incidence of the bis -terpyrine iron(II) complex.

Presumably this was due to the disproportionation 25 of the newly-formed iron(II) mono - complex before reaction with the second terpyridine ligand could occur.

3+ 2+

O OH O N N O N N H O O 2 HO OH N Fe H2O + N R N Fe N R H2O 3:2 acetone/water acetone/ice bath N N N N

Scheme 5.1 General Schematic for the Synthesis of Asymmertic bis -

terpyridyl Iron Complexes

To suppress this side reaction, the synthesis was conducted at lower temperatures.

In order to accomplish this, the solution of mono -terpyridine iron(III) and the ascorbyl

135

palmitate/substituted terpyridine solution were cooled in a water/ice bath for at least 15

minutes prior to use and during dropwise addition. The reaction is shown in Scheme 5.1.

Ascorbyl palmitate was chosen as ascorbic acid precipitated upon cooling. Although

there was some clouding of the acetone solution from precipitation of the ascorbyl

palmitate, the resulting products were found to be nearly totally asymmetric, as opposed

to the products synthesized with ascorbic acid, which contained appreciable amounts of

the bis -terpyridine complex under the same conditions.

The fraction of the desired asymmetric complex was determined by NMR, fast

atom bombardment mass spectrometry (FAB-MS), and thin layer chromatography

(TLC). In the NMR spectrum, the integration of the integrated intensity single 4’ proton

peak was compared to the integrations for the other pyridyl protons, showing a 1:1 ration

of the unsubstituted ligand to the substituted ligand. While not a definite measure of the

percentage of the heteroleptic complex in the reaction mixture, NMR provided

circumstantial evidence of the asymmetry of the complex. TLC of the compounds was

performed on silica plate using acetonntrile which had been washed with a saturated

sodium nitrate solution, a modification of a previously used eluent. 26 While not quantitative, the TLC characterization of these compounds was effective for identifying symmetrical impurities in the product.

Mass spectroscopy was perhaps the most effective technique for the characterization of the asymmetrical complexes. The fast atom bombard technique was used due to its suitability for use with metal-polypyridine complexes. 27-29 The degree of

asymmetry in the complex was determined from the peaks representing the complex

- minus one PF 6 counterion. Using FAB mass-spec, with qualitative supporting evidence

136

from TLC, it was determined that complexes (2) and (4) were >97% heteroleptic. The

NMR spectra of the compounds are shown in Figure 5.1.

Figure 5.1 NMR spectra of the Iron(II) Complexes in CD 3CN. The purity of theasymmetric complexes were determined from the proton peak from the 4’ position of unsubstituted terpyridine centered around 8.60-8.70 ppm and the 3’/ 5’ proton peak centered around 9.02 ppm for the chloro substituted terpyridines and 8.03 ppm for the amino-substituted terpyridines.

• Electronic Spectra

The optical absorbance spectra of the complexes were recorded at a concentration of 0.0275 mM in acetonitrile, and are shown in Figure 5.2. The spectrum of the unsubstituted complex (3) agreed with previously published. 30 Complexes (1), (2) , and

(3) displayed three main peaks in the range 250-300nm, with the two peaks in the range

275-285nm being overlapped.

137

Figure 5.2 UV-Vis Spectrum of Terpyridine Complexs investigated in Chapter 5.

For the disubstituted amino complex (5), there was no peak in the 310-320 nm, there was however, a strong absorption at 281nm, with shoulders at 273nm, 292nm, and

303nm. The asymmetric amino-substituted complex (4) did have overlapping absorptions at 306nm 319nm, but it was weakened in intensity as compared to the complexes previously mentioned.

• Cyclic Voltammetry

The cyclic voltammery data obtained in acetonitrile for the complexes is shown in Table

5.2. The values recorded for (1) and (3) are in good agreement with those previously published, while the values for (5) are near those published for similar amino-substituted complexes. 31 The peak separation of all complexes was within the range of 60-80 mV,

138

which is characteristic of reversible behavior and is in good agreement with literature

results on similar complexes under comparable conditions. 21,32 Here, redox behavior of

the Fe(III/II) couple strongly depend on the type and number of substituents present. For

the chloro-substitited complexes (1) and (2) E1/2 was determined to be 1.196V and

1.167V, respectively. The differences in the Fe(III/II) from the unsubstitued complex (3)

(E 1/2 = 1.137V) were E1/2 = 0.059 for complex (1) , and E1/2 = 0.031 for complex (2) .

Interestingly, each chlorine substiutent shifts the redox potential by approximately 0.03V.

For the amino-substituted complexes (4) and (5) , E 1/2 was found to be 0.889V and

0.713V, respectively. Relative to that of (3) E1/2 = -0.248V for complex (4) , and E1/2

= -0.424V for complex (5) . This corresponds to a E1/2 per amino group was found to be

-0.21V, which is in good agreement with the value of -0.18V previously published for diethylamino substitutents. 21 However, in this case the second substiuent has less of an effect than the first, suggesting an interaction between the two ligands which is absent in the chlorine-substituted complexes. The value obtained for propylamino substituents is in good agreement with the value of From the data gathered, it is shown that electron- withdrawing chloro groups systematically drive the Fe(III/II) couple to more positive values, while the electron-donating amino groups systematically drove E 1/2 to more negative values. The stepwise dependence of the Fe(III/II) couple on the substituent groups provides further evidence for complex asymmetry.

139

Table 5.1 Electrochemical Properties of the Terpyridine Complexes Synthesized in

Chapter 5. The E 1/2 values listed are for the Fe(III/II) couple. As can be determined from the table, each chloro-substitutent adds approximatly 30mV to the redox potential while each amino substitutent subtracts approximatly 23mV.

Complex R1 R2 E1/2 (V) E 1/2 (V) (1) Cl Cl 1.196 0.059 (2) H Cl 1.167 0.0305 (3) H H 1.137 0 (4) H -NH 2(CH 2)2CH 3 0.889 -0.248

(5) -NH 2(CH 2)2CH 3 -NH 2(CH 2)2CH 3 0.713 -0.424

• Kinetics Measurements

The kinetic scans of all compounds were tested by measuring the disappearance of the MLCT bands at 550-570 nm over 70,000 s in 1:1 acetonitrile/ethanol, using

NiCl 24(H 2O) as the decomposing metal ion. The scans were fitted with a mono- exponential decay for the full duration of the experiment. The results of the kinetics experiments are shown in Table 5.1

Table 5.2 Kinetic Stability of the Complexes Synthesized in this Chapter 5. It can be clearly determined from this data that the electron-withdrawing chloro-groups have a large effect on stability, with each chloro group speeding the decay of the bis complex significantly. The amino groups, on the other hand, do not have much effect on kinetic stability. -1 5 Complex R1 R2 k-2 s * 10 (1) Cl Cl 46.8 (2) H Cl 16.6 (3) H H 2.26

(4) H -NH 2(CH 2)2CH 3 5.50

(5) -NH 2(CH2)2CH 3 -NH 2(CH 2)2CH 3 2.30

As can be surmised from Table 5.1 and the decay scans shown in Figure 5.3, the

electron-withdrawing chlorine substituent groups had a significant effect on the decay

rate, as compared to the unsubstituted complex. As expected based on the electron-

140

withdrawing nature of the chloro group, the dichloro complex (1) decayed the most

rapidly, at 4.68*10 -4 s-1, while the monochloro complex (2) decayed less quickly, at

1.66*10 -4 s-1. The unsubstituted complex (3) , as was measured in our previous study, 22

decayed at a rate of 2.26*10 -5 s-1, nearly a full order of magnitude more slowly than complex (2) .

Figure 5.3. Kinetic Stability of Terpyridine complexes. The bis compelex of cl-terpy (1) clearly shows the most rapid decay of the complexes measured, closely followed -4 -1 - by the terpy/cl-terpy mixed-ligand complex (2) at k-2 = 4.86*10 s and k-2 = 1.68*10 4 s-1, respectively. The bis -terpyridine complex (3) decays nearly an order of -5 -1 magnitude more slowly that (2) ( k-2 = 2.26*10 s ) , with a very similar decay rate to -5 -1 the bis -npa-terpy complex (5) ( k-2 = 2.30*10 s ). For currently unknown reasons the npa-terpy.terpy mixed-ligand complex (4) decays about twice as rapidly as (3) -5 -1 and (5) ( k-2 = 5.50*10 s ).

The monoamino substituted complex (4) decayed nearly two times more rapidly than the unsubstituted complex, at a rate of 5.11*10 -5 s-1, as would be expected based on our previous studies. 22 Unexpectedly, the diamino complex (5) decayed at nearly the same rate as the unsubstituted complex (3) , at a rate of 2.30*10 -5 s-1. While in water it appears that an overabundance of electron density imparted by the amino groups

141

decreased stability of terpyridine-based complexes of iron(II). It would be expected that

the amino groups would have a large effect on the decay kinetics based on the CV scans

of the complexes, show that the amino groups donate electron density to complex to a

greater magnitude than the chloro groups withdrawal. The current study suggests that

effect of substituent groups on complex stability is solvent dependant, as solvent

interaction has been shown to have a dramatic effect on the decay rate of terpyridine

complex (see Chapter 4). The amino substitutents most likely change how the solvent

interacts with the ligands, thereby altering the decay rate.

5.5 Conclusions

In this study, it has been shown that asymmetric bis-terpyridine iron(II)

complexes can be readily synthesized by in situ reduction of [terpy]FeCl 3 in the presence of a substituted terpyrine ligand at lowered temperatures. Using this technique, we were able to synthesize monochloro and monoamino substituted bis-terpyridinyl complexes, the asymmetry of which were proven using NMR and FAB-mass spec, along with qualitative evidence gained from TLC.

In order to determine the effect of the substituent groups on the kinetic stability of asymmetric compounds the complexes where decomposed with NiCl 26H 2O in 1:1 v/v

acetonitrile/ethanol. It was found that the electron withdrawing chloro-groups decreased

stability as compared to the unsubstituted complex, with the dichloro complex decaying

the most rapidly. For the amino-substituted complexes, the monosubstituted complex

decayed more rapidly than the unsubstituted complex, while the disubstituted complex

was as stable as the unsubstituted complex. Cyclic voltammetry scans of the complexes

142

showed a sequential increase in the voltage of the Fe(III/II) couple with each electron-

withdrawing chloro-substitution, while each electron-donating amino group lowered the

voltage of the couple.

In summation, we have described a technique to synthesize asymmetric bis - terpyridine iron(II) complexes with a high purity and good yield. The physical properties of these complexes have shed new light on the role of functional group substitution on metal-ligand complexes. It is hoped that this study will lay the groundwork for controlled hierarchical synthesis of more advanced metal-ligand structures.

5.6 References

(1) Morgan, G. T.; Burstall, F. H. Journal of the Chemical Society 1932 , 20- 30. (2) Melson, G. A.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1963 , 2662-2672. (3) Sillen, S. L.; Martell, A. E.; Bjerrum, J. Stability Constants of Metal-Ion Complexes ; 2 ed.; Chemical Society: London, 1964; Vol. 17. (4) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-625. (5) Schwarz, G.; Bodenthin, Y.; Geue, T.; Koetz, J.; Kurth, D. G. Macromolecules 2009 , 43 , 494-500. (6) Guillet, P.; Fustin, C.-A.; Wouters, D.; Hoeppener, S.; Schubert, U. S.; Gohy, J.-F. Soft Matter 2009 , 5. (7) Shunmugam, R.; Gabriel, G. J.; Aamer, K. A.; Tew, G. N. Macromolecular Rapid Communications , 31 , 784-793. (8) Aamer, K. A.; Tew, G. N. Macromolecules 2007 , 40 , 2737-2744. (9) Sullivan, B. P.; Calvert, J. M.; Meyer, T. J. Inorganic Chemistry 1980 , 19 , 1404-1407. (10) Allen, G. H.; Sullivan, B. P.; Meyer, T. J. Journal of the Chemical Society, Chemical Communications 1981 , 793-795. (11) Collin, J. P.; Guillerez, S.; Sauvage, J. P. Journal of the Chemical Society- Chemical Communications 1989 , 776-778. (12) Collin, J. P.; Guillerez, S.; Sauvage, J. P. Inorganic Chemistry 1990 , 29 , 5009-5010. (13) Gohy, J.-F. o.; Lohmeijer, B. G. G.; Varshney, S. K.; Décamps, B.; Leroy, E.; Boileau, S.; Schubert, U. S. Macromolecules 2002 , 35 , 9748-9755.

143

(14) Lohmeijer, B. G. G.; Schubert, U. S. Angewandte Chemie International Edition 2002 , 41 , 3825-3829. (15) Lohmeijer, B. G. G.; Schubert, U. S. Journal of Polymer Science Part A: Polymer Chemistry 2003 , 41 , 1413-1427. (16) Gohy, J.-F.; Lohmeijer, B. G. G.; Schubert, U. S. Macromolecular Rapid Communications 2002 , 23 , 555-560. (17) Aamer, K. A.; De Jeu, W. H.; Tew, G. N. Macromolecules 2008 , 41 , 2022-2029. (18) Mugemana, C.; Guillet, P.; Hoeppener, S.; Schubert, U. S.; Fustin, C.-A.; Gohy, J.-F. Chemical Communications , 46 , 1296-1298. (19) Hanabusa, K.; Nakamura, A.; Koyama, T.; Shirai, H. Die Makromolekulare Chemie 1992 , 193 , 1309-1319. (20) Gagliardo, M.; Perelaer, J.; Hartl, F.; van Klink, G. P. M.; van Koten, G. European Journal of Inorganic Chemistry 2007 , 2007 , 2111-2120. (21) Hathcock, D. J.; Stone, K.; Madden, J.; Slattery, S. J. Inorganica Chimica Acta 1998 , 282 , 131-135. (22) Henderson, I. M.; Hayward, R. C. In Preparation. (See Chapter 4). (23) Cotton, S. A.; Franckevicius, V.; Fawcett, J. Polyhedron 2002 , 21 , 2055- 2061. (24) Pazderski, L.; Pawlak, T.; Sitkowski, J.; Kozerski, L.; Szlyk, E. Magnetic Resonance in Chemistry , 49 , 237-241. (25) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society 1962 , 341-&. (26) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; Thompson, A. Inorganic Chemistry 1995 , 34 , 2759-2767. (27) Liang, X.; Suwanrumpha, S.; Freas, R. B. Inorganic Chemistry 1991 , 30 , 652-658. (28) Strouse, G. F.; Anderson, P. A.; Schoonover, J. R.; Meyer, T. J.; Keene, F. R. Inorganic Chemistry 1992 , 31 , 3004-3006. (29) Denti, G.; Serroni, S.; Sindona, G.; Uccella, N. Journal of The American Society for Mass Spectrometry 1993 , 4, 306-311. (30) Braterman, P. S.; Song, J. I.; Peacock, R. D. Inorganic Chemistry 1992 , 31 , 555-559. (31) Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.; Slattery, S. J. Inorganica Chimica Acta 2006 , 359 , 2400-2406. (32) Ayers, T.; Turk, R.; Lane, C.; Goins, J.; Jameson, D.; Slattery, S. J. Inorganica Chimica Acta 2004 , 357 , 202-206.

