Utilizing Metallosupramolecular Polymers As Smart

UTILIZING METALLOSUPRAMOLECULAR POLYMERS AS SMART

MATERIALS

JUSTIN R. KUMPFER

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Dissertation Adviser: Dr. Stuart J. Rowan

Department of Macromolecular Science and Engineering

CASE WESTERN RESERVE UNIVERSITY

May, 2012

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Justin Richard Kumpfer ______

Doctor of Philosophy candidate for the______degree.*

Dr. Stuart Rowan (signed)______

Dr. João Maia ______

Dr. Mohan Sankaran ______

Dr. David Schiraldi ______

March 20th, 2012 (date of defense)______

*We also certify that written approval has been obtained for any proprietary material contained therein.

Table of Contents

List of Figures vii

Acknowledgments xv

Abstract xvi

Chapter 1: Introduction 1

1.1 Metal-Ligand Interactions 2

1.2 Metallosupramolecular Polymers 7

1.2.1 Main-Chain Metallosupramolecular Polymers 11

1.2.2 Side-Chain Metallosupramolecular Polymers 23

1.3 Thesis Scope 26

1.4 References 28

Chapter 2: Stimuli-Responsive Europium-Containing Metallosupramolecular

Polymers 32

2.1 Introduction 33

2.2 Results and Discussion 37

2.2.1 Optical Properties of the Metallosupramolecular Polymer Films 40

2.2.2 Thermomechaincal Properties 44

2.2.3 Stimuli-Responsive Behavior 48

2.3 Conclusions 53

2.4 Experimental Methods 53

2.5 Acknowledgments 59

iii

2.6 References 60

Chapter 3: Influence of Metal-Ion and Polymer Core on the Melt Rheology of

Metallosupramolecular Films 63

3.1 Introduction 64

3.2 Results and Discussion 65

3.2.1 Influence of Eu3+ Content 67

3.2.2 Influence of the Core Polymer 76

3.2.3 Response at High Strain 83

3.3 Conclusions 87

3.4 Experimental Methods 88

3.5 Acknowledgments 90

3.6 References 91

Chapter 4: Thermo-, Photo-, and Chemo-Responsive Shape-Memory Properties

from Photocrosslinked Metallosupramolecular Polymers 95

4.1 Introduction 96

4.2 Results and Discussion 101

4.2.1 Material Properties 101

4.2.2 Thermally-Induced Shape-Memory Properties 106

4.2.3 Light-Induced Shape-Memory Properties 108

4.2.4 Effect of Different Metals and Counterions on the Light-Induced

Shape-Memory Behavior 115

4.2.5 Localized Shape-Memory Response 119

4.2.6 Chemo-Responsive Shape-Memory Effect 121

4.3 Conclusions 123

4.4 Experimental Methods 123

4.5 Acknowledgments 129

4.6 References 130

Chapter 5: Vapochromic and Piezochromic Films From Platinum Complex Polymer

Blends 135

5.1 Introduction 136

5.2 Results and Discussion 139

5.2.1 Spectroscopic Studies 142

5.2.2 Structural Properties 150

5.2.3 Mechanochromic behavior 153

5.3 Conclusions 157

5.4 Experimental Methods 158

5.5 Acknowledgments 163

5.6 References 163

Chapter 6: Fabrication of Platinum Nanoparticles Utilizing Metallosupramolecular

Polymers as Templates 169

6.1 Introduction 171

6.2 Results and Discussion 173

6.2.1 Morphological Studies 175

6.2.2 Reduction of Metallosupramolecular Polymers using Extracted

Discharge 182

6.2.3 Nanoparticle Growth and Assembly 187

6.3 Conclusions 191

6.4 Experimental Methods 191

6.5 Acknowledgments 195

6.6 References 195

Chapter 7: Self-Assembly of Metallosupramolecular Polymers in the Melt 197

7.1 Introduction 198

7.2 Results and Discussion 200

7.2.1 Polymer Synthesis and Characterization 200

7.2.2 Self-Assembly in the Melt 205

7.3 Conclusions 215

7.4 Experimental 215

7.5 Acknowledgments 221

7.6 References 221

Bibliography 225

LIST OF FIGURES

Figure 1.1 Examples of different metal-ligand coordination geometries. 3

Figure 1.2 a) Schematic representation of the equilibrium thermodynamic and kinetic properties of a 2:1 complex and b) schematic of selective ligand exchange. 5

Figure 1.3 Schematic representation of a metal-ligand complex for chemical sensing applications. 6

Figure 1.4 Schematic overview of some of the potential properties of metallosupramolecular polymers. 7

Figure 1.5 Influence of stoichiometry in main-chain supramolecular polymers on degree of polymerization as a function of concentration. 10

Figure 1.6 Example of dynamic ligand exchange between nonemissive compounds 1.4 and 1.5 to yield a new emissive polymer. 13

Figure 1.7 Structure of 1.7 used to prepared water soluble metallosupramolecular polymers and schematic of assembly with b) Nd3+ and c) Nd3+:Zn2+ in a 3:1 ratio. 15

Figure 1.8 Schematic illustration of the fabrication of a conducting metallosupramolecular polymer/electrode device for a nitric oxide sensor by measuring the resistance. 18

Figure 1.9 a) Images of stimuli-responsive metallosupramolecular gels which respond to electrical oxidation and reduction and b) the material properties as a function of the order of metal addition. 19

Figure 1.10 a) Schematic representation of the mechanism of optical healing of a metallosupramolecular polymer and images of self-healing with UV irradiation by b) AFM and c) visible recovery. 22

Figure 1.11 Example of side-chain metallosupramolecular polymers with lanthanides which demonstrate irreversible and/or reversible thermo-responsive changes in emission. 25

Figure 2.1 Synthesis of macromonomer 2. 37

vii

Figure 2.2 a) 1H NMR of macromonomer 2 and b) plot of molecular weight of 2 vs polymerization time. 39

Figure 2.3 UV-vis titration curves of 2 with a) Eu(ClO4)3 and c) Zn(ClO4)2 along with determination of the binding ratios for b) Eu(ClO4)3 and d) Zn(ClO4)2. e) Schematic representation of the self-assembly of 2 with Zn2+ and Eu3+. 41

Figure 2.4 Picture before and after stretching of a Zn2+:Eu3+ 70:30 film with 2. 42

Figure 2.5 a) Images of the fluorescence of a series of films with Zn2+:Eu3+ ratios ranging from 100:0 to 50:50, b) the corresponding PL spectra, and c) schematic of the antenna effect. 43

Figure 2.6 DMTA temperature sweeps of films with Zn2+:Eu3+ 100:0-50:50 ratios a) annealed and b) unannealed. 45

Figure 2.7 DMTA temperature sweeps of films with Zn2+:Eu3+ 80:5-80:20 ratios. 46

Figure 2.8 a) WAXS diffractograms of select films with Zn2+:Eu3+ ratios of 100:0, 90:10, 70:30, and 50:50 and b) a schematic of the change in ordering as the amount of Eu3+ is increased. 48

Figure 2.9 a) Pictures of the thermal stimuli-responsive sensing capability of a Zn2+:Eu3+ 70:30 film. Influence on the PL spectra with temperature focusing on the Eu3+ metal-based emission as b) temperature is increased and c) temperature is cycled between room temperature and 120 °C. 50

Figure 2.10 Pictures of a Zn2+:Eu3+ 70:30 film a) as cast, b) dipped in a 10 mM solution of triethyl phosphate in hexanes for 1 min., c) exposed to triethyl phosphate vapor for 24 hrs., and d) dipped directly into triethyl phosphate along with d) the corresponding change in the PL spectra. 52

Figure 2.11 13C NMR of 2. 56

Figure 2.12 MALDI-TOF of 2. 56

Figure 2.13 MALDI-TOF-TOF of 2. 57

Figure 2.14 PL spectra of 2 with Zn2+:Eu3+ 100:0-50:50 in solution. 58

Figure 2.15 TGA of films of 2 with Zn2+:Eu3+ 100:0-50:50. 58

viii

Figure 3.1 Chemical structures of macromonomers 2 and 3. 66

Figure 3.2 Master curves of films prepared with 2 containing Zn2+:Eu3+ 100:0-50:50 ratios. 69

Figure 3.3 Time-temperature superposition curves of tan δ vs ω a) for films prepared with 2 and Zn2+:Eu3+ 100:0-50:50 with temperatures ranging from 30 °C to 110 °C and b) a Zn2+:Eu3+ 60:40 film with 2 from -30 °C to 70 °C. 71

Figure 3.4 Shift factors a) horizontal and b) vertical for films prepared with 2 with Zn2+:Eu3+ 100:0-50:50 ratios. 73

Figure 3.5 Time-temperature superposition plot of complex viscosity vs ω for films prepared with 2 with Zn2+:Eu3+ 100:0-50:50 ratios. 76

Figure 3.6 TEM micrographs of a) 100% Zn(NTf2)2 and b) 100% La(NTf2)3 films prepared with 3. 78

Figure 3.7 SAXS of a) films prepared with 2 with Zn2+:Eu3+ 100:0-50:50 ratios and b) comparison of Zn2+:Eu3+ 100:0 films prepared with 2 and 3. c) Schematic of ordered self-assembly of 2 and 3 upon addition of Zn2+ and/or Eu3+. 79

Figure 3.8 Master curves comparing film prepared with 2 vs 3 for Zn2+:Eu3+ ratios of a) 100:0, b) 50:50, and c) 0:100. 81

Figure 3.9 a) Horizontal and b) vertical shift factors of film prepared with 2 and Zn2+:Eu3+ 0:100 and prepared with 3 with Zn2+:Eu3+ 100:0, 50:50, and 0:100. 82

Figure 3.10 Strain sweeps of films prepared with 2 and Zn2+:Eu3+ ratios a) 100:0, b) 90:10, c) 80:20, d) 70:30, e) 60:40, and f) 50:50 along with images of the stress response both inside and outside the viscoelastic region. 84

Figure 4.1 a) Example of typical shape-memory polymer behavior and b) an example of a typical one-way shape-memory curve. 97

Figure 4.2 Schematic of the mechanism of light-induced shape-memory behavior for films prepared with 4. 100

Figure 4.3 a) Synthesis of macromonomer 4 and b) 1H NMR of macromonomer 4. 102

Figure 4.4 Schematic of photocrosslinking of films prepared with 4 and Eu3+ using a tetrathiol crosslinker. 103

Figure 4.5 Pictures of using photocrosslinking to prepare a 4.1∙Eu(NTf2)3 film with a permanent spiral shape. 104

Figure 4.6 a) SAXS of an uncrosslinked film of 4∙Eu(NTF2)3 and b) DMTA temperature sweeps of 4-4.3∙Eu(NTf2)3 films. 105

Figure 4.7 Thermal one-way shape-memory cycles for 4.1-4.3∙Eu(NTf2)3 films. 108

Figure 4.8 Pictures of light-induced shape-memory behavior of a 4.1∙Eu(NTf2)3 film. 109

2 Figure 4.9 Pictures of a 4.2∙Eu(NTf2)3 film a) under low intensity (70 mW/cm ) UV light and b) selectively irradiating half of the film with high intensity (1000 mW/cm2) UV light. Controlled force shape-memory experiments on c) 4.1∙Eu(NTf2)3, d) 4.2∙Eu(NTf2)3, and e) 4.3∙Eu(NTf2)3 films. 111

Figure 4.10 Strain fixing and recovery of films of 4.1-4.3∙Eu(NTf2)3. 112

Figure 4.11 a) Controlled force shape-memory experiment of a 4.2∙Eu(NTf2)3 film using a cutoff (<400 nm) filter and b) the effect of crosslinking on the creep rate for 4.1-4.3∙Eu(NTf2)3 films. 114

* Figure 4.12 SAXS of a 4.1∙Eu(NTf2)3 film. a) 1D plot of q in the permanent, strained, and recovered state and b) 2D scattering plot. 115

Figure 4.13 Controlled force shape-memory experiments of a) 4.2∙Zn(NTf2)2 and b) 4.2∙Zn(OTf)2. Strain fixing and recovery of 4.2 with Eu(NTf2)3, Zn(NTf2)2, and Zn(OTf)2. 117

Figure 4.14 DMTA temperature sweeps of a) a 4.2∙Zn(NTf2)2 film, b) a 4.2∙Zn(OTf)2 film, and c) uncrosslinked films of 4∙Eu(NTf2)3, 4∙Zn(NTf2)2, and 4∙Zn(OTf)2. 119

Figure 4.15 Pictures of localized shape-memory behavior with UV light of a 4.2∙Eu(NTf2)3 film by selectively opening sides of the film in a box shape. 120

Figure 4.16 Pictures of chemo-responsive shape-memory behavior of a 4.2∙Eu(NTf2)3 film with methanol vapor. 122

Figure 4.17 13C NMR of 4. 126

Figure 4.18 MALDI-TOF of 4. 127

Figure 4.19 UV-vis titration of 4 with Zn(ClO4)2. 127

Figure 5.1 a) Synthesis of 5 and complexation with Pt2+ and 1H NMR of b) 5 and c) [Pt(5)Cl](PF6). 140

Figure 5.2 Images of a) the vapochromic response of 10% w/w [Pt(5)Cl](PF6) in PMMA upon exposure to acetonitrile vapor, POM of b) 2.5% and c) 10% w/w [Pt(5)Cl](PF6) in PMMA, and FM of d) 2.5% and e) 10% w/w [Pt(5)Cl](PF6) in PMMA. 142

Figure 5.3 UV-vis curves of 10% w/w [Pt(5)Cl](PF6) in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA demonstrating a vapochromic response to acetonitrile vapor and recovery after 24 hrs. 143

Figure 5.4 Kinetics of the change in UV-vis absorption upon acetonitrile vapor exposure and recovery after 24 hrs or heating to 100 °C of a 10% w/w [Pt(5)Cl](PF6) in PEMA film. 145

Figure 5.5 Images of the emission of thin films of 10% [Pt(5)Cl](PF6) w/w in PMMA a) as cast and b) after exposure to acetonitrile. c) PL of the emission change with time upon exposure. 147

Figure 5.6 Normalized PL spectra of the vapochromic behavior of films prepared with 10% w/w [Pt(5)Cl](PF6) in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA. 148

Figure 5.7 Unnormalized PL spectra of heating a 10% w/w [Pt(5)Cl](PF6) in PMMA film. 149

Figure 5.8 WAXS diffractograms of films of PMMA containing a) 2.5, b) 5, c) 10, and d) 20% w/w [Pt(5)Cl](PF6). 150

Figure 5.9 WAXS diffractograms of the neat polymer matrix and 10% w/w [Pt(5)Cl](PF6) in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA before and after acetonitrile exposure. 152

Figure 5.10 Mechanochromic response of [Pt(5)Cl](PF6) on filter paper demonstrating a) yellow color and b) emission and c) orange color and d) emission after scraping. e) PL spectrum of [Pt(5)Cl](PF6) before and after mechanical stimulation. 154

Figure 5.11 Images of the mechanoluminescent change of a 10% w/w [Pt(5)Cl](PF6) in PMMA film upon a) scratching or b) hitting and c) the PL spectrum of the compressed film. 156

13 Figure 5.12 C NMR of 5 in CDCl3. 160

Figure 5.13 NALDI of 5. 160

13 Figure 5.14 C NMR of [Pt(5)Cl](PF6) in (CD3)2SO. 161

Figure 5.15 NALDI of [Pt(5)Cl](PF6). 162

Figure 6.1 Synthesis of [Pt2(2)Cl2](Cl2), [Pt2(3)Cl2](Cl2), and [Pt2(6)Cl2](Cl2). 171

Figure 6.2 Schematic of the reduction of the metallosupramolecular polymer thin films by extracted discharge to fabricate nanoparticles. 173

Figure 6.3 a) DMTA of [Pt2(3)Cl2](Cl2). b) Images of films prepared from b) [Pt2(3)Cl2](Cl2) and c) [Pt2(2)Cl2](Cl2). 175

Figure 6.4 Morphological characterization of a [Pt2(3)Cl2](Cl2) film using a) SAXS and b) WAXS and c) SAXS and d) WAXS of a film of [Pt2(2)Cl2](Cl2). 177

Figure 6.5 Schematic showing the partial short range structures of metallosupramolecular polymers. 179

Figure 6.6 Experimental setup of the reduction of [Pt2(2)Cl2](Cl2), [Pt2(3)Cl2](Cl2), or [Pt2(6)Cl2](Cl2) by extracted discharge from a microplasma. 180

Figure 6.7 TEM images of platinum nanoparticles formed from [Pt2(2)Cl2](Cl2) after exposure times of a) 0, b) 0.5, c) 1, and d) 3 hrs. 181

Figure 6.8 TEM of [Pt2(2)Cl2](Cl2) after attempts at reduction by a) extracted discharge (200 µA), b) microplasma (0 µA), c) UV light, and ethylene glycol reflux. 183

xii

Figure 6.9 Particle size distributions for [Pt2(2)Cl2](Cl2), [Pt2(3)Cl2](Cl2), and [Pt2(6)Cl2](Cl2) with exposure time. 185

Figure 6.10 Effect of exposure time and polymer core on a) average particle size and b) particle density. 186

1 Figure 6.11 H NMR of a) 2 and [Pt2(2)Cl2](Cl2). 191

Figure 7.1 Schematic representation of self-assembly of a metallosupramolecular network at the interface between two polymer layers. 200

Figure 7.2 a) Synthesis of 7x and b) 9x. 201

Figure 7.3 GPC of 710 and 750. 202

1 Figure 7.4 H NMR of 710 and 910. 203

13 Figure 7.5 C NMR of 8, 710, and 910. 204

Figure 7.6 TGA of a) 710 and b) 910. 205

Figure 7.7 UV-vis before and after formation of a metallosupramolecular network with a) 910 (10% w/w), b) 910 (25% w/w), c) 950 (10% w/w), and d) 950 (25% w/w). 207

Figure 7.8 PL before and after formation of a metallosupramolecular network with a) 910 (10%), b) 910 (25%), c) 950 (10%), and d) 950 (25%). 209

Figure 7.9 a) Unnormalized and b) normalized variable-temperature PL spectra of a 2+ two-layer film of 910:PS / Zn :PMMA (10% w/w). 209

Figure 7.10 a) Image of the change in emission color and intensity upon formation of a 2+ metallosupramolecular network from 910:PS / Zn :PMMA (10% w/w) and b) image of the phosphorescence observed in the same film. 211

2+ Figure 7.11 Change in UV-vis and PL with contact time for 910:PS / Zn :PMMA (10% and 25% w/w). 212

Figure 7.12 Confocal microscopy of the self-assembled polymers prepared from 2+ 2+ 910:PS / Zn :PMMA (10% and 25% w/w) and 950:PS / Zn :PMMA. 215

xiii

Figure 7.13 1H NMR of 8. 219

Figure 7.14 13C NMR of 8. 219

Figure 7.15 NALDI of 8. 220

xiv

ACKNOWLEDGMENTS

“In his heart a man plans his course, but the LORD determines his steps.” Proverbs 16:9

I would like to thank first and foremost my Lord and Savior Jesus Christ for giving me

His grace, love, strength, wisdom, and guidance. Without Him I would not be the person

I am today.

I would like to thank my amazing wife, Sarah, for her never-ending support, love, and optimism and also my family for their love and constant support.

Finally I would like to thank my advisor, Stuart, for his guidance and advice and also my fellow group members for their help and companionship.

Utilizing Metallosupramolecular Polymers as Smart Materials

Abstract

JUSTIN R. KUMPFER

This dissertation focuses on the preparation and properties of a number of

metallosupramolecular polymers, which combine the processability and mechanical

properties of polymers with the functionality of metals. Ditopic 2,6-bis(N- methylbenzimidazol-2ʹ-yl)pyridine (Mebip) ligand end-capped metallosupramolecular polymers coordinated with Zn2+ and Eu3+ metal-ions and their resulting stimuli-

responsive properties were initially investigated. These polymers were found to exhibit

thermo- and chemo-responsive properties directly related to the Eu3+ content. Structure-

property relationships were investigated using detailed rheological and morphological

studies which revealed that the nature of the polymer core strongly influences the

mechanical properties and degree of phase separation of the metal-ligand complexes.

This work was expanded upon using a crosslinkable polymer core to form metallosupramolecular polymer films which show shape-memory properties. The shape- memory behavior was able to be triggered using temperature, UV light, or solvents and the shape fixing was tailorable using different metal-ions and counterions. Continued investigations into the use of different metal-ions yielded mechanically stable

xvi

vapochromic materials by blending small-molecule Mebip:Pt2+ complexes into a series of

polymethacrylates. The use of a polymer matrix revealed that the films also displayed

piezochromic properties. Pt2+ was then used to form metallosupramolecular polymers with the ditopic macromonomers previously studied to give films which assemble via Pt

– Pt interactions. Thin films of the Pt2+-containing polymers were utilized as templates in the formation of Pt nanoparticles with <5 nm diameter by reduction using electrons generated from an extracted discharge from an atmospheric microplasma. The particle size and density was found to be directly related to the polymer core and exposure time.

Finally, a side-chain Mebip-functionalized polymer blended into a polymer matrix was compression molded with a polymer blend containing Zn2+ to form a

metallosupramolecular polymer network at the polymer-polymer interface. Spectroscopy

was used to confirm that formation of the metallosupramolecular polymer occurred

within 5 mins at the processing temperature (210 °C) and microscopy confirmed the

location of the metallosupramolecular polymer to be at the interface between the two

polymer layers.

xvii

CHAPTER 1

Introduction

The interest and development of supramolecular materials1-3 has grown

tremendously in the past two decades beginning with the Nobel Prize winning work of

Lehn, Pederson, and Cram in 1987 on host-guest interactions with crown ethers.4 The use

of supramolecular chemistry has in large part grown because of the vast potential of

possible assemblies resulting from a wide variety of available interactions, including

hydrogen bonding,5-7 π – π interactions,8 and metal coordination.9-11 Supramolecular

polymers differ from conventional polymers as their formation results from self-assembly which requires no use of initiators or catalysts, and that the noncovalent interactions can be reversible in nature which has led to their use as stimuli-responsive materials which are more susceptible to various environmental factors (heat, pH, solvent, etc…).

A class of supramolecular polymers which have received increasing attention are metallosupramolecular polymers12-14 which utilize noncovalent interactions between metals and ligands as the self-assembling motif. Metallosupramolecular polymers are interesting as they offer the potential to combine the mechanical properties and processability of polymers with the inherent functionalities of metals.

1.1 METAL-LIGAND INTERACTIONS

A compelling aspect to using metal-ligand interactions is that there is a huge collection of ligands available which have the potential to be combined with the majority of transition metals and lanthanide metals. The characteristics of the metal-ligand assembly, such as geometry, stoichiometry, reversibility, and stability, can be tailored by judicious choice of the ligand and metal-ion used which allows for the development of materials with targeted properties.

The geometry of the assembly can be influenced by the ligand (monodentate vs

bidentate vs terdentate) or the number of coordination sites of the metal-ion. Along with

this, there can be homo-complexes, in which all the ligands are the same, or hetero-

complexes, which utilize two or more types of ligands. An example of the difference in

geometry with simple changes to the ligand and/or metal is shown in Figure 1.1 for

homo-complexes with either bipyridine or terpyridine. A hexacoordinate metal (ex. Zn2+,

Fe2+, etc…) is able to form 3:1 ligand:metal-ion complexes when a bidentate ligand is

used, however, if the same metal is coordinated with a terdentate ligand, a 2:1

ligand:metal-ion complex will be obtained. Furthermore, changing the metal-ion to a lanthanide which has more coordination sites (8-10), a 3:1 complex can be created using a terdentate ligand. This demonstrates just a simple example of the variability in designing metal-ligand architectures using complexes containing the same ligands and the use of multiple different ligands allows for the creation of much more structurally complex architectures. Using this, it is easy to envision fine-tuning the metal-ligand complex to meet desired criteria by careful selection of metal and ligand.

Figure 1.1 Examples of different metal-ligand geometries by changing the ligand (bidentate or terdentate) or the metal (MII vs MIII).

Other than geometry, two important factors in designing materials that use metal-

ligand interactions are the kinetics (labile vs inert) and the thermodynamic stability. The

kinetic parameters (k1, k-1, k2, k-2, etc…) relate to the rate of

complexation/decomplexation as shown in Figure 1.2a. Along with the kinetics, the

thermodynamic properties, which are expressed in terms of the individual or overall

binding constant (K1 or Kn), relate to the stability of the metal-ligand complex, with

stronger binding systems being more stable. An example of the large differences in the

thermodynamic and kinetics properties is shown by an examination of complexes formed

with terpyridine ligands and various metals.15 By simply changing the metal-ion used

from Fe2+ to Co2+, the overall binding constant decreases by approximately three orders

of magnitude. Likewise, a dramatic change in the dissociation rates were observed with a

change from ~10-3 s-1 for a 2:1 terpyridine:Co2+ complex to ~10-7 s-1 for the same

complex with Fe2+. Of course the thermodynamic and kinetic parameters can be

influenced by their environment as stated before which can also allow for further fine-

tuning of the metal-ligand properties.

An example of utilizing the differences in the binding constants of different

ligands in the same metal-ligand complex to produce stimuli-responsive behavior has

been shown by Collin and coworkers.16 They synthesized a hetero-complex by coordinating Ru2+ with one terpyridine ligand, one phenanthroline ligand, and a range of monodentate ligands. An example of the metal-ligand complex formed using pyridine as the monodentate ligand is shown in Figure 1.2b. They found that irradiating the complex with white light while in acetonitrile resulted in a ligand exchange in which the monodentate ligand was replaced by the acetonitrile solvent. As the terpyridine and

phenanthroline coordination with Ru2+ is essentially inert,17,18 the monodentate ligand,

which has the weakest binding constant, with Ru2+ is selectively replaced by a

coordinating solvent molecule.