144

CHAPTER 6

TUNABLE ORGANOGELS UTILIZING TERPYRIDINE-BASED

CROSSLINKS

6.1 Overview

In this chapter we discuss the synthesis and rheological properties of poly( N- isopropylacrylamide) (poly(NIPAAm)) organogels with supramolecular crosslinks composed of terpyridine complexes of copper, cobalt, nickel, and iron. The gels were synthesized by dissolving poly(NIPAAm) with pendent terpyridine groups in DMF followed by the addition of the metal salt complexed with either bipyridine or phenanthroline as a sacrificial ligand. As the complexes of the sacrificial ligand dissociated, the metal was bound to the more strongly binding terpyridine ligands, thus forming the crosslinking terpyridine bis complexes. The frequency and stress relaxation properties of the gels could be altered dramatically by variation of the crosslinking metal.

Interestingly, the characteristic bulk relaxation rate β was found to be several orders of magnitude lower than the measured or expected dissociation rates of the free bis complexes in solution. This phenomenon is attributed to cooperative binding of the ligands, combined with the limited availability of ligand with which to exchange.

6.2 Introduction

Physical gels rely on a wide variety of interactions to form crosslinks, including hydrogen bonding 1, aggregation, 2 and metal-ligand interactions 3. Gels crosslinked via metal-ligand interactions are particularly interesting due to the wide range in kinetic and

145

thermodynamic properties obtainable with metal-ligand complexes. 4 One example of the

wide variety of obtainable properties is the work of the Craig research group, in which

crosslinking via pincer complexes were tuned by varying the metal ions complexed and

the steric hindrance provided by substituent groups on the ligand, both are factors that

influence exchange kinetics of the ligands. 4,6 Additionally Craig and coworkers have

studied how, the bulk rheological properties were determined by the exchange rates of the

ligands. 7 In this study, the bulk values were the same order of magnitude as the exchange rates of the free complex with a nearly linear dependence on the inverse of the relaxation rate on crosslinker concentration. Another interesting series of studies conducted by

Rowan et al. involves the formation of gels from telechelic ligand-functionalized oligo- ethlylene oxide. The gels were formed by mixing the telechelic oligo-ethylene oxide species with Zn(II) or Co(II), which formed a supramolecular polymer through bis complexation, and a small amount of La(III) or Eu(III), which formed crosslinking tris compexes. 8-12 Tris metal-ligand complexes were also employed by Fraser et al. in the formation of PEO-based hydrogels with iron(II) tris -bipyridine crosslinks. 13

In terms of tunability, terpyridine complexes represent ideal candidates for the construction of supramolecular crosslinks. Terpyridine forms bis complexes with a variety of transition metals, and the properties of these complexes show a wide variability of physical properties depending on the metal. In a series of kinetics studies, Wilkins and coworkers determined the kinetic stabilities of bis complexes of terpyridine varied from

10 -7 s-1 for iron(II) complexes to exchange rate greater than 10 s -1 for Copper(II). 14 In

Chapter 3, we have determined that substitutions in the 4’ positions of the ligand can have a large effect on the stability of terpyridine complexes, as well as the solvent in

146

which the complex is dissolved, which is discussed in Chapter 4. The variation of these

conditions provides means to modify the kinetic stability of the complexes, and thereby

tunig the bulk properties of the resulting gels. 15,16

In this study we have synthesized poly(NIPAAm) gels with crosslinks formed from bis complexes of terpyridine with Fe(II), Co(II), Ni(II), and Co(II).

Poly(NIPAAm) with randomly incorporated terpyridine pendent groups were gelled by addition of the metal in DMF. Shear rheology and stress relaxation studies revealed a wide variety of properties by variation of the metal ion. The gels exhibited a range of bulk relaxation rates from 10 -5 s-1 to 10 2 s-1, depending on the metal used to form the gel.

It was also found that the bulk relaxation rates of the gels were much slower than the decay rate of the free bis complexes in solution, indicative of gels which follow the limited free association model.

6.3 Experimental

• Materials and Methods

Spectrometery grade N,N’-Dimethylformamide (DMF) and HPLC water were obtained from Fisher Scientific. Acryloyl chloride (97%) and N-isopropyl acrylamide

(97%) were obtained from Sigma-Aldrich. Iron(II) chloride tetrahydrate (99+%) 2-

(methylamino)ethanol (99%), Copper(II) bromide (99%), and 4’-chloro-2,2’:6’,2”- terpyridine (99%), and Nickel(II) chloride hexahydrate (99.9999%) were obtained from

Acros Organics. Cobalt(II) chloride hexahydrate ( ≥98%) was obtained from Fluka, and

4’-(N-2-hydroxyethyl-N-methylamino)-2,2’:6’,2”-terpyridine was prepared as in our

previously study. All compounds were used as obtained without further purification.

147

All NMR spectra were recorded on a Bruker Avance 300 MHz spectrometer,

rheological measurements were performed on a TA instruments AR 2000 rheometer

using a 40 mm cone with a 2 o cone angle. All gels were measured at 100 m gap width.

Decay studies were completed using a Hitachi U-3010 spectrophotometer equipped with an Applied Photophysics RX2000 rapid mixing unit. The acryloyl-terpy complex was present at a 0.25 mM concentration, while the decomposing metal ion was present at a 10 mM concentration. The iron complex was decayed with Ni(II), and the signal at 570 nm

(MLCT band of the iron complex) was measured over 10 min, being fitted by a single exponential to obtain the decay constant. The cobalt complex was decayed using Fe(II), and the formation of the resultant iron(II) complex was tracked at 570 nm for 1 h. The first 60 s of these scans were fitted in order to establish the decay rate of the bis complex.

• Syntheitic Procedures

Synthesis of 4‘-(N-2-arcoyloxyethyl-N-methylamino)-2,2‘:6‘,2‘‘-terpyridine

(acryloyl-terpy). To an oven-dried 50 mL round bottom flask was added 1.45 g of 4’-

(N-2-hydroxyethyl-N-methylamino)-2,2’:6’,2”-terpyridine and a stir bar, and the flask was fitted with a septum and purged with nitrogen for 10 min. Dry THF (18 mL) was added to the flask, and the mixture was sparged with nitrogen for 10 min, while cooling in an ice/water bath. To the flask, 2.5 mL of acryloyl chloride was added via syringe, and the mixture was stirred for 10 min. At the end of this period of time, 2.5 mL of dry triethylamine was added though the septum via syringe, and the reaction was allowed to warm to room temperature, and stirred for 12 h. After this time, the yellow pale reaction mixture was added to 200 mL of water and stirred. The aqueous solution was adjusted to

148

pH 10 using aqueous sodium hydroxide, causing a white precipitate to form. The

precipitate was filtered and dried in vacuo , resulting in a white powder. The product was

determined to be of sufficient purity by NMR to be further synthetic steps. Yield: 0.70g

1 3 3 (41.0%) H NMR (CDCl 3): δ 8.70 (d, 2H, J = 4.90Hz, H6,H6” ), 8.64 (d, 2H, J =

3 3 7.91Hz, H3,H3” ), 7.86 (t, 2H, J = 7.10Hz, H4,H4” ), 7.85 (s, 2H, H3’,H5’ ), 7.32 (t, 2H, J =

2 3 3 5.91Hz), 6.39 (dd, 1H, J = 1.32Hz, J = 16.01Hz CH==C H2 cis), 6.10 (dd, 1H, J = 7.02

3 2 3 Hz, J = 10.48Hz, C H==CH 2), 5.79 (dd, 1H, J = 1.29Hz, J = 10.41Hz CH==C H2 trans),

3 3 4.47 (t, 2H, J = 5.61Hz, C H2OC(O)), 3.93 (t, 2H, J = 5.61Hz, CH2CH2OC(O)), 3.27 (s,

13 3H, NC H3) C NMR (CDCl 3): δ 166.04, 156.93, 155.85, 155.38, 148.85, 136.71,

131.33, 128.05, 123.48, 121.39, 103.87, 61.93, 50.14, 38.65

Synthesis of poly(NIPAAm) with pendent terpyridine groups. To a 50 ml RBF flask was added a stir bar 0.812 g of N-isopropyl acylamide, 0.120 g of aryloyl-terpy, and

0.004 g of AIBN. The flask was capped with a septum, and purged with nitrogen for 15 min, before 5 mL of dioxane was added and the mixture was sparged with nitrogen for a further 15 min. The solution was heated to 80 oC for seven hours under nitrogen while stirring. At the end of this time, the product was precipitated by dropwise addition to

300ml diethyl ether. The product was filtered, and allowed to dry in air. Yield: 0.806 g

1 (86.1%) H NMR (D 6-DMSO): δ 8.69 (d, 2H, ), 8.59 (d, 2H, ), 7.96 (t, 2H, ), 7.77 (t, 2H)

7.44 (t, 2H), 3.85 (broad, 19H), 1.96 (broad, 19H), 1.48 (broad, 38H), 1.04 (broad,

114H).

• Gel Synthesis

Gels were synthesized by dissolving 0.070 g of poly(NIPAAm) in DMF, followed by addition of 0.075 mL a 0.126 M aqueous solution of the complex of the metal

149

ion/sacrificial ligand complex. This gave 9.4*10 -6 moles of total metal in solution, as compared to 2.3*10 -5 moles of crosslinker in solution. Since formation of the bis complex requires that one metal ion binds two ligands, a theoretical maximum 82% of the ligands can be occupied. The sacrificial ligands were necessary to slow formation of the gel so that it could be loaded into the rheometer before gelation occurred. After the gel was initially mixed, it was sufficiently fluid to handle via syringe. Upon sitting, the terpyridine complex crosslinks formed via ligand exchange with the metal/sacrificial ligand complex. During this setup time, the gel was subjected to 12 h of pre-shear at 1

Hz. For cobalt and iron, phenanthroline was used as the sacrificial ligand. For nickel, bipyridine was used, as the tris phenanthroline complex was stable enough to inhibit gel formation for several days. The general synthesis of gels is shown in Scheme 6.1. The cobalt gels set rapidly even using sacrificial ligands, because of this, the viscous solution formed upon the addition of the cobalt complex was compressed between the cone and the plate of the rheometer and heated to 70 o C until the normal force registered by the rheometer was near zero and unchanging. After this, the gel was allowed the normal setup and pre-shear procedures.

150

Scheme 6.1 The Synthesis of Supramolecular Gels Used in this Study.

The metal/sacrificial ligand complexes were prepared by dissolving the metal in water with 3 equivalents of the ligand followed by 20 min of sonication, then filtration.

The synthesis of copper gels did not require use of a sacrificial ligand, as the exchange of the copper complexes was rapid enough so that the gel solution was liquid upon formation. The gels were set up in the rheometer for 12 h under a 1 Hz pre-shear.

6.4 Results and Discussion

Incorporating pendent ligands into the side chain of poly(NIPAAm) has allowed us to create polymer organogels with tunable metal-ligand crosslinks, resulting in a gel with tunable rheological and relaxation properties based on the metal used to crosslink the gel. In the following section we discuss the dynamic mechanical and stress relaxation properties of the gels, along with a discussion on theoretical versus observed modulus, and a comparison of the decay rate of the free ligand versus the rate of dissociation of crosslinks determined from stress relaxation and rheology data.

151

For iron and nickel gels, the oscillatory rheological data shows gels with elastic

properties in the frequency range studied. The storage moduli for both nickel and iron

gels is shown in Figure 6.1. The plateau storage modulus G’ of the nickel gel was

measured to be around 1.25 kPa, while the modulus for the iron gel was measured to be

between 0.45 and 0.5 kPa. The equation

G = nRT /V

where G is the modulus, n is the number of crosslinks, R is the universal gas costant, T is

the temperature, and V is the unit volume, was originally published by Flory to relate the

modulus of networks to the crosslink density. 17 Using this equation, the theoretical maximum modulus of the gels based on the metal concentration of the gels is 21.3 kPa.

Using this calculation, it is apparent that the iron and nickel gels are forming only 2.3% and 5.9% of the theoretical maximum of crosslinks, respectively. Iron-croslinked gels synthesized with twice the concentration of polymer and metal (13 weight %, up from 6.5 weight %) showed a plateau modulus of 3.15 kPa, more than a six-fold increase. While this is a six fold increase in plateau modulus, it still only represents about 15% of the theoretical maximum modulus. The lower than expected modulus, and the six fold increase in modulus arising from a two fold increase in concentration suggest that the polymer is present below-overlap concentration. This means that intramolecular crosslinks inside the polymer globule are the predominant species, rather than the intermolecular crosslinks which would be formed at concentrations above the overlap.

This supposition is supported by rheological results obtained for acryamide gels at similar molar concentrations. 18

152

Figure 6.1 Storage Moduli (G’) of Iron and Nickel Gels. The 6.5 wt% iron gels ( ) had a storage modulus of approximately 0.5 kPa, the 6.5wt% nickel gels ( ) showed a storage modulus of 1.25 kPa, the 13 wt% iron gel ( ) had a storage modulus of 3.15KPa, more than a 6x increase over the 6.5 wt% gel.

The oscillatory scans of the iron and nickel gels were indicative of an elastic gel, with no transitions present in the frequency range of the experiments. Since the oscillatory rheological experiments could not determine the characteristic decay rate ( β) of the iron and nickel gels, stress relaxation was employed in order to record β for these gels. The stress relaxation data for all the gels tested is shown in Table 6.1. A relaxation rate of β = 5.6*10 -5 s-1 was recorded for the nickel gels, while for the iron gels the decay rate was found to be β = 9.2*10 -5 s-1. The fact that the iron gels have a slightly higher decay rate than that of the nickel gels is surprising due to the fact that the bis complexes of nickel with terpyridine decay approximately an order of magnitude more rapidly than iron in aqueous solution. 14,19 It must be remembered however, that the ligands in these gels are not in water, and are substituted in the 4’ position, both factors which have an effect on the decay rate of these complexes. 153

Table 6.1 β Values of the Gels by Rheology and Stress Relaxation Metal β (rheology) in s -1 Β (stress relaxation) in s -1 Fe - 9.2*10 -5 Co - 5.7*10 -4 Ni - 5.6*10 -5 Cu 13 - Co/Cu 1.4*10 -3/13 5.5*10 -3

The gels crosslinked by copper bis -complexes of terpyridine formed very fluid gels on account of the exchange of the copper(II) terpyridine complexes. Previously,

Hogg and Wilkins attempted to measure the exchange of terpyridine complexes of

-1 14 copper(II) and found only that the rate of decay of the bis complex was k-2 > 10 s , a rate much more rapid than that of iron, cobalt and nickel. Besides the fluidity of the gel, another side product of the comparatively rapid exchange of copper(II) complexes was the absence of need for use of sacrificial ligands. The frequency as determined by oscillatory rheological testing was found to be β = 13 s -1 from a fit of the data to the

Maxwell model. The plot of the storage and loss moduli for the copper gel is shown in

Figure 6.2.

154

Figure 6.2 Storage ( ) and Loss ( ) Moduli of the Copper Gel. The data showed -1 a β value of 13 s .

Cobalt based gels showed a β transition in between that of the elastic iron and nickels gels and the fluid copper gel. Stress relaxation measurements returned a value of

β = 5.7*10 -4, which is in fairly good agreement with the oscillatory rheological data, which shows a transition near the edge of the tested range, which was 1*10 -4 Hz. The storage modulus for the cobalt showed a plateau at approximately 0.22 kPa, approximately 1% of the theoretical maximum.

In addition to gels crosslinked with a single metal, a gel composed of both copper and cobalt was synthesized. Interestingly, for a 1:1 mixture of copper/cobalt the gel shows two characteristic frequencies. Maxwell fitting of these data at low frequencies returned a value of β = 1.4*10 -3 Hz for cobalt, along with a second transition at higher frequencies attributed to the copper crosslinks. The data for both the cobalt and mixed gels are shown in Figure 6.3. The stress relaxation data for the mixed metal gels showed a

β value of 5.5*10 -3 s-1, which is in fairly good agreement with the Maxwell fit for cobalt

155

(the relaxation of the copper portion of the gel could not be measured due to the rapidity

of the decay).

Figure 6.3 The Storage and Loss Moduli of Cobalt and Mixed Gels. The cobalt gel (G’ = blue triangles, G” = blue squares) showed a plateau storage modulus of 0.22 kPa, β =5.7*10 -4 Hz. The mixed copper/cobalt gel (G’ = red triangles, G” = red squares) showed a β value of 1.4*10 -3 s-1 for the cobalt portion, and β ≈ 13 s-1 for the copper portion.

• Relation of Gel Dynamic to Terpyridine Dissociation

In order to determine how the β values of the gels were related to the dissociation

rate of the crosslinking terpyridine complexes, the decay constants for the bis complexes

of acryloyl-terpy with iron and cobalt were measured in DMF, using the techniques

outlined by Holyer, Hubbard, and Wilkins, 19 and the studies which we have completed

-2 -1 -3 previously. These studies returned a value of k-2 = 2.5*10 s for iron and k-2 = 5.4*10

s-1 for cobalt; unfortunately decay rates could not be measured for the nickel and copper

complexes. These values are much larger than the β values obtained by stress relaxation and oscillatory rheology methods by nearly two orders of magnitude for cobalt and

156 approximately three orders of magnitude for iron, which stands in stark constrast to previous studies on metal-crosslinked supramolecular gels. 6

For the iron, nickel, and pure copper gels, it appears the model presented by

Rubinstein and Semenov, 15,16 is valid. This model states that if the bond energy of the crosslink is high enough, and the opportunities for recombination are limited, the lifetime of the bulk network association will be much greater than the lifetime of the individual associations, as the associations break and reform several times before dissociating permanently during stress relaxation.

Two factors contribute to slowed relaxation: high equilibrium constant of binding, and the limited number of ligands on the polymer chain. In a system such as the

PVP/pincer complex gel reported by Craig, every monomer can act as a ligand, giving the recently dissociated metal center many options for complex reformation in the immediate vicinity. In our gels, the local options for re-binding are limited due to the low molar content of the ligand. This would encourage the complex to reform with the same ligand after the initial dissociation, which, along with the high binding energy, explains the difference between the β values measured for the gels and the decay of the crosslinkers in solution. In the case of copper gels, the free-ligand decay rate is unknown in DMF, and the data available for water is only an approximation. Because of this, no conclusions could be drawn as to differences between gel stress relaxation and free ligand exhnage for the copper gels. In the case of the mixed metal gels, the labile crosslinks formed by copper provide a “fluid” framework with rapid interchanges, allowing the recently dissociated cobalt centers to recombine with binding sites made available by rapid copper

157

exchange. For this reason the β values for the mixed metal gels are much closer to the

decay constants of the acryloyl-terpy complexes in solution.

6.5 Conclusions

In this study we have synthesized organogels with metal-ligand crosslinks based

on terpyridine complexes, the charcacteristic frequencies ( β) of which are varied based on

the nature of the metal (or metals) used to form the complexes. The β values for each gel

were measured by stress relaxation and/or oscillatory rheology. In all cases except for

copper, for wich there is insufficient data to form an expectation, the β values were much lower than the expected (or measured) decay rates of the terpyridine-based crosslinker complexes in solution. This presumably corresponds to a limited free association model, as described by Rubinstein and Semenov, wherein the supramolecular bonds are broken and reformed in the same binding site many times before migration to a new binding site by the metal occurs.

In mixed cobalt/copper systems, exchange of the copper portion of the gel provided an abundance of free ligands, which allowed the cobalt portion of the gel to have a β value more closely resembling that of the decay of the free ligand in solution.

These interesting results lay the groundwork for further studies on these systems in the future, as variation of the solvent, linking chemistry, ligand density, and further variation of the metal could potentially further increase the tunability of these systems.

6.6 References

(1) Feldman, K. E.; Kade, M. J.; Meijer, E. W.; Hawker, C. J.; Kramer, E. J. Macromolecules 2009 , 42 , 9072-9081. (2) Cho, Y.; Lee, J. H.; Jaworski, J.; Park, S.; Lee, S. S.; Jung, J. H. New Journal of Chemistry , 36 . 158

(3) Nair, K. P.; Breedveld, V.; Weck, M. Macromolecules , 44 , 3346-3357. (4)Wilkins, R. G. (1974). The Study of Kinetics and Mechanism of Reactions of Transition Metal Complexes . Boston, Allyn and Bacon. (5) Yount, W. C.; Loveless, D. M.; Craig, S. L. Angewandte Chemie International Edition 2005 , 44 , 2746-2748. (6) Xu, D.; Hawk, J. L.; Loveless, D. M.; Jeon, S. L.; Craig, S. L. Macromolecules , 43 , 3556-3565. (7) Yount, W. C.; Loveless, D. M.; Craig, S. L. Journal of the American Chemical Society 2005 , 127 , 14488-14496. (8) Kumpfer, J. R.; Jin, J.; Rowan, S. J. Journal of Materials Chemistry , 20 . (9) Weng, W.; Jamieson, A. M.; Rowan, S. J. Tetrahedron 2007 , 63 , 7419- 7431. (10) Weng, W.; Beck, J. B.; Jamieson, A. M.; Rowan, S. J. Journal of the American Chemical Society 2006 , 128 , 11663-11672. (11) Rowan, S. J.; Beck, J. B. Faraday Discussions 2005 , 128 . (12) Zhao, Y.; Beck, J. B.; Rowan, S. J.; Jamieson, A. M. Macromolecules 2004 , 37 , 3529-3531. (13) Fiore, G. L.; Klinkenberg, J. L.; Pfister, A.; Fraser, C. L. Biomacromolecules 2008 , 10 , 128-133. (14) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society (Resumed) 1962 . (15) Rubinstein, M.; Semenov, A. N. Macromolecules 1998 , 31 , 1386-1397. (16) Rubinstein, M.; Semenov, A. N. Macromolecules 2001 , 34 , 1058-1068. (17) Flory, P. J. Priciples of Polymer Chemistry ; Cornell University Press: Ithica, New York, 1953. (18) Yeung, T.; Georges, P. C.; Flanagan, L. A.; Marg, B.; Ortiz, M.; Funaki, M.; Zahir, N.; Ming, W.; Weaver, V.; Janmey, P. A. Cell Motility and the Cytoskeleton 2005 , 60 , 24-34. (19) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-625.

159

CHAPTER 7

SUMMARY AND OUTLOOK

7.1 Overview

The focus of this work has been to increase the understanding of the effects governing the physical properties of terpyridine complexes, to develop novel functionalization techniques aimed specifically at tailoring ligand and complex properties, and to apply the knowledge so gained in the design and synthesis of supramolecuar materials. Synthetically, the studies presented in this work have advanced the field of terpyridine chemistry by providing new routes of functionalization, including the direct addition of polystyryllithium to terpyridine and the addition of amino-linked molecules to mesyl-PEO, thus forming PEO-N(Me)-Terpy. In addition to these organic synthetic techniques, a novel technique was introduced in which heteroleptic terpyridine complexes of iron could be synthesized.

The kinetic studies undertaken in this work have advanced the understanding of physical inorganic chemistry by elucidating the effect of substituent groups and solvent conditions on the stability of bis complexes of terpyridine. It was determined that substituents in the 4’ position of terpyridine do indeed affect the kinetic stability of complexes of terpyridine which, in some cases, speeds the decay by more than an order of magnitude. While the effects of substituents on kinetic stability were appreciable, the effect of solvent on kinetic stability of the complexes was quite pronounced and varied over several orders of magnitude.

160

Finally, the knowledge gained in the synthetic and kinetic studies on these complexes was applied to the creation of an organogel with supramolecular crosslinks, with the aim of tuning the bulk properties of the gel based on variation of the metal used to crosslink the gel. It was found that the physical properties of the gel did indeed depend on the metal used; however, the gel did not necessarily have the expected properties, as other effects besides metal kinetics played a role in determining the bulk physical properties. In the following section, an overview of the general topics covered in this writing is presented along with suggestions for potentially useful future studies.

7.2 Synthesis of Novel Terpyridine-Functionalized Polymers

In the course of this study, a few novel terpyridine-functionalized polymers were synthesized with different intended purposes. In chapter 2, a novel technique for the functionalization of polystyrene via anionic polymerization is presented, the technique being based the nucleophilic substitution of polystyryllithium to the 6 position of terpyridine. A simplified reaction scheme for this process is displayed in Scheme 7.1.

Li N N N

Scheme 7.1 Functionalization of Polystyryllithium with Terpyrdine

The reaction required a large excess of terpyridine to be successful, a marked disadvantage in this process. The process did allow for direct functionalization that did not require the synthesis of precursors, however, and also allowed for the production of

161

novel 6-polystyryl terpyridine. Furthermore, it was determined that the product could

easily be purified to nearly 100% functionalization using flash-column chromatography.

While some work was done to prove that this polymer complexed with iron, the

behavior of metal complexes of this class of polymers is virtually unknown. In the

future, studies on complexation of 6-polystyryl terpyridine with a variety of metal iron

would be beneficial in understanding the effects of steric bulk on terpyridine complexes.

Previous studies by Constable, et. al. have shown that steric bulk at the 6/6” position is

capable of changing the spin state, and therefore the magnetic and optical absorption

properties, of terpyridine complexes with iron and cobalt. 1,2 It would be interesting to know how a polymer chain would contribute to steric bulk in various solvents and if variation of solvents could, in turn, have an effect on spin-state related properties.

Another possible way to extend the study of these anionic polymers would be to synthesize functionalized polymers other than polystyrene. Isoprene, butadiene, vinyl pyridines, and various substituted styrene are all capable of anionic polymerization and very little modification of the current technique would be needed in order to produce terpyridine-functionalized version of these polymers. This new anionic polymer toolbox could then be used to produce such structures as supramolecular block copolymers.

A final aspect of the anionic terpyridine functionalized polymers is the formation of ligand-centered block copolymers. This was briefly touched on in Appendix 2.B, providing information proving that such copolymers of polystyrene/polyisoprene and polystyrene/poly( p-methylstyrene) were possible. The ability of these copolymers to complex with metal ions remains unexplored, however, and would be an interesting subject of study. Whether or not the ligand-centered block copolymers can form a bis

162 complex is yet unknown. If the ligands are too hindered to form bis complexes, the unoccupied coordination sites may allow the complex to act as a catalyst, which would not be possible in the case of the bis complex.

Aside from anionic polymers, a few novel polymers based on PEO were also produced. The reaction of PEO-methyl ether with Cl-terpy is well known, producing ether-linked terpyridine-functionalized polymers. 3 For the kinetics studies, an amino linkage between the terpyridine and the PEO chain was desired, and this goal was accomplished by functionalizing mesyl-PEO with 4′-(N-2-Hydroxyethyl-N- methylamino)-2,2 ′:6 ′,2 ′′ -terpyridine . In addition to this, thiol-linked PEO-terpy was created by the reaction of terpy-SH with mesyl-PEO. In turn, sulfonyl-linked PEO-terpy was produced by oxidation of the thioether-linked species.

While all these PEO-based structures are, to the best of our knowledge, novel, our interest in them was the stability of complexes formed with Co(II) and Fe(II); little study was done beyond the completion of this goal. Worthwhile studies could be completed into characterization of complexes of other metal irons and different properties. For example, Co(II), Fe(II), Ni(II), and Ru(II) complexes of these polymers could be characterized by GPC using techniques previously outlined by Shcubert et. al. 4-6 in order to characterize the relative stabilities of the complexes, based on each metal-linking chemistry combination. Furthermore, studies could be completed on the magnetic and spectral properties as well, as we did no characterization of emission spectra or magnetic susceptibilities.