Figure 1.2 a) Schematic representation of the equilibrium thermodynamic (K1, K2) and kinetic

(k1, k-1, k2, k-2) properties of a 2:1 ligand:metal-ion complex. b) Example of selective ligand

exchange of the pyridine in complex 1.1 with acetonitrile to form complex 1.2 using

photoirradiation.

The use of competitive binders to disrupt or alter a metal-ligand complex is another method to achieving stimuli-responsive behavior. The optical properties of a dialkoxyarylethynylene functionalized 2,6-bis(N-methylbenzimidazol-2ʹ-yl)pyridine ligand (1.3) complex with Eu3+ were used as a method to detect the presence of chemical

warfare agent vapors.19,20 When 1.3 is complexed with Eu3+, it displays an intense red

colored emission as a result of the “antenna effect” in which 1.3 absorbs UV light and

transfer that energy to the Eu3+ metal-ion to give a distinctive metal-based emission. In

this study, this highly emissive metal-ligand complex was exposed to organophosphates

which act as chemical warfare agent mimics. The organophosphates are able to displace

the 1.3 and bind the Eu3+ as shown in Figure 1.3. The decomplexation of 1.3:Eu3+

produces a change in the emission color from the red metal-based complex emission to that of the free ligand and the organophasphate:Eu3+ (nonemissive) which in the case of

1.3 is green. The study also found that the response could be tailored by using different metal-ions (La3+ and Zn2+) leading to the use of arrays which could selectively sense

organophosphates and some amines.

Figure 1.3 Proposed displacement mechanism for the supramolecular sensors which

utilize the change in emission from the red metal-based complex emission to a green

ligand emission.

The dynamic binding of metal-ligand complexes offers the possibility for a wide

range of stimuli-responsiveness with the potential to fine-tune the desired response by

proper ligand and metal selection. However, small molecule metal-ligand complexes

suffer from a lack of mechanical properties which limit their usefulness. The use of

metal-ligand complexes to create metallosupramolecular polymers affords an elegant

6 approach to retaining the characteristics of the small molecule metal-ligand properties while adding the mechanical properties of polymers.

1.2 METALLOSUPRAMOLECULAR POLYMERS

The aim of creating metallosupramolecular polymers is to design materials with new and interesting properties. Therefore, careful consideration needs to be taken when choosing each element in order to select for the desired properties. An example of just a small portion of the available properties is shown in Figure 1.4. As has been already discussed, the choice of appropriate ligand and metal will determine the dynamic properties of the polymer, but the characteristics of the polymer are also extremely important. Many conventional polymers can now be used in conjunction with metal-ligand interactions so there is a huge library of available properties.

Figure 1.4 Schematic overview of some of the potential properties of metallosupramolecular polymers.

One of the biggest differences between metallosupramolecular polymers and

conventional, covalent polymers is the molecular weight dependence on the binding

constant, solvent, temperature, etc… Typically, metallosupramolecular polymers are

assembled from low molecular weight monomers or oligomers which upon introduction

of the metal-ion form high molecular weight polymers (assuming the metal-ligand

complex is chosen to have a binding geometry of two or more ligands per metal). For an

ideal system in which there is no ring formation, the degree of polymerization (DP) for

reversible metallosupramolecular polymers is given by

~ ( [ ]) / (Eq. 1.1) 1 2 where퐷푃 K퐾 is푀 the overall binding constant and [M] is the monomer concentration.21 From

this equation it is apparent that in order to achieve high molecular weight assemblies both

the binding constant and the monomer concentration must be large. Furthermore, if the

concentration is low the probability of forming macrocycles rather than linear polymer

chains increases which decreases the overall degree of polymerization.

Another extremely important consideration in the formation of

metallosupramolecular polymers is the stoichiometry of the system. In order to attain chain extension, a 2:1 ligand:metal-ion ratio is necessary as a 1:1 ligand:metal-ion ratio

will result in chain termination. Thus, for a ditopic monomer which contains the metal- chelating ligands on the end-groups, self-assembly to high molecular weight is produced using a 1:1 ratio of monomer:metal-ion. While the reversibility of the metal-ligand interaction does allow for some self-correcting, small deviations from a 1:1 ratio will have large impacts on the size of the assembly. A theoretical example of this effect can be seen in Figure 1.5.22 In this example, the average number of ditopic monomers is

plotted as a function of concentration for different stoichiometries using log[K1] and

2+ log[K2] for an Fe :terpyridine complex where log[K1] << log[K2]. These results show that for an exactly 1:1 stoichiometry, the size of the metallosupramolecular polymer will be very large (theoretically forming a single high molecular weight polymer chain).

However, if the stoichiometry is off by just 1% the polymer growth is stopped as a result of a deficit of one component and only reaches a plateau at higher concentrations (see

Figure 1.5a). Figure 1.5b more clearly demonstrates the effect of stoichiometry and concentration on the degree of polymerization. While higher concentrations can give large degrees of polymerization, it also makes the assembly more susceptible to slight changes in stoichiometry. An interesting feature of the curves presented in Figure 1.5b is that the curves are asymmetric with respect to stoichiometry factor, y, and that an excess of metal results in longer assemblies than excess monomer. Likewise, Dormidontova and coworkers23 used Monte Carlo simulations to study equilibrium metallosupramolecular

polymers which also confirmed that the degree of polymerization is very stoichiometry

dependent.

Figure 1.5 a) Average number of ditopic monomers per assembly as a function of

concentration (C) for different stoichiometries (y) using stability constants of Fe2+:terpyridine. b)

Plot of vs y for different monomer concentrations (a) 10-3 mol∙L-1, (b) 10-4 mol∙L-1, and (c)

10-5 mol∙L-1. c) Degree of association for rings and chain as a function of stoichiometry.

When designing a metallosupramolecular polymer, the location of the metal- ligand complex must also be considered. Using a ditopic monomer with the ligands on the polymer chain-ends will result in a metallosupramolecular polymer in which the metal-ion is located in the polymer backbone, to yield a “main-chain” polymer.24 Other

polymeric architectures25 which can be formed include side-chain,26,27 branched,28

crosslinked,29 star-shaped,30,31 dendritic,32,33 or helical.34,35 The work presented herein

primarily utilizes main-chain, branched, and side-chain functionalized

metallosupramolecular polymers and as such those architectures will be introduced in

more detail focusing on stimuli-responsive polymers.

1.2.1 Main-Chain Metallosupramolecular Polymers

Main-chain metallosupramolecular polymers are those in which the metal-ligand motif is

incorporated directly into the polymer chain. This is typically accomplished by the

addition of the appropriate metal-ion to a ditopic ligand end-capped monomer

(sometimes referred to as macromonomer if the spacer is oligomeric or polymeric) in a

1:1 ratio resulting in a metallosupramolecular polymer with many metal-ligand complexes in the backbone. These polymers typically exhibit more dynamic behavior than covalent polymers as the reversibility of the metal-ligand complex allows for polymerization/depolymerization and ring-chain equilibria among others which has been used to create stimuli-responsive materials.

The first report of a main-chain metallosupramolecular polymer utilizing dynamic metal-ligand complexes was published by Rehahn and coworkers.36 In it, they

synthesized a ditopic monomer with phenanthroline ligand end-groups and formed

metallosupramolecular polymers by the addition of Ag+ and/or Cu+. Their report demonstrated the influence of choice of solvent on the assembly of high molecular weight polymers. Polymerization with metal-ions was shown to result in stabile polymeric molecules when accomplished in a noncoordinating solvent, however, it was found that the presence of even a small amount of a coordinating solvent, such as acetonitrile or pyridine, lead to highly labile metal-ligand complexes that gave high rates of exchange to produce low molecular weight species and cyclics. They argued that the solvent molecules could competitively bind the metal-ion, increasing the rate of decomplexation and produced chain cleavage at the location of the metal-ligand complex.

The difference in behavior based on the environment (solvent) is an example of the

potential of using metal-ligand complexes to obtain stimuli-responsive behavior. To

further emphasize the use of solvent to influence the metal-ligand dynamics, Rehahn and

coworkers also looked at the mixing behavior of two separate polymers. One polymer

prepared with only Ag+ was combined with another polymer prepared only with Cu+.

When mixed in a noncoordinating solvent, the polymers were found to be stable for days

with no evidence of mixing between the two polymers from decomplexation from one

and recomplexation to the other. Only upon adding a coordinating solvent (acetonitrile)

did decomplexation occur which yielded polymers that incorporated a mixture of Ag+ and

Cu+. This well-designed example of using the metallosupramolecular polymer’s

environment to induce a change in the characteristics provides for evidence of the

stimuli-responsiveness that can be easily achieved in metal-ligand containing systems.

While Rehahn and coworkers were able to control the labile nature of the metal- ligand interactions with solvent, Lehn and coworkers37,38 explored their use in the solid-

state in order to change a material’s properties. Neutral main-chain

metallosupramolecular polymers were investigated using ligands based on

pyridylhydrazone and acylhydrazone units on the chain-ends of low moleculer weight

poly(dimethyl siloxane) (PDMS) assembled using Zn2+ or Ni2+. They found that tailoring

the ligand functionality allowed for them to prepare a set of materials with very different

optical and mechanical properties. In an elegant example of using the dynamic nature of

the metal-ligand interactions to generate novel properties, they found that simply placing two different metallosupramolecular polymers together and heating (50 °C, 24 hrs) they could get ligand exchange between the different polymers without the use of solvent or catalysts as shown schematically in Figure 1.6a. For example, a polymer blend (1.6) was

prepared using two homopolymers 1.4, which is a very weakly emissive gum, and 1.5, a

nonemissive film (see Figure 1.6b). When stacked and heated, ligand exchange between the homopolymers creates a new polymer blend which has properties not found in either of the starting homopolymers. Polymer blend 1.6 displayed mechanical properties which fell in-between polymers 1.4 and 1.5, however the blend now displayed a strong emission not apparent in 1.4 or 1.5. This study illustrates the potential of using the dynamic characteristics of metallosupramolecular polymers in solid films to create novel materials properties.

Figure 1.6 a) Example of dynamic ligand exchange between neutral homopolymers 1.4 and 1.5 to form the hetropolymer 1.6 without the need for solvent or catalysts. b) Images the change in emission from 1.4 (left) and 1.5 (middle) which are nonemissive to the emissive polymer 1.6

(right) under excitation at 365 nm. The emissive hetero-complex is highlighted.

Schubert and coworkers39 have reported on the synthesis and characterization of a

series of main-chain metallosupramolecular polymers by combining a ditopic terpyridine

end-capped poly(ethylene glycol) with a variety of different transition metal-ions to

create high molecular weight polymers. Detailed studies were conducted to determine the

influence of different metal-ions on the degree of polymerization based on the differences

in the thermodynamic stability of each complex. Using concentration-dependent viscosity

measurements, they were able to directly relate the observed relative viscosities to the

degree of polymerization for metallosupramolecular polymers formed with Fe2+, Co2+,

Ni2+, Cd2+, and Cu2+. They found that the trend in viscosity followed with the stability

constants for the small molecule biscomplexes as Fe2+ > Ni2+ > Co2+ > Cu2+ > Cd2+. They also found that upon self-assembly films were able to be formed by solution-casting from either methanol or chloroform as a demonstration of the ability to form high molecular weight polymers from low molecular weight starting materials.

One advantage that polymers formed with metal-ligand interactions have over hydrogen-bonded systems is that the proper choice of metal-ion and ligand can be used to create water-soluble polymers, which is not possible when using hydrogen bonding as they are subject to disruption by the coordination of water molecules. Cohen Stuart and coworkers have prepared multiple examples of water-soluble metallosupramolecular polymers using a short poly(ethylene glycol) PEG chain end-capped with pyridine-2,6-

dicarboxylate ligands (1.7).40,41 Aqueous solutions of reversible coordination polymers

prepared with Nd3+ were studied using rheological techniques to determine the equilibrium chain structure. As Nd3+ is able to coordinate up to three ligands, it may be

expected that the resulting metallosupramolecular polymers have a highly branched or

crosslinked structure. Interestingly, however, examination of the elastic modulus, zero-

shear viscosity, and relaxation time with concentration match well with predictions for

the dynamics of linear equilibrium polymers. With this information, they proposed a

linear chain structure which contained small cyclic species as shown in Figure 1.7 as a

branched structure would result in increased chain-ends, thus the equilibrium of the

system is shifted toward linear chains. In an effort to justify this claim they explored

mixed metal systems containing Zn2+ along with Nd3+. They found that the mixed metal

systems showed liquid-like viscosity which they claim is primarily a result of the

incorporation of Zn2+ into the cyclics which, as Zn2+ forms 1:2 complexes with the

ligand, result in chain terminators. They also found that the 100% Nd3+ polymer was able

to form a thermal-responsive gel where elevated temperatures induced liquid-like

behavior and that the viscoelastic properties returned upon cooling.

Figure 1.7 a) Structure of ligand 1.7. Proposed structure of b) 100% Nd3+ (black circles) polymer with 1.7 and c) example of chain termination by addition of 25% Zn2+ (gray circles).

Block copolymers represent another type of main-chain supramolecular polymer

architecture which has generated interest in recent years.42,43 Diblock copolymers with

the structure A-[M]-B and triblock copolymers with a A-[M]-B-[M]-A structure have been reported by Gohy and Schubert44,45 where A represents one polymer, B represents a

different polymer, -[ is the ligand, and M represents the metal-ion. In the diblock copolymer, only one metal-ligand complex is present in the polymer backbone, however its presence allows for these block copolymers to behave differently than traditional block copolymers as the choice of ligand, metal, and counterion will impact the assembly behavior. Construction of metal-ion containing block copolymers is not as straight forward as mixing of the two components and metal will result in a mixture of homopolymers and copolymers. Rather, one must use an appropriate metal, such as Ru3+,

which forms stable monocomplexes, followed by reduction of the metal to Ru2+ in the presence of the second polymer. Schubert and coworkers studied the phase separated morphology of a polystyrene-block-poly(ethylene oxide) connected with a 2:1 terpyridine:Ru2+ complex in the melt. It was found that in this system the counterion used

- played a large role in differentiating between a lamellar or spherical morphology. PF6

counterions gave spherical morphologies resulting from aggregation of the metal-ligand

- complexes while a bulkier BPh4 counterion yielded a lamellar morphology with the metal-ligand complexes at the lamellar interfaces. The ability to tailor properties with counterions just gives another tool for fine-tuning the behaviors of metallosupramolecular polymers.

One of the key aspects of metallosupramolecular polymers is their use as stimuli- responsive materials. As synthetic pathways to many metallosupramolecular polymers

are now available, more focus has been placed on their use as functional materials. A

field in which this is apparent is that of sensors. Swager and coworkers46 have reported

the use of an electropolymerizable metallosupramolecular polymer as a sensor for nitric

oxide (NO) by utilizing the polymer conductivity. The polymer was synthesized using a

Co2+ complex with a salen ligand functionalized with 3,4-(ethylenedioxy)thiophene

groups (1.8) by electropolymerization when between two interdigitated microelectrodes

(see Figure 1.8). The choice of metal-ligand complex was done such that the orbital

energies matched very closely to that of the poly(3,4-(ethylenedioxy)thiophene), allowing

the polymer to be highly conductive. The Co2+:salen complex leaves the Co2+ unsaturated

in coordination sites and thus is able to interact with small coordinating molecules. In the presence of nitric oxide, coordination occurred with the Co2+ which was found to result in

a change in the orbital energies of the complex relative to the polymer backbone. This

change decreases the conductivity and was thus monitored by measuring the increase in

resistance relative to NO concentration. Amazingly this system was able to detect NO gas

at 1 ppm concentrations, was reversible, and was selective (no response was detected to

O2, CO or CO2).

Figure 1.8 Schematic illustration of the fabrication of conducting metallosupramolecular polymer/electrode devices by the electropolymerization of 1.8 across interdigitated microelectrodes leading to a chemoresistive response upon NO gas exposure.

Electro- and mechanically-sensitive stimuli-responsive metallosupramolecular

polymer gels have been prepared by Terech and coworkers.47 In this study, a small

molecular weight gelator which contained two different types of metal-ion receptor sites

(terpyridine end-groups with a cyclam core) was used. They found that polymers prepared with Cu2+ formed by initial complexation with the cyclam unit at 1:1

stoichiometry, followed by complexation with the terpyridine units at a 2:1

metal:monomer ratio and that the opposite was found using Co2+ or Ni2+ as these metal-

ions have larger binding constants with the terpyridine. Interestingly, they found that

organogels could be formed, however gelation was affected by the counterion present (by

means of differences in electrostatic interactions and/or solubility, see Figure 1.9a) and/or

the addition order of certain metal-ions (see Figure 1.9b). They found that the gels were

stimuli-responsive to a variety of different stimuli, including mechanically-induced gel-

sol transitions (thixotropic) and electrical stimulus. The cyclam-metal complex acts as a

redox responsive site which Terech and coworkers have used to electrochemically switch

between a gel and liquid state. They report that the main contributing factor to this

stimuli-responsive behavior is the change in the solubility of the metallosupramolecular

polymer which is vital to gel formation where oxidation of the cyclam-metal complex

generates additional positive charges along the polymer backbone which affects the solubility.

Figure 1.9 a) Images of the formation of solutions or gels in CH3CN as a function of counterion

(1) CoCl2, (2) Co(NO3)2, (3) NiCl2, and (4) Ni(NO3)2 and an example of the redox-controlled reversible gel-to-liquid transition with Co2+ in DMF (2.15 mM). b) Chart of the resulting materials properties as a function of the addition order of metal-ions in DMF.

A demonstration of mechano-responsive metallosupramolecular polymers formed

using ditopic phosphine end-capped poly(tetrahydrofuran) (pTHF) coordinated with

palladium metal-ions were prepared by Sijbesma and coworkers.48 Coordination within

the polymer was found to be inert in the presence of a small molecule analogue with the

same binding motif as the dissociation rate of the Pd2+:phosphine ligand complex is very

slow. To overcome this, the metallosupramolecular polymers were sonicated with the

nonpolymeric complexes, resulting in the formation of new hetrocomplexes and a

dramatic reduction in molecular weight. They postulated that the exchange occurs as a

result of chain scission between the Pd2+:phosphine bond caused by shear-induced forces

from the collapse of gas bubbles which are formed during the sonication process. Taking

this idea to the next level, the same group also created mechano-responsive gels using

either Ir2+ or Rh2+ which are able to bind 3-4 phosphine ligands, creating a crosslinked

network architecture.49 Sonication of the gels produced chain scission leading to the

transition from gel to sol. Using the difference in the ligand exchange kinetics they were

able to tailor the time required to reform the gels. The use of Rh2+ gave liquid-like

properties after three minutes of sonication and the system returned to the gel state after

only one minute. Gels prepared with Ir2+ also became liquid-like after three minutes

however regelation required multiple hours to days depending on the temperature. The

use of a mechno-response is intriguing as it allows for new processing methods of the polymer and additionally offers the potential of easily creating catalytic sites.

An excellent example of using the dynamic nature of the metal-ligand interaction

within a metallosupramolecular polymer to generate a macroscopic stimuli-response in

polymer films was carried out by Rowan and coworkers.50 In this work mechanically

stable, elastomeric films were prepared using a ditopic macromonomer consisting of two

2,6-bis(N-methylbenzimidazoyl)pyridine (Mebip) ligands on the ends of a poly(ethylene-

co-butylene) polymer core with either Zn2+ or La3+ metal-ions. Phase separation of the

ionic metal-ligand complexes from the polymer core results in physical crosslinking to

yield good mechanical properties and detailed studies on the morphology revealed a

lamellar structure for all films. Using a noninvasive technique, Rowan and coworkers

capitalized on the absorption properties of the metal-ligand complexes to develop self-

healing materials when exposed to UV light (320-390 nm). It was found that the metal-

ligand complexes would absorb the light, and while some of that energy was converted into emission, more importantly, some was converted into localized heating. Localized heating, which occurs primarily at the metal-ligand interaction, results in increased rates of decomplexation as shown schematically in Figure 1.10a. A film that has been scratched could then be healed by this mechanism as the increased decomplexation leads to a reduction of the viscosity, allowing the polymer to flow into the scratch to fill it. A demonstration of this is shown by AFM in Figure 1.10b. Removal of the light source allows for recomplexation to yield a healed film (see Figure 1.10c) and it was found that large healing efficiencies could be obtained depending on the metal-ion and stoichiometry of the systems. Films prepared with Zn2+ required a deficit of metal-ions

(~30%) in order to achieve large healing efficiencies, which detrimentally affected the

mechanical properties, whereas the use of La3+ gave good healing efficiencies while

maintaining the mechanical properties at or near stoichiometry. One advantage to using

light as a stimulus is that it allowed for localized healing where only the irradiated

21 portion of the film showed self-healing capabilities without affecting the entirety of the film.

Figure 1.10 a) Schematic representation of the mechanism of optical healing of a phase separated metallosupramolecular polymer. b) AFM height images of a damaged film prepared with 100%

La3+ before healing (left) and after partial (middle) and complete (right) healing. c) Images of the optical healing of a polymer film prepared with 70% Zn2+ using UV light (320-390 nm).

1.2.2 Side-Chain Metallosupramolecular Polymers

The bulk of stimuli-responsive metallosupramolecular polymers contain the metal-ligand motif in the polymer backbone, however polymers containing the metal in side-chains have also become increasingly studied for their use in materials. An example of a chemo- responsive side-chain metallosupramolecular polymer network was recently given by

Craig and coworkers.51 To create polymer networks, poly(4-vinylpyridine) was crosslinked using a bimetallic (Pt2+ or Pd2+) pincer compound by coordination with the polymeric pyridine side-chains in DMSO. They found that these networks could be used to demonstrate chemo-responsive viscosity switching by way of a sol-gel transition near the percolation threshold.52 Below a critical crosslinker concentration the networks displayed free-flowing sol behavior; however above this concentration the viscosity was observed to dramatically increase because of formation of weak gels and the concentration range at which this occurred was very small. In fact they found these systems to be very sensitive to the amount of crosslinker added as, in one case, the mechanical properties were shown to improve by five orders of magnitude over only a

0.09 weight % increase of the crosslinker. As the mechanical properties of these metallosupramolecular networks was found to be directly related to the dissociation kinetics it is possible to design systems with even more dramatic changes mechanical properties if a crosslinker is used with a slower dissociation rate.

An important property of metal-containing polymers is that they can display interesting optical and electronic characteristics. Metal-ions from the lanthanide series are particularly of interest because of their characteristic intense emissions with distinctive colors depending on the choice of metal. Tew and coworkers53,54 investigated the optical

properties of poly(methyl methacrylates) (PMMA) functionalized with terpyridine groups

on the side-chains coordinated with different lanthanides (1.9) shown in Figure 1.11a.

Films prepared by coordination with Eu3+ gave a characteristic red metal-based emission

color while those prepared with Tb3+ appear characteristically green in color. In an

elegant approach to using the dynamic metal-ligand properties in these films, Tew and coworkers showed that mixing two homopolymers (one prepared with only Eu3+ and the

other with only Tb3+) in a 1:1 ratio gave an emission profile that was essentially the

superposition of the two individual homopolymers (which appeared green as the emission

intensity of the Tb3+ complex is much stronger than that of Eu3+). However, heating the

combined homopolymers together and subsequently cooling resulted in an irreversible

change in emission color from green to yellow as shown in Figure 1.11b. The new

emission profile was speculated to be a result of either the formation of a bridging

- bimetallic complex (possibly formed by the counterion, NO3 or one of the pyridine rings

from the ligand) or increased π – π interactions between the terpyridine complexes. They

also noted that the new yellow-emitting polymer now displayed a thermo-reversible

change in emission to an orange/red color. They explained this stimuli-responsive behavior as a result of quenching of the Tb3+ complex emission at elevated temperature to

give emission similar to that of the Eu3+ complex.

Figure 1.11 a) Structures of PMMA functionalized with either one type of terpyridine:Ln3+

complexes in the side-chain (Ln3+ = Eu3+ or Tb3+) (1.9) or a combination of lanthanides in a 1:1 ratio (1.10). b) Images (λex = 350 nm) of an irreversible thermo-responsive change in emission by heating two homopolymers to give a new blend and c) new thermo-reversible emission change of the new mixed metal system resulting from quenching of the green Tb3+ emission at elevated temperatures.

1.3 THESIS SCOPE

The goal of this dissertation is to highlight the preparation and characterization of a

variety of stimuli-responsive metallosupramolecular polymers. Chapter 2 introduces a

study on the influence of introducing increasing amounts of a lanthanide (Eu3+) metal-ion

into metallosupramolecular polymer films assembled with a transition metal (Zn2+).

Mechanically stable, elastomeric films were prepared using Mebip end-capped poly(tetrahydrofuran) (pTHF) with varying ratios of Zn2+:Eu3+ (100:0 – 50:50). The addition of Eu3+ resulted in increased stimuli-responsive behavior which allowed for their

use as fluorescent temperature and chemical sensors.

Chapter 3 further investigates the metallosupramolecular films prepared in

Chapter 2 by examining their structure-property relationships. Films prepared in the same

manner with Mebip end-capped poly(ethylene-co-butylene) were also studied to

determine the impact of the polymer core on the observed properties. Detailed melt

rheological studies were used to characterize the increased thermoresponsive behavior

with increasing Eu3+ content and related to the morphology. A simplified network theory is also introduced to explain the observed sinusoidal response in the “nonlinear”

viscoelastic region in the strain sweeps resulting from large relaxation times.