163

7.3 Kinetics Studies of Terpyridine Complexes

The work done on the kinetic stability in this dissertation has been focused on

determining the effects of substituent groups and solvent conditions on the stability of bis complexes of Fe(II) and Co(II). A reaction scheme of the decay of the polymer complexes in water is shown in Scheme 7.2. Prior to these studies, only the rate constants for unsubstituted terpyridines in aqueous solution were widely known. 7,8 The

studies we have conducted have provided additional knowledge about how structure

effects kinetics, i.e. that both amino and ether substituents on the 4’ position of the

terpyridine speed decay by an order of magnitude for iron, while amino-substituted

terpyridine ligands facilitate the rapid oxidation of cobalt(II) to cobalt(III). We have also

shown that it is possible to vary the stability of terpyridyl iron and cobalt complexes by

several orders of magnitude simply by changing the solvent in which the complex was

dissolved.

N N N H2O N k-2 R N M N R R N M H2O + N R H2O N N N H2O N

O O R = O N M = Co, Fe

Schem 7.2 Decay of Terpyridine Complexes

In the subtituent group studies it was found that for all substituents tested (ether,

thioether, amino, and sulfonyl linkages), the rate of decay for iron increased by at least an

order of magnitude over the unsubstituted complex. The substituents whose complexes

were the most stable were amino and thioether linkages, the least stable being the ether

and sulfonyl linkages. As current terpyrdine-functionalized polymers generally rely on

164

ether linkages, these amino and thioether linkages provide a more stable alternative when

forming supramolecular structures.

For the cobalt complexes of the ligands listed above, the dependence of the decay

rate on the ligand was not as extreme as for the iron complexes. The largest difference

observed was for complexes of sulfonyl-substitued ligands, which decayed approximately

five times faster than the unsubstituted complexes. The complexes of the 5 kg/mol ether

and amino-linked PEO ligands decayed at rates comparable to the unsubstituted complex

while the thioether-linked complex decayed approximately three times more rapidly than

the unsubstituted ligand.

0.14 Interestingly, for the iron complexes, a power law relationship of k-2 ~ N was found to exist between the number of repeat units and the decay rate. These results were consistent with the inverse of the theoretical rate dependence derived for chemically

controlled linking reactions. These are the only results of which we are aware to show a

progressive dependence of molecular weight to the reverse reaction of two end-

functionalized polymers.

The molecular weight dependence on cobalt complex decay was not so

straightforward; the lower molecular weight PEO-based ligands decay with decreasing

rate until the 5 kg/mol ligand, which had the slowest decay rate of the ether-linked

ligands. The decay rate then increased with increasing molecular weight, though the

number of data points was insufficient to gain a decisive fit for the dependence of decay

rate on molecular weight. These results further highlight the differences in the decay

kinetics of the iron and cobalt complexes, as the distortion of the d 7 cobalt complexes

likely gives rise to the differences.

165

The effect of solvent on the decay rate was also measured. As many studies of

terpyridine complexes do not use water as a solvent, it was considered worthwhile to

determine the stability of the complexes in organic solvents. For the iron complexes, a

definite correlation was found between the ability of the solvent to dissolve the

terpyridine ligand and the decay rate, with the decay kinetics spanning more than 5 orders

of magnitude. In contrast, the difference in the decay rates of cobalt complexes spanned

only one order of magnitude, with no discernible dependence on the solubility of the

ligand versus decay rate.

In conducting these studies, we have outlined several factors that influence the

stability of terpyridine complexes, all of which are relevant to the formation of

terpyridine based polymer supramolecular structures. In the future, exploring the effects

of more susbtituent groups would be highly beneficial, especially those not requiring the

use of an electropositive or electronegative heteroatom. Coupling techniques using Cl-

terpy, such as Heck coupling 9 or Sonogashira coupling 10 could be employed in the synthesis of terpyridines with sp 2 and sp hybridized carbon substituents, while a

Grignard-type reaction could be employed in the synthesis of sp 3 hybridized carbon substituents. This is important as the Hammett constant of many carbon substituents are very close to 0 (the Hammett constant of the hydrogen group), giving the potential for substituted ligand decay rates which are comparable to the unsubstituted ligands.

Variation of the metal groups beyond cobalt and iron would also be very beneficial, further expanding the knowledge of solvent and substituent groups on metal- ligand complexes. In Chapter 3, nickel was neglected in the decay studies, as the complexes could not be decayed using techniques similar to those employed for iron and

166

cobalt. It would be ideal to study the decay of nickel complexes, as they are more

thermodynamically stable than iron or cobalt complexes. 11 In the future, techniques such

as exchange studies with isotopically labeled ligands or rates of acid induced

dissociation 12 may be employed in the elucidation of the decay rates of nickel complexes.

These techniques may also be employed in the characterization of the stability of complexes of other metal ions as well.

7.4 Synthesis of Heteroleptic Terpyridine-Based Complexes of Iron(II)

In Chapter 4, a convenient synthesis of heteroleptic complexes of iron by the oxidation of the mono terpy-iron(III) in the presence of a second terpyridine ligand is

described. This technique was used to create monosubstituted terpyridine complexes, the

cyclic voltammetry and kinetic decay properties of which were characterized and

compared against the disubstituted and unsubstituted complexes. This technique is

important in that it is the first truly successful synthesis of asymmetric terpyridine

complexes of iron(II).

Chapter 4 represents the results of an initial study, one that can be expanded upon

in future optimization studies. The yields of the reaction could be improved and other

solvent systems could be employed with the aim of making the reaction more compatible

to a wider range of molecular structures. Ligands besides substituted terpyridines may be

used as the ligand added oxidation. Likewise, a tridentate ligand other than terpyridine

may be used as the starting ligand, as other tridentate complexes are capable of forming a

mono complex with iron(III). 13

167

7.5 Organogels with Supramolecular Terpyridne-Based Crosslinks

In Chapter 6 of this dissertation, the synthesis of a poly(NIPAAm)-based gel in

DMF with terpyridine complex-based crosslinks was discussed. The gels were synthesized by first synthesizing poly(NIPAAm) with pendant terpyridine groups, then dissolving the polymer in DMF and adding a solution of the desired metal ion, either as the free salt or as a complex with sacrificial ligands. While other studies have been done on supramolecular gels, this is the first we of aware of to utilize terpyridine, which is useful in creating solid gels because of its binding strength. It was found that the bulk relaxation properties of the gels could be tuned by variation of the metal used, though the relaxation rates did not match the decay rate of the free complex in most cases.

Future work for this project could potentially entail combining the terpyridine crosslinks with covalent crosslinks or other supramolecular crosslinks to create multifunctional gels. Also studies can be completed on the swelling of the gels and the dissolution of the gels with competing ligands. Variations of the polymer and solvent could potentially provide methods of further tuning the gel.

7.6 Conclusions

In this dissertation, several ways of modifying the physical properties of terpyridine complexes have been presented. In supramolecular chemistry, terpyridine complexes are often utilized for their perceived stability, using kinetics and thermodynamic values obtained in water and not necessarily reflecting the conditions used to make the desired structures. We have provided basic information into the

168

physical properties of complexes of substituted terpyridines and those in organic solvents.

studying addition to providing a useful body of work, these studies demonstrate that

substitutions and solvents must be taken into account when considering the stabilities of

supramolecular complexes.

In addition to the physical data, we have introduced novel terpyridine and

bipyridine functionalized polystyrene by anionic polymerization. We have also created

organogels with terpyridine complex-based crosslinks. These gels could be tuned by

variation of the metal ligand. These studies provide a valuable contribution to the

understanding of terpyridine chemistry, establishing a base of knowledge which can be

used to create tunable supramolecular structures in the future. .

7.7 References

(1) Constable, E. C.; Baum, G.; Bill, E.; Dyson, R.; van Eldik, R.; Fenske, D.; Kaderli, S.; Morris, D.; Neubrand, A.; Neuburger, M.; Smith, D. R.; Wieghardt, K.; Zehnder, M.; Zuberbühler, A. D. Chemistry – A European Journal 1999 , 5, 498-508. (2) Constable, E. C.; Housecroft, C. E.; Kulke, T.; Lazzarini, C.; Schofield, E. R.; Zimmermann, Y. Journal of the Chemical Society, Dalton Transactions 2001 . (3) Gohy, J. F.; Lohmeijer, B. G. G.; Schubert, U. S. Macromolecules 2002 , 35 , 4560-4563. (4) Meier, M. A. R.; Hofmeier, H.; Abeln, C. H.; Tziatzios, C.; Rasa, M.; Schubert, D.; Schubert, U. S. E-Polymers 2006 . (5) Ott, C.; Wouters, D.; Thijs, H. M. L.; Schubert, U. S. Journal of Inorganic and Organometallic Polymers and Materials 2007 , 17 , 241-249. (6) Chiper, M.; Meier, M. A. R.; Wouters, D.; Hoeppener, S.; Fustin, C. A.; Gohy, J. F.; Schubert, U. S. Macromolecules 2008 , 41 , 2771-2777. (7) Hogg, R.; Wilkins, R. G. Journal of the Chemical Society 1962 , 341-&. (8) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-&. (9) Heck, R. F.; Nolley, J. P. The Journal of Organic Chemistry 1972 , 37 , 2320-2322. (10) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Letters 1975 , 4467- 4470. (11) Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorganic Chemistry 1966 , 5, 622-625.

169

(12) Farina, R. D.; Hogg, R.; Wilkins, R. G. Inorganic Chemistry 1968 , 7, 170- 172. (13) Cotton, S. A.; Franckevicius, V.; Fawcett, J. Polyhedron 2002 , 21 , 2055- 2061.

170

BIBLIOGRAPHY

Aamer, K. A., W. H. De Jeu, et al. (2008). "Diblock copolymers containing metal complexes in the side chain of one block." Macromolecules 41 (6): 2022-2029.

Aamer, K. A. and G. N. Tew (2004). "Synthesis of Terpyridine-Containing Polymers with Blocky Architectures." Macromolecules 37 (5): 1990-1993.

Aamer, K. A. and G. N. Tew (2007). "Supramolecular Polymers Containing Terpyridineâˆ’Metal Complexes in the Side Chain." Macromolecules 40 (8): 2737- 2744.

Aamer, K. A. and G. N. Tew (2007). "Synthesis, dynamic light scattering, and luminescence properties of copolymers containing iridium(III) bisterpyridine in the side chain." Journal of Polymer Science Part A: Polymer Chemistry 45 (6): 1109-1121.

Aihara, M., H. Kishita, et al. (1975). "Polarographic Studies of Bis(2,2',2''-Terpyridine) Complexes of Manganese(Ii), Cobalt(Ii), Nickel(Ii) and Europium(Iii) in Acetonitrile." Bulletin of the Chemical Society of Japan 48 (2): 680-683.

Allen, G. H., B. P. Sullivan, et al. (1981). "Charge-transfer luminescence from mixed- ligand, 2,2',2"-terpyridine-phosphine, complexes of osmium(II)." Journal of the Chemical Society, Chemical Communications (15): 793-795.

Allmann, R., W. Henke, et al. (1978). "Presence of a static Jahn-Teller distortion in copper(II) terpyridine complexes. 1. Crystal structure of diterpyridinecopper(II) nitrate." Inorganic Chemistry 17 (2): 378-382.

Andres, P. R., H. Hofmeier, et al. (2003). "Synthesis of 4'-Functionalized 2,2':6',2"- Terpyridines via the Pyridone Route: Symmetric and Asymmetric Bis-Complex Formation." Synthesis 2003 (18): 2865,2871.

Andres, Philip R., R. Lunkwitz, et al. (2003). "New 4 ′-Functionalized 2,2 ′:6 ′,2 ′′ - Terpyridines for Applications in Macromolecular Chemistry and Nanoscience." European Journal of Organic Chemistry 2003 (19): 3769-3776.

Andres, P. R. and U. S. Schubert (2004). "Formation of Metallo-Polymers and - Macrocycles by Complexation of Alkyl-Linked Di-Terpyridines with Iron(II) Ions." Synthesis 2004 (08): 1229-1238.

171

Atkinson, G. and J. E. Bauman (1962). "The Thermodynamics of Complexes Formed with 2,2'-Bipyridine and Transition Metal Ions." Inorganic Chemistry 1(4): 900- 904.

Atkinson, G. and J. E. Bauman (1963). "The Chelate Effect Studied by Calorimetry, Potentiometry, and Electron Spin Resonance." Inorganic Chemistry 2(1): 64-67.

Ayers, T., S. Scott, et al. (2000). "Redox and spin state control of Co(II) and Fe(II) N- heterocyclic complexes." Inorganica Chimica Acta 307 (1-2): 7-12.

Ayers, T., R. Turk, et al. (2004). "Tuning redox and spin state properties of Fe(II) N- heterocyclic complexes via electronic/steric influence on metalâ€“ligand binding." Inorganica Chimica Acta 357 (1): 202-206.

Bailar, J. C. (1956). Chemistry of Coordination Compounds . New York, Reinhold Publishing Company.

Bakker, W. I. I., M. Haas, et al. (1994). "A General-Method for the Determination of the Kinetic Stability of Macrocyclic Alkali-Metal Complexes with Rates of Decomplexation Below 10 -3 s-1." Journal of the Chemical Society-Perkin Transactions 2 (1): 11-14.

Barr, D., R. Snaith, et al. (1988). "Characterization of the Bis(Pyridine) Complex of the Butyllithium-Pyridine Adduct, 2-Bun(C5H5N)Li.(C5H5N)2 - Its Decomposition to 2-Butylpyridine Via a Dihydropyridine." Polyhedron 7(8): 665-668.

Beattie, J. K., R. A. Binstead, et al. (1995). "Dynamics of cobalt(II) spin-equilibrium complexes." Inorganica Chimica Acta 235 (1-2): 245-251.

Bertini, I., C. Luchinat, et al. (1996). "NMR of paramagnetic substances." Coordination Chemistry Reviews 150 : R7-+.

Bjerrum, J., J. P. McReynolds, et al. (2007). Hexamminecobalt(III) Salts , John Wiley & Sons, Inc.

Blandamer, M. J., J. Burgess, et al. (1984). "Initial state and transition state contributions to solvent effects on rate constants for racemisation and dissociation of the 2+ [Fe(phen) 3] cation in DMSO-water mixtures." Transition Metal Chemistry 9(9): 347-350.

Brandt, W. W. and J. P. Wright (1954). "The Iron-2,2',2"-Terpyridine System1." Journal of the American Chemical Society 76 (11): 3082-3083.

Brandt, W. W. and J. P. Wright (1954). " METAL AMINE COORDINATION COMPOUNDS .1. COPPER (II) COMPLEXES." Journal of the American Chemical Society 76 : 3082-3083.

172

Braterman, P. S., J. I. Song, et al. (1992). "Electronic absorption spectra of the iron(II) complexes of 2,2'-bipyridine, 2,2'-bipyrimidine, 1,10-phenanthroline, and 2,2':6',2''-terpyridine and their reduction products." Inorganic Chemistry 31 (4): 555-559.

Brennan, L. and D. L. Turner (1997). "Paramagnetic NMR shifts in cyanoferricytochrome c: Investigation of thermal stability and deviations from Curie law behaviour." Biochimica et Biophysica Acta (BBA) - Protein Structure and Molecular Enzymology 1342 (1): 1-12.

Burazerovic, S., J. Gradinaru, et al. (2007). "Hierarchical Self-Assembly of One- Dimensional Streptavidin Bundles as a Collagen Mimetic for the Biomineralization of Calcite." Angewandte Chemie International Edition 46 (29): 5510-5514.

Burnworth, M., D. Knapton, et al. (2007). "Metallo-Supramolecular Polymerization: A Route to Easy-To-Process Organic/Inorganic Hybrid Materials." Journal of Inorganic and Organometallic Polymers and Materials 17 (1): 91-103.

C. Constable, E., C. E. Housecroft, et al. (1997). "The ditopic ligands 4']- (diphenylphosphino)-2,2' : 6',2"-terpyridine L 1 and 4;-(oxodiphenylphosphanyl)- 2,2':6',2 "-terpyridine L 2: co-ordination to iron(II), ruthenium(II), cobalt(II) and 2 palladium(II); crystal structures of [RuL 2][PF6] 2•H 2O•MeCN and trans- 1 [PdCl 2L 2]•2.5CH 2Cl 2." Journal of the Chemical Society, Dalton Transactions (14): 2427-2434.

Calì, R., E. Rizzarelli, et al. (1979). "Thermodynamics of 2,2 ′,2 ′'-terpyridinecopper(II) complexes in aqueous solution." Transition Metal Chemistry 4(5): 328-332.

Chambers, J., B. Eaves, et al. (2006). "Inductive influence of 4'-terpyridyl substituents on redox and spin state properties of iron(II) and cobalt(II) bis-terpyridyl complexes." Inorganica Chimica Acta 359 (8): 2400-2406.

Chan, Y.-T., X. Li, et al. (2011). " Design, Synthesis, and Traveling Wave Ion Mobility Mass Spectrometry Characterization of Iron(II)- and Ruthenium(II)-Terpyridine Metallomacrocycles " Journal of the American Chemical Society 133 (31): 11967- 11976.

Chattopadhyay, P. K. and J. F. Coetzee (1973). "Solvent dependence of ligand substitution kinetics of nickel(II)." Inorganic Chemistry 12 (1): 113-120.

Chaurin, V., E. C. Constable, et al. (2006). "What is the coordination number of copper(II) in metallosupramolecular chemistry?" New Journal of Chemistry 30 (12): 1740-1744.

Chiper, M., D. Fournier, et al. (2008). "Thermosensitive and Switchable Terpyridine- Functionalized Metallo-Supramolecular Poly(N-isopropylacrylamide)." Macromolecular Rapid Communications 29 (20): 1640-1647.

173

Chiper, M., M. A. R. Meier, et al. (2007). "New Insights into Nickel(II), Iron(II), and Cobalt(II) Bis-Complex-Based Metallo-Supramolecular Polymers." Macromolecular Chemistry and Physics 208 (7): 679-689.

Chiper, M., M. A. R. Meier, et al. (2008). "Supramolecular Self-Assembled Ni(II), Fe(II), and Co(II) ABA Triblock Copolymers." Macromolecules 41 (8): 2771-2777.

Cho, E. J., S. Jung, et al. (2010). "Fluorescent receptor-immobilized silica-coated magnetic nanoparticles as a general binding agent for histidine-tagged proteins." Chemical Communications 46 (35): 6557-6559.

Cho, Y., J. H. Lee, et al. (2012). "The influence of ultrasound on porphyrin-based metallogel formation: efficient control of H- and J-type aggregations." New Journal of Chemistry 36 (1): 32-35.

Chow, H. S. (2005). Metal Complexes of 4'-Substituted 2,2':6',2"-Terpyridines in Supramolecular Chemistry. Chemisrty . Basel, Switzerland University of Basel. Doctor of Philisophy .

Chow, H. S., E. C. Constable, et al. (2005). "When electron exchange is chemical exchange-assignment of 1H NMR spectra of paramagnetic cobalt(II)-2,2':6',2"- terpyridine complexes." Dalton Transactions (2): 236-237.

Cock, P. A., C. E. Cottrell, et al. (1972). "Rates of Reaction of Nickel (II) with Terpyridine in Dimethyl Sulfoxide." Canadian Journal of Chemistry 50 (3): 402- 411.

Collin, J. P., S. Guillerez, et al. (1989). "Ruthenium and Osmium Complexes of 2,2' - 6',2''-Terpyridine Covalently Linked to Electron-Acceptor and Electron-Donor Groups." Journal of the Chemical Society-Chemical Communications (12): 776- 778.

Collin, J. P., S. Guillerez, et al. (1990). "Synthesis of Functionalized Asymmetrical Bis(Terpyridine)Osmium(Ii) Complexes under Mild Conditions." Inorganic Chemistry 29 (24): 5009-5010.

Constable, E. C. (2007). "2,2 ': 6 ',2 ''-terpyridines: From chemical obscurity to common supramolecular motifs." Chemical Society Reviews 36 (2): 246-253.

Constable, E. C., G. Baum, et al. (1999). "Control of iron(II) spin states in 2,2 ': 6 ',2 ''- terpyridine complexes through ligand substitution." Chemistry-a European Journal 5(2): 498-508.

Constable, E. C., G. Baum, et al. (1999). "Control of Iron(II) Spin States in 2,2 ′:6 ′,2 ″- Terpyridine Complexes through Ligand Substitution." Chemistry – A European Journal 5(2): 498-508.

174

Constable, E. C. and H. J. Emeléus (1986). The Coordination Chemistry of 2,2':6',2''- Terpyridine and Higher Oligopyridines. Advances in Inorganic Chemistry , Academic Press. 30: 69-121.

Constable, E. C., B. A. Hermann, et al. (2005). "2,2':6',2"-Terpyridine-4'(1'H)-thione: a missing link in metallosupramolecular chemistry." New Journal of Chemistry 29 (11): 1475-1481.

Constable, E. C., C. E. Housecroft, et al. (2001). "Redistribution of terpy ligands- approaches to new dynamic combinatorial libraries." Journal of the Chemical Society, Dalton Transactions (19): 2864-2871.

Constable, E. C., C. E. Housecroft, et al. (2008). "Metal-mediated thiol-disulfide interconversion-a new tool for metallosupramolecular chemistry." Dalton Transactions (29): 3795-3797.

Constable, E. C., C. E. Housecroft, et al. (1997). "The ditopic ligands4 ′- (diphenylphosphino)-2,2 ′:6 ′,2 ″-terpyridine L1 and4 ′-(oxodiphenylphosphanyl)- 2,2 ′:6 ′,2 ″-terpyridine L2: co-ordination to iron(II),ruthenium(II), cobalt(II) andpalladium(II); crystal structures of[RuL22][PF6]2·H2O·MeCN andtrans- [PdCl2L12]·2.5CH2Cl2." Journal of the Chemical Society, Dalton Transactions (14): 2427-2434.

Constable, E. C., C. E. Housecroft, et al. (2004). "A Convenient Synthesis of Multitopic 2,2':6',2"-Terpyridine Ligands." Synthesis 2004 (EFirst): 869,874.

Constable, E. C. and A. M. W. C. Thompson (1995). "Strategies for the assembly of homo- and hetero-nuclear metallosupramolecules containing 2,2':6',2"-terpyridine metal-binding domains." Journal of the Chemical Society, Dalton Transactions (10): 1615-1627.

Constable, E. C. and M. D. Ward (1990). "Synthesis and Coordination Behavior of 6',6''- Bis(2-Pyridyl)-2,2'-4,4''-2'',2'''-Quaterpyridine - Back-to-Back 2,2'-6',2''- Terpyridine." Journal of the Chemical Society-Dalton Transactions(4): 1405- 1409.

Constable, E. C., M. D. Ward, et al. (1988). "A convenient, high yield synthesis of 2,2':6',2"-terpyridine and its iron(II) complex." Inorganica Chimica Acta 141 (2): 201-203.

Constable*, E. C., C. E. Housecroft*, et al. (2000). "Programmed assembly of heteromultinuclear complexes using 4'-diphenylphosphino-2,2':6',2"-terpyridine." Inorganica Chimica Acta 300-302 (0): 49-55.

Cossy, C., L. Helm, et al. (1987). "High-Pressure Nmr-Study .33. Dimethylformamide- Exchange Mechanism on Some 1st-Row Divalent Transition-Metal Ions." Helvetica Chimica Acta 70 (6): 1516-1525.

175

Cotton, F. A. and G. Wilkinson (1962). Advanced Inoeganic Chemistry . New York, Wiley Interscience.

Cotton, S. A., V. Franckevicius, et al. (2002). "Syntheses and structures of iron(III) complexes of simple N-donor ligands." Polyhedron 21 (20): 2055-2061.

Cox, B. G., J. Garciarosas, et al. (1981). "Solvent Dependence of the Stability of Cryptate Complexes." Journal of the American Chemical Society 103 (6): 1384-1389.

Cusumano, M., A. Giannetto, et al. (1998). "Kinetics of ligand replacement, in N- substituted ethylenediamine palladium(II) complexes." Polyhedron 17 (1): 125- 129.

Danil de Namor, A. F., M. L. Zapata-Ormachea, et al. (1997). "Thermodynamics of Alkaliâˆ’Metal Cations and Macrocycles (18-Crown-6, Ethyl p-tert- butylcalix(6)arenehexanoate, Cryptand 222) in Solution and in the Solid State." The Journal of Physical Chemistry B 101 (34): 6772-6779.

Denti, G., S. Serroni, et al. (1993). "Redox processes of ruthenium (II) polypyridine complexes induced by fast-atom bombardment mass spectrometry." Journal of The American Society for Mass Spectrometry 4(4): 306-311.

Dimitris, K. and G. Gilbert (1987). Mechanisms of Inorganic Reactions . New York, John Wiley and Sons.

Dobrawa, R. and F. Würthner (2005). "Metallosupramolecular approach toward functional coordination polymers." Journal of Polymer Science Part A: Polymer Chemistry 43 (21): 4981-4995.

Ducommun, Y., K. E. Newman, et al. (1980). "High-pressure oxygen-17 NMR evidence for a gradual mechanistic changeover from Ia to Id for water exchange on divalent octahedral metal ions going from manganese(II) to nickel(II)." Inorganic Chemistry 19 (12): 3696-3703.

Duncan, T. V., T. Ishizuka, et al. (2007). "Molecular Engineering of Intensely Near- Infrared Absorbing Excited States in Highly Conjugated Oligo(porphinato)zinc- (Polypyridyl)metal(II) Supermolecules." Journal of the American Chemical Society 129 (31): 9691-9703.

Dunn, T. M., D. S. McLure, et al. (1965). Some Aspects of Crystal Field Theory . New York, Harper and Row.

El-ghayoury, A., H. Hofmeier, et al. (2003). "Combining covalent and noncovalent crosslinking: A novel terpolymer for two-step curing applications." Macromolecules 36 (11): 3955-3959.

176

Elias, H., C. Hasserodt-Taliaferro, et al. (1985). "Effect of the coordination geometry and of substituent shielding on the kinetics of ligand substitution in copper(II) chelates." Inorganic Chemistry 24 (20): 3192-3198.

Ellis, P., R. Hogg, et al. (1959). "671. Exchange studies of certain chelate compounds of the transitional metals. Part VII. Substituent effects on the dissociation rates of nickel(II)-phenanthroline and -bipyridyl complexes." Journal of the Chemical Society (Resumed) : 3308-3313.

Elsbernd, H. and J. K. Beattie (1972). "Nmr-Spectra of Terpyridine and Bis-Terpyridine Complexes of Cobalt(III) and Iron(II)." Journal of Inorganic & Nuclear Chemistry 34 (2): 771-&.

Enachescu, C., I. Krivokapic, et al. (2007). "Optical investigation of spin-crossover in cobalt(II) bis-terpy complexes." Inorganica Chimica Acta 360 (13): 3945-3950.

End, N., K.-U. Schoening, et al. (2006). COLORING KERATIN FIBERS WITH METAL COMPLEXES. E. P. Office. Appl. No. WO2006EP61990 20060503

Epperson, J. D., L.-J. Ming, et al. (2001). "Paramagnetic Cobalt(II) as an NMR Probe of Dendrimer Structure: Mobility and Cooperativity of Dendritic Arms." Journal of the American Chemical Society 123 (35): 8583-8592.

Farina, R. D., R. Hogg, et al. (1968). "Rate-pH profile for the dissociation of iron(II)- and cobalt(II)-2,2',2''-terpyridine complexes." Inorganic Chemistry 7(1): 170-172.

Feldman, K. E., M. J. Kade, et al. (2009). "Model Transient Networks from Strongly Hydrogen-Bonded Polymers." Macromolecules 42 (22): 9072-9081.

Fetters, L. J. and E. M. Firer (1977). "Partial Chain Coupling of Anionically Prepared Polymers by Air Termination." Polymer 18 (3): 306-307.

Fiore, G. L., J. M. Edwards, et al. (2007). "Ruthenium(II) tris(bipyridine)-centered poly(ethylenimine) for gene delivery." Biomacromolecules 8(9): 2829-2835.

Fiore, G. L., J. L. Klinkenberg, et al. (2008). "Iron Tris(bipyridine)-Centered Poly(ethylene glycol)-Poly(lactic acid) Star Block Copolymers." Macromolecules 41 (23): 9397-9405.

Fiore, G. L., J. L. Klinkenberg, et al. (2008). "Iron Tris(bipyridine) PEG Hydrogels with Covalent and Metal Coordinate Cross-Links." Biomacromolecules 10 (1): 128- 133.

Flory, P. J. (1953). Priciples of Polymer Chemistry . Ithica, New York, Cornell University Press.

177

Fragouli, P. G., H. Iatrou, et al. (2004). "Synthesis and characterization of linear tetrablock quarterpolymers of styrene, isoprene, dimethylsiloxane, and 2- vinylpyridine." Journal of Polymer Science Part A: Polymer Chemistry 42 (3): 514-519.