In Chapter 4, the lessons learned in Chapters 2 and 3 on tailoring the reversibility

of the metal-ligand interactions are used to design a metallosupramolecular shape- memory polymer. Using the phase separated metal-ligand interactions as a reversible crosslinking motif in conjunction with a covalently crosslinked poly(butadiene) core allowed for the formation of films which displayed thermal, light, and chemically-

induced shape-memory properties which could be tailored by the choice of metal or

counterion used.

Chapter 5 addresses the use of vapochromic square-planer Mebip:Pt2+ complexes

in a polymer matrix to design polymer films which can be used as organic vapor sensors.

Exposure of the films to solvent vapor produced both a visible (yellow to red) and

fluorescent (yellow to orange) color change as a result of increased Pt – Pt interactions.

The use of Mebip:Pt2+ complexes in polymer matrices also resulted in films which

displayed mechanochromic properties consistent with force-induced structural

rearrangement.

Metallosupramolecular polymers assembled with Pt2+ and Mebip end-capped polymers were used for the fabrication of platinum nanoparticles in Chapter 6. Exposing thin films of the metallosupramolecular polymers to electrons from an extracted discharge generated by microplasma lead to reduction of the Pt2+ to Pt0, resulting in well-

dispersed, unagglomerated nanoparticles with diameters under 5 nm. The size of the

nanoparticles was found to be directly related to the polymer morphology and the

exposure time.

Chapter 7 addresses the ability to form metallosupramolecular networks in the

melt. When Mebip side-chain functionalized polystyrene blended into a polystyrene

matrix was brought into contact with Zn2+ blended into poly(methylmethacryalte) above

Tg, metallosupramolecular networks formed as the components diffused toward the

interphase. It was found that self-assembly occurred quickly, typically within five minutes when monitored spectroscopically.

1.4 REFERENCES

1. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

2. ten Brinke, G.; Ruokolainen, J.; Ikkala, O. Adv. Polym. Sci., 2007, 207, 113–177.

3. Serpe, M. J.; Craig, S. L. Langmuir, 2007, 23, 1626–1634.

4. Lehn, J. –M. Chem. Soc. Rev., 2007, 36, 151-160.

5. Armstrong, G.; Buggy, M. J. Mater. Sci., 2005, 40, 547-559.

6. Aida, T.; Meijer, E. W.; Stupp, S. I. Science, 2012, 335, 813-817.

7. Appel, W. P. J.; Nieuwenhuizen, M. M. L.; Meijer, E. W. Multiple Hydrogen-Bonded Supramolecular Polymers, in Supramolecular Polymer Chemistry, ed. Harada, A.; Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2011.

8. Burattini, S.; Colquhoun, H. M.; Fox, J. D.; Friedmann, D.; Greenland, B. W.; Harris, P. J. F.; Hayes, W.; Mackay, M. E.; Rowan, S. J. Chem. Commun., 2009, 6717-6719.

9. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176- 188.

10. McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. Rotello, V. M.; Thayumanavan, S. John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178.

11. Whittell, G. R.; Manners, I. Adv. Mater., 2007, 19, 3439-3468.

12. Schubert, U. S.; Eschbaumer, C. Angew. Chem. Int. Ed., 2002, 41, 2892-2926.

13. Dobrawa, R.; Würthner, F. J. Polym. Sci. Part A: Polym. Chem., 2005, 43, 4981- 4995.

14. Friese, V. A.; Kurth, D. G. Coordin. Chem. Rev., 2008, 252, 199-211.

15. Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorg. Chem., 1966, 5, 622-625.

16. Bonnet, S.; Collin, J. –P.; Gruber, N.; Sauvage, J. –P.; Schofield, E. R. Dalton Trans., 2003, 4654-4662.

17. Kelch, S.; Rehahn, M. Macromolecules, 1999, 32, 5818-5828.

18. Eisenbach, C. D.; Schubert, U. S. Macromolecules, 1993, 26, 7372-7374.

19. Burnworth, M.; Rowan, S. J.; Weder, C. Chem. Eur. J., 2007, 13, 7828–7836.

20. Knapton, D.; Burnworth, M.; Rowan, S. J.; Weder, C. Angew. Chem., Int. Ed., 2006, 45, 5825–5829.

21. Ciferri, A. Macromol. Rapid Commun., 2002, 23, 511-529.

22. Kurth, D. G.; Higuchi, M. Soft Matter., 2006, 2, 915-927.

23. Chen, C. C.; Dormidotova, E. E. J. Am. Chem. Soc., 2004, 126, 14972-14978.

24. Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565-578.

25. Yan, Y.; Huang, J. Coordin. Chem. Rev., 2010, 254, 1072-1080.

26. Pollino, J. M.; Weck, M. Chem. Soc. Rev., 2005, 34, 193-207.

27. Hammond, M. R.; Andreopoulou, A. K.; Pefkianakis, E.; Kallitsis, J. K.; Mezzenga, R. Chem. Mater., 2009, 21, 2169-2172.

28. Weng, W.; Beck, J. B.; Jamieson, A. M.; Rowan, S. J. J. Am. Chem. Soc., 2006, 128, 11663-11672.

29. Sievers, T. K.; Vergin, A.; Möhwald, H.; Kurth, D. G. Langmuir, 2007, 23, 12179- 12184.

30. Hoogenboom, R.; Schubert, U. S. Chem. Soc. Rev., 2006, 35, 622-629.

31. Payne, S. J.; Fiore, G. L.; Fraser, C. L.; Demas, J. N. Anal. Chem., 2010, 82, 917-921.

32. Astruc, D.; Chardac, F. Chem. Rev., 2001, 101, 2991-3024.

33. Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev., 2010, 110, 1857-1959.

34. Meudtner, R. M.; Ostermeier, M.; Goddard, R.; Limberg, C.; Hecht, S. Chem. Eur. J., 2007, 13, 9834- 9840.

35. Meudtner, R. M.; Hecht, S. Macromol. Rapid Commun., 2008, 29, 347-351.

36. Lahn, B.; Rehahn, M. Macromol. Symp., 2001, 163, 157-176.

37. Chow, C. F.; Fujii, S.; Lehn, J. –M. Chem. Asian J., 2008, 3, 1324-1335.

38. Chow, C. F.; Fujii, S.; Lehn, J. –M. Angew. Chem. Int. Ed., 2007, 46, 5007-5010.

39. Schmatloch, S.; van den Berg, A. M. J.; Alexeev, A. S.; Hofmeier, H.; Schubert, U. S. Macromolecules, 2003, 36, 9943-9949.

40. Vermonden, T.; van Steenbergen, M. J.; Bessling, N. A. M.; Marcelis, A. T. M.; Hennink, W. E.; Sudhölter, E. J. R.; Cohen Stuart, M. A. J. Am. Chem. Soc., 2004, 126, 15802-15808.

41. Vermonden, T.; van der Gucht, J.; de Waard, P.; Marcelis, A. T. M.; Bessling, N. A. M.; Sudhölter, E. J. R.; Fleer, G. J.; Cohen Stuart, M. A. Macromolecules, 2003, 36, 7035-7044.

42. Yang, S. K.; Ambade, A. V.; Weck, M. Chem. Soc. Rev., 2011, 40, 129-137.

43. Yang, S. K.; Ambade, A. V.; Weck, M. J. Am. Chem. Soc., 2010, 132, 1637-1645.

44. Guillet, P.; Fustin, C. –A.; Lohmeijer, B. G. G.; Schubert, U. S.; Gohy, J. –F. Macromolecules, 2006, 39, 5484-5488.

45. Fustin, C. –A.; Guillet, P.; Schubert, U. S.; Gohy, J. –F. Adv. Mater., 2007, 19, 1665- 1673.

46. Holliday, B. J.; Stanford, T. B.; Swager, T. M. Chem. Mater., 2006, 18, 5649-5651.

47. Gasnier, A.; Royal, G.; Terech, P. Langmuir, 2009, 25, 8751-8762.

48. Paulusse, J. M. J.; Huijbers, J. P. J.; Sijbesma, R. P. Chem. Eur. J., 2006, 12, 4928- 4934.

49. Paulusse, J. M. J.; van Beek, D. J. M.; Sijbesma, R. P. J. Am. Chem. Soc., 2007, 129, 2392-2397.

50. Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334–337.

51. Yount, W. C.; Loveless, D. M.; Craig, S. L. J. Am. Chem. Soc., 2005, 127, 14488- 14498.

52. Loveless, D. M.; Jeon, S. L.; Craig, S. L. J. Mater. Chem., 2007, 17, 56-61.

53. Shunmugam, R.; Tew, G. N. J. Am. Chem. Soc., 2005, 127, 13567-13572.

54. Shunmugam, R.; Tew, G.N. Macromol. Rapid Commun., 2008, 29, 1355-1362.

CHAPTER 2

Stimuli-Responsive Europium-Containing

Metallosupramolecular Polymers

Adapted from: Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145–151.

2.1 INTODUCTION

The ability to utilize supramolecular interactions1 toward the design of new stimuli- responsive materials has gained a great amount of interest in recent years.2 For example,

main-chain supramolecular polymers, whose polymer backbone consists of both covalent

bonds and noncovalent interactions, can result in large (e.g. mechanical) responses to

small changes in the environment.3 Reversible metal–ligand coordination is one class of

non-covalent interaction that has been widely utilized to access main-chain

supramolecular polymers.4 Most of these metallosupramolecular polymers rely on ligand

coordination with transition metal-ions. Such systems have been targeted, at least in part,

on account of the strength of the metal–ligand interaction and the fact that 2 : 1 ligand to

metal complexes (required for linear chain extension) can be easily accessed with either

terdentate ligands bound to hexacoordinate metal-ions or bidentate ligands bound to

tetracoordinate metal-ions.5 A series of metal-ions that have seen much less attention in

this field are the lanthanide metals,6 which can be of special interest as their complexes

have been shown to have many potential applications that include (bio)sensing,7 optical fiber lasers and amplifiers,8 liquid crystals9 and light-emitting diodes.10 One of the most

interesting and unique properties of certain lanthanide ion complexes is their excellent

luminescence characteristics that include extremely sharp emission bands and high

photoluminescence (PL) efficiency.11

Combining the distinctive properties of lanthanide complexes with the mechanical

properties of polymers presents an interesting method of creating stimuli-responsive metal-containing materials. We have shown that 2,6-bis-(10- methylbenzimidazolyl)pyridine (Mebip) ligands can be placed at the ends of ditopic

macromonomers, which consist of either conjugated12 or polyether cores,13 and produce polymeric materials upon addition of an appropriate transition metal-ion (e.g. Fe2+, Co2+,

or Zn2+) in a 1:1 ratio of metal to ditopic macromonomer. Such transition metals bind the

terdentate Mebip in a 2:1 ligand to metal ratio that allows for the formation of linear high

molecular weight metallosupramolecular polymers resulting in a significant enhancement

in the solid-state mechanical properties of these materials compared to the unbound

macromonomer.

Pioneering work by Piguet and Bünzli has shown that planar aromatic terdentate

ligands, like Mebip, will coordinate with lanthanide ions to form metallosupramolecular

complexes.14 Some lanthanide ion complexes, e.g. Eu3+ complexes, can show an intense

metal-centered luminescence in the presence of an appropriate UV absorbing ligand via

the so-called ‘‘antenna effect’’.15 This is in effect a light conversion process which occurs

by absorption of the light by the ligand, followed by a ligand-to-metal energy transfer

process finally resulting in the metal-ion based emission. The Mebip ligand has been

shown16 to act as an effective ‘‘antenna’’ for Eu3+ ions and as such this ‘‘antenna effect’’

can be used to demonstrate whether the ligand is complexed to the Eu3+ ion. More

recently, Piguet and co-workers have taken an in-depth look at the interaction between

lanthanides and terdentate aza-ligands in both the solid-state and in solution.17 With their

preference for high coordination numbers (8–10),18 lanthanides, such as Eu3+, can

coordinate with Mebip in a 3:1 ligand to metal ratio. Piguet and co-workers demonstrated

that the nature of the complexes formed between the lanthanides and the terdentate aza-

ligands is very sensitive to the counterion present. For example, europium nitrates

(Eu(NO3)3) were found to bind such ligands predominately in a 1:1 ligand to metal ratio

because of competitive binding between the nitrate and the ligand. Conversely, europium

perchlorate (Eu(ClO4)3) will bind in a 3:1 Mebip ligand to metal ratio as the Mebip

ligands now have much stronger interactions with the metal-ion than the weakly binding

perchlorate counterions. This suggests that incorporating lanthanide perchlorates into a

supramolecular polymer should result in the creation of dynamic branching or

networking points19 that could lead to changes in mechanical properties.

Previously,20 we have prepared stimuli-responsive gels in which Mebip ligands

were placed on the ends of a pentaethylene glycol core.21 The addition of Zn2+ ions with

or without varying amounts of lanthanide ions produced gels in acetonitrile that displayed

interesting stimuli-responsive behavior. The lanthanide containing gels were found to

show enhanced thermo-, mechanical- and chemo-responsiveness which in part was a

consequence of the Mebip binding more weakly to the lanthanide ion as compared to the

Zn2+ ions.

Along the same lines, complexes of low molecular weight Mebip derivatives have

demonstrated the capability of sensing small amounts of certain organophosphates22 that

were used as pesticides and nerve gas agent mimics.23 These ligands were complexed

with La(NO3)3 and Eu(NO3)3 to form highly emissive 1:1 complexes. When exposed to

the organophosphates, either in solution or vapor, these materials showed a change in

color from red to blue. In this case the organophosphate binds to the Eu3+ ion, displacing

the Mebip ligand and thus converts the red emissive Eu3+–Mebip complex to the free

Mebip ligand (blue emission) and a non-emissive Eu3+–organophosphate complex. It is

interesting to note that stronger binding Zn2+ complexes of similar Mebip ligands did not show Mebip displacement in the presence of organophosphates.

Prior work has already shown that ditopic macromonomer 2 (Figure 2.1) derived

from low molecular weight poly(tetrahydrofuran) (p(THF)) cores (Mn < 5000 g/mol)

form thermoplastic elastomers once complexed with Zn2+ ions. It is believed that the

elastomeric behavior occurs as a consequence of phase separation between the hard ionic

metal–ligand complexes and the soft poly(tetrahydrofuran) segments. Thus with the goal of developing stimuli-responsive thermoplastic elastomers we report herein the effects of

replacing a percentage of the Zn2+ ion complexes with the more responsive Eu3+

complexes in these metallosupramolecular polymers. From a structural point-of-view this is also interesting as conceptually 3:1 complexes will result in the formation of dynamic- crosslinking points. Thus the addition of both Eu3+ and Zn2+ ions to 2 should result in a crosslinked metallosupramolecular polymer in solution and in the solid-state. Of course, the presence of physical crosslinking in the solid phase on account of the phase separation between metal–ligand complexes and the polymer core adds an additional level of complexity to this system.

Figure 2.1 Synthesis of macromonomer 2 where the polymer core is grown via a cationic, ring- opening, living polymerization of tetrahydrofuran and subsequently end-capped with 4-hydroxy- 2,6-bis(10-methylbenzimidazoyl)pyridine ligands.

2.2 RESULTS AND DISCUSSION

Macromonomer 2 has been previously synthesized using Mitsunobu chemistry by attaching 4-hydroxy- 2,6-bis(10-methylbenzimidazoyl)pyridine ligands to the end of a telechelic, dihydroxyl poly(tetrahydrofuran). For this work, a different approach was taken to synthesize 1 as the prior method involved complicated purification to remove the side products generated in the Mitsunobu reaction, but also because the previous method was limited in the molecular weight that could be obtained, by using a commercially available preformed, hydroxyl-terminated poly(tetrahydrofuran) core. Rather, compound

2 was made using a cationic, living ring-opening polymerization of tetrahydrofuran. The

reaction was carried out using trifluoromethanesulfonic anhydride as a bidirectional

initiator and 4-hydroxy-2,6-bis(10-methylbenzimidazoyl)pyridine as the end-capping

chain terminator as is shown in Figure 2.1. This versatile method allows for high purity

(see Figure 2.2a) and control of the polymer molecular weight by simply varying the

reaction time and/or initiator concentration as shown in Figure 2.2b, as well as allowing the reaction to be carried out on relatively large scales (routinely run on a 15 g scale).

1 Figure 2.2 a) 600 MHz H NMR of 2 in CDCl3. b) Plot of molecular weight of 2 after purification for different polymerization times. The molecular weights nearly follow a linear growth with time (represented by the red line) which is indicative of a living polymerization. This can also be used to synthesize 2 with a desired molecular weight depending on the time of polymerization.

2.2.1 Optical Properties of the Metallosupramolecular Polymer Films

The coordination of Zn(ClO4)2 and Eu(ClO4)3 to 2 was initially studied using UV-vis

spectroscopy (Fig. 2.3). Titrating Eu3+ into a solution of 2 leads to complexation, which results in an increase in the ligand’s absorbance from approximately 330 nm to 370 nm and a decrease in the absorbance at 315 nm, which corresponds to the λmax of the free

ligand as shown in Figure 2.3a. Plotting the absorbance at 340 nm vs. the ratio of metal to

2 (Fig. 2.3b) shows that Eu(ClO4)3 initially binds three of the polymeric Mebip ligands

(as shown by the slope change at a metal:monomer ratio of 1:3), and continued addition of Eu(ClO4)3 results in complete binding evidenced by another more subtle slope change

at a metal:monomer ratio of 2:3 which corresponds to the theoretical ratio for the

polymeric metallosupramolecular polymer with 100% Eu3+. This confirms that the

presence of the polymer does not significantly alter the binding ratio of Mebip to Eu3+.

Similarly, 2 binds Zn2+ in a 1:1 ratio (or 2 ligands per metal-ion) as shown in Figures

2.3c-d. Knowing how each metal-ion will bind with 2 allows for the creation of high molecular weight metallosupramolecular polymers (shown schematically in Figure 2.3e), which should be able to be processed into mechanically stable films as a result of the phase separation of the metal-ligand complexes from the polymer core as has been shown in films prepared with Zn2+ only.

Figure 2.3 UV-vis titration of 2 with a) Eu(ClO4)3 and c) Zn(ClO4)2 (25µM). Plotting the absorbance at λ = 340 nm vs the molar ratios of 2 to b) Eu3+ or d) Zn2+ shows that the polymer and that Eu3+ binds in a 2:3 metal:2 ratio while Zn2+ binds in a 1:1 metal:2 ratio. e) Schematic representation of self-assembly into a phase separated metallosupramolecular polymer.

Films of the metallosupramolecular polymers with the metal perchlorate salts

were obtained by solution casting. Six films were studied containing different ratios of

Zn2+ to Eu3+ ranging from only Zn2+ and no Eu3+ (100:0) to 50% Zn2+ and 50% Eu3+

(50:50). The percentages refer to the percent of ligands in the system which are bound to

that metal-ion, assuming a 2:1 ratio for Zn2+ and 3:1 ratio for Eu3+ (although we cannot

rule out the presence of a certain percentage of 2:1 and even 1:1 Mebip:Eu3+ complexes).

All of these ratios produced optically clear, mechanically stable, elastomeric films (Fig.

2.4). Interestingly, attempts to cast films containing greater than 50% of the ligands complexed to Eu3+ resulted in mechanically unstable films, suggesting that there is a limit

to the amount of the weaker binding Eu3+ that can be used in order to get films with good

mechanical properties (see Chapter 3 for a more detailed description of this).

Figure 2.4 A Zn2+:Eu3+ 80:20 film stretched by hand which shows the elastomeric nature of the films caused by physical crosslinks formed from phase separation of the metal-ligand complexes from the polymer core.

All six of the films studied were optically clear under visible light and were

fluorescent upon exposure to UV light. Fig. 2.5 shows the fluorescence of the six films

under irradiation with 365 nm light. The film containing no Eu3+ (100:0) gives a blue

2+ 3+ fluorescence (λmax = 410 nm) on account of the Mebip:Zn complex. As Eu is

introduced into the supramolecular polymer, more of the Eu3+:Mebip complex emission is observed and as a result the film’s emission changes from blue to red/pink. For example, the film containing Zn2+:Eu3+ 80:20 appears light pink/purple from the

combination of the blue emission and the red emission. These effects are shown in the photoluminescence spectra (excited at 365 nm) of the films in Figure 2.5. The addition of

Eu3+ results in a decrease in the emission band at 410 nm and the appearance of the

distinctive Eu3+ emission bands at 575, 590, 615, and 645 nm. These bands have been

5 7 24 3+ identified as the D0→ F0–3 transitions. Upon increasing the concentration of Eu

within the films there is a decrease in the emission band at 410 nm and the distinctive

Eu3+-based emission bands at 590 and 615 nm appear.

Figure 2.5 a) Pictures of solution cast films of 2 with varying amounts of Zn2+ and Eu3+ under UV light (365 nm) demonstrating the change in the fluorescence from blue (Mebip:Zn2+ complex emission) to red/pink (metal-based Mebip:Eu3+ complex emission) as the Eu3+ content is increased. b) Photoluminescence spectra of films of 2 and varying ratios of Zn2+ and Eu3+ ranging 2+ 3+ from Zn :Eu 100:0 to 50:50 in 10% increments. λex = 365 nm. The small peak at 470 nm is an instrument artifact. c) Schematic representation of the antenna effect.

2.2.2 Thermomechanical Properties

The thermomechanical properties of the films were then investigated with the goal of

elucidating the effect of varying the Zn2+:Eu3+ ratio of these metallosupramolecular

polymers. Thus dynamic mechanical thermal analysis (DMA) studies were carried out on

the series of films that range from Zn2+:Eu3+ 100:0 to Zn2+:Eu3+ 50:50. Figure 2.6a shows

the DMA data for the six films that were annealed at -35 °C for 60 min prior to being

cooled to -110 °C, at which point the run was started. If this annealing step is not carried

out and the samples are cooled to -110 °C directly from room temperature then an increase in modulus is seen around -35 °C (Figure 2.6b) during the experiment, which is attributed to the cold crystallization of the p(THF) core. Annealing at -35 °C for 60 min before the experiment removes most of this effect. Two other major thermal transitions are observed in all six of the films. The first transition corresponds to the glass transition temperature of poly(tetrahydrofuran) which is seen at ca. -80 °C, while the second transition is attributed to the Tm of the low molecular weight p(THF) segment that occurs

around 0 °C. The presence of both the peaks is consistent with the p(THF) being phase

separated from the metal complexes. The major difference in the thermomechanical properties of these films comes at higher temperatures. As a general trend the lower the

Zn2+:Eu3+ ratio the weaker and more thermally sensitive the films become. For example,

2+ after the Tm transition the modulus of the 100% Zn film stays relatively constant at ca.

30 MPa (the relative independence of modulus with temperature in this range is

consistent with the presence of physical crosslinking due to phase separation) with

increasing temperature, only dropping off significantly above 120 °C. The Zn2+:Eu3+

50:50 film on the other hand shows a gradual decrease in modulus, starting at ca. 30 MPa

after the melting point of p(THF), with increasing temperature and showing a significant drop in modulus above 40 °C.

Figure 2.6 Dynamic mechanical thermal analysis of solution cast films made from 2 with different ratios of Zn2+:Eu3+ ranging from 100:0 to 50:50. a) Samples were annealed at -35 °C prior to being run. b) Samples cooled directly to -110 °C from room temperature show an increase in modulus at -35 °C from cold crystallization. All samples display a glass transition of ca. -80 °C and a melting temperature of ca. 0 °C which are attributed to the poly(tetrahydrofuran) core.

The trend toward more thermally responsive films upon increasing Eu3+ content can be rationalized in a number of different ways. One aspect is the relative binding

strength of Eu3+ vs Zn2+ to the ligand. It has already been shown that Eu3+ is a weaker binder of Mebip than Zn2+ ions. Thus, it is expected that these films will be more

thermally responsive.

To see if the presence of Eu3+ has any effect on the mechanical properties of these

films, a series of four films that ranged from 80:5 to 80:20 (Zn2+:Eu3+) were prepared and

investigated. Interestingly, all four supramolecular polymers formed mechanically stable

stand-alone films, even though for three of them a significant number of chain-ends, in

the form of free ligand, must be present. The presence of physical crosslinks caused by

phase separation explains why these materials are not as sensitive to metal–ligand

stoichiometry as may be expected.25 DMA of these films, which are seen in Figure 2.7, do show, however, that the presence of the Eu3+ ions does enhance their

thermomechanical properties. There is a small improvement in the films’ moduli as free

ligands are bound to Eu3+, and an improvement in the temperature at which the sample

yields (the slight increase in Eʹ right before the sample yields is presumably from slippage at the clamps).

Figure 2.7 Dynamic mechanical thermal analysis of solution cast films with varying ratios of Zn2+:Eu3+ ranging from 80:5 to 80:20, showing the enhancement of mechanical properties with addition of Eu3+ ions over films prepared with free chain-ends.