Fraser, C. L. and A. P. Smith (2000). "Metal complexes with polymeric ligands: Chelation and metalloinitiation approaches to metal tris(bipyridine)containing materials." Journal of Polymer Science Part a-Polymer Chemistry 38 : 4704-4716.

Frausto da Silva, J. J. R. (1983). "The chelate effect redefined." Journal of Chemical Education 60 (5): 390-393.

Friese, V. A. and D. G. Kurth (2009). "From coordination complexes to coordination polymers through self-assembly." Current Opinion in Colloid & Interface Science 14 (2): 81-93.

Funahashi, S. and R. B. Jordon (1977). "Solvent-exchange kinetics in iron(II) solution of N,N-dimethylformamide and dimethyl sulfoxide studies by nuclear magnetic resonance line broadening." Inorganic Chemistry 16 (6): 1301-1306.

Füssl, A., A. Schleifenbaum, et al. (2006). "Cellular Uptake of PNA-Terpyridine Conjugates and Its Enhancement by Zn 2+ Ions." Journal of the American Chemical Society 128 (18): 5986-5987.

Gagliardo, M., J. Perelaer, et al. (2007). "Mono- and Heterodimetallic FeII and RuII Complexes Based on a Novel Heteroditopic 4 ′-{Bis(phosphanyl)aryl}-2,2 ′:6 ′,2 ″- terpyridine Ligand." European Journal of Inorganic Chemistry 2007 (15): 2111- 2120.

Gohy, J. F., B. G. G. Lohmeijer, et al. (2002). "Metallo-supramolecular block copolymer micelles " Macromolecules 35 (12): 4560-4563.

Gohy, J.-F., B. G. G. Lohmeijer, et al. (2004). " Cylindrical micelles from the aqueous self-assembly of an amphiphilic poly(ethylene oxide)-b-poly(ferrocenylsilane) (PEO-b-PFS) block copolymer with a metallo-supramolecular linker at the block junction " Chemistry: A European Journal 10 : 4315-4323.

Gohy, J.-F., B. G. G. Lohmeijer, et al. (2002). "Reversible Metallo-Supramolecular Block Copolymer Micelles Containing a Soft Core." Macromolecular Rapid Communications 23 (9): 555-560.

Gohy, J.-F., B. G. G. Lohmeijer, et al. (2003). "From Supramolecular Block Copolymers to Advanced Nano-objects " Chemistry: A European Journal 9: 3472-3479.

Gohy, J.-F. o., B. G. G. Lohmeijer, et al. (2002). "Stimuli-Responsive Aqueous Micelles from an ABC Metallo-Supramolecular Triblock Copolymer." Macromolecules 35 (26): 9748-9755.

178

Göritz, M. and R. Krämer (2005). "Allosteric Control of Oligonucleotide Hybridization by Metal-Induced Cyclization." Journal of the American Chemical Society 127 (51): 18016-18017.

Gosnell, A. B., J. A. Gervasi, et al. (1969). "Anionic Grafting of Polystyrene and Polyisoprene to Living Poly(2-vinylpyridine)." Journal of Polymer Science Part C: Polymer Symposia 22 (2): 611-620.

Graessley, W. W. (1980). "Polymer-Chain Dimensions and the Dependence of Viscoelastic Properties on Concentration, Molecular-Weight and Solvent Power." Polymer 21 (3): 258-262.

Guerrero-Sanchez, C., B. G. G. Lohmeijer, et al. (2005). "Synthesis of terpyridine- terminated polymers by anionic polymerization." Macromolecules 38 (25): 10388- 10396.

Guillet, P., C. A. Fustin, et al. (2008). "Tuning block copolymer micelles by metal-ligand interactions." Soft Matter 4(11): 2278-2282.

Guillet, P., C. A. Fustin, et al. (2009). "Amphiphilic brushes from metallosupramolecular block copolymers." Soft Matter 5(7): 1460-1465.

Guillet, P., C.-A. Fustin, et al. (2008). "Tuning block copolymer micelles by metal-ligand interactions." Soft Matter 4(11): 2278-2282.

Haensch, C., M. Chiper, et al. (2008). "Reversible Supramolecular Functionalization of Surfaces: Terpyridine Ligands as Versatile Building Blocks for Noncovalent Architectures." Langmuir 24 (22): 12981-12985.

Hanabusa, K., A. Nakamura, et al. (1992). "SYNTHESIS, POLYMERIZATION, COPOLYMERIZATION, AND TRANSITION-METAL COORDINATION OF 4-(2,2'-6',2''-TERPYRIDIN-4'-YL)STYRENE AND ITS POLYMERS AND COPOLYMERS " Die Makromoluculare Chemie 193 : 1309-1319.

Hanabusa, K., A. Nakamura, et al. (1992). "Synthesis, polymerization, copolymerization, and transition-metal coordination of 4-(2,2 ′:6 ′,2 ″-terpyridin-4′-yl)styrene and its polymers and copolymers." Die Makromolekulare Chemie 193 (6): 1309-1319.

Hanabusa, K., K. Nakano, et al. (1990). "SYNTHESIS AND PROPERTIES OF POLYMERIC IRON(II)-COMPLEXES CONTAINING TERPYRIDINYL GROUPS " Die Makromoluculare Chemie 191 : 391-396.

Hanabusa, K., K. Nakano, et al. (1990). "Synthesis and properties of polymeric iron(II)- complexes containing terpyridinyl groups." Die Makromolekulare Chemie 191 (2): 391-396.

179

Hancock, R. D. and G. J. McDougall (1977). "The role of electronic delocalization over the chelate ring in stabilizing complexes of 2,2'-bipyridyl." Journal of the Chemical Society, Dalton Transactions (1): 67-70.

Harris, M. G. and R. Stewart (1977). "Amino group acidity in aminopyridines and aminopyrimidines." Canadian Journal of Chemistry 55 (21): 3800-3806.

Hathcock, D. J., K. Stone, et al. (1998). "Electron donating substituent effects on redox and spin state properties of iron(II) bis-terpyridyl complexes." Inorganica Chimica Acta 282 (2): 131-135.

Heck, R. F. and J. P. Nolley (1972). "Palladium-catalyzed vinylic hydrogen substitution reactions with aryl, benzyl, and styryl halides." The Journal of Organic Chemistry 37 (14): 2320-2322.

Hein, O., C. .Eschbaumer, et al. (2000). "Functionalized polymers with metal complexing segments: a simple and high-yield entry towards 2,2 ': 6 ',2 ''-terpyridine-based oligomers " Macrommolecular Rapid Communications 21 : 1156-1161.

Helm, L. and A. E. Merbach (2005). "Inorganic and Bioinorganic Solvent Exchange Mechanisms." Chemical Reviews 105 (6): 1923-1960.

Henderson, I. M. and R. C. Hayward (2010). "Synthesis of End-Functionalized Polystyrene by Direct Nucleophilic Addition of Polystyryllithium to Bipyridine or Terpyridine." Macromolecules 43 (7): 3249-3255.

Henderson, I. M. and R. C. Hayward (2012). "Substituent Effects on the Stabilities of Polymeric and Small Molecule bis-Terpyridine Complexes." Polymer Chemistry 3(5): 1221-1230

Hitoshi, M. (2001). "Metalloaromaticity." Coordination Chemistry Reviews 219-221 (0): 957-992.

Hofmeier, H., S. Schmatloch, et al. (2003). "Linear terpyridine-ruthenium(II) poly(ethylene glycol) coordination polymers." Macromolecular Chemistry and Physics 204 (18): 2197-2203.

Hofmeier, H. and U. S. Schubert (2003). "Supramolecular Branching and Crosslinking of Terpyridine-Modified Copolymers: Complexation and Decomplexation Studies in Diluted Solution." Macromolecular Chemistry and Physics 204 (11): 1391-1397.

Hogg, R. and R. G. Wilkins (1962). "57. Exchange studies of certain chelate compounds of the transitional metals. Part VIII. 2,2’,2”-terpyridine complexes." Journal of the Chemical Society (Resumed) : 341-350.

Holyer, R. H., C. D. Hubbard, et al. (1965). "Kinetics of Replacement Reactions of Complexes of Transition Metals with 1,10-Phenanthroline and 2,2'-Bipyridine." Inorganic Chemistry 4(7): 929-&.

180

Holyer, R. H., C. D. Hubbard, et al. (1966). "The Kinetics of Replacement Reactions of Complexes of the Transition Metals with 2,2',2"-Terpyridine." Inorganic Chemistry 5(4): 622-625.

Hu, B., A. Fuchs., et al. (2006). "Supramolecular magnetorheological polymer gels " Journal of Applied Polymer Science 100 : 2464-2479.

Huang, W. and C. D. Han (2006). "Ruthenium(II) Complex-Induced Dispersion of Montmorillonite in a Segmented Main-Chain Liquid-Crystalline Polymer Having Side-Chain Terpyridine Group." Macromolecules 39 (24): 8207-8209.

Huang, Z., X. Chen, et al. (2011). "Connecting BINOL and terpyridine: the electronic effects of the linkage." Tetrahedron Letters 52 (42): 5497-5500.

Ireland, R. E. and R. S. Meissner (1991). "Convenient Method for the Titration of Amide Base Solutions." Journal of Organic Chemistry 56 (14): 4566-4568.

Ito, M., T. Tsukatani, et al. (2005). "Preparation and characterization of gold nanoparticles with a ruthenium-terpyridyl complex, and electropolymerization of their pyrrole-modified metal nanocomposites." Journal of Materials Chemistry 15 (9): 960-964.

Ivano Bertini, Claudio Luchinat, et al. (2001). Solution NMR of Paramagnetic Molecules: Applications to Metallobiomolecules and Models , Elsevier.

Jameson, D. L. and L. E. Guise (1991). "AN IMPROVED, 2-STEP SYNTHESIS OF 2,2'-6',2''-TERPYRIDINE " Tetrahedron Letters 32 : 1999-2002.

Jari, H. (2004). "Synthesis of 4'-substituted 2,2':6',2"-terpyridines via a Mitsunobu reaction." Tetrahedron Letters 45 (29): 5707-5709.

Jenner, G. (1975). "Polymerization of Conjugated Dienes at High Pressures. X. Influence of Pressure upon the Microstructure of Anionic Polyisoprene." Journal of Macromolecular Science: Part A - Chemistry 9(1): 83-93.

Ji, Z., Y. Li, et al. (2011). "One-Photon Photophysics and Two-Photon Absorption of 4- [9,9-Di(2-ethylhexyl)-7-diphenylaminofluoren-2-yl]-2,2 ′:6 ′,2 ′′ -terpyridine and Their Platinum Chloride Complexes." Chemistry – A European Journal 17 (8): 2479-2491.

Johansson, O. (2006). "Pd-Catalyzed Amination of Chloro-Terpyridine for the Preparation of Amine-Containing Ruthenium(II) Complexes." Synthesis 2006 : 2585-2589.

Judge, J. S. and W. A. Baker (1967). "On the spin equilibrium in bis(2,2',2''-terpyridine) cobalt(II) salts." Inorganica Chimica Acta 1: 68-72.

181

Kanan, M. W. and D. G. Nocera (2008). "In situ formation of an oxygen-evolving catalyst in neutral water containing phosphate and Co(2+)." Science 321 (5892): 1072-1075.

Kanan, M. W., Y. Surendranath, et al. (2009). "Cobalt-phosphate oxygen-evolving compound." Chemical Society Reviews 38 (1): 109-114.

Kawaguchi, S., G. Imai, et al. (1997). "Aqueous solution properties of oligo- and poly(ethylene oxide) by static light scattering and intrinsic viscosity." Polymer 38 (12): 2885-2891.

Khalatur, P. G., A. S. Pavlov, et al. (1984). "Effect of Excluded Volume Interactions on the Strength of Chemical-Bonds in Polymers." Doklady Akademii Nauk Sssr 276 (6): 1431-1434.

Khan, M. S. (2006). "Aggregate formation in poly(ethylene oxide) solutions." Journal of Applied Polymer Science 102 (3): 2578-2583.

Khokhlov, A. R. (1981). "Influence of excluded volume effect on the rates of chemically controlled polymer-polymer reactions." Die Makromolekulare Chemie, Rapid Communications 2(9-10): 633-636.

Kjaer, A. M. and J. Ulstrup (1986). "Effects of reactant ionic charges and nonlocal dielectric screening on electron transfer between bis(terpyridine)cobalt(II) and tris(phenanthroline) tris(bipyridine) and analogous, electrically neutral sulfonated cobalt(III) complexes." Inorganic Chemistry 25 (5): 644-651.

Kleinberg, J., R. K. Murmann, et al. (1965). Mechanisms of Inorganic Reactions , AMERICAN CHEMICAL SOCIETY.

Kremer, S., W. Henke, et al. (1982). "High-spin-low-spin equilibriums of cobalt(2+) in the terpyridine complexes Co(terpy)2X2.nH2O." Inorganic Chemistry 21 (8): 3013-3022.

Kumpfer, J. R., J. Jin, et al. (2010). "Stimuli-responsive europium-containing metallosupramolecular polymers." Journal of Materials Chemistry 20 (1): 145-151.

Lamba, J. J. S. and C. L. Fraser (1997). "Synthesis of metal-centered star-shaped polyoxazolines using Fe(II) and Ru(II) tris-bipyridine derivatives as multifunctional initiators." Journal of the American Chemical Society 119 (7): 1801-1802.

Lehn, J.-M. (1993). "Supramolecular chemistry." Science 260 (5115): 1762-1762.

Lehn, J.-M. (1995). Supramolecular Chemistry: Concepts and Perspectives . Weinheim, Wiley-VCH.

182

Lehn, J. M. (1981). "Chemistry Beyond the Molecule - Organic Supramolecular Structures and Functions." Abstracts of Papers of the American Chemical Society 182 (AUG): 5.

Lewandowski, A. and J. Malinska (1996). "Stability of Cu + and Cu 2+ complexes with cryptands in aprotic media." New Journal of Chemistry 20 (6): 653-657.

Liang, X., S. Suwanrumpha, et al. (1991). "Fast atom bombardment tandem mass spectrometry of (polypyridyl)ruthenium(II) complexes." Inorganic Chemistry 30 (4): 652-658.

Lohmeijer, B. G. G., H. Schlaad, et al. (2003). "Synthesis and thermal properties of diblock copolymers utilizing non-covalent interactions." Macromolecular Symposia 196 (1): 125-135.

Lohmeijer, B. G. G. and U. S. Schubert (2001). "Synthesis of metallo-supramolecular block-copolymers." Polym. Mater. Sci. Eng. 85 : 460.

Lohmeijer, B. G. G. and U. S. Schubert (2002). "Supramolecular Engineering with Macromolecules: An Alternative Concept for Block Copolymers." Angewandte Chemie International Edition 41 (20): 3825-3829.

Lohmeijer, B. G. G. and U. S. Schubert (2003). "Playing LEGO with macromolecules: Design, synthesis, and self-organization with metal complexes." Journal of Polymer Science Part A: Polymer Chemistry 41 (10): 1413-1427.

Lohmeijer, B. G. G. and U. S. Schubert (2003). "Water-soluble building blocks for terpyridine-containing supramolecular polymers: Synthesis, complexation, and pH stability studies of poly(ethylene oxide) moieties." Macromolecular Chemistry and Physics 204 (8): 1072-1078.

Lohmeijer, B. G. G. and U. S. Schubert (2005). "The LEGO toolbox: Supramolecular building blocks by nitroxide-mediated controlled radical polymerization." Journal of Polymer Science Part a-Polymer Chemistry 43 (24): 6331-6344.

Lowe, G., A. S. Droz, et al. (1999). "Cytotoxicity of (2,2':6',2"-Terpyridine)platinum(II) Complexes to Leishmania donovani, Trypanosoma cruzi, and Trypanosoma brucei." Journal of Medicinal Chemistry 42 (6): 999-1006.

Lowe, G., A.-S. Droz, et al. (1999). "The Design and Synthesis of Bis-[4'-Azido- 2,2':6',2''-Terpyridine Platinum(II)] Complexes with Rigid and Extended Linkers for Studying the Topology of DNA by Photoaffinity Labeling." Bioorganic Chemistry 27 (6): 477-486.

Maanpaa, L., V. Luzet, et al. (2009). "Bimetallic Cu 2+ complexes of bis-terpyridine ligands as catalysts of the cleavage of mRNA 5[prime or minute]-cap models. The effect of linker length and base moiety." New Journal of Chemistry 33 (9): 1853- 1858.

183

Maestri, M., N. Armaroli, et al. (1995). "Complexes of the Ruthenium(II)-2,2':6',2''- terpyridine Family. Effect of Electron-Accepting and -Donating Substituents on the Photophysical and Electrochemical Properties." Inorganic Chemistry 34 (10): 2759-2767.

Majouga, A., R. Romashkina, et al. (2010). "New organic ligands of the terpyridine series: modification of gold nanoparticles, preparation of coordination compounds with Cu(I), catalysis of oxidation reactions." Chemistry of Heterocyclic Compounds 46 (9): 1076-1083.

Maslen, E. N., C. L. Raston, et al. (1974). "Crystal structure of bis-(2,2' :6',2"- terpyridyl)cobalt(II) bromide trihydrate." Journal of the Chemical Society, Dalton Transactions (16): 1803-1807.

Matwiyoff, N. A. (1966). "Proton Magnetic Resonance Studies of the Solvation of Cations in N,N-Dimethylformamide Solutions of Cobalt(II) and Nickel(II) Perchlorates and the Direct Determination of the Solvation Number of Cobalt(II)." Inorganic Chemistry 5(5): 788-795.

Mavroudis, A., A. Avgeropoulos, et al. (2003). "Synthesis and Morphological Behavior of Model Linear and Miktoarm Star Copolymers of 2-Methyl-1,3-Pentadiene and Styrene." Chemistry of Materials 15 (10): 1976-1983.

McAlvin, J. E. and C. L. Fraser (1999). "Polymerization of 2-ethyl-2-oxazoline using di-, tetra-, and hexafunctional ruthenium tris(bipyridine) metalloinitiators." Macromolecules 32 (21): 6925-6932.

McAteer, C. H. and P. Moore (1983). "Rates and activation parameters for dimethyl sulphoxide exchange with hexakis(dimethyl sulphoxide)-nickel(II), -cobalt(II), and -iron(III) ions." Journal of the Chemical Society, Dalton Transactions (2): 353-357.

McDaniel, D. H. and H. C. Brown (1958). "An Extended Table of Hammett Substitutent Constants Based on the Ionization of Substituted Benzoic Acids." The Journal of Organic Chemistry 23 (3): 420-427.

McMurry, J. (2004). Organic Chemistry . Belmont, CA, Brooks/Cole-Thomas Learning.

Meier, M. A. R., H. Hofmeier, et al. (2006). "First GPC results of terpyridine based chain extended supramolecular polymers: comparison with viscosity and analytical ultracentrifugation." E-Polymers .

Meier, M. A. R., B. G. G. Lohmeijer, et al. (2003). "Relative binding strength of terpyridine model complexes under matrix-assisted laser desorption/ionization mass spectrometry conditions." Journal of Mass Spectrometry 38 (5): 510-516.

184

Meier, M. A. R. and U. S. Schubert (2003). "Terpyridine-modified poly(vinyl chloride): Possibilities for supramolecular grafting and crosslinking." Journal of Polymer Science Part A: Polymer Chemistry 41 (19): 2964-2973.

Melson, G. A. and R. G. Wilkins (1963). "496. Factors influencing the rates of dissociation of metal complexes. Part V. The effect of increased chelation on the dissociation of nickel(II)-polyamine complexes." Journal of the Chemical Society (Resumed) : 2662-2672.

Miller, R. R. and W. W. Brandt (1954). "Colorimetric Determination of Cobalt with 2,2',2"-Terpyridine." Analytical Chemistry 26 (12): 1968-1969.

Miyachi, M., M. Ohta, et al. (2008). "Surface Bottom-up Fabrication of Porphyrin- terminated Metal Complex Molecular Wires with Photo-electron Conversion Properties on ITO." Chemistry Letters 37 (4): 404-405.

Morgan, G. and F. Burstall (1932). "Dehydrogenation of pyridine of anhydrous ferric chloride." Journal of the Chemical Society : 20-30.

Morgan, G. and F. H. Burstall (1937). "347. Researches on residual affinity and coordination. Part XXXVII. Complex metallic salts containing 2 : 6-di-2'- pyridylpyridine (2 :2':2"-tripyridyl)." Journal of the Chemical Society (Resumed) : 1649-1655.

Morgan, G. T. and F. H. Burstall (1932). "Dehydrogenation of pyridine of anhydrous ferric chloride." Journal of the Chemical Society : 20-30.

Morgan, G. T. and H. D. K. Drew (1920). "CLXII.-Researches on residual affinity and co-ordination. Part II. Acetylacetones of selenium and tellurium." Journal of the Chemical Society, Transactions 117 : 1456-1465.

Morton, M. and L. J. Fetters (1964). "Homogeneous Anionic Polymerization .V. Association Phenomena in Organolithium Polymerization." Journal of Polymer Science Part a-General Papers 2(7PA): 3311-&.

Moss, M. and M. Mellon (1943). "Colorimetric Determination of Cobalt with Terpyridyl." Industrial & Engineering Chemistry Analytical Edition 15 (1): 74-75.

Moss, M. L. and M. G. Mellon (1942). "Colorimetric Determination of Iron with 2,2'- Bipyridyl and with 2,2',2'-Terpyridyl." Industrial & Engineering Chemistry Analytical Edition 14 (11): 862-865.

Mountford, S. J., E. M. Campi, et al. (2011). "Synthesis of N-heterocyclic ligands for use in affinity and mixed mode chromatography." Tetrahedron 67 (2): 471-485.

Mugemana, C., P. Guillet, et al. (2010). "Metallo-supramolecular diblock copolymers based on heteroleptic cobalt(iii) and nickel(ii) bis-terpyridine complexes." Chemical Communications 46 (8): 1296-1298.

185

Munro, D. (1977). "Misunderstandings Over the Chelate Effect." Chemstry in Britain : 13 (3): 100-105.

Muro, M. L., C. A. Daws, et al. (2008). "Microarray pattern recognition based on PtII terpyridyl chloride complexes: vapochromic and vapoluminescent response." Chemical Communications (46): 6134-6136.

Musumeci, S., E. Rizzarelli, et al. (1973). "Low valence states of metal chelates: III. bis(2,2',2''-terpyridine) cobalt(II) perchlorate." Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 46 (1): 109-117.

Mytides, P., S. Rozou, et al. (2006). "2,2":6',2'-Terpyridine[hydroxypropyl-Î²- cyclodextrin] 1:3 complex used as chelating agent for the determination of iron with a sensitive, selective and fast liquid chromatographic method." Analytica Chimica Acta 566 (1): 122-129.

Nair, K. P., V. Breedveld, et al. (2011). "Multiresponsive Reversible Polymer Networks Based on Hydrogen Bonding and Metal Coordination." Macromolecules 44 (9): 3346-3357.

Ndoni, S., C. M. Papadakis, et al. (1995). "Laboratory-Scale Setup for Anionic- Polymerization under Inert Atmosphere." Review of Scientific Instruments 66 (2): 1090-1095.

Newkome, G. R., R. Güther, et al. (1995). "Wege zu dendritischen Netzwerken: Bis- Dendrimere durch Verknüpfung von Kaskadenmolekülen über Metallzentren." Angewandte Chemie 107 (18): 2159-2162.

Newkome, G. R. and E. He (1997). "Nanometric dendritic macromolecules: stepwise assembly by double(2,2':6',2"-terpyridine)ruthenium(I) connectivity." Journal of Materials Chemistry 7(7): 1237-1244.

Nicholson, R. S. and I. Shain (1964). "Theory of Stationary Electrode Polarography. Single Scan and Cyclic Methods Applied to Reversible, Irreversible, and Kinetic Systems." Analytical Chemistry 36 (4): 706-723.

Olmstead, W. N., Z. Margolin, et al. (1980). "Acidities of water and simple alcohols in dimethyl sulfoxide solution." The Journal of Organic Chemistry 45 (16): 3295- 3299.

Orland, W. K. (1978). "Medium Effects upon Solvent Exchange Kinetics for Iron(II) and Nickel(II) Perchlorates." Transactions of the Kansas Academy of Science (1903-) 81 (4): 303-309.

Ostermeier, M., M.-A. Berlin, et al. (2010). "Complexes of Click-Derived Bistriazolylpyridines: Remarkable Electronic Influence of Remote Substituents on Thermodynamic Stability as well as Electronic and Magnetic Properties." Chemistry – A European Journal 16 (33): 10202-10213.

186

Ott, C., G. M. Pavlov, et al. (2009). "Alternating Terpyridine-Endfunctionalized Copolymers of Styrene and Diphenylethylene via Anionic Polymerization Techniques: A Detailed Characterization Study." Journal of Polymer Science Part a-Polymer Chemistry 47 (14): 3691-3701.

Ott, C., D. Wouters, et al. (2007). "New preparation and purification methods for metallo-supramolecular block copolymers." Journal of Inorganic and Organometallic Polymers and Materials 17 (1): 241-249.

Park, J., A. N. Pasupathy, et al. (2002). "Coulomb blockade and the Kondo effect in single-atom transistors." Nature 417 (6890): 722-725.

Park, M. K., H. S. Kim, et al. (2005). " New oligomeric ether plasticizers for solid polymer electrolytes: Synthesis and electrical properties of oligomeric PEO having bis(five-membered cyclic carbonate)s at chain ends." Journal of Industrial and Engineering Chemistry 11 (2): 222-227.

Pazderski, L., T. Pawlak, et al. (2011). "1H, 13C, 15N NMR coordination shifts in Fe(II), Ru(II) and Os(II) cationic complexes with 2,2 ′:6 ′,2 ″-terpyridine." Magnetic Resonance in Chemistry 49 (5): 237-241.

Pearson, R. G. (1959). "The Nature of Substitution Reactions in Inorganic Chemistry." The Journal of Physical Chemistry 63 (3): 321-326.

Pedersen, C. J. (1967). "Cyclic polyethers and their complexes with metal salts." Journal of the American Chemical Society 89 (26): 7017-7036.

Potts, K. and D. Usifer (1988). "Polymers and polymer-metal complexes containing pendent 2,2':6',2"-terpyridinyl ligands." Macromolecules 21 : 1985-1991.

Potts, K. T., M. J. Cipullo, et al. (1981). "Ketenedithioacetals as synthetic intermediates. A versatile synthesis of pyridenes, polypyridinyls, and pyrylium salts." Journal of the American Chemical Society 103 (12): 3585-3586.

Potts, K. T., M. J. Cipullo, et al. (1982). "Synthesis of 2,6-disubstituted pyridines, polypyridinyls, and annulated pyridines." The Journal of Organic Chemistry 47 (16): 3027-3038.

Potts, K. T. and D. Konwar (1991). "Synthesis of 4'-vinyl-2,2':6',2''-terpyridine." The Journal of Organic Chemistry 56 (15): 4815-4816.

Potts, K. T., P. Ralli, et al. (1986). "2,2' : 6',2' -TERPYRIDINE " Organic Synthesis 64 : 189.

Powell, D. H., H. Lothar, et al. (1991). " 17 O nuclear magnetic resonance in aqueous solutions of Cu 2 + : The combined effect of Jahn--Teller inversion and solvent exchange on relaxation rates." The Journal of Chemical Physics 95 (12): 9258- 9265.

187

Quirk, R. P. and W. C. Chen (1984). "Functionalization of Polymeric Organo-Lithium Compounds - Oxidation." Journal of Polymer Science Part a-Polymer Chemistry 22 (11): 2993-3000.

Quirk, R. P. and S. Corona-Galvan (2001). "Controlled anionic synthesis of polyisoprene-poly(2-vinylpyridine) diblock copolymers in hydrocarbon solution." Macromolecules 34 (5): 1192-1197.

R. Newkome, G., E. He, et al. (1999). "Neutral highly branched metallomacromolecules: incorporation of a (2,2’:6’,2”-terpyridine)ruthenium(II) complex without external counterions." Chemical Communications (1): 27-28.

Rablen, D. and G. Gordon (1969). "Water exchange rate of the mono(2,2',2"- terpyridine)nickel(II) ion by oxygen-17 nuclear magnetic resonance." Inorganic Chemistry 8(2): 395-397.

Rao, J. M., M. C. Hughes, et al. (1976). "Voltammetry of Terpyridine and Terosine Complexes of Cobalt(Ii) and Iron(Ii)." Inorganica Chimica Acta 16 (JAN): 231- 236.

Reichhardt, C. and T. Welton (2011). Solvent Effects in Organic Chemistry . Weinheim, Wiley-VCH.