It can be expected that in the solid-state a significant percentage of the Eu3+ ions

are complexed to three Mebip ligands and thus can potentially act as dynamic-

crosslinking sites. However, as mentioned before, the mechanical properties of these films seem to be more related to the phase separation between the core and the metal

3+ complexes. Thus it is also possible that the incorporation of the (Eu:Mebip3) complexes

2+ results in a disruption of the ordered (Zn:Mebip2) crosslinking hard phase. To examine

this aspect in more detail wide angle X-ray scattering (WAXS) studies were carried out

on selected films which contain different ratios of Zn2+ to Eu3+ (Figure 2.8a). As we have

seen before the Zn2+ only films show two main peaks, one at d = 9.8 Å (2θ = 9.0) and

another at d = 60.2 Å (2θ = 1.5). These peaks have previously been assigned to the metal-

to-metal distance within the hard phase (d = 9.8 Å) and the metal-to-metal distance

between different hard phases (d = 60.2 Å) (Figure 2.8b). As the Eu3+ content is increased in these films the peak (at d = 9.8 Å) corresponding to the order within the hard phase starts to decrease (as shown in Figure 2.8a) suggesting some sort of disruption of the

ordered packing within the hard phase. However, the peak at d = 60.2 Å is still observed

even in the Zn2+:Eu3+ 50:50 material. These results are consistent with the presence of some phase separation in the Eu3+-containing films but with disorder occurring within the

hard phase, likely because of trying to introduce the much bulkier 3:1 Mebip:Eu3+

complexes, which may contribute to a slight weakening of the films. Further structural

data, specifically looking at the phase separation is detailed in the next chapter.

Figure 2.8 a) Wide angle X-ray scattering (WAXS) of selected films of 2 with varying ratios of Zn2+:Eu3+ which demonstrate the loss of order in the hard phase (d = 9.8 Å, 2θ = 9) as the ratio of Zn2+:Eu3+ decreases. b) Schematic of the ordering of the metal-ligand complexes in these metallosupramolecular films in which an increase in Eu3+ does not affect the phase separation but does however decrease the ordering within each hard phase.

2.2.3 Stimuli-Responsive Behavior

Having demonstrated the increased thermomechanical sensitivity of the Eu3+-containing polymers, an investigation of the optical response of these films was undertaken upon exposure to either thermal or chemical stimuli. For example, upon heating the 70:30

(Zn2+:Eu3+) film to about 120 °C on a glass slide the fluorescence changes from pink to blue which can be seen in Figure 2.9a. The photoluminescence data (Figure 2.9b)

confirms that the increase in temperature results in the disappearance of the Eu3+-based

emission peaks (at 575, 590, and 615 nm). This provides further back-up evidence that, at least in part, the increase in the thermomechanical sensitivity of these films is on account of the more thermally sensitive Eu3+ complexes where an increase in temperature promotes an increased rate of decomplexation of the complexes formed with Eu3+. As the

temperature is returned to ambient (by simply removing the film from the heat source)

the pink color reappears, consistent with the recomplexation of the Eu3+ to Mebip

ligands. It should be noted that while there is a slight difference in emission color

between the film that is solution cast and the one that has been exposed to high

temperatures, a subsequent heating and cooling cycle did not show any significant change

in the emission spectra (Figure 2.9c) suggesting that once annealed this is a reversible

process. This phenomenon suggests that these materials may be used as temperature

sensors with reversible characteristics.26

Figure 2.9 a) Pictures of fluorescent temperature sensing behavior of a Zn2+:Eu3+ 70:30 film under UV light (365 nm). A sample heated to 120 °C results in quenching of the Eu3+ metal-based fluorescence. The original fluorescence returns upon cooling back to room temperature. b) The intensity of the Eu3+ metal-based emission peaks (575, 590, and 615 nm) are decreased as the temperature is elevated with complete quenching occurring at 120 °C. c) Photoluminescence spectra of the above films showing the reversible disappearance of the Eu3+ metal-based emission

upon heating and cooling (λex = 377 nm).

Along with being temperature sensors, the materials’ ability to act as chemical sensors was also investigated. As mentioned before, previous work has shown that small molecules containing the Mebip ligand have the ability to detect nerve gas agent mimics.

Using this knowledge, a 70:30 (Zn2+:Eu3+) film was exposed to triethyl phosphate, the

same nerve gas agent mimic used in the previous study. The first study simply dipped the

film into triethyl phosphate directly upon which the fluorescence changed visibly from

the red Eu3+ complex metal-based emission (Figure 2.10a) to the blue emission of the

Zn2+ complex and/or free ligand only (Figure 2.10b), however the polymer is slightly

soluble so exposure must be quick in order to avoid compromising the film. Similarly, a

solution of triethyl phosphate in hexanes (10 mM) produced the same quenching of the metal-based emission from Eu3+ complexes. As can be seen in Fig. 2.10c, the Eu3+-based

fluorescence is quenched only in the part of the film that was exposed to the solution.

This is consistent with the triethyl phosphate displacing the Eu3+ ion which results in the combined blue fluorescence from the Zn2+:Mebip complex and free ligand.

We also wanted to explore the potential for these films to detect organophosphate

vapors. To do this, a film was exposed to a triethyl phosphate enriched environment.

After 24 h exposure the film showed quenching of the Eu3+-based emission (Figure

2.10d). A film exposed for only 1 h showed similar changes, although the film’s red

fluorescence quickly returned upon removal from the triethyl phosphate-rich atmosphere, highlighting the reversibility of this process if the exposure times are kept short. These studies show that the incorporation of Eu3+ into these metallosupramolecular polymers also allows the films to be used as chemical sensors for detection of liquid, solution, and gaseous organophosphates by way of fluorescence quenching as demonstrated in Figure

2.10e.

Figure 2.10 Pictures of the fluorescence of a Zn2+:Eu3+ 70:30 film a) as cast, b) dipped into a 10 mM solution of triethyl phosphate in hexanes for 1 min, c) exposed to triethyl phosphate vapor for 24 hrs, and d) dipped directly into triethyl phosphate under UV light (365 nm). All films exposed to triethyl phosphate show quenching of the red Eu3+ metal-based emission as the Eu3+ is competitively bound by the phosphate. b) Photoluminescence spectra of the above films. All 3+ films show a decrease in the characteristic Eu emission peaks at 590 and 615 nm (λex = 377 nm).

2.3 CONCLUSIONS

The combination of Zn2+ and Eu3+ metal-ions with 4-oxy-2,6-bis-(10-

methylbenzimidazolyl)pyridine (Mebip) ditopic end-capped poly(tetrahydrofuran) results

in the formation of a metallosupramolecular polymer that yields stimuli-responsive

thermoplastic elastomer films. A series of these metallosupramolecular polymer films were studied with differing ratios of Zn2+ to Eu3+ in order to determine the effect that

increasing the Eu3+:Zn2+ ratio has on the photoluminescent, thermomechanical and

structural properties of these films. Increasing the films’ Eu3+ content results in a change

in fluorescence of the films from blue to red, which occurs as the blue ligand-based

emission from the Mebip:Zn2+ complexes is replaced by the more intense red Eu3+-based

emission from the Eu3+:Mebip complexes. The films’ mechanical properties show an

enhanced sensitivity to temperature with increasing Eu3+ content and the Eu3+-based

emission of these films is also sensitive to temperature as well as to chemicals known to

displace the Eu3+ from the ligand. This stimuli-responsive behavior combined with the

relatively good mechanical properties of these films offer potential as flexible

fluorescence-based temperature and chemical sensors.

2.4 EXPERIMENTAL METHODS

2.4.1 Materials

All reagents and solvents were purchased from Aldrich Chemical Co. Reagents were used without further purification. Solvents were distilled from suitable drying agents.

Spectrophotometric grade chloroform and acetonitrile were used for all experiments.

2.4.2 Instruments

NMR spectra were recorded on a Varian 600 NMR spectrometer. Dynamic mechanical

thermal analysis experiments were performed using a TA Instruments DMAQ800 under

-1 N2 with liquid N2 cooling and heated at a rate of 10 °C min . UV-vis spectra were

obtained by a Perkin-Elmer Lambda 800 UV-vis spectrometer. Titration experiments

were performed in quartz cuvettes with scanning in the range of 250–400 nm with an

integration time of 0.24 s. Fluorescence spectra were obtained with a SPEX Fluorolog 3

(model FL3-12); corrections for the spectral dispersion of the Xe-lamp, the instrument

throughput, and the detector response were applied. Temperature and chemical dependent

photoluminescence spectra were acquired on an Ocean Optics ACD1000-USB

spectrometer (λex = 377 nm) through the use of a Y-shaped optical fiber in conjunction

with a Gel Instrumente AG hotstage with a TC2 temperature controller. The broad peak

centering around 490 nm that is observed in some measurements is the product of a

reflection from the slide the films were placed on. Molecular weights of the materials

were measured by mass spectrometry on a Bruker AUTOFLEX III MALDI-TOF/TOF

mass spectrometer using HABA [2-(4’-hydroxybenzeneazo) benzoic acid] as the matrix

with a sodium trifluoroacetate additive. X-Ray measurements were conducted using a

Rigaku SA-HF3 X-ray generator for the D/MAX2000/PC series diffractometer. All

samples for X-ray study were prepared as films and placed on a glass cover slide aligned

in the path of the wide-angle diffractometer.

2.4.3 Preparation of 2

In flame dried glassware, a mixture of 4-hydroxy-2,6-bis(10- methylbenzimidazoyl)pyridine27 (HOMebip, 5.03 g, 14.1 mmol) and N,N- diiospropylethylamine (DIEA, 5.00 mL, 28.7 mmol) was suspended in 350 mL of freshly distilled tetrahydrofuran (THF). In a separate flask, 350 mL of freshly distilled THF were

cooled with stirring to 5 °C in an ice bath. Trifluoromethanesulfonic anhydride (1 mL,

5.9 mmol) was added to the THF and allowed to polymerize for 20 min. After 20 min, the

suspension of HOMebip and DIEA was added to terminate the growing polymer chain

and stirred overnight. The solvent was removed under vacuum to give a viscous purple

oil which was then dissolved in CHCl3 and washed with 1M NaOH (× 3). The precipitated unreacted HOMebip was filtered and the organic phase was stirred with decolorizing carbon to further remove excess HOMebip. After filtering off the carbon, the solvent was again removed under vacuum and the product was purified by column chromatography (CHCl3 : MeOH, 100 : 0, 99 : 1, 98 : 2) to give macromonomer 2 (7.9 g,

67%). δH(600 MHz; CDCl3) 7.92 (s, 4H, Ar), 7.86 (d, 4H, Ar), 7.46 (d, 4H, Ar), 7.36 (m,

8H, Ar), 4.26 (t, 4H, MebipOCH2), 4.23 (s, 12H, CH3), 3.40 (m, 256H, OCH2CH2), 1.61

(m, 260H, OCH2CH2); δC(100 MHz; CDCl3) 166.4, 150.9, 150.2, 142.3, 137.0, 123.3,

122.6, 119.9, 111.6, 109.8, 70.4, 32.4, 26.3. m/z (MALDI TOF-TOF) (matrix: HABA):

Mn = 3500. PDI: 1.03.

13 Figure 2.11 100 MHz C NMR of 2 in CDCl3.

Figure 2.12 MALDI-TOF of 2.

Figure 2.13 MALDI-TOF-TOF of 2 showing fragmentation of only the 4-hydroxy-2,6- bis(10-methylbenzimidazoyl)pyridine ligand (354 m/z) from the parent peak (3250 m/z).

2.4.4 Typical Sample Preparation of Metallosupramolecular Polymers

Example procedure for 2 with Zn2+:Eu3+ 90:10. A solution containing 180 mg (0.045

mmol) of 2 in 1 mL of chloroform was mixed with 0.3 mL (0.003 mmol) of Eu(ClO4)3

from a 10 mM stock solution in acetonitrile then with 2.025 mL (0.0405 mmol) of

Zn(ClO4)2 from a 20 mM stock solution in acetonitrile. Once everything had dissolved, the solvent was removed under vacuum and the resulting polymer was redissolved in 1.5 mL of chloroform. This solution was then cast into an aluminum-walled casting dish with

a Teflon sheet bottom. The complex was allowed to air-dry overnight and then was vacuum dried in an oven for 6 h at 40 °C. Films of other metal ratios were prepared by

altering the amount of the Eu3+ and Zn2+ stock solutions appropriately. All films studied

had a thickness of 200 ± 20 µm.

Figure 2.14 Photoluminescence spectra of 2 and varying ratios of Zn2+ and Eu3+ ranging from 2+ 3+ Zn :Eu 100:0 to 50:50 in 10% increments in solution. λex = 365 nm. The small peaks at 470 nm and 495 are instrument artifacts.

Figure 2.15 Thermogravimetric analysis of the series of films prepared with 2 and varying ratios of Zn2+:Eu3+. The films are very thermally stable and > 5% weight loss is only observed above 300 °C, however there is a small amount of degradation of the perchlorate anion beginning around 150 °C.

2.4.5 Temperature and Chemical Responsiveness Experiments

A Zn2+:Eu3+ 70:30 film was used for all experiments. The response to temperature was

examined by placing the film on a glass slide at room temperature and heating to 120 °C before cooling to room temperature by simply removing the film from the heat source; this cycle of heating and cooling was then repeated. The PL spectra were recorded at different temperatures using an optical fiber setup (Ocean Optics).

Experiments for the detection of triethyl phosphate were carried out using a

Zn2+:Eu3+ 70:30 film. The solution detection was carried out by dipping the film into a 10

mM triethyl phosphate solution in hexanes. Color changes were observed after short

periods of time (10 s) with the full quenching of the red Eu3+-based emission occurring after 1 min. This was also seen in 1 mM solutions, although the time required for detection was greater. Vapor detection was accomplished by creating an analyte-rich environment by heating a beaker of triethyl phosphate at 60 °C for 2 h in a closed container. The temperature was then dropped back to room temperature (to avoid any thermal response) and the film was placed into the container for the desired amount of time.

2.5 ACKNOWLEDGMENTS

This material is based upon work supported by the National Science Foundation under

Grant no. CHE-0704026, DMR-0602869, DMR-0423914, and MRI-0821515 (for the purchase of the MALDI-TOF/TOF). Special thanks to Jihzu Jin for his initial work on these metallosupramolecular polymers.

2.6 REFERENCES

1. (a) Lehn, J. -M. Chem. Soc. Rev., 2007, 36, 151–160. (b) Serpe, M. J.; Craig, S. L. Langmuir, 2007, 23, 1626–1634. (c) Ciferri, A. Macromol. Rapid Commun., 2002, 23, 511–529.

2. (a) ten Brinke, G.; Ruokolainen, J.; Ikkala, O. Adv. Polym. Sci., 2007, 207, 113–177. (b) Rieth, S.; Baddeley, C.; Badjić, J. D. Soft Matter, 2007, 3, 137–154. (c) Mendes, P. M. Chem. Soc. Rev., 2008, 37, 2512–2529. (d) Gunnlaugsson, T.; Leonard, J. P. Chem. Commun., 2005, 3114–3131.

3. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

4. (a) Friese, V. A.; Kurth, D. G. Coord. Chem. Rev., 2008, 252, 199–211. (b) Dobrawa, R.; Würthner, F. J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4981–4995. (c) Hofmeier, H.; Schubert, U. S. Chem. Soc. Rev., 2004, 33, 373–399. (d) McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. Rotello, V. M.; Thayumanavan, S. John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178. (e) Friese, V. A.; Kurth, D. G. Curr. Opin. Colloid Interface Sci., 2009, 14, 81–93. (f) Schmatloch, S.; van den Berg, A. M. J.; Alexeev, A. S.; Hofmeier, H.; Schubert, U. S. Macromolecules, 2003, 36, 9943–9949.

5. Kurth, D. G.; Higuchi, M. Soft Matter, 2006, 2, 915–927.

6. (a) Vermonden, T.; van Steenbergen, M. J.; Besseling, N. A. M.; Marcelis, A. T. M.; Hennink, W. E.; Sudhölter; E. J. R.; Cohen Stuart, M. A. J. Am. Chem. Soc., 2004, 126, 15802–15808. (b) Bender, J. L.; Corbin, P. S.; Fraser, C. L.; Metcalf, D. H.; Richardson, F. S.; Thomas, E. L.; Urbas, A. M. J. Am. Chem. Soc., 2002, 124, 8526– 8527.

7. Handl, H. L.; Gillies, R. J. Life Sci., 2005, 77, 361–371.

8. Kuriki, K.; Koike Y.; Okamoto, Y. Chem. Rev., 2002, 102, 2347–2356.

9. (a) Binnemans, K. Liquid-crystalline lanthanide complexes in Inorganic Chemistry in Focus II, ed. Meyer, G.; Naumann, D.; Wesemann, L. Wiley-VCH, Weinheim, 2005, ch. 15, pp. 267–291. (b) Binnemans, K.; Goerller-Walrand, C. Chem. Rev., 2002, 102, 2303–2345.

10. (a) Kido, J.; Okamoto, Y. Chem. Rev., 2002, 102, 2357–2368. (b) Curry, R. J.; Gillin, W. P. Curr. Opin. Solid State Mater. Sci., 2001, 5, 481–486.

11. Bünzli, J. -C. G.; Piguet, C. Chem. Soc. Rev., 2005, 34, 1048–1077.

12. (a) Iyer, P.; Beck, J. B.; Rowan, S. J.; Weder, C. Chem. Commun., 2005, 3, 319–321. (b) Knapton, D.; Rowan, S. J.; Weder, C. Macromolecules, 2006, 39, 651–657.

13. Beck, J. B.; Ineman J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060–5068.

14. Piguet, C.; Williams, A. F.; Bernardinelli, G.; Bünzli, J. -C. G. Inorg. Chem., 1993, 32, 4139–4149.

15. (a) Lis, S. J. Alloys Compd., 2002, 341, 45–50. (b) Sabbatini, N.; Guardigli, M.; Lehn, J. -M. Coord. Chem. Rev., 1993, 123, 201–228.

16. Petoud, S.; Bünzli, J. -C. G.; Glanzman, T.; Piguet, C.; Xiang, Q.; Thummel, P. J. Lumin., 1999, 82, 69–79.

17. Escande, A.; Guénée, L.; Buchwalder, K. -L.; Piguet, C. Inorg. Chem., 2009, 48, 1132–1147.

18. Piguet, C.; Bünzli, J.- C. G.; Bernardinelli, G.; Hopfgartner, G.; Williams, A. F. J. Alloys Compd., 1995, 225, 324–330.

19. (a) Chen, C. -C.; Dormidontova, E. E. J. Am. Chem. Soc., 2004, 126, 14972–14978. (b) Wang, S. H.; Chen, C.- C.; Dormidontova, E. E. Soft Matter, 2008, 4, 2039–2053.

20. (a) Beck, J. B.; Rowan, S. J. J. Am. Chem. Soc., 2003, 125, 13922– 13923. (b) Zhao, Y.; Beck, J. B.; Rowan, S. J.; Jamieson, A. M. Macromolecules, 2004, 37, 3529–

3531. (c) Weng, W.; Beck, J. B.; Jamieson, A. M.; Rowan, S. J. J. Am. Chem. Soc., 2006, 128, 11663–11672.

21. (a) Kuroiwa, K.; Shibata, T.; Takada, A.; Nemoto, N.; Kimizuka, N. J. Am. Chem. Soc., 2004, 126, 2016–2021. (b) Kim, H. -J.; Zin W. -C.; Lee, M. J. Am. Chem. Soc., 2004, 126, 7009–7014. (c) Kawano, S.; Fujita, N.; Shinkai, S. J. Am. Chem. Soc., 2004, 126, 8592–8593. (d) Kishimura, A.; Yamishita, T.; Aida, T. J. Am. Chem. Soc., 2005, 126, 179–183.

22. Knapton, D.; Burnworth, M.; Rowan, S. J.; Weder, C. Angew. Chem., Int. Ed., 2006, 45, 5825–5829.

23. Burnworth, M.; Rowan, S. J.; Weder, C. Chem. Eur. J., 2007, 13, 7828–7836.

24. Godlewska, P.; Macalik, L.; Hanuza, J. J. Alloys Compd., 2008, 451, 236–239.

25. Vermonden, T.; van der Gucht, J.; de Waard, P.; Marcelis, A. T. M.; Besseling, N. A. M.; Sudhölter, E. J. R.; Fleer, G. J.; Cohen Stuart, M. A. Macromolecules, 2003, 36, 7035–7044.

26. Kunzelman, J.; Chung, T.; Mather, P. T.; Weder, C. J. Mater. Chem., 2008, 18, 1082– 1086.

27. Beck, J. B.; Rowan, S. J. Faraday Discuss., 2005, 128, 43–53

CHAPTER 3

Influence of Metal-Ion and Polymer Core on the Melt

Rheology of Metallosupramolecular Films

Adapted from: Kumpfer, J. R.; Wie, J. J.; Swanson, J. P.; Beyer, F. L.; Mackay, M. M.; Rowan,

S. J. Macromolecules, 2012, 45, 473-480.

3.1 INTRODUCTION

As mentioned in the previous chapter, supramolecular polymers utilize reversible,

noncovalent interactions to achieve high molecular weight materials that combine the

properties of typical covalent polymers with those of low molecular weight molecules.1- 6

These polymers have gained increasing interest from a materials’ properties standpoint as

they can have mechanical properties similar to covalent polymers but have much greater

temperature-dependant viscosities as a result of the reversible interactions, which allows

easy processing and/or the material to exhibit stimuli-responsive behaviors.7- 11 An

additional benefit of these types of polymers is that their properties are widely tailorable

simply by modifying the reversible interaction used. In this chapter we explore a deeper

understanding of the structure-property relationships of these materials. We combine an

in-depth look at the melt rheological properties and behavior of the series of materials in

Chapter 2 along with small-angle X-ray scattering (SAXS) experiments to further investigate the impact of changing the metal interaction that is used.

A number of supramolecular polymers using different noncovalent interactions have been studied, with the largest focus being on hydrogen bonding systems.12- 15

System utilizing metal-ligand coordination16- 20 are growing more popular as they allow

for a wide range of properties as many different ligands and metal-ions can be used.

While detailed studies of the melt rheological properties of supramolecular polymers

assembled with hydrogen bonds have been carried out,21- 26 the same is not the case for metallosupramolecular polymers. This is likely a result of there being a much smaller reported number of metallosupramolecular polymers that display the ability to be processed into mechanically stable films. As such, in-depth solution rheological studies

64 of metal-coordination polymers have been reported,27-29 and similarly there have been investigations into the rheological and stimuli-responsiveness of metallosupramolecular gels.30- 33 However, to date there have been no studies performed exploring the solid- state, melt rheological properties of metallosupramolecular polymer films.

In the previous chapter we have shown that elastomeric films can be prepared from ligand end-capped low Tg polymers assembled with zinc and/or lanthanides and that these films show interesting stimuli-responsive properties that are highly dependent on the metal-ion content. In Chapter 2, metallosupramolecular polymers prepared from a

2,6-bis(N-methylbenzimidazoyl)pyridine (MeBip) endcapped poly(tetrahydrofuran) (1) with zinc and europium ions yield temperature and chemo-responsive films,34,35 while those prepared with MeBip endcapped poly(ethylene-co-butylene) (2) yield thermally and photo-responsive rehealable elastomers with either zinc or lanthanide ions.36 In order to achieve a better understanding of these metallosupramolecular polymers, we report a detailed rheological study of both these materials to examine the role of the metal-ions and the nature of the polymer backbone on the materials’ properties.

3.2 RESULTS AND DISCUSSION

Macromonomer 1 was again prepared via a cationic, ring-opening living polymerization of poly(THF) (see Chapter 2) while macromonomer 2 was prepared using Mitsunobu chemistry to attach the Mebip ligand to the terminal hydroxyl groups on a poly(ethylene- co-butylene) core as previously reported.35,36 As described in Chapter 2, these macromonomers were then self-assembled in solution with Zn2+ and Eu3+ perchlorate salts to yield the metallosupramolecular polymers which were then solution cast to yield 65 elastomeric films. For mixed metal-ion systems the percentages reported represent the idealized percentage of the total MeBip ligands that are coordinated with each metal-ion, assuming a 2:1 Mebip:Zn2+ binding and a 3:1 Mebip:Eu3+. For example, in a mixed

Zn2+:Eu3+ film of 70:30, 70% of the MeBip ligands can be bound to a Zn2+ ion and 30% of the ligand can bind to the Eu3+. The metal:macromonomer ratio was calculated to ensure all ligands can bind to a metal-ion and all metal-ions are fully coordinated.

Specifically in this study we were interested in examining the effect of two different parameters on the rheological properties of these materials, namely the effect of Zn2+ ions versus Eu3+ ions and the effect of the macromonomer core.

Figure 3.1 Chemical structures of Mebip end-capped macromonomers studied with poly(tetrahydrofuran) (2) and poly(ethylene-co-butylene) (3) cores.

3.2.1 Influence of Eu3+ Content

We have shown in the previous chapter, using dynamic mechanical thermal analysis

(DMA), that films of the metallosupramolecular polymers of 2 with higher ratios of the

Eu3+ to Zn2+ exhibit enhanced temperature sensitivity which leads to more thermally

sensitive stimuli-responsive films. Thus, to examine this effect in more detail a series of

films of 2 with varying Zn2+:Eu3+ ratios were prepared ranging from 100:0 to 50:50 in

10% increments. As we are also interested in the effects of changing the polymer core, films were also prepared with macromonomer 2, shown in Figure 3.1, with Zn2+:Eu3+

ratios of 100:0, 50:50, and 0:100. Metallosupramolecular polymers of 2 with less than

50% of the ligand bound to Zn2+ ions did not form mechanically stable films, yielding

instead only oils, however films made with 100% Eu3+ and 3 form elastomeric,

mechanically stable films.

Frequency sweeps were carried out over a range of temperatures for the films

prepared with 2 and then combined using time-temperature superposition to obtain

master curves which are shown in Figure 3.2. The master curves are useful in helping to

give insight into the structure of these metallosupramolecular polymers and how the

structure changes as more and more Eu3+ is introduced into the films. From the master

curves in Figure 3.2 it can be concluded that there is a change in the structure as the ratio

of Zn2+ to Eu3+ decreases. Films composed of 100% Zn2+ display typical rheological behavior of high molecular weight polymers after observation of the low frequency region. In the case of the high Zn2+ content films (100:0-70:30), the terminal (flow)

region cannot be reached over the range of temperatures and frequencies studied, and

only the plateau region can be observed. Surprisingly, no minimum in the tan δ curve (vs

ω) is observed that would correlate to a plateau modulus as shown in Figure 3.3. Even at lower temperatures, ca. -50 °C, no minimum is found for tan δ as seen in Figure 3.3b which may result from the fact that even at -50 °C the polymer is still well above Tg of

the poly(tetrahydrofuran) core (ca. -80 °C) .

Figure 3.2 Master curves of films of 2 prepared with varying ratios of Zn2+:Eu3+. a) 100:0, b) 90:10, c) 80:20, d) 70:30, e) 60:40, f) 50:50. Storage modulus (blue triangles), loss modulus (green circles), and complex viscosity (red squares) vs oscillatory angular frequency.

Tref = 30 °C.

As the Eu3+ content in the films increases, the frequency at which the behavior

changes from the plateau region to the terminal region shifts to higher values. This is

indicative of a decrease in the effective molecular weight of the assembled

macromonomers and/or a reduction of entanglements as a result of the molecular weight

reduction. This gives insight into what is changing structurally as the Eu3+ content is

increased. For a Zn2+:Eu3+ 100:0 film, it would be expected to form a linear, high

molecular weight polymer as the 2:1 Mebip:Zn2+ complex will just yield chain extension.

Once Eu3+ is added, it can bind in a 3:1 Mebip:Eu3+ ratio which can result in branching along the metallosupramolecular polymer. Thus, as more Eu3+ is introduced, the amount

of branching will increase, which will effectively reduce the molecular weight of the

linear polymer segments, which are formed by chain extension with Zn2+ complexes

between crosslinks, since the amount of the Zn2+ is reduced, while the total molecular

weight increases. Concordantly, it can be expected that the polydispersity of the linear

polymer segments formed with Zn2+ will also increase as the Eu3+ content is increased

because the Eu3+, which is responsible for branching, can be inserted anywhere within a

linear polymer segment. This is backed up by examination of the G" vs. ω curves. In the

high content Zn2+ materials (100:0 and 90:10) both a minima and maxima are observed as

ω is decreased in the G" curve, however this behavior is replaced by a more linear one as

more Eu3+ is added, consistent with an increase in the polydispersity of the material.37

Figure 3.3 Time-temperature superposition curves of tan δ vs. ω of a) films prepared with 2 with Zn2+:Eu3+ ratios ranging from 100:0 to 50:50 from temperatures ranging from 30 °C to 110 °C and b) a 60:40 film with 2 with temperatures ranging from -30 °C to 70 °C. In all cases no minima are observed which would correlate to a plateau modulus (Ge).

Studies into the relaxation time, which is roughly defined as the terminal viscosity

divided by the plateau modulus, are also found to greatly depend on the Eu3+ content

within the films. It is not possible to determine the terminal viscosity for films containing

0% to 30% Eu3+, as the terminal region was not reached, but it is obvious that there is a

trend of increasing relaxation times with decreasing Eu3+ content (there is a shift in the

crossover point between Gʹ and Gʺ to larger ω). For the films containing 40% and 50%

Eu3+, the terminal viscosities, which were estimated at 3.3×1010 and 6.3×109 Pa∙s, are

used with estimations of the plateau modulus (3.79 and 3.97 MPa) to give relaxation

times of 8700 and 1600 seconds, respectively. As mentioned earlier no minimum in tan δ

was observed in these films so the estimations of the plateau modulus were taken from

the lowest measured tan δ value. Thus for these two films the estimated plateau modulus is lower than the real value. None-the-less these results show that the relaxation times of

these films are extremely dependent on the metal ratios, where a 10% increase in Eu3+

leads to a five-fold decrease in relaxation time. The increase in the estimated plateau

modulus from the 40% to 50% Eu3+ is only 5% which is within experimental error and

this suggests that the decrease in the terminal viscosity is primarily responsible for the

dramatic decrease in relaxation times as more Eu3+ is added. This is likely on account of

the increasing amount of Eu3+ increasing branching which will play a role in decreasing the viscosity. It is also possible that the decrease in terminal viscosity in the higher concentration Eu3+ materials is on account of the easier decomplexation of the weaker

bound Eu3+ complexes relative to the Zn2+ complexes resulting in more facile

depolymerization in the system.

The effect on the temperature-sensitivity with changing metal ratios can be

elucidated by examining the vertical and horizontal shift factors of the master curves

shown in Figure 3.4. The horizontal shift factors (Figure 3.4a) change in both slope and intensity which implies that these materials are quite temperature dependent. The vertical shift factors (Figure 3.4b) also show the dramatic change in the temperature sensitivity, particularly at high temperature, with decreasing the Zn2+:Eu3+ ratio. For example, the

vertical shift factors for the Zn2+:Eu3+ 100:0 film change little with temperature, while the

Zn2+:Eu3+ 50:50 film exhibits much different behavior in the temperature range studied.

A trend is observed with changing metal ratio and as the amount of Eu3+ is increased the

vertical shift factors move from a linear response to a larger and larger nonlinear response

with temperature. This data is consistent with the previously published DMA results

which show the increasing temperature sensitivity with increasing amounts of Eu3+ in the

films.

Figure 3.4 a) Horizontal and b) vertical shift factors of the series of films prepared with 2. The

difference in aT at high temperatures suggest these films are relatively frequency dependent, 3+ whereas changes in bT with increasing Eu display the increased temperature sensitivity. Error bars are shown for the Zn2+:Eu3+ 50:50 sample only for clarity; however, they apply to all of the data sets in b). The error results from a curve fit minimizing an objective function to overlay the curves.

While the temperature dependence of the vertical shift factor bT may seem

unusual it can be explained quite simply by remembering its origin. Consider network

38 theory where the storage and loss modulus can both be written as nkBT ƒ(λω) where n is the number of crosslinks per unit volume, kB, Boltzmann’s constant, T, temperature,

and ƒ(λω), a function describing the modulus frequency dependence (ω) which is

normalized by a relaxation time (λ). Of course, the functional form is different for the

two moduli. The number of crosslinks can be interpreted as the number of molecules per

unit volume or entanglements per unit volume depending on model specifics for the

system at hand. For linear polymer melts n can be written as ρNA/M where ρ is the mass

density, NA, Avogadros number and M is molecular weight, allowing one to write the

vertical shift factor as 73

[ρN A / M ] k BT ρT bT = = (Eq. 3.1) [ρ0 N A / M 0 ] k BT0 ρ0T0 where the subscript 0 is used to denote “at the reference temperature” and the molecular weight is assumed constant in the later form of the equation. Since the melt density does not change very much with temperature and the absolute temperature similarly changes little, plus as temperature rises the density falls, bT will typically change by at most 10%.

The metallosupramolecular materials studied here have higher temperature sensitivity, especially those containing larger amounts of Eu3+. This is consistent with

Eu3+ decomplexing from the Mebip end groups as the temperature is increased. This leads to bT falling by a factor of two in the case of the 50:50 material suggesting the effective molecular weight changes with temperature. This may violate the inherent assumption of thermorheological simplicity, invalidating the use of shift factors, however, the effect is relatively mild in that the entire material does not decomplex and we believe that one can use shift factors to obtain an approximation to the full rheological response for the extended frequency range. Using the shift factors we can then estimate the activation energy (Ea) of the system which will be a mean of the activation energy for reputation and breaking of the chains. The activation energies increase with increasing

Eu3+ content ranging from ~55 kJ/mol for the 100:0 film to ~68 kJ/mol for the 50:50 film.

These values are relatively close to that given by Cohen Stuart and coworkers for a metallosupramolecular polymer prepared from a ditopic pyridine-2,6-dicarboxylate end- capped poly(ethylene glycol) with 100% Nd3+ of 49 kJ/mol. Yet, this caveat should be kept in mind; the higher Eu3+ content materials may change in effective molecular weight and the terminal region, obtained from higher temperature data, may have a smaller

molecular weight than data collected at lower temperatures. Indeed we should expect this

since at higher temperatures the materials obtain low viscosity through more complete decomplexation allowing them to heal if need be.

The complex viscosity is also directly affected by the metal-ion composition of

the films. As the ratio of Zn2+:Eu3+ is decreased, the viscosity is observed to decrease,

most noticeably upon approaching the terminal region. As mentioned before, the

transition to the terminal region is only observed for the films comprised of Zn2+:Eu3+

60:40 and 50:50 which can also be seen in the complex viscosity data. It is necessary to

remember that the effective molecular weight for these films should be reduced as

temperature is increased as a result of metal-ligand decomplexation of the more thermally responsive Eu3+ complexes. Increasing the Eu3+ content will also result in increased branching as mentioned earlier. This suggests that the viscosity decrease is most likely a result of depolymerization at high temperatures and/or increased branching of the metallosupramolecular polymer. The probability of forming rings should also increase

with increasing Eu3+ which may also contribute to the reduction in viscosity, although no

evidence directly supports this idea. The film with the lowest complex viscosity

(Zn2+:Eu3+ 50:50) has a terminal viscosity of approximately 5×109 Pa·s. This value is

higher than any previously reported for a supramolecular polymer and may in part be

related to the microphase separated morphology of these materials (which is looked at in

detail later on in this chapter). Figure 3.5 shows that the complex viscosities are also

shown to increase dramatically as the Eu3+ content is decreased as there is a five-fold estimated increase in the terminal viscosity of the 60:40 over the 50:50 film. For films consisting of Zn2+:Eu3+ ratios of 100:0 to 70:30, a terminal viscosity is not reached in the

temperature-frequency range studied. All of these films show viscosities ranging from

5×1011 to 1×1012 Pa·s with terminal viscosities expected to be even larger. This is

extremely surprising for thermoplastic elastomers, as one definition of a glass is a

material exhibiting a viscosity of 1×1012 Pa·s.39 These films show viscosities just below that defined transition, suggesting that these films are behaving as extremely viscous liquids.

Figure 3.5 Time-temperature superposition plot of complex viscosity vs. oscillatory angular frequency. Films of 2 with Zn2+:Eu3+ ratios of 60:40 and 50:50 approach terminal viscosities while films containing less Eu3+ have nonterminal viscosities of order 1012 Pa·s which is near the defined viscosity of a glass. The temperature range studied was 30 – 110 °C.

3.2.2 Influence of the core polymer

In order to study the impact of the polymer backbone on the melt rheology behavior of

these metallosupramolecular polymers select films (Zn2+:Eu3+ 100:0, 50:50, and 0:100)

were prepared from the end-capped poly(ethylene-co-butylene) macromonomer (3) and compared with the previous results for the same Zn2+:Eu3+ ratios with 2. Both cores have 76

low glass transition temperatures (ca. -80 and -25 °C for 2 and 3 respectively), however the poly(tetrahydrofuran) core is more polar and the oxygen atoms in the backbone have the potential to act as weak metal-ion coordinating sites, which is not possible in the hydrocarbon core of 3.

Meijer and coworkers40 have shown that the polarity of the core polymer in

telechelic supramolecular polymers assembled using hydrogen-bonding of benzene-1,3,5-

tricarboxamide motifs influences the self-assembly process. Polar polymers were found

to disrupt hydrogen bonding and dramatically reduce the mechanical properties compared

to the nonpolar polymers which yielded highly phase separated materials. It was expected

that these metallosupramolecular polymers would behave in a similar fashion so that the

poly(ethylene-co-butylene) core would enhance the material properties of the films as it

should allow for greater phase separation from the charged metal-ligand complexes, whereas the poly(tetrahydrofuran) monomer would be more miscible with the complexes and has the potential to weakly coordinate the metal-ions, especially the Eu3+ ions.

Examples of the long-range ordering of lamellae resulting from phase separation in films

prepared with 3 are shown in Figures 3.6a-b.

Figure 3.6 a) Representative TEM micrographs showing the lamellar morphology of films of 3 with a) 100% Zn(NTf2)2 and b) 100% La(NTf2)3.

This anticipated difference in the microphase separation behaviors of 2 and 3 was

studied using small-angle X-ray scattering (SAXS) on films of the different materials, as

was recently shown for solution cast films of 3 with Zn2+. Microphase separated

morphologies with appreciable long-range order give rise to constructive interference

between scattered X-rays producing Bragg diffraction maxima; materials with good long-

range order exhibit a higher number of Bragg maxima than those with poor long-range

order. Figure 3.7a shows the SAXS data for the series of films prepared with 2. For all

samples, two distinct Bragg diffraction maxima are observed, characteristic of moderate

long-range order. The second order peaks are located at twice the spacing of the primary

maxima (q*), indicative of a lamellar morphology in which the poly(tetrahydrofuran) core

and metal-ligand complexes form alternating layers of soft and hard phases, respectively,

and consistent with those reported previously for metallosupramolecular polymers. The

lamellar period, d, where d = 2π/q*, is approximately 8.2 nm for all samples, and gives

the center-to-center distance between hard phases in the case of a lamellar morphology. 78

Figure 3.7 a) SAXS of the series of films prepared with 2 showing a typical pattern for a lamellar morphology. The presence of two distinct scattering peaks represent moderate long-range ordering of the metal-ligand phases. b) Comparison of Zn2+:Eu3+ 100:0 films made with 2 (bottom, red) and 3 (top, gold). Films prepared with macromonomer 3 show much greater long- range ordering (four distinct scattering peaks) as a result of better phase separation. c) Schematic representation of phase separated lamellar morphology upon self-assembly.

When compared to the films prepared with 3, which all showed at least three

Bragg diffraction maxima, it is apparent that the degree of ordering in the films based on

2 is significantly reduced (Fig. 3.7b). This is consistent with the more polar poly(tetrahydrofuran) showing less phase separation from the charged metal-ligand phase and also being able to interfere with the ability of the Zn2+ to form metal-ligand

complexes with the Mebip ligand, resulting in reduced long-range order relative to the

100% Zn2+ sample based on poly(ethylene-co-butylene) (3). Interestingly, the degree of

long-range order is unaffected by the replacement of Zn2+ with Eu3+ in the materials

based on poly(tetrahydrofuran) macromonomer 2. Therefore, in the case of the polar

poly(tetrahydrofuran) cores, the effect of the polar core on the long-range order strongly

dominates over the effect of metal-ion.

As expected because of the increased lamellar ordering and better phase

separation, the films prepared with 3 show dramatic enhancements in material properties

over films prepared with 2 as shown in Figure 3.8. A direct comparison of the Zn2+:Eu3+

100:0 materials (Fig. 3.8a) show an order of magnitude enhancement in the modulus and

viscosity just by changing the polymer core from 2 to 3. The mechanical enhancement is most obvious when comparing the 50:50 and 0:100 films, focusing primarily on the lower frequency behavior. As shown earlier, films made with 2 at this metal ratio essentially

displayed a terminal region, whereas films of the same metal ratio with 3 are still in the

plateau region at the same frequencies and temperatures as shown in Figure 3.8b-c. In

fact, the Zn2+:Eu3+ 50:50 film with 3 demonstrates very similar behavior with the 100:0 film prepared with 2 suggesting that the polymer core plays a larger role in determining the materials’ properties than the two different metal-ions in the mixed Zn2+ and Eu3+

films. This effect can, in part, be explained by the increased microphase separation and

long-range morphology observed in the metallosupramolecular polymers of 3 when

compared to 2. As mentioned before, films of 2 with Zn2+:Eu3+ 0:100 do not form stable

films and result in oils. Interestingly, however, films prepared with 100% Eu3+ and 3 do

form mechanically stable films with modulus values similar to the 50% Eu3+ films

(though they do show an increased temperature dependence). This is consistent with the

increased phase separation imparted by the nonpolar core of macromonomer 3 and/or the

ability of the poly(THF) oxygens in 2 to weakly coordinate with the Eu3+ and compete

with the ligand complexes.

Figure 3.8 Direct comparison of the master curves obtained for films prepared with 2 (open) vs. 3 (closed) containing Zn2+:Eu3+ ratios of a) 100:0, b) 50:50, and c) 0:100. Films prepared with 3 show enhancement in G’ (blue triangles), G” (green circles), and η* (red squares) which is a result of increased phase separation between the polymer core and metal-ligand complexes.

Observation of the shift factors of the films prepared with 3 from the master

curves shows that they have similar trends to those presented earlier for the films made

with 2. As more Eu3+ is introduced into these materials they show increased frequency

and temperature dependence, especially at higher temperatures. Unlike the films prepared

with 2, however, even at a Zn2+:Eu3+ ratio of 50:50 films of 3 show no sign of reaching 81

the terminal region. In fact it is only when observing the 0:100 film prepared with 3 that

we observe the terminal region at higher temperatures. As shown in Figure 3.9a, aT of the

0:100 film with 3 is decreased by an order of magnitude compared to the 100:0 and 50:50

film and rough estimations of the activation energies range from ~60 kJ/mol for the 100:0 film to ~73 kJ/mol for the 0:100 film. Similarly, bT (Figure 3.9b) shows little change over

the temperatures studied in the 100:0 and 50:50 films, while the 0:100 film shows a very

nonlinear response in comparison. This behavior of the shift factors is consistent with

what was reported earlier for the films prepared with 2 and increased phase separation

observed in the films with 3 is believed to be the cause of the slightly decreased

temperature and frequency dependence.

Figure 3.9 a) Horizontal and b) vertical shift factors of films prepared with 3 with Zn2+:Eu3+ ratios of 100:0, 50:50, and 0:100 (closed) and 2 with 0:100 (open). Similar to the films prepared with 2, these films are relatively frequency dependent and have increased temperature sensitivity with increasing Eu3+ content.

3.2.3 Response at high strain

Strain sweeps are generally used to determine what strain magnitude can be used in order

to stay within the linear viscoelastic region which is verified as a linear response in G',

G", and tan δ. Once the stress no longer follows a linear trend the material is said to be

outside of the viscoelastic region. This can also be seen in the stress waveform which

follows a sinusoidal pattern within the linear viscoelastic region while outside of this

region the waveform becomes non-sinusoidal having multiple frequency components

easily visualized by a Fourier transform of the stress wave. Interestingly, for all Zn2+:Eu3+

films a sinusoidal waveform is observed even though the material is no longer in the

linear viscoelastic region as shown in Figure 3.10. The large amplitude oscillatory strain

experiments were performed with a sampling frequency of 500 data points per second

which will not influence data analysis since the frequency used in this study, 10 rad/s, are

well below the sampling rate. Furthermore, a Fourier transform of the output torque

showed no higher order harmonics consistent with the model proposed below. For all

films, a strain of greater than 15% is well within the apparent non-linear region as the

storage modulus has typically decreased by a factor of approximately two; however, a linear stress (torque) response is still seen. This is very uncommon and we believe this to be a result of the extremely long relaxation times that are observed for these materials.

Figure 3.10 Strain sweeps of films of 2 with Zn2+:Eu3+ a) 100:0 b) 90:10 c) 80:20 d) 70:30 e) 60:40 f) 50:50. The waveform at 1% strain (i) shows a typical sinusoidal response in the linear region of the strain sweep. At increased strains (ii) the material is no longer in the viscoelastic region; however the waveform still shows a linear response as a result of the extremely long relaxation times which prevent them from relaxing within the experiment timeframe.

As mentioned above, the relaxation times for the two samples that displayed a

terminal region were 1000-10,000 s in value. These are quite large and because of this it

is believed that the system cannot relax during a strain cycle. This is expected because of

the phase separated morphology prohibiting molecular motion greatly affecting the

destruction or creation of the network segment terms in the network theory frequently

used to describe non-linear rheological behavior.41 The term can be written as a function

of the strain or stress, which are both functions of time, exemplifying that as the strain or

stress is increased the rate of segment destruction increases and the segment creation

decreases, but they are also functions of time. Here we assume the network creation rate,

0 0 L, can be written as L0 exp(aγ ), where L0 and a are constants and γ is the strain amplitude making the creation rate only a function of the strain amplitude and not time.

0 Similarly the destruction rate, 1/λ, is given by 1/λ0 exp(bγ ), is again dependent on the

strain amplitude but not time. Of course, the strain in this experiment is given by, γ(t) =

γ°sin(ωt), where ω is the frequency.

We postulate that there is minimal relaxation during a strain cycle since the relaxation times are so large in these materials, as mentioned above, and so assume the destruction and creation rates are related to the strain amplitude. When this is done the equations developed by Ahn and co-workers are remarkably simple to solve. The stress is a pure sinusoid of the form sin(ωt+δ), where δ is the phase lag, so one easily finds for a singular relaxation mode

o e[a−b] γ ω 2 G′ = o (Eq. 3.2) e2bγ + ω 2

and

o eaγ ω G′′ = o (Eq. 3.3) e2bγ + ω 2

From this model we can draw two conclusions. Firstly, it is possible to have a pure sinusoid in the “non-linear” viscoelastic region as we believe that the material cannot relax during a period of oscillation. In true polymeric materials relaxation occurs via segmental Brownian motion. This does not seem to occur in our materials and it appears that structural recovery is slow and related to the materials’ morphology, the metal-ions coordinating with the end groups, and in the case of 2, with the oxygen atoms present in

the polymeric backbone. The thermal energy is insufficient to allow rapid recovery for

the conditions we have chosen to use.

Secondly, equations 3.2 and 3.3 are extremely robust in their predictions. The

relative magnitude of a and b can change the strain response of the material to generate

rheological phase diagrams suggested by Ahn and co-workers. The results in Figure 3.9

suggest a > b since both G' and G" decrease with γ°. Of course, detailed modeling would

be required to satisfactorily fit the moduli data over all frequencies and strains, however

little would be gained as we would have to make arbitrary assumptions concerning

multiple relaxation modes, specifically which mode would be represented by equations

3.2 and 3.3 and whether a and b are the same for each relaxation mode. This model is

very powerful though and rich in its predictions, and lends credence to our hypothesis

that relaxation during a deformation cycle does not occur.

As far as we know this is the first observation of this effect where a pure sinusoid

is observed in the so-called non-linear region. The materials synthesized in this work are

86 obviously unusual when compared to conventional covalent polymers; however, we do not believe that this effect is unique. If another system cannot relax during an oscillation period then similar behavior could be observed. Obviously our system has extremely large relaxation times allowing us to observe this behavior.

3.3 CONCLUSIONS

A detailed study of the rheological behavior of a series of metallosupramolecular films in the melt was carried out to determine the impact of changing metals and the macromonomers. It was found that in films with varying ratios of zinc and europium, increasing the europium content resulted in a dramatic change in the temperature response which allowed the films to reach the terminal region. It was also determined that the polymeric core plays a significant role in the materials’ properties where a non-polar poly(ethylene-co-butylene) core displayed dramatic enhancement of G', G" and η* relative to the more polar poly(tetrahydrofuran) core. All of the films studied had terminal complex viscosities greater than 1x109 Pa·s which is higher than any previously reported for a supramolecular polymer. The films exhibited a sinusoidal waveform in the

“nonlinear” viscoelastic region of the strain sweeps as a result of extremely long relaxation times which prevents the material from relaxing during a period of oscillation.

This was used to develop a simplified network model which can be applied to other materials that are unable to relax during an oscillation period.

3.4 EXPERIMENTAL METHODS

3.4.1 Materials

Compounds 2 and 3 were synthesized according to previously published methods35,36

with molecular weights of 4200 g/mol and 4000 g/mol respectively. All reagents and

solvents were purchased from Aldrich Chemical Co. and used without further

purification. Spectrophotometric grade chloroform and acetonitrile were used for all

experiments.

3.4.2 Sample film preparation

Example procedure for 2 with Zn2+:Eu3+ 70:30

952 µL (0.01 mmol) of a Eu(ClO4)3 solution in acetonitrile (10 mM) and 1666.7 µL

(0.03 mmol) of a Zn(ClO4)2 solution in acetonitrile (20 mM) was mixed with 200 mg

(0.05 mmol) of 2 in 5 mL of chloroform. The solvent was removed under vacuum and the

metallosupramolecular polymer was redissolved in 2 mL of chloroform, cast into an

aluminum-walled casting dish with a Teflon base, and allowed to air-dry overnight. The

films were further dried in a vacuum oven at 40 °C for 6 hrs. to remove any residual

solvent before use. Films made with 3 and of other metal ratios were prepared by varying

the amounts of Eu(ClO4)3 and Zn(ClO4)2 appropriately. Typical film thicknesses were

0.25 mm ± 0.05 mm.

3.4.3 Rheological Measurements

All experiments were carried out on a TA Instruments ARES G2 rheometer with 8 mm

parallel plates. Circular discs of the films were obtained using an 8 mm diameter steel

punch. Prior to each experiment, the samples were heated under compression for short

periods of time (5 mins., see below for more details) before cooling to 30 °C. This step is

required for good adhesion between the films and the plates. It also erases any thermal

history that may have resulted from crystallization of the poly(tetrahydrofuran) core of

macromonomer 2. Films of composition Zn2+:Eu3+ 100:0 to 80:20 were heated to 130 °C

under a normal force of 5 N. Films with higher europium contents required slightly lower

temperatures and pressures to prevent the film from flowing out of the plates. As such,

the 70:30 film was pressed at 110 °C and 5 N, the 60:40 film at 90 °C and 3 N, and the

50:50 film was pressed at 70 °C and 3 N. Strain sweeps were performed in order to

determine the linear viscoelastic region for each film. Samples were strained at a constant

frequency of 10 rad/s and ramped from 0.1% strain until the samples lost adhesion, which

was typically at strains larger than 30%. All frequency sweep experiments were

performed at 1% strain to ensure that all responses were in the viscoelastic region.

Frequency sweeps were carried out over a frequency range of 0.1 to 100 rad/s with

temperatures from 30 °C to 110 °C in 20 °C increments. Temperatures never exceeded

130 °C to avoid decomposition of the metal salts. The frequency sweeps for each film

were then combined using time-temperature superposition to generate master curves.

The reference temperature was 30 °C for all master curves.

3.4.4 Small-angle X-ray scattering (SAXS) characterization

SAXS data were collected using a customized, pinhole collimated SAXS camera. X-rays

having wavelength (λ) of 1.542 Å were generated at 45 kV and 100 mA with a Rigaku

Ultrax18 rotating Cu anode and filtered using Ni foil. Two-dimensional data sets were

collected using a Molecular Metrology 120 mm diameter area detector, then corrected for

background noise and sample absorption. The corrected data were then azimuthally averaged to yield one-dimensional intensity, I, as a function of scattering vector magnitude, q, where q = 4πsin(θ)/λ and 2θ is the scattering angle. The instrument was calibrated using Ag behenate and the data placed on an absolute scale (cm-1) using type-2

glassy carbon, previously calibrated at the Advanced Photon Source, Argonne National

Laboratory, as a secondary intensity standard. Each sample was characterized at two

sample-to-detector distances, 150 cm and 50 cm, and the reduced data combined into one

continuous data set. All SAXS data processing, manipulation, and analysis was

performed using Wavemetrics IGOR Pro and procedures available from Argonne

National Laboratory.42

3.5 ACKNOWLEDGMENTS

This material is based upon work supported by the U.S. Army Research Office

(W911NF-09-1-0288), the National Science Foundation under Grants CHE-0704026,

DMR-0602869, and MRI-0821515 (for the purchase of the MALDI-TOF/TOF), and the

Case School of Engineering. Special thanks goes to Jeong Wie and John Swanson for

their contributions to the rheology experiments, Fredrick Beyer for his help with the

SAXS measurements, and Michael Mackay for his invaluable help in interpreting the

rheological data and proposed simplified network model.

3.6 REFERENCES

1. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

2. De Greef, T. F. A.; Smulders, M. M. J.; Wolffs, M.; Schenning, A. P. H. J.; Sijbesma, R. P.; Meijer, E. W. Chem. Rev., 2009, 109, 5687–5754.

3. Lehn, J. -M. Prog. Polym. Sci., 2005, 30, 814–831.

4. Weck, M. Polym. Int., 2007, 56, 453–460.

5. Harada, A.; Takashima, Y.; Yamaguchi, H. Chem. Soc. Rev., 2009, 38, 875–882.

6. Courtois, J.; Baroudi, I.; Nouvel, N.; Degrandi, E.; Pensen, S.; Ducouret, G.; Chanéac, C.; Bouteiller, L.; Creton, C. Adv. Funct. Mater., 2010, 20, 1803–1811.

7. Wojtecki, R. J.; Meador, M. A.; Rowan, S. J. Nature Mater., 2011, 10, 14.

8. Kim, H. -J.; Lim, Y. -B.; Lee, M. J. Polym. Sci., Part A: Polym. Chem., 2008, 46, 1925–1935.

9. Rieth, S.; Baddeley, C.; Badjić, J. D. Soft Matter, 2007, 3, 137–154.

10. Nair, K. P.; Breedveld, V.; Weck, M. Macromolecules, 2011, 44, 3346–3357.

11. (a) Yuen, F.; Tam, K. C. Soft Matter, 2010, 6, 4613–4630. (b) Liu, J.; Chen, G.; Goo, M.; Jiang, M. Macromolecules, 2010, 43, 8086–8093. (c) Sholrollahi, P.; Mirzadeh, H.; Scherman, O. A.; Huck, W. T. S. J. Biomed. Mater. Res., Part A, 2010, 95A, 209– 221. (d) Matsumoto, S.; Yamaguchi, S.; Ueno, S.; Komatsu, H.; Ikeda, M.; Ishizuka, K.; Iko, Y.; Tabata, K. V.; Aoki, H.; Ito, S.; Noji, H.; Hamachi, I. Chem.—Eur. J., 2008, 14, 3977–3986.

12. Shimizu, L. S. Polym. Int., 2007, 56, 444–452.

13. Bouteiller, L. Adv. Polym. Sci., 2007, 207, 79–112.

14. Armstrong, G.; Buggy, M. J. Mater. Sci., 2005, 40, 547–559.

15. Binder, W. H.; Zirbs, R. Adv. Polym. Sci., 2007, 207, 1–78.

16. Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565–578.

17. Moughton, A. O.; O’Reilly, R. K. Macromol. Rapid Commun., 2010, 31, 37–52.

18. McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers; Rotello, V. M.; Thayumanavan, S., Eds.; John Wiley and Sons: Hoboken, NJ, 2008; Chapter 7, p 157.

19. Li, Z.; Djohari, H.; Dormidontova, E. E. J. Chem. Phys., 2010, 133, 184904-1-9.

20. Xu, D.; Craig, S. L. Macromolecules, 2011, 44, 5465–5472.

21. van Beek, D. J. M.; Spiering, A. J. H.; Peters, G. W. M.; te Nijenhuis, K.; Sijbesma, R. P. Macromolecules, 2007, 40, 8464–8475.

22. Sivakova, S.; Bohnsack, D. A.; Mackay, M. E.; Suwanmala, P.; Rowan, S. J. J. Am. Chem. Soc., 2005, 127, 18202–18211.

23. Elkins, C. L.; Park, T.; Mckee, M. G.; Long, T. E. J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4618–4631.

24. Herbst, F.; Schröter, K.; Gunkel, I.; Gröger, S.; Thurn-Albrecht, T.; Balbach, J.; Binder, W. H. Macromolecules, 2010, 43, 10006–10016.

25. Wietor, J.-L.; van Beek, D. J. M.; Peters, G. W.; Mendes, E.; Sijbesma, R. P. Macromolecules, 2011, 44, 1211–1219.

26. Woodward, P. J.; Merino, D. H.; Greenland, B.W.; Hamley, I.W.; Light, Z.; Slark, A. T.; Hayes, W. Macromolecules, 2010, 43, 2512–2517.

27. Xu, D.; Craig, S. L. J. Phys. Chem. Lett., 2010, 1, 1683–1686.

28. Vermonden, T.; van Steenbergen, J.; Besseling, N. A. M.; Marcelis, A. T. M.;Hennink, W. E.; Sudhölter, E. J. R.; Cohen Stuart, M. A. J. Am. Chem. Soc., 2004, 126, 15802–15808.

29. Serpe, M. J.; Craig, S. L. Langmuir, 2007, 23, 1626–1634.

30. Weng, W.; Li, Z.; Jamieson, A. M.; Rowan, S. J. Soft Matter, 2009, 5, 4647–4657.

31. Lee, J. H.; Lee, H.; Seo, S.; Jaworski, J.; Seo, M. L.; Kang, S.; Lee, J. Y.; Jung, J. H. New J. Chem., 2011, 35, 1054–1059.

32. Adarsh, N. N.;Dastidar, P. Cryst. Growth. Des., 2011, 11, 328–336.

33. (a) Zhao, Y.; Beck, J. B.; Rowan, S. J.; Jamieson, A. M. Macromolecules, 2004, 37, 3529–3531. (b) Batabyal, S. K.; Leong, W. L.; Vittal, J. J. Langmuir, 2010, 26, 7464– 7468. (c) Grondin, P.; Roubeau, O.; Castro, M.; Saadaoui, H.; Colin, A.; Clérac, R. Langmuir, 2010, 26, 5184–5195. (d) Weng, W.; Beck, J. B.; Jamieson, A. M.; Rowan, S. J. J. Am. Chem. Soc., 2006, 128, 1163–11672.

34. Beck, J. B.; Ineman, J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060–5068.

35. Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145–151.

36. Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334–337.

37. Ferry, J. D. Viscoelastic Properties of Polymers, 3rd ed.; John Wiley & Sons, Inc.: New York, 1980.

38. Larson, R. G. Constitutive Equations for Polymer Melts and Solutions; Butterworths: Boston, 1988.

39. Fluegel, A.;Varshneya, A. K.; Earl, D. A.; Seward, T. P.; Oksoy, D. Ceram. Trans., 2005, 170, 129–143.

40. Mes, T.; Smulders, M. M. J.; Palmans, A. R. A.; Meijer, E. W. Macromolecules, 2010, 43, 1981–1991.

41. Sim, H. G.; Ahn, K. H.; Lee, S. J. J. Non-Newtonian Fluid Mech., 2003, 112, 237– 250.

42. Ilavsky, J.; Jemian, P. R. J. Appl. Cyrstallogr., 2009, 42, 347.

CHAPTER 4

Thermo-, Photo-, and Chemo-Responsive Shape-

Memory Properties from Photocrosslinked

Metallosupramolecular Polymers

Adapted from: Kumpfer, J. R.; Rowan, S. J. J. Am. Chem. Soc., 2011, 133, 12866-12874.

4.1 INTRODUCTION

Equipped with a better understanding of how the metal-ion and polymer core can be used

to tailor the optical, mechanical, and structural properties of metallosupramolecular films, we aim to combine that knowledge and apply it to design a series of stimuli-responsive materials. As has been shown, the phase separation of the metal-ligand complexes act as physical crosslinks which are able to give the films elastomeric properties, yet these crosslinks are susceptible to various stimuli including ultraviolet light, elevated temperatures, and chemicals. If another set of permanent crosslinks were introduced to these films along with the noncovalent metal-ligand crosslinks, we should be able to create films which show shape-memory properties.

Materials are said to exhibit shape-memory properties if they are able to fix a temporary shape and recover back to their “remembered” permanent shape when exposed to an external stimuli.1- 4 These materials typically consist of two distinct types of

crosslinking: a reversible “crosslink” (usually in the form of a thermal transition such as

Tg, Tm, or clearing point of a liquid crystalline material) responsible for holding the

temporary shape and a nonreversible crosslink (which can be either a covalent or a

physical crosslink) used to fix the permanent shape. In a polymer that exhibits thermal

shape-memory properties, the material is heated above a temperature at which the

reversible crosslinks break and deformed (stretched, twisted, folded, etc…). Upon

cooling, the reversible crosslinks reform and lock in the “temporary” deformed shape.

The original shape is recovered by simply heating the sample again to release the

reversible crosslinks and the material returns to the entropically favored shape as a result

of releasing the entropic strain placed on the stretched polymer backbone. An example of

a thermal shape-memory polymer is shown in Figure 4.1a where a PMMA-PEG semi- interpenetrating network with shape-memory properties which was fixed into a temporary spiral shape recovers to its permanent strait shape when heated and held at 100 °C. These

types of materials have gained increasing interest as a result of their many possible

applications, ranging from biomedical5- 8 to textiles.9,10

Figure 4.1 a) Thermal shape recovery of a PMMA-PEG semi-interpenetrating network held at 100 °C11 and b) a typical one-way shape-memory cycle experiment.

While shape-memory effects brought about by direct heating12- 15 are the most

common, other stimuli such as light16-18 have become increasingly attractive as a

consequence of the ability to utilize reduced operating temperatures, be applied remotely,

and result in localized shape-memory effects.19 Generally, light-induced shape-memory is achieved either via photochemistry or through light-induced heating. For example,

Lendlein and co-workers used UV light to facilitate shape-memory effects by way of a photoreversible cycloaddition reaction.20 Using polymers that contained cinnamic groups

they were able to selectively crosslink the polymer via cycloaddition with UV light (λ >

260 nm) to fix a temporary shape and photocleave (λ < 260 nm) to recover the shape.

Along with light, remote shape-memory effects have been developed utilizing IR

irradiation21,22 or magnetic fields23,24 as the stimuli; however, polymers typically have

very low thermal conductivity which requires the use of fillers with a much larger

thermal conductivity to enhance heat transfer and facilitate the remote heating. These

systems also tend not to have localized effects, which can limit their usefulness.

Along with heat and light, shape-memory polymers have also been developed to

be chemo-responsive.25 Polyurethanes displaying shape-memory properties have been shown to return to their permanent states after immersion in water.26 This results from

plasticization of the polyurethanes, which decreases the polymers’ Tg to below that of

room temperature, upon which the samples can recover. Other solvents such as DMF,

ethyl acetate, and ethyl formate have been used with the same mechanism to show shape-

memory recovery of a thermosetting styrene-based resin.27

In this chapter, phase separated reversible supramolecular crosslinks, in the form

of metal-ligand coordination bonds,28 is the key component that holds the material in its

temporary shape. Using this motif we have developed a class of materials that not only

respond to heat, but also to light and chemicals such as methanol, acetone, and amines.

The use of light here is particularly attractive as it does not require additional fillers and

allows for remotely activated, localized shape-memory effects. There are very few

examples of shape-memory polymers that utilize noncovalent, supramolecular

interactions to “fix” the temporary shape. Those that have include polyelectrolyte phase

separation,29- 32 inclusion complexes,33 and hydrogen-bonded systems.34-36 Work done by

Anthamatten and co-workers37 used a covalently crosslinked network that also contained

the hydrogen-bonding moiety, 2-ureido-4[1H]-pyrimidone (UPy), as side groups. They

were able to utilize the thermally sensitive hydrogen-bonded UPy dimer to achieve

reversible fixing of the temporary shape. Similarly, Guan and co-workers38 synthesized a

linear polymer containing UPy groups within the polymer backbone. They found that

disruption of the hydrogen bonding with heating, and subsequent cooling, resulted in

formation of intermolecular dimers, which imparted shape-memory properties to the films.

There is a growing amount of interest in using reversible bonds (either covalent or noncovalent) to access macroscopically responsive materials.39-41 We and others42,43 have

recently shown that certain metal-ligand interactions are not only sensitive to heat but can

also be sensitive to certain chemicals (such as nerve gas agent mimics or amines),44,45 as

well as light46 through a photo-thermal conversion process. Thus, the inherent properties

of the metal-ligand complexes allow for them to act as dynamic, reversible crosslinks that

are able to absorb UV light and transfer the energy to localized heating without the need

for additional fillers. We were therefore interested in seeing if we could use such

reversible metal-ligand interactions in the fixing process in shape-memory polymers.

Furthermore, we were also interested in using light to form the permanent crosslinks.

This would then allow us to cast a film of a metallosupramolecular polymer, process it

into the desired form, and then use light to induce the covalent crosslinking and create the

permanent shape.

Our proposed approach is schematically shown in Figure 4.2 and involves placing

metal-chelating ligands at the ends of a low molecular weight crosslinkable

(poly(butadiene)) polymer core. The addition of a metal salt results in a

metallosupramolecular polymer where the metal-ligand complexes phase separate from

the polymer core in the solid state. These “hard” metal-ligand phases become the

reversible crosslinks in the shape memory polymer. Creation of the fixed crosslinks can

be achieved by the photocrosslinking of the poly(butadiene) core with a tetra-functional

thiol through a photoinduced thiol-ene reaction.47 The metal-ligand complexes are able to

absorb UV light, and this absorbed energy is converted into fluorescence and localized

heating. This localized heating results in softening of the hard phase and increased

decomplexation or rate of exchange of the metal-ligand complexes. Thus, the material can be deformed while exposed to UV light (or heat), and upon removal of the light the sample cools, the metal-ligand interactions re-engage, and the phase separation reforms, locking in the temporary deformed state. Further exposure to light will again break up the phase separation, and the stored strain is released and the material returns to its entropically favored, “remembered” shape. Similarly, exposure to chemicals or solvents that are able to plasticize the hard phase and/or competitively bind the metals should be

able to bring about shape-memory behavior.

Figure 4.2 Proposed mechanism of shape-memory behavior using light as a stimulus: (a) UV light is absorbed by the metal-ligand complexes and is converted to localized heat, which disrupts the phase separation; (b) the material can then be deformed; (c) removal of the light while the material is deformed allows the metal-ligand complexes to reform and lock in the temporary shape; and (d) additional exposure to and subsequent removal of UV light allows for a return to the permanent shape.

100

4.2 RESULTS AND DISCUSSION

4.2.1 Material Properties

In the previous chapters we have shown that low molecular weight polymers in which a

low Tg core has been end-capped with 4-hydroxy-2,6-bis(10- methylbenzimidazoyl)pyridine (HOMebip) are able to self-assemble into mechanically stable, elastomeric films when coordinated with transition metals,48 and/or

lanthanides.46,49 In this study, hydroxyl-terminated poly(butadiene) (HTPB) was chosen as the core polymer for its ability to be easily crosslinked through its 1,2-vinyl groups.

The HTPB was end-capped with HOMebip to give a ditopic macromonomer, 4, using the

Mitsunobu reaction (Figure 4.3). Following purification, the 1H NMR spectrum of

macromonomer 4 showed no evidence of the presence of residual hydroxyl groups that

would result from incomplete end-capping with the ligand. NMR was also used to determine the molecular weight of the macromonomer via end-group analysis, which gave Mn = 4300 g/mol. The molecular weight was further confirmed via MALDI-TOF

mass spectrometry and via monitoring the UV-vis signals while titrating in

45,50 Zn(ClO4)2.

101

Figure 4.3 a) Synthesis of metal-chelating ligand end-capped polybutadiene macromonomer (4) 1 via a Mitsunobu reaction and (b) 600 MHz H NMR of 4 in CDCl3.

Mechanically stable films were prepared by dissolving the macromonomer, the

desired amount of a tetrathiol crosslinker (pentaerythritol tetrakis(3-

mercaptopropionate)), a photoinitiator (Irgacure 819), and a stoichiometric amount of

metal salt to ensure full complexation (i.e., at ratios of 1:3 Eu3+:ROMebip or 1:2

Zn2+:ROMebip) and then solution cast. After drying, the films are photocrosslinked using

102

320-390 nm UV light, and any residual photoinitiator is removed by an extraction with

dichloromethane to yield the crosslinked metallosupramolecular polymer films, 4.1-

3∙Eu(NTf2)3 with a targeted 3, 8, and 14 crosslinks/chain, respectively, as shown in

Figure 4.4.

Figure 4.4 Schematic representation of photocrosslinking of films. Films containing different amounts of tetrathiol crosslinker and photoinitiator are solution cast and the dried films are photocrosslinked to yield 4.1-3∙Eu(NTf 2)3 with a targeted 3, 8, and 14 crosslinks/chain, respectively.

The ability to also use light as the means of covalently crosslinking opens up the

possibility of easily creating interesting and complex permanent shapes. An example of

this is shown in Figure 4.5, where an uncrosslinked metallosupramolecular film was set

into a permanent spiral shape by simply twisting around a substrate and irradiating with

UV light. The ability to form more complex permanent shapes via photocrosslinking

should allow for uses in a wider variety of applications and devices.

103

Figure 4.5 An advantage of photocrosslinking is that solution-cast uncrosslinked films can be

fixed into a variety of permanent shapes (a) a strip of uncrosslinked 4∙Eu(NTf2)3 film is wrapped around a cylinder to give a spiral shape and (b) irradiation with low intensity UV light (~70 mW/cm2) initiates photocrosslinking to yield a film with a permanent spiral shape.

Films of a metallosupramolecular polymer made with 4 and europium

bistriflimide (Eu(NTf2)3) were initially studied. We have shown previously that mechanically stable films can be formed with ditopic macromonomers based on poly(ethylene-co-butylene) (i.e., hydrogenated poly(butadiene)) using La(NTf2)3 as a

result of phase separation of the “hard” metal-ligand complex from the “soft” polymer core. Up to three terdentate ligands can bind to Eu3+, and upon binding the free ligand-

based emission (blue) is converted into the metal-based emission (red) of the Eu3+

complex.51 This occurs on account of the “antenna effect”, which is in effect a light

conversion process that occurs by absorption of the light by the ligand, followed by a

ligand-to-metal energy transfer resulting in the metal-ion based emission. Thus, this

“antenna effect” can be used to demonstrate whether the ligand is complexed to the Eu3+

ion (red emission) or is unbound (blue emission).44,45,52 Europium also has a weaker binding constant with terdentate ligands than transition metals, which makes the complexes more dynamic at lower temperatures.

Small angle X-ray scattering (SAXS) was used to probe the phase separation responsible for the physical, reversible crosslinks and to determine the films’

104

morphology. A representative 1D scattering plot is shown in Figure 4.6 for an

uncrosslinked film prepared with Eu(NTf2)3 containing enough tetrathiol crosslinker and

photoinitiator for a film targeting eight crosslinks per chain. A primary Bragg diffraction

peak (q*) at q = 0.72 corresponding to a distance of 8.70 nm is observed, which correlates

to the distance between the metal-ligand “hard” phases. This value matches closely with

previously published SAXS data on metallosupramolecular films of similar molecular weight. A secondary peak is also observed at q = 1.44 (d = 4.35 nm), one-half that of the primary peak (2q*), which is typical of a lamellar morphology that we have also seen in

the hydrogenated version of the material. The SAXS then tells us that the morphology of

the film before crosslinking is unchanged even while containing all of the components

required for crosslinking, and that the films have a lamellar morphology with alternating

metal-ligand and polymer phases.

Figure 4.6 (a) SAXS of an uncrosslinked film of 4∙Eu(NTf2)3 containing enough tetrathiol and photoinitiator to target eight crosslinks per chain displaying a primary peak at 8.7 nm and a secondary peak of 4.3 nm, indicative of a lamellar morphology. (b) Dynamic mechanical thermal analysis of films of the uncrosslinked 4∙Eu(NTf 2)3 and crosslinked 4.1-3∙Eu(NTf2)3. Increased

crosslinking results in a suppression of the Tg and mechanical enhancement above 40 °C at which the uncrosslinked film flows as a result of disruption of the phase separated morphology.

105

The mechanical properties of the photocrosslinked films 4.1-3∙Eu(NTf2)3 were investigated using dynamic mechanical thermal analysis (DMA). A control film of uncrosslinked 4∙Eu(NTf2)3 was studied along with films of varying crosslinking density.

In Figure 4.6b, the mechanical properties of the control film show two distinct transitions.

The first is the Tg of the poly(butadiene) core at ca. -60 °C, and the second occurs above

40 °C and relates to the disruption of the hard phase, which in turn causes the sample to

yield/flow similar to the behavior seen in Chapter 2 as Eu3+ was introduced. As covalent

crosslinks are introduced to the poly(butadiene) core, both transitions are depressed, and

above 50 °C the samples no longer flow and appear to approach an elastic plateau region

above 150 °C. In the lightly crosslinked film, a decrease in the storage modulus is still observed at Tg although not as much as the uncrosslinked sample. As the crosslinking

increases, the Tg is replaced by a very broad and weak transition. The maximum for this

transition is around 60 °C, after which the mechanical properties are predominantly a

result of the permanent crosslinking. Not surprisingly, in general the modulus also

increases with increasing crosslinking.

4.2.2 Thermally-Induced Shape-Memory Properties

To deduce the shape-memory characteristics of these films, standard thermal shape- memory experiments using DMA were carried out. The strain-fixing ratio (Rf) and the

strain-recovery ratio (Rr) are the most common values used to determine the efficiency of the shape-memory response.53 The strain-fixing ratio is given as

u Rf = × 100% (Eq. 4.1) εm ε

106

where εm is the maximum strain after equilibration and εu is the strain after unloading,

and the strain-recovery ratio is given as

- R = m r × 100% r - (Eq. 4.2) εm εp ε ε where εr is the recovered strain and εp is the initial, permanent strain, which is set as 0 in

the experiment. As the Eu3+:ROMebip complex has been shown to be responsive to temperatures above 50 °C, it stands to reason that these metallosupramolecular polymer

films should be able to thermally fix a temporary strain by heating above 50 °C, applying

a fixed force to strain the material, and cooling below 50 °C before removing the force.

Shape-memory would then be achieved by reheating the sample without any applied

force. To determine the validity of this, films 4.1-3∙Eu(NTf2)3 were studied with a

typical thermal one-way shape-memory cycle. The results are shown in Figure 4.7 for

4.1-3∙Eu(NTf2)3. The sample was heated to 100 °C, and a 0.5 N force was applied. For

the thermal shape-memory tests, forces above 0.5 N caused the samples to break when

held at elevated temperatures for extended time periods. After being cooled to 25 °C, the

load was removed, and the sample demonstrated fixing of the strained shape. Subsequent

heating to 100 °C recovers the majority of the strain, and cooling back to 25 °C recovers

the remaining strain that results from thermal expansion in the film. The initial strain

fixing values (Rf) are high for all films 4.1-3∙Eu(NTf2)3 (87%, 87%, and 90%,

respectively), and all show recovery values (Rr) above 90%, typically >97%.

107

Figure 4.7 Standard thermal one-way shape-memory cycle for films of a) 4.1, b) 4.2, and c) 4.3 with Eu(NTf2)3. The films are deformed at 100 °C to a set force and the temperature is reduced while the force is held constant. The force is removed and the strain fixing is determined. Heating the sample back up to 100 °C followed by cooling to room temperature allows the material to recover to its original shape.

4.2.3 Light-Induced Shape-Memory Properties

Having demonstrated the thermal shape-memory properties of these materials, we went on to investigate their photoresponsive properties where we utilize the ability of the metal-ligand complexes to absorb light and convert some of that energy to localized heating. An example of light-induced shape-memory behavior of a permanent shaped spiral of 4.1∙Eu(NTf2)3 is shown in Figure 4.8. The ability to trigger heating using UV

light is interesting as it allows for remote activation of the shape-memory properties

108

while also being able to control the response as the heating will be localized to only the areas exposed.

An example of the localized heating effect is shown in Figure 4.9. A

3+ 4.2∙Eu(NTf2)3 film displays the typical red Mebip:Eu complex metal-based emission

(Figure 4.9a) when observed under low intensity UV light (70 mW/cm2). When the left

side of the film is selectively irradiated with high intensity UV light (1000 mW/cm2)

(Figure 4.9b), the energy is converted into localized heating, which results in decomplexation of the Eu3+ that in turn converts the film to the blue, free Mebip

emission. The red emission (as observed under low intensity light) returns over a period

of a few seconds after removal of the high intensity light.

Figure 4.8 Images demonstrating the shape-memory behavior of a 4.1∙(EuNTf2)3 film. The film was set into a permanent spiral shape, (a) uncurled and irradiated with 320-390 nm UV light to fix the temporary shape and (b) reirradiated with 320-390 nm UV light to recover the permanent shape.

109

The shape-memory response using UV light as the stimulus was characterized by

placing the samples in a DMA instrument and irradiating them with light. Films 4.1-

3∙Eu(NTf2)3 were studied to elucidate the effect of crosslinking on the mechanical and

shape-memory properties. The light-induced shape-memory behavior can be seen in

Figure 4.9c-e for films of 4.1, 4.2, and 4.3 with Eu(NTf2)3 respectively. Controls were also performed without any exposure to UV light to act as a comparison. To prove that the mechanism for light-induced shape-memory behavior is a result of localized heating

from absorbance of 320-390 nm UV light by the metal-ligand complexes and not from

inadvertent heating from the light source, a 4.2∙Eu(NTf2)3 film was tested using the same

conditions with a filter that blocks UV light below 400 nm (see Figure 4.11a). By doing this, the sample is still irradiated by the light source, but not in wavelengths absorbed by the metal-ligand hard phase. The film shows strains, fixing, and recovery values very similar to the nonirradiated controls, showing that there are little to no thermal effects from the light source and that the observed shape-memory behavior is a result of a

photothermal conversion process.

110

Figure 4.9 Picture of a film of 4.2∙Eu(NTf2)3 in its permanent shape a) under low intensity (70 mW/cm2) UV light in which the typical red metal-based emission from the Eu3+ complex is seen and b) selectively irradiating the left side of the film with high intensity (1000 mW/cm2) UV light which yields the free ligand emission. Controlled force shape-memory experiment on c) 4.1, d)

4.2, and e) 4.3 with Eu(NTf2)3. The sample was strained to a set force of 1N while irradiated with light (UV light exposure highlighted in blue), allowed to equilibrate and the force was removed to determine the initial shape fixing. Additional exposure to light removes the stored strain and the sample returns to the permanent shape.

The Rf and Rr values for the different crosslink density films are shown in Figure

4.10. All samples exposed to UV light display fixing values of over 80% of the deformed

strain. This is a drastic improvement over the control films (which were strained and

allowed to recover without exposure to light at room temperature) where fixing never

111

exceeds 40%. The final recovered strain is >95% for both the irradiated samples and the

controls, but the response time is much different. Samples irradiated with UV light show

strain recovery in a matter of seconds, whereas the controls require approximately 1 h to

completely recover. These initial fixing and recovery values match well with those reported above for the thermally activated shape-memory behavior.

Figure 4.10 Strain fixing and recovery of films of 4.1-3∙Eu(NTf 2)3 which have different levels of permanent crosslink densities. Films exposed to light show greatly increased strain-fixing over the non-irradiated control films while the recovery remained consistent for all. The dynamic nature of the reversible crosslinks is responsible for the additional 10-15 % decrease in the strain fixing 1 hr. after the initial unloading. Rf, i is the initial fixing value, Rf, 1hr is the fixing after 1 hr.,

and Rr is the recovery.

While Rf is typically defined using the strain immediately after unloading, this

may not be the best way to report fixing values where the strain fixing is a result of a

supramolecular interaction. Metal-ligand interactions are dynamic in nature, which is a

result of constant complexation/decomplexation, even in the solid state. This can result in

a slow recovery of the strain after the initial unloading. This is observable in Figure 4.9c-

112

e between the unloading and recovery with light; however, it is apparent that the strain

recovery decreases with time. Thus, it may be more accurate to describe Rf in terms of a

fixing value after a given time period (εt) rather than the initial fixing after unloading.

This new fixing value was determined by monitoring the strain recovery after unloading

for 1 h, after which the strain was measured (ε1h) and used in the place of εu as

1h Rf 1h = × 100% (Eq. 4.3) ε m ε A comparison of the initial fixing values with the fixing after 1 h is also shown in Figure

4.10. After 1 h, the strain recovery of all films had slowed to less than 2.5 × 10-3 %/min,

and the strain fixing for all 4.1-3∙Eu(NTf2)3 films was still typically over 70% with the

films having the lowest initial fixing also having the most strain recovery.

Figure 4.10 also shows that while the recovery stays relatively constant with crosslink density, the fixing increases in the higher crosslinked films. To further probe this effect, creep experiments were performed on each sample by straining the films using a 1N set force without exposure to UV light and observing the strain increase over 30 min

(see Figure 4.11b). The rough creep rates were calculated as the increase in strain over 30

min under a 1N load. All of the materials show a small amount of creep with the highest creep rate, 4.6 × 10-2 %/min (vs ca. 3.2 × 10-3 %/min for the highest crosslinked

materials), being observed with the lowest crosslinked material. This is consistent with

the soft phase not being fully crosslinked, which allows the materials to relax more after

fixing.

113

Figure 4.11 a) Controlled force shape-memory experiment of a 4.2∙Eu(NTf 2)3 film using a filter to cutoff λ < 400 nm. There is very little strain fixing which demonstrates that the 320-390 nm UV light is predominantly responsible for the shape-memory properties as the light is converted into localized heating. b) Effect of crosslinking on the creep rates for films with different crosslink densities. The creep rate decreases with increased crosslinking which accounts for the higher strain fixing values.

The structural effects of straining, fixing, and recovering were also determined

using SAXS. Figure 4.12a shows the primary Bragg diffraction peak for a film of

4.1∙Eu(NTf2)3 in the permanent shape, after straining while irradiating with UV light to

fix the temporary shape, and recovery after further irradiation. The film in its permanent

shape has a peak corresponding to 8.6 nm (q = 0.73 nm-1) as mentioned earlier, which is

assigned to the distance between the metal-ligand “hard” phases. After straining the sample and fixing with light, the peak shifts to a smaller scattering vector (q = 0.64 nm-1), which relates to an increase in distance to 9.8 nm; however, this number is likely slightly exaggerated as some orientation was∼ observed in the 2D SAXS image for the strained film as shown in Figure 4.12b. None-the-less the distance between metal-ligand phases is increased after stretching and fixing the shape. Further exposure with light results in recovery to 8.7 nm, which is essentially the value for the permanent shape.

114

Thus, the SAXS confirms the changes in morphology during the shape-memory process

are consistent with the process outlined in Figure 4.2.

Figure 4.12 a) 1D SAXS of a film of 4.1∙Eu(NTf 2)3 in its permanent, strained, and recovered states. The peaks correspond to the distance between the “hard” metal-ligand phases and are 8.6 nm (permanent), 9.7 nm (strained) and 8.7 nm (recovered). b) 2D SAXS scattering pattern of the strained 4.1∙Eu(NTf 2)3 film which shows evidence of some orientation (slightly elongated top- to-bottom).

4.2.4 Effect of Different Metals and Counterions on the Light-Induced Shape-Memory

Behavior

In an effort to try to elucidate the role of the metal salt on the shape-memory properties,

films were made in which the metal and/or counterion were varied. All films studied

were prepared targeting 8 crosslinks per chain (4.2). Films of 4.2 with Zn(NTf2)2 and

Zn(OTf)2 (4.2∙Zn(NTf2)2 or 4.2∙Zn(OTf)2) were prepared in the same manner as the

Eu(NTf2)3 films, and their shape-memory behavior was studied to determine the

influence of these metal salts on Rf and Rr.

115

Figure 4.13 shows the results of the shape-memory testing on the two zinc films

(4.2∙Zn(NTf2)2 or 4.2∙Zn(OTf)2) as compared to 4.2∙Eu(NTf2)3. It is obvious that the

metal and counterion have little to no effect on the recovery (Rr), as recovery is a result

of the fixed phase (the crosslinked poly(butadiene)) restoring the stored strain. Changing

the metal-ion from Eu3+ to Zn2+ and keeping the counterion the same, however, results in

a slight decrease in the initial fixing of approximately 5% and an even bigger decrease in

the fixing value over 1 h (only ca. 60% fixing vs 77% for 4.2∙Eu(NTf2)3). This result is consistent with the inability to achieve full decomplexation of the Zn2+ from the ligand

during exposure to UV light, presumably as a result of more energy being required to disrupt the more strongly bound phase separated zinc complexes as compared to those with europium. Along the same lines, changing the counterion from bistriflimide to triflate results in a much more dramatic decrease in the fixing values when compared to the 4.2∙Eu(NTf2)3 film. There is approximately a 16% decrease in the initial fixing value

comparing the films made with Zn(OTf)2 to Zn(NTf2)2 and a 21% decrease when

compared to the Eu(NTf2)3 film. The fixing values are even more dramatic after 1 h, with

4.2∙Zn(OTf)2 exhibiting a fixing value of only 40%. This result implies that the

counterion has a large influence on the packing of the “hard” metal-ligand phase and that

by changing from a bulky bistriflamide to a triflate counterion, much higher temperatures

are necessary to disrupt the phase separation. Thus, for the given exposure time, only

partial disruption of the reversible phase occurs, and the metal-ligand complexes that are

not broken now act with the covalent crosslinks to recover the strain. This means that the

initial fixing value also reflects the recovery rate after 1 h where films with higher initial

fixing values have slower strain recovery with time.

116

Figure 4.13 Controlled force shape-memory experiment on a) 4.2∙Zn(NTf 2)2 and b)

4.2∙Zn(OTf)2. The sample was strained to a set force of 1N while irradiated with light (UV light exposure highlighted in blue), allowed to equilibrate and the force was removed to determine the initial shape fixing. Additional exposure to light removes the stored strain and the sample returns to the permanent shape. c) Strain fixing and recovery of films of 4.2 with different metal salts. Recovery is unaffected by the metal salt used, however the fixing value decreases when changing - - from europium to zinc and when the counterion is changed from NTf2 to OTf.

To further probe the effect the metal salts have on the properties of these

materials, the mechanical properties of uncrosslinked films of 4 with Eu(NTf2)3,

Zn(NTf2)2, and Zn(OTf)2 and crosslinked films of 4.2∙Zn(NTf2)2 and 4.2∙Zn(OTf)2

117

were explored. As shown in Figure 4.14b-c, the uncrosslinked films display very

different thermal behavior. At temperatures above Tg of the poly(butadiene) core, the

mechanical properties are primarily a result of the phase separated morphology between

the hard metal-ligand complexes and the soft polymer core. Once a temperature is

achieved that is sufficient to induce a breakup of the metal-ligand hard phase and an

increase in metal-ligand decomplexation rate, the material will flow. Figure 4.14 shows

that for the Eu(NTf2)3 film, the temperature required to disrupt the phase separation is

only slightly above room temperature, with the materials flowing above 40 °C.

Conversely, the Zn(OTf)2 films, while having slightly lower modulus values, are

mechanically stable above 150 °C, and thus much higher temperatures must be achieved

to disrupt the phase separation. The Zn(NTf2)2 film also requires temperature in excess

of 100 °C to achieve enough decomplexation in order to flow. This dramatic change in thermal stability of these metallosupramolecular films with different metal salts is probably related to the Tg (or Tm) and/or order-disorder transition of the resulting metal- ligand hard phase in the soft polymer matrix, although we have not been able to definitively identify such a thermal transition in differential scanning calorimetric (DSC) studies of these materials. Nevertheless, these results are consistent with the degree of fixing in the shape-memory materials being related to the thermal stability of the hard phase, which in turn is controlled by the different metals and counterions.

118

Figure 4.14 Dynamic mechanical thermal analysis of crosslinked films containing of 4.2 with a)

Zn(NTf2)2 and b) Zn(OTf)2. c) DMA of uncrosslinked films of 4 with Eu(NTf2)3, Zn(NTf2)2 and

Zn(OTf)2. An increasing amount of thermal energy is required to disrupt the metal-ligand hard – – phase going from europium to zinc and from NTf2 to OTf counterions.

4.2.5 Localized Shape-Memory Response

As mentioned earlier, light-activated systems are advantageous as they offer the

possibility of producing a localized shape-memory response because of the low thermal

conductivity of most polymers. To demonstrate this, we prepared a 1 cm3 cube from a

4.2∙Eu(NTf2)3 film. The permanent, “remembered” shape is a cross template, and by

folding each side and irradiating with UV light (320-390 nm), a 3D box temporary shape

119 was obtained as shown in Figure 4.15. Irradiating the top edge only allows the box top to open while the sides maintain their temporary shape. Figure 4.15 shows the result of selectively irradiating each of the vertices of the box, one at a time, to open each side of the box until the box has returned to its permanent cross template. The localization effect is only limited by the size of the UV light source and could be expected to work on much smaller scales, which would be useful for a variety of applications.

Figure 4.15 Demonstration of the localized shape-memory effect in a film of 4.2∙Eu(NTf2)3. Using UV light, films can be fixed into complex shapes such as a box. The sides of the box can then be selectively opened by irradiating at the apex (shown with arrows) to return to the permanent shape.

120

4.2.6 Chemo-Responsive Shape-Memory Effect

The data suggest that the thermal stability of the metal-ligand hard phase is critical to the

shape-memory properties of these films. Therefore, conceptually any chemical that can

impact the thermal stability of this hard phase (e.g., via plasticization or decomplexation

of the metal-ligand complex) could be used to induce shape-memory properties. Thus, we

examined the potential of a few solvents to induce recovery of the permanent shape of a

4.2∙Eu(NTf2)3 film. The film, which had a permanent rectangular shape, was initially

fixed using light, as described earlier, to bend the film in half. The temporarily fixed film

was suspended in a sealed container above the appropriate solvent. Figure 4.16 shows the

effect of a methanol atmosphere on the film. After only 2 min of exposure, the film

showed near full recovery back to the permanent shape. Films of 4.2∙Zn(NTf2)2 and

4.2∙Zn(OTf)2 also showed shape recovery when exposed to a methanol environment.

Taken together, these results are more consistent with the methanol plasticizing the

metal-ligand hard phase rather than a competitively binding process as the Bip:Zn2+

complexes are stable in the presence of alcohols.54 The film also showed some shape- memory behavior when exposed to other chemicals such as acetone and triethylamine; however, samples exposed to triethylamine display only partial shape-recovery and discoloration resulting from oxidation. As a result, we were able to demonstrate a third shape-memory stimulus for these films in the form of chemoresponsiveness.

121

Figure 4.16 Demonstration of chemo-responsive shape-memory behavior. A 4.2∙Eu(NTf2)3 film was fixed into a bent shape using UV light and suspended above methanol in a sealed container. Images were taken every 30 seconds to monitor the shape-memory behavior. The film recovers its permanent shape after only a couple of minutes because methanol affects the stability of the metal-ligand complexes which results in decreased phase separation that leads to recovery of the permanent shape.

122

4.3 CONCLUSIONS

Utilizing a metallosupramolecular polymer with a covalently crosslinkable core, shape- memory films have been made. We used a covalent photocrosslinking process that allows the possibility of accessing complex permanent shapes and a metal-ligand hard phase as the reversible phase used to fix the temporary shape. Therefore, any stimulus (thermo-, photo-, and chemo-) that results in softening of the hard phase and increased decomplexation or rate of exchange of the metal-ligand complexes can be used to access the shape-memory properties of these films. In addition, the low thermal conductivity of

the polymer allows light to selectively target a localized shape-memory response of the temporary shapes. Varying crosslink densities of the soft phase were studied, and all were found to have excellent shape-memory properties with initial strain-fixing values greater than 80% and strain-recovery over 95%. Finally, the use of different metal salts, which impact the properties of the hard phase, allows the fixing behavior of these materials to be tailored. Thus, we have demonstrated that a combination of dynamic metal-ligand complexes phase separated into a hard phase is a flexible platform with which to create shape-memory materials that have the ability to respond to multiple stimuli.

4.4 EXPERIMENTAL METHODS

4.4.1 Materials

All solvents and reagents were purchased from Aldrich Chemical Co. and used without further purification. The hydroxyl-terminated poly(butadiene) (HTPB) was purchased from Scientific Polymer Products, Inc. with a reported Mn of 3000 g/mol and 20% cis,

20% vinyl, and 60% trans composition. All solvents were distilled with suitable drying

123

agents, and spectroscopic grade chloroform was used to cast all films. Bistriflimide metal

salts were prepared according to the literature.55

4.4.2 Instruments

NMR spectra were recorded on a Varian 600 NMR spectrometer. Dynamic mechanical

thermal analysis experiments were performed using a TA Instruments DMAQ800 under

N2 with liquid N2 cooling and heated at a rate of 3 °C/min. UV-vis spectra were obtained on a Perkin-Elmer Lambda 800 UV-vis spectrometer. Titration experiments were performed in Quartz cuvettes scanning in the range of 250-400 nm with an integration time of 0.24 s. Molecular weights of the materials were measured by mass spectrometry on a Bruker AUTOFLEX III MALDI TOF/TOF mass spectrometer using dithranol as the matrix. The UV lamp used for photocrosslinking and shape-memory experiments was a

Bluepoint 4 Ecocure from Honle UV America Inc. All experiments were carried out with a 320-390 nm filter. Heat produced directly from the lamp was not observed to exceed 60

°C during all experiments. Small-angle X-ray scattering (SAXS) measurements were conducted using a Rigaku S-MAX 3000 SAXS system. Cu KR X-rays from a MicroMax-

002+ sealed tube source (λ = 0.154 nm) were collimated through three pinhole slits to yield a final spot size of 0.7 mm at the sample position. Two-dimensional (2D) SAXS data were collected using a Rigaku multiwire area detector with a circular active area of

133 mm and a spatial resolution of 1024 × 1024 pixels. The sample-to-detector distance and the scattering vector, q, were calibrated using a silver behenate (AgBe) standard with a characteristic (001) peak position at q = 1.076 nm-1. The calculated sample-to-detector distance was 1.5 m. Typical exposure times were 1 h.

124

4.4.3 Preparation of 4

In dried glassware, 4-hydroxy-2,6-bis(10-methylbenzimidazoyl)pyridine508 (HOMebip,

5.17 g, 14.5 mmol) and triphenylphosphine (TPP, 7.75 g, 29.5 mmol) were suspended in

75 mL of distilled THF under Ar. The suspension was cooled to 80 °C in an

isopropanol-dry ice bath with stirring. To this was slowly added 7.5 mL∼ of a 40 wt %

solution of diethyl azodicarboxylate in toluene (DEAD, 34 mmol), and the reaction

turned a clear, orange-brown color. In a separate flask, hydroxy-terminated

poly(butadiene) (HTPB, 15.5 g, 5.2 mmol) was dissolved in 50 mL of distilled THF. The

HTPB solution was added to the reaction flask at 80 °C with stirring. The reaction was

left stirring for 24 h while gradually warming to room temperature. The solvent was then

removed under vacuum to give a viscous oil. The product was redissolved in

dichloromethane and extracted with a 1 M NaOH solution to help to remove some of the

excess HOMebip, which crashes out of solution and is removed by filtration. The organic

fraction was washed three more times with NaOH solution, dried with Na2SO3, filtered,

stirred with decolorizing carbon to further remove the excess HOMebip, filtered, and the

solvent was removed under vacuum. The product was then purified twice by column

chromatography using CHCl3:MeOH, 100:0, 99:1, 98:2 to remove the remaining

HOMebip and diethyl ether to remove any triphenylphosphine oxide. To remove any

excess DEAD, the product was dissolved in 100 mL of dichloromethane and stirred

vigorously for 3 days with 300 mL of 2 M NaOH solution. After extracting the organic

layer, the solvent was removed under vacuum to give the macromonomer 4 (14.8 g,

63%). δH (600 MHz; CDCl3) 7.93 (s, 4H, Ar), 7.852 (d, 4H, Ar), 7.44 (d, 4H, Ar), 7.34

(m, 8H, Ar), 5.54 (m, 14H, CH2CHCH), 5.39, 5.36 (m, 110H, CH2CHCHCH2), 4.95 (m,

125

28H, CH2CH), 4.73 (d, 1.6H, cis-CHCH2O), 4.22 (d, 3.2H, trans-CHCH2O), 4.21 (s,

12H, CH3), 4.16 (m, 1.3H, vinyl-CH2CHCHCH2O), 2.06 (m, 14H, CH2CHCH2), 2.01

(m, 220H, CH2CHCHCH2), 1.42, 1.25, (m, 28H, CH2CH2CHCH2); δC (100 MHz;

CDCl3) 166.4, 151.3, 150.6, 142.9, 137.4, 131.5, 129.6, 128.6, 123.7, 123.0, 120.4,

1 114.4, 112.2, 110.1, 43.7, 38.4, 34.2, 32.9, 30.3, 27.6, 25.1.Mn = 4300 g/mol by H NMR and UV-vis titration with Zn(ClO4)2. m/z (MALDI TOF-TOF; matrix: dithranol) Mn =

3400. PDI: 1.03.

13 Figure 4.17 100 MHz C NMR of 4 in CDCl3.

126

Figure 4.18 MALDI-TOF of 4.

Figure 4.19 UV-vis titration of 4 with Zn(ClO4)2. Inset: Plot of absorbance at 341 nm vs the molar ratio of metal to 4 which shows that Zn2+ binds 1:1 with 4; evidence of complete end- capping.

127

4.4.4 Preparation of 4.1-3·Eu(NTf2)3

The materials were covered with aluminum foil for all steps to prevent the polymer from

photocrosslinking before it had been cast into a film. 293 μL of Eu(NTf2)3 in CH3CN

(100 mM) was added to a solution of 4 (200 mg in 5 mL of CHCl3), stirred for 5 min, and

the solvent removed under vacuum at 40 °C to avoid potential thermal crosslinking. In a

separate vial, the tetrathiol crosslinker (pentaerythritol tetrakis(3-mercaptopropionate))

and a photoinitiator (Iragacure 819) were dissolved in 4 mL of CHCl3. The crosslinking density was controlled by the amount of the tetrathiol crosslinker added. Films were targeted to have 3, 8, or 14 crosslinks per chain by adding 17 mg (0.035 mmol), 45.5 mg

(0.093 mmol), or 79.5 mg (0.162 mmol), respectively, of the tetrathiol crosslinker and 3.0 mg (0.009 mmol), 8.7 mg (0.02 mmol), or 15.2 mg (0.04 mmol), respectively, of photoinitiator. This solution is then used to dissolve the metallosupramolecular polymer

4∙Eu(NTf2)3. Once dissolved, the solution is cast into an aluminum walled casting dish

with a Teflon sheet bottom. The solvent was allowed to evaporate at room temperature

for 24 h to yield mechanically stable uncrosslinked films.

4.4.5 Photocrosslinking of the Films

The uncrosslinked solution-cast films (at different loading of the tetrathiol crosslinker)

were placed between two glass slides to avoid bending of the film during the crosslinking

process. All films were irradiated using 320-390 nm light at an intensity of 70 mW/cm2

for 1 h on each side to induce reactions of the 1,2 vinyl groups (predominantly) with the

thiols. The resulting crosslinked films have gel fractions of >95% and thicknesses of 220

± 20 μm. After the initial crosslinking, the films were exposed to higher intensity light

128

(900-1000 mW/cm2) to ensure maximum crosslinking and no unreacted thiol end-groups.

Any residual photoinitiator was removed by three extractions in dichloromethane. The

resulting films, 4.1-3∙Eu(NTf2)3, have targeted 3, 8, and 14 crosslinks/chain, respectively

as mentioned above.

4.4.6 Shape-Memory Experiments

Strips of the films with approximate dimensions of 10 mm × 5 mm × 0.2 mm were placed

in the DMTA. After the initial length was recorded, the films were stretched to a set force

of 1 N at 2 N/min. At the same time, the films were irradiated with UV light (320-390

nm, 900-1000 mW/cm2). After the films reached the set force (1 N), the light was

removed and the samples were allowed to equilibrate for 5 min, after which the force was

removed. After waiting to observe the relaxation behavior, the films were irradiated with

UV light. Experiments were performed a minimum of five times per sample to get the

average values for fixing and recovery.

4.5 ACKNOWLEDGMENTS

This material is based upon work supported by the National Science Foundation under

Grant nos. CHE-0704026 and MRI-0821515 (for the purchase of the MALDI-TOF/TOF) and by the Case School of Engineering and the Kent H. Smith Charitable Trust. Special thanks also go to Brian Michal for help with the SAXS experiments.

129

4.6 REFERENCES

1. Behl, M.; Razzaq, M. Y.; Lendlein, A. Adv. Mater., 2010, 22 3388–3410.

2. Lendlein, A.; Shastri, V. P. Adv. Mater., 2010, 22, 3344–3347.

3. Ratna, D.; Kocsis, J. K. J. Mater. Sci., 2008, 43, 254–269.

4. Liu, C.; Qin, H.; Mather, P. T. J. Mater. Chem., 2007, 17, 1543–1558.

5. Small, W., IV; Metzger, M. F.; Wilson, T. S.; Maitland, D. J. IEEE J. Sel. Top. Quantum Electron., 2005, 11, 892–901.

6. Sokolowski, W.; Metcalfe, A.; Hayashi, S.; Yahia, L.; Raymond, J. Biomed. Mater., 2007, 2, S23–S27.

7. Buckley, P. R.; McKinley, G. H.; Wilson, T. S.; Small, W., IV; Benett, W. J.; Bearinger, J. P.; McElfresh, M. W.; Maitland, D. J. IEEE Trans. Bio-Med. Electron., 2006, 53, 2075–2083.

8. Small, W., IV; Wilson, T. S.; Benett, W. J.; Loge, J. M.; Maitland, D. J. Opt. Express, 2005, 13, 8204–8213.

9. Meng, Q. H.; Hu, J. L.; Zhu, Y.; Lu, J.; Liu, Y. Smart Mater. Struct., 2007, 16, 1192– 1197.

10. Meng, Q. H.; Hu, J. L. Composites, Part A, 2009, 40, 1661–1672.

11. Liu, G.; Ding, X.; Cao, Y.; Zheng, Z; Peng, Y. Macromol. Rapid Commun., 2005, 26, 649-652.

12. Mather, P. T.; Luo, X.; Rousseau, I. A. Annu. Rev. Mater. Res., 2009, 39, 445–471.

13. Ahn, S.; Deshmukh, P.; Kasi, R. M. Macromolecules, 2010, 43, 7330–7340.

14. Zhou, J.; Schmidt, A. M.; Ritter, H. Macromolecules, 2010, 43, 939–942.

130

15. Ishida, K.; Yoshie, N. Macromolecules, 2008, 41, 4753–4757.

16. Jiang, H.; Kelch, S.; Lendlein, A. Adv. Mater., 2006, 18, 1471–1475.

17. Yu, Y.; Ikeda, T. Macromol. Chem. Phys., 2005, 206, 1705–1708.

18. Chen, M.; Xing, X.; Liu, Z.; Zhu, Y.; Liu, H.; Yu, Y.; Cheng, F. Appl. Phys. A: Mater. Sci. Process., 2010, 100, 39–43.

19. Leng, J.; Zhang, D.; Liu, Y.; Yu, K.; Lan, X. Appl. Phys. Lett., 2010, 96, 111905/1– 111905/3.

20. Lendlein, A.; Jiang, H.; Jünger, O.; Langer, R. Nature, 2005, 434, 879–882.

21. Leng, J.; Wu, X.; Liu, Y. J. Appl. Polym. Sci., 2009, 114, 2455–2460.

22. Koerner, H.; Price, G.; Pearce, N. A.; Alexander, M.; Vaia, R. A. Nat. Mater., 2004, 3, 115–120.

23. Weigel, T.; Mohr, R.; Lendlein, A. Smart Mater. Struct., 2009, 18, 1–9.

24. Yakacki, C. M.; Satarkar, N. S.; Gall, K.; Likos, R.; Hilt, J. Z. J. Appl. Polym. Sci., 2009, 112, 3166–3176.

25. Leng, J.; Lan, X.; Liu, Y.; Du, S. Prog. Mater. Sci., 2011, 56, 1077–1135.

26. Huang, W. M. Open Med. Devices J., 2010, 2, 11–19.

27. Lu, H.; Liu, Y.; Leng, J.; Du, S. Smart Mater. Struct., 2009, 18, 1–5.

28. (a) Paulusse, J. M. J.; van Beek, D. J. M.; Sijbesma, R. P. J. Am. Chem. Soc., 2004, 126, 15802–15808. (b) Yount, W. C.; Loveless, D. M.; Craig, S. L. J. Am. Chem. Soc., 2005, 127, 14488–14496. (c) Kim, H.-J.; Lee, J.-H.; Lee, M. Angew. Chem., Int. Ed., 2005, 44, 5810–5814. (d) South, C. R.; Pinon, V.; Weck, M. Angew. Chem., Int. Ed., 2008, 47, 1425–1428. (e) Xu, D. H.; Liu, C. Y.; Craig, S. L. Macromolecules,

131

2011, 44, 2343–2353. (f) Nair, K.; Breedveld, V.; Weck, M. Macromolecules, 2011, 44, 3346–3357.

29. Han, S.-I.; Gu, B. H.; Nam, K. H.; Im, S. J.; Kim, S. C.; Im, S. S. Polymer, 2007, 48, 1830–1834.

30. Guan, Y.; Cao, Y.; Peng, Y.; Xu, J.; Chen, A. S. C. Chem. Commun., 2001, 1694– 1695.

31. Weiss, R. A.; Izzo, E.; Mandelbaum, S. Macromolecules, 2008, 41, 2978–2980.

32. Merline, J. D.; Nair, C. P. R.; Gouri, C.; Shrisudha, T.; Ninan, K. N. J. Mater. Sci., 2007, 42, 5897–5902.

33. Zhang, S.; Yu, Z.; Govender, T.; Luo, H.; Li, B. Polymer, 2008, 49, 3205–3210.

34. Chen, S.; Hu, J.; Zhuo, H.; Yuen, C.; Chan, L. Polymer, 2010, 51, 240–248.

35. Chen, S.; Hu, J.; Liu, Y.; Liem, H.; Zhu, Y.; Liu, Y. J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 444–454.

36. Chen, S.; Cao, Q.; Jing, B.; Cai, Y.; Liu, P.; Hu, J. J. Appl. Polym. Sci., 2006, 102, 5224–5231.

37. Li, J.; Viveros, A.; Wrue, M. H.; Anthamatten, M. Adv. Mater., 2007, 19, 2851–2855.

38. Kushner, A. M.; Vossler, J. D.; Williams, G. A.; Guan, Z. J. Am. Chem. Soc., 2009, 131, 8766–8768.

39. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

40. Wojtecki, R. J.; Meador, M. A.; Rowan, S. J. Nat. Mater., 2011, 10, 14–27.

41. Hsu, L.; Weder, C.; Rowan, S. J. J. Mater. Chem., 2011, 21, 2812–2822.

132

42. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176–188.

43. Dobrawa, R.; Würthner, F. J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4981– 4995.

44. Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145–151.

45. Knapton, D.; Burnworth, M.; Rowan, S. J.; Weder, C. Angew. Chem., Int. Ed., 2006, 45, 5825–5829.

46. Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334–337.

47. Kade, M. J.; Burke, D. J.; Hawker, C. J. J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 743–750.

48. Beck, J. B.; Rowan, S. J. Faraday Discuss., 2005, 128, 43–53.

49. Kumpfer, J. R.; Wie, J. J.; Swanson, J. P.; Beyer, F. L.; Mackay, M. M.; Rowan, S. J. Macromolecules, 2012, 45, 473-480.

50. Beck, J. B.; Ineman, J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060–5068.

51. (a) Lis, S. J. Alloys Compd., 2002, 341, 45–50. (b) Sabbatini, N.; Guardigli, M.; Lehn, J.-M. Coord. Chem. Rev., 1993, 123, 201–228.

52. Petoud, S.; Bünzli, J.-C. G.; Glanzman, T.; Piguet, C.; Xiang, Q.; Thummel, P. J. Lumin., 1999, 82, 69–79.

53. Lendlein, A.; Kelch, S. Angew. Chem., Int. Ed., 2002, 41, 2034–2057.

54. Weng, W.; Li, Z.; Jamieson, A. M.; Rowan, S. J. Macromolecules, 2009, 42, 236– 246.

133

55. Nie, J.; Kobayashi, H.; Sonoda, T. Catal. Today, 1997, 36, 81–84.

134

CHAPTER 5

Vapochromic and Piezochromic Films From Platinum

Complex Polymer Blends

5.1 INTRODUCTION

The combination of reversible metal-ligand interactions with polymeric materials1- 6 has

gained increasing interest as a way to access an interesting class of stimuli-responsive

materials,7- 11 which have shown promise as sensors,12- 15 re-healable materials,16 shape- memory materials,17 and light-emitting diodes.18,19 Recently, square-planer platinum(II) coordination complexes have gained a great amount of attention, largely on account of their optoelectronic properties.20- 27 Particular interest has been paid to their electronic

absorption and photoluminescent properties,28- 32 which have been reported to be directly

related to the intermolecular interactions between adjacent platinum atoms.33-35 These Pt

– Pt interactions can be strongly influenced by the coordinating ligand and its substituents

as well as the counterion used, which can allow for tuning of the spectroscopic

properties.36- 39 Studies involving the use of platinum complexes with either terpyridine40-

44 or 2,6-bis(benzimidazol-2ʹ-yl)pyridine45,46 have gained interest largely as a consequence of their tendency to exhibit strong photoluminescence originating from

lowest excited states having significant ligand charge-transfer and/or metal-to-ligand charge-transfer character. The absorption and emission properties of these complexes typically are quite sensitive to external stimuli in solution and as solids.

An interesting property of certain square-planer platinum complexes is the

tendency to change color and/or photoluminescence when exposed to volatile organic

compounds (VOCs).47- 52 For example, it has been shown that salts of Pt(MeBip)Cl+

(Mebip = 2,6-bis(N-methylbenzimidazol-2ʹ-yl)pyridine) are vapochromic and that simply

changing the counter-anion provides a means of modulating the spectroscopic properties

and vapor selectivity.53,54 Studies on the single-crystal structures before and after vapor

136

exposure suggest that vapor sorption results in a rearrangement of the complexes such

that there is a decrease in length of the Pt – Pt interactions. In a detailed study, Sun and

co-workers55 investigated the vapochromic behavior and subsequent selective response to

VOCs of platinum(II) complexes of 2,6-bis(N-alkyl benzimidazol-2ʹ-yl)pyridine ligands with varying N-alkyl chain lengths and counterions. They observed that both the alkyl chain length and the counterion influence the vapochromic behavior and speculated that variations in these components influence the stabilities of both the vapor-absorbed and non-absorbed materials, as well as barriers to their interconversion.

Another type of stimuli-responsiveness which can result in color and emission changes is mechanochromism,56- 60 and selected platinum(II) complexes have also shown

this interesting behavior.61- 65 Grinding or rubbing (crystalline) solids of these complexes results in strong red-shifted emissions and analogous color changes. The mechanism behind the mechanochromism has been shown to be related to a force-induced structural rearrangement, which result in an increased number of stronger intermolecular Pt – Pt interactions. These materials could be interesting as piezochromic sensors; however they are limited by the poor mechanical properties of the crystalline solids.

While the majority of studies performed on square-planer platinum containing complexes have focused on the spectroscopic properties in solution or in the crystalline form, less attention has been focused on incorporating them into polymer matrices.66- 69

However, it is well known that the optoelectronic properties of metal complexes can be

impacted by their micro-environment, as in the case of vapochromic materials composed

of luminescent platinum(II) complexes intercalated into zirconium phosphate layered

materials.70

137

Nafion films imbedded with platinum(II) complexes (2,2ʹ:6ʹ,2ʺ:6ʺ,2ʹʺ-

quaterpyridine or 4ʹ-(4-methoxy-phenyl)-2,2ʹ:6ʹ,2ʺ-terpyridine) have been studied as singlet oxygen photosensitizers,69-72 however reports of their spectroscopic properties are

scarce. In one example which investigates the spectroscopic properties, it was found that incorporation of a non-emissive Pt(2,2ʹ:6ʹ,2ʺ:6ʺ,2ʹʺ-quaterpyridine)2+ complex into inverse micelles of water swollen Nafion produced strongly luminescent materials resulting from increased concentration and intermolecular interactions of the complexes

within the micelles. In related studies of platinum complexes with a cyclometalated 4,6- diphenyl-2,2ʹ-bipyridine ligand immobilized in Nafion or covalently attached to silica

supports, Che et al. demonstrated that the spectroscopic response to guest solvent/vapor

molecules was dependent on the nature of the support.73 Notably, the Nafion films

exhibited a solvatochromatic shift in ethanol solution but not aprotic solvents, whereas

the silica substrates were responsive to pentane vapor but not ethanol. These earlier

studies clearly demonstrate that the colorimetric and luminescence properties of platinum

complexes are very sensitive to the surrounding microenvironment. While the majority of

studies on stimuli-responsive square-planer platinum containing complexes have focused on the vapochromic behavior, the ability of select platinum complexes to exhibit mechanochromic responsiveness61,63,65 has been much less studied and little attention was

given to the impact of mechanical manipulation on the spectroscopic properties of the

supported metal complexes in polymers.

In this work, platinum complexes with 4-dodecyloxy-2,6-bis(N- methylbenzimidazoyl)pyridine (5) were prepared and dispersed within a series of

methacrylate polymers with the goal of creating mechanically stable films which exhibit

138

stimuli-responsive behavior. We report the spectroscopic properties and vapochromic

response of these materials. We also investigate the influence of the polymer matrix on

the absorption, emission, and reversibility of the response and show that these

polymer/platinum complex blends display interesting mechanochromic and

mechanoluminescent behaviors when scratched, compressed, or stretched.

5.2 RESULTS AND DISCUSSION

Platinum complexes of the ligand 4-dodecyloxy-2,6-bis(N-methylbenzimidazoyl)pyridine

(5) were targeted as the spectroscopic and vapochromic properties of the Pt(Mebip)Cl+

53,54 - are well known. It has been established that crystals of the PF6 salt of this complex produces a strong color change from yellow to red upon exposure to acetonitrile. A ligand with a dodecyloxy substituent was used to aid the solubility of the ligand and complex and dispersibility of the complex in the polymer matrices. Compound 5 was

synthesized using typical Williamson ether conditions to couple 1-bromododecane with

4-hydroxy-2,6-bis(N-methylbenzimidazoyl)pyridine74 as shown in Scheme 5.1a. The

purified 4-dodecyloxy-2,6-bis(N-methylbenzimidazoyl)pyridine ligand was fully

characterized using NMR (see Figure 5.1b (1H) and 5.12 (13C)) and mass spectroscopy

(Figure 5.13). The platinum complex ([Pt(5)Cl](Cl)) was prepared using a modification

of a literature procedure where the ligand, 5, and Pt(DMSO)2Cl2 were allowed to react in

a 3:1 CHCl3:MeOH solution at 50 °C under inert atmosphere for one day. Anion

metathesis was accomplished by addition of a saturated aqueous NH4PH6 solution. NMR

of the purified product showed no evidence of uncomplexed species (Figure 5.1c (1H)

and 5.14 (13C)) and gratifyingly crystals of this compound are yellow and exhibit

139 vapochromic behavior, turning red in the presence of acetonitrile (MeCN), as has been shown for similar complexes.35,46

1 Figure 5.1 a) Synthesis of platinum complex [Pt(5)Cl](PF6) and 600 MHz H NMR spectra of b)

5 in CDCl3 and c) [Pt(5)Cl](PF6) in CD3CN.

A series of polymethacrylates were selected in order to determine the effect of the matrix on the responsiveness of the Pt(II) complex blends. The polymers were poly(methyl methacrylate) (PMMA), poly(ethyl methacrylate) (PEMA), poly(butyl-co- isobutyl methacrylate) (PBcIBMA), and poly(butyl methacrylate) (PBMA), which have glass transition temperatures (Tgs) of 90, 63, 35, and 15 °C, respectively. The blends were prepared by adding an appropriate amount of solid [Pt(5)Cl](PF6) to a chloroform solution of the polymer and sonicating for 16 hrs. The resulting dispersion was then solution cast and dried to yield mechanically stable yellow [Pt(5)Cl](PF6) blended

140

polymer films. Gratifyingly, exposure of these films to either acetonitrile liquid or vapor

resulted in the dramatic change in color from yellow to orange/red (Figure 5.2a shows the

color change of the 10% w/w [Pt(5)Cl](PF6) in PMMA). Polarized optical microscopy

(POM) and fluorescent microscopy of the 2.5 and 10% w/w [Pt(5)Cl](PF6) films show

that the [Pt(5)Cl](PF6) complexes form small crystallites (ca. 1 µm) which are well

dispersed within the polymer matrix (Figure 5.2b-e). Thus the mechanism for the vapochromic response in these films is expected to be similar to those proposed previously for the crystalline [Pt(ligand)Cl](PF6) solids, in which the solvent vapor is

able to diffuse into, and alter the packing of adjacent complexes, resulting in shorter Pt –

Pt interactions.70 It should be noted that at loadings above 20% by weight a significant

decrease in mechanical stability was observed.

141

Figure 5.2 a) Images showing the vapochromic response of a 10% w/w [Pt(5)Cl](PF6) containing blend in PMMA. The yellow, as cast film changes color to red/orange after exposure to acetonitrile liquid or vapor. Polarized optical microscopy images of drop cast films in PMMA containing b) 2.5% w/w [Pt(5)Cl](PF6) and c) 10% w/w [Pt(5)Cl](PF6) which show well- dispersed micron-sized crystallites. Fluorescent microscopy shows that the fluorescence

originates from the [Pt(5)Cl](PF6) crystals in the d) 2.5% w/w and e) 10% w/w films.

5.2.1 Spectroscopic Studies

Solid-state UV-visible absorption spectroscopy was used to study the vapochromic

response of the blends prepared with 10% w/w of [Pt(5)Cl](PF6) in the four different

polymer matrices (PMMA, PEMA, PBcICMA, and PBMA). All the blends yielded

yellow films upon solution casting from chloroform. The films were cast onto glass slides

for all the UV-vis measurements. As shown in Figures 5.3, the as cast, yellow films

display absorption spectra similar to those previously reported for the Pt complexes with

142

45,55 related ligands, including an intense band (λmax = 342 nm) corresponding to a π – π*

transition of 5 and a weak shoulder at ~435 nm attributed to a metal-to-ligand (π*)

charge-transfer (MLCT) transition. During exposure to acetonitrile vapor, a new

structured band with maxima at 510 and 538 nm appears in the electronic spectrum,

accompanied by a decrease in intensity in the π – π* region. The new band results from a

metal-to-metal-to-ligand charge transfer (MMLCT) transition as a consequence of

increased Pt – Pt interactions, where the HOMO derives in large part from the

antibonding combination of the 5dz2(Pt) orbitals of adjacent complexes and the LUMO has predominantly π*(L) character.

Figure 5.3 UV-vis absorption spectra of a 10% w/w [Pt(5)Cl](PF6) containing blend in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA. Exposure to acetonitrile vapor results in

MMLCT band appearing at 510 and 538 nm. The recovery of the response can be tuned by the Tg

of the polymer matrix where PBMA films (Tg = 15 °C) show near full recovery after 24 hrs

whereas PMMA (Tg = 90 °C) requires several days.

143

Figure 5.3 shows the changes in the electronic spectra of thin films of the 10% w/w PMMA (Tg = 90 °C), PEMA (Tg = 63 °C), PBcIBMA (Tg = 35 °C), and PBMA (Tg

= 15 °C) blends upon exposure to acetonitrile. After acetonitrile exposure all films

exhibit the appearance of very similar λmax and relative intensities of the MMLCT

transition peaks. These results show that the solvent vapor is able to penetrate the

crystallites and induce a structural change with stronger Pt – Pt interactions. While the

spectroscopic response and rate of vapor absorption is similar for all blends, the color

recovery appears to be related to the Tg of the polymer matrix. Blends prepared with

matrices with Tgs above room temperature (PMMA, PEMA, and PBcIBMA) show only a slight reversal when left at room temperature in the absence of acetonitrile vapor for 24

hrs; the films still appear red/orange in color and the UV-vis spectra only show a slight decrease in the MMLCT peaks. The films retain their red color even after being left for several days at room temperature, and only recover to a light orange color after a month.

However, blends prepared with PBMA, which has a Tg below room temperature, display

nearly full recovery to the yellow film 24 hrs after being exposed to MeCN, as assessed

by UV-vis absorption spectroscopy. Notably, the electronic spectrum shows no evidence

of the MMLCT band after 24 hrs, and the MLCT and π – π* features have nearly returned

to their original intensities before exposure (Fig. 5.3d). Thus the recovery of thin films

(~50 µm) of these vapochromic blends can be tailored simply by changing the Tg of the polymer matrix. In thicker films (ca. 200 µm), the color change is still evident after one month in all the samples, with the higher Tg polymer blends (PMMA and PEMA) still

appearing red and the low Tg blends (PBcIBMA and PBMA) appearing orange.

144

Figure 5.4 Absorption change of a 10% w/w [Pt(5)Cl](PF6) blend in PEMA upon CH3CN exposure showing no change after 1 min, partial recovery after 24 hrs, and full recovery after heating to 100 °C.

Figure 5.4 shows the change in the UV spectra of the 10% w/w PEMA film over time. It was found that after only one minute of exposure the films display the typical

MMLCT peaks between 500 and 550 nm and further exposure did not significantly change the absorption. As mentioned, after leaving the film for first 24 hrs at room temperature only a slight recovery of the absorbance is observed, however heating the films to elevated temperatures results in full recovery. This behavior contrasts sharply with that of the crystalline-solid platinum complexes, which typically recover the original color within minutes to hours. Thus, the ability to tune the response time and even retain the exposed color over long time periods, in combination with the potential to erase the response when necessary by exposing the film to elevated temperature suggests that these vapochromic blends offer greater flexibility in sensor applications than salts of the platinum complexes alone.

145

The exposure of the [Pt(5)Cl](PF6) containing blends to acetonitrile also results in

significant changes to the emission as shown in Figure 5.5a. The as cast films display a weakly structured broad emission band maximizing near 575 nm with a full-width at half maximum (FWHM) of ~3050 cm-1. For each film, upon exposure to acetonitrile vapor,

the emission maximum shifts to 600 nm, and the intensity increases dramatically,

reaching 3-5x that of the intensity of the unexposed film. The emission band becomes

increasingly symmetric, loses evidence of structure, and narrows to give FWHM values

of ~1800 cm-1. The overall behavior is consistent with the vapochromic response and electronic structures of Pt(Mebip)Cl+ salts and related compounds,47,55,56,75 and

accordingly, the emission is assigned as originating from a lowest 3MMLCT state. The

resulting emission profiles are very similar for the four different blends, indicating that

the emissive vapor-exposed platinum crystallites have essentially the same architectures

for this selection of polymers. Interestingly, when exposed to an acetonitrile-saturated air atmosphere, the change in color of the films from yellow to red appears complete by eye within 3 mins and there is no noticeable difference in the absorbance spectra after only 1 min of exposure (see Figure 5.4). By contrast, emission spectra recorded at 30 s intervals indicate that equilibrium is reached in approximately 15 mins and that only ~30% of the

total emission change occurs within the first 3 mins. Thus, while the absorption and

emission changes are consistent with low-lying MMLCT states resulting from vapor-

induced structural changes within crystallites, the timescales of these changes are

incommensurate. The underlying reasons for this behavior are not yet fully understood.

146

Figure 5.5 Images showing the luminescence from a 10% w/w [Pt(5)Cl](PF6) in PMMA film under UV light (365 nm) a) before and b) after exposure to acetonitrile vapor. c) Emission spectra

(λex = 377 nm) recorded during exposure to acetonitrile vapor. Spectra were recorded at 30 s intervals over a 900 s period. Features at 377 and 754 nm are instrumental artifacts.

147

Figure 5.6 Normalized emission spectra of the vapoluminescent behavior of 10% w/w

[Pt(5)Cl](PF6) blend in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA (λex = 377 nm) and the recovery with time and/or temperature.

Heating the films to drive off the solvent returns the film color to a visibly yellow

color (Figure 5.4), however, while the absorption spectra appear to fully recover, the

emission profile does not when samples are heated above 100 °C as shown in Figure 5.6

and 5.7. After heating the samples to 100 °C for 5 mins and then cooling to room

temperature, the λmax of the emission profile is essentially identical to that of the original

vapor-exposed material as shown in Figure 5.6. This behavior is consistent with the

presence of a small number of defect sites within the yellow crystallites that still give rise to the characteristic MMLCT emission, even after heating. As was demonstrated in

Figure 5.5, the red, MMLCT emission is much more intense than the yellow emission.

Thus, even the presence of a small amount of defects could be responsible for the

148

apparent irreversible emission response as the spectrum is dominated by the intensity of

the MMLCT emission. Figure 5.7 shows the unnormalized spectra of 10% w/w

[Pt(5)Cl](PF6) blend in PMMA after exposure to MeCN at different temperatures which

highlights the difference in emission intensities of the film. The key observation is the

recovery of some of the red peak emission upon cooling down to room temperature from

150 °C. This observation, combined with the fact that in the absorbance there is

essentially no MMLCT bands present after heating (Figure 5.4), suggests that the number

of the MMLCT emitting defects is small, yet their strong emission intensity prevents a

full recovery in λmax to that of the yellow film.

Figure 5.7 Emission change of a 10% w/w [Pt(5)Cl](PF6) blend in PMMA upon heating a

CH3CN exposed film showing a dramatic decrease in intensity and blue-shift in λmax when measured at 150 °C. Cooling the film to room temperature yields an emission profile with the

same λmax as the exposed film but with decreased intensity, consistent with a small number of complexes displaying MMLCT emission.

149

5.2.2 Structural Properties

Previous studies52,53 have shown that exposure to solvent vapor alters the packing of

[Pt(ligand)](X) crystallites from alternating “long” and “short” Pt – Pt distances (ca. 5.27

and 3.38 Å) for the yellow crystal form and only the “short” Pt – Pt distance in the red

crystallites. To determine if a similar mechanism underlies the vapochromic