Reiff, W. M. (1974). "Orbital ground states of low-spin iron(III) compounds. Bis(2,2',2''- terpyridine)iron(III) perchlorate." Journal of the American Chemical Society 96 (12): 3829-3834.

Reiff, W. M., W. A. Baker, et al. (1968). "Binuclear, oxygen-bridged complexes of iron(III). New iron(III)-2,2',2"-terpyridine complexes." Journal of the American Chemical Society 90 (18): 4794-4800.

Romeo, R., G. Arena, et al. (1992). "Steric and electronic factors influencing the reactivity of tertiary phosphines toward platinum(II) complexes." Inorganic Chemistry 31 (23): 4879-4884.

Rounaghi, G. H. and F. Mofazzeli (2005). "Study of Complex Formation Between Dicyclohexano-18-Crown-6 (DCH18C6) with Mg 2+ , Ca2 +, Sr 2+ , and Ba ; Cations in Methanol–Water Binary Mixtures Using Conductometric Method." Journal of Inclusion Phenomena and Macrocyclic Chemistry 51 (3): 205-210.

Rowan, S. J. and J. B. Beck (2005). "Metal-ligand induced supramolecular polymerization: A route to responsive materials." Faraday Discussions 128 : 43- 45.

Ruben, M., J.-M. Lehn, et al. (2006). "Addressing metal centres in supramolecular assemblies." Chemical Society Reviews 35 (11): 1056-1067.

188

Rubinstein, M. and A. N. Semenov (1998). "Thermoreversible Gelation in Solutions of Associating Polymers. 2. Linear Dynamics." Macromolecules 31 (4): 1386-1397.

Rubinstein, M. and A. N. Semenov (2001). "Dynamics of Entangled Solutions of Associating Polymers." Macromolecules 34 (4): 1058-1068.

Ruloff, R., G. v. Koten, et al. (2004). "Novel heteroditopic chelate for self-assembled gadolinium(III) complex with high relaxivity." Chemical Communications (7): 842-843.

Schmatloch, S., M. F. González, et al. (2002). "Metallo-Supramolecular Diethylene Glycol: Water-Soluble Reversible Polymers." Macromolecular Rapid Communications 23 (16): 957-961.

Schubert, U. S., C. Eschbaumer, et al. (2001). "2,2':6',2''-Terpyridine metal complexes as building blocks for extended functional metallo-supramolecular assemblies and polymers." Synthetic Metals 121 (1-3): 1249-1252.

Schubert, U. S., C. Eschbaumer, et al. (2001). "4'-Functionalized 2,2':6',2"terpyridines as building blocks for supramolecular chemistry and nanoscience." Tetrahedron Letters 42 (28): 4705-4707.

Schubert, U. S., O. Hien, et al. (2000). "Functionalized polymers with metal complexing segments: a simple and high-yield entry towards 2,2 ': 6 ',2 ''-terpyridine-based oligomers." Macromolecular Rapid Communications 21 (16): 1156-1161.

Schubert, U. S., H. Hofmeier, et al. (2006). Modern Terpyridine Chemistry . Weinheim, Wiley-VCH.

Schwarz, G., Y. Bodenthin, et al. (2009). "Structure and Properties of Dynamic Rigid Rod-Like Metallo-Supramolecular Polyelectrolytes in Solution." Macromolecules 43 (1): 494-500.

Schweitzer, G. K. and J. M. Lee (1952). "Racemization Rate Studies on Tris-(2,2'- dipyidyl)-nickel(II) Chloride." The Journal of Physical Chemistry 56 (2): 195-198.

Seiden, L., F. Basolo, et al. (1959). "Mechanism of Racemization of Complex Ions. V. The Dissociation and Racemization of Tris-(2,2'-bipyridine)-iron(II) and Tris- (1,10-phenanthroline)-iron(II) Ions in Water-Methanol and Methanol Solutions1,2." Journal of the American Chemical Society 81 (15): 3809-3813.

Shunmugam, R., G. J. Gabriel, et al. (2010). "Metal–Ligand-Containing Polymers: Terpyridine as the Supramolecular Unit." Macromolecular Rapid Communications 31 (9-10): 784-793.

Shunmugam, R. and G. N. Tew (2007). "Dialing in color with rare earth metals: facile photoluminescent production of true white light." Polymers for Advanced Technologies 18 (11): 940-945.

189

Shunmugam, R. and G. N. Tew (2008). "Terpyridine–Lanthanide Complexes Respond to Fluorophosphate Containing Nerve Gas G-Agent Surrogates." Chemistry – A European Journal 14 (18): 5409-5412.

Shunmugam, R. and G. N. Tew (2008). "White-light emission from mixing blue and red- emitting metal complexes." Polymers for Advanced Technologies 19 (6): 596-601.

Sillen, S. L., A. E. Martell, et al. (1964). Stability Constants of Metal-Ion Complexes . London, Chemical Society.

Sisley, M. J., Y. Yano, et al. (1982). "N-14 Magnetic-Resonance Studies of the Effect of Pressure and Temperature on the Rate of Exchange of Acetonitrile Solvent on Iron(Ii) and Manganese(Ii) Perchlorates." Inorganic Chemistry 21 (3): 1141-1145.

Sonogashira, K., Y. Tohda, et al. (1975). "Convenient Synthesis of Acetylenes - Catalytic Substitutions of Acetylenic Hydrogen with Bromoalkenes, Iodoarenes, and Bromopyridines." Tetrahedron Letters 16 (50): 4467-4470.

Spahni, W. and G. Calzaferri (1984). "Synthesis of Para-Substituted Phenyl-Terpyridine Ligands." Helvetica Chimica Acta 67 (2): 450-454.

Spike, C. G. and R. W. Parry (1953). "Thermodynamics of Chelation. I. The Statistical Factor in Chelate Ring Formation1." Journal of the American Chemical Society 75 (11): 2726-2729.

Stepto, R. F. T. (2009). "Dispersity in Polymer Science." Pure and Applied Chemistry 81 (2): 351-353.

Strouse, G. F., P. A. Anderson, et al. (1992). "Synthesis of polypyridyl complexes of ruthenium(II) containing three different bidentate ligands." Inorganic Chemistry 31 (14): 3004-3006.

Sullivan, B. P., J. M. Calvert, et al. (1980). "Cis-Trans Isomerism in (Trpy)(Pph3)Rucl2 - Comparisons between the Chemical and Physical-Properties of a Cis-Trans Isomeric Pair." Inorganic Chemistry 19 (5): 1404-1407.

Swift, T. J. and R. E. Connick (1962). "Nmr-Relaxation Mechanisms of O17 in Aqueous Solutions of Paramagnetic Cations and Lifetime of Water Molecules in First Coordination Sphere." Journal of Chemical Physics 37 (2): 307-&.

Tew, G. N., K. A. Aamer, et al. (2005). "Incorporation of Terpyridine into the Side Chain of Copolymers to Create Multi-Functional Materials." Polymer 46 : 8440-8447.

Thaler, A., B. G. Cox, et al. (2003). "Stability constants of aza-oxa-crown ether complexes with silver(I) in nonaqueous polar solvents." Inorganica Chimica Acta 351 (0): 123-132.

190

Thompson, A. (1997). "The synthesis of 2,2 ': 6 ',2 ''-terpyridine ligands - versatile building blocks for supramolecular chemistry." Coordination Chemistry Reviews 160 : 1-52.

Thummel, R. P. and Y. Jahng (1985). "N-Oxides of 2,2':6',2"-terpyridine." The Journal of Organic Chemistry 50 (19): 3635-3636.

Van Meter, F. M. and H. M. Neumann (1976). "The rates of racemization and dissociation of the tris(1,10-phenanthroline)iron(II) cation in various solvents." Journal of the American Chemical Society 98 (6): 1388-1394.

Vigee, G. S. and P. Ng (1971). "Line Broadening and Contact Shift Studies of Some Metal Ion-Dmso Complexes." Journal of Inorganic & Nuclear Chemistry 33 (8): 2477-2489.

Vitello, L. and J. Erman (2001). "Oxidation of bis(terpyridine)cobalt(II) chloride by cytochrome c peroxidase compounds I and II." Journal of Biological Inorganic Chemistry 6(5): 578-589.

Wade, L. G. (2003). Organic Chemistry . Upper Saddle River, NJ, Pearson Education, inc.

Wang, Q., P. Day, et al. (2006). "Synthetic strategies for preparing BEDT-TTF derivatives functionalised with metal ion binding groups." New Journal of Chemistry 30 (12): 1790-1800.

Webb, G. A. (2004). Nuclear Magnetic Resonance , Royal Society of Chemistry.

Weng, W., J. B. Beck, et al. (2006). "Understanding the Mechanism of Gelation and Stimuli-Responsive Nature of a Class of Metallo-Supramolecular Gels." Journal of the American Chemical Society 128 (35): 11663-11672.

Weng, W., A. M. Jamieson, et al. (2007). "Structural origin of the thixotropic behavior of a class of metallosupramolecular gels." Tetrahedron 63 (31): 7419-7431.

Werner, A. (1893). "Beitrag zur Konstitution anorganischer Verbindungen." Z. Anorg. Chem. 3: 267-330.

Whittle, B., S. R. Batten, et al. (1996). "Ruthenium(II) complexes of some new polynucleating ligands incorporating terpyridyl and macrocyclic aza-crown binding sites." Journal of the Chemical Society, Dalton Transactions (22): 4249- 4255.

Wilkins, R. G. (1970). "Mechanisms of ligand replacement in octahedral nickel(II) complexes." Accounts of Chemical Research 3(12): 408-416.

Wilkins, R. G. (1974). The Study of Kinetics and Mechanism of Reactions of Transition Metal Complexes . Boston, Allyn and Bacon.

191

Winter, A. and U. Schubert (2007). "New Polyester-Based Terpyridine Macroligands and their Blue Iron(II) Complexes." Macromolecular Chemistry and Physics 208 : 1956-1964.

Winter, A., A. Wild, et al. (2009). "Azido- and Ethynyl-Substituted 2,2':6',2"- Terpyridines as Suitable Substrates for Click Reactions." Synthesis 2009 (09): 1506-1512.

Worsfold, D. J. and S. Bywater (1964). "ANIONIC POLYMERIZATION OF ISOPRENE." Canadian Journal of Chemistry 42 (12): 2884-2892.

Wu, X. F., J. E. Collings, et al. (2001). "Ruthenium tris(bipyridine)-centered linear and star-shaped polystyrenes: Making atom transfer radical polymerization and metal complex initiators compatible." Macromolecules 34 (9): 2812-2821.

Wu, X. F. and C. L. Fraser (2000). "Architectural diversity via metal template-assisted polymer synthesis: A macroligand chelation approach to linear and star-shaped polymeric ruthenium tris(bipyridine) complexes." Macromolecules 33 (11): 4053- 4060.

Xu, D., J. L. Hawk, et al. (2010). "Mechanism of Shear Thickening in Reversibly Cross- Linked Supramolecular Polymer Networks." Macromolecules 43 (7): 3556-3565.

Yeung, T., P. C. Georges, et al. (2005). "Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion." Cell Motility and the Cytoskeleton 60 (1): 24-34.

Yoshida, H., T. Ozawa, et al. (1993). "Kinetics and Mechanism of Complex-Formation Reaction of Iron(Iii) with 4-Methyl-8-(Carboxymethyl)Aminomethyl Umbelliferrone and Related Ligands - Role of Substituent Groups on Nitrogen Donor Atoms on the Reaction." Polyhedron 12 (11): 1319-1328.

Yount, W. C., D. M. Loveless, et al. (2005). "Small-Molecule Dynamics and Mechanisms Underlying the Macroscopic Mechanical Properties of Coordinatively Cross-Linked Polymer Networks." Journal of the American Chemical Society 127 (41): 14488-14496.

Yount, W. C., D. M. Loveless, et al. (2005). "Strong Means Slow: Dynamic Contributions to the Bulk Mechanical Properties of Supramolecular Networks." Angewandte Chemie International Edition 44 (18): 2746-2748.

Zhang, L. W., Y. H. Zhang, et al. (2006). "Synthesis of bis(2,2 ': 6 ',2 ''-terpyridine)- terminated telechelic polymers by RAFT polymerization and ruthenium-polymer complexation thereof." European Polymer Journal 42 (10): 2398-2406.

Zhang, Z., S. Mikkola, et al. (2007). "Regio-selective synthesis of polyazacyclophanes incorporating a pendant group as potential cleaving agents of mRNA 5’-cap structure." Tetrahedron 63 (47): 11716-11723.

192

Zhao, Y., J. B. Beck, et al. (2004). "Rheological Behavior of Shear-Responsive Metallo- Supramolecular Gels." Macromolecules 37 (10): 3529-3531.

Zhou, G. and I. I. Harruna (2005). "Synthesis and Characterization of Bis(2,2':6',2"- terpyridine)ruthenium(II)-Connected Diblock Polymers via RAFT Polymerization." Macromolecules 38(10): 4114-4123.

Zhou, G. C. and Harruna, II (2004). "Synthesis and characterization of tris(2,2 '- bipyridine)ruthenium(II)-centered polystyrenes via reversible addition- fragmentation chain transfer (RAFT) polymerization." Macromolecules 37 (19): 7132-7139.

Zhou, G. C., Harruna, II, et al. (2005). "Ruthenium-centered thermosensitive polymers." Polymer 46 (24): 10672-10677.

Zhou, G. C., J. B. He, et al. (2007). "Self-assembly of amphiphilic tris(2,2 '- bipyridine)ruthenium-cored star-shaped polymers." Journal of Polymer Science Part a-Polymer Chemistry 45 (18): 4204-4210.

Zhou, Z., G. H. Sarova, et al. (2006). "Fullerene Polypyridine Ligands: Synthesis, Ruthenium Complexes, and Electrochemical and Photophysical Properties." Chemistry – A European Journal 12 (16): 4241-4248.

Ziegler, K. and H. Zeiser (1930). "Research on alkalic organic compounds, VII. Annoucement: Alkali metal alkyle and pyridin (Preliminary announcement)." Berichte Der Deutschen Chemischen Gesellschaft 63 : 1847-1851.

Ziegler, K. and H. Zeiser (1931). "Examinations on alkali-organic compounds. VII. Reactions between lithium alkyls, pyridine and condensated pyridine systems." Justus Liebigs Annalen Der Chemie 485 : 174-192.

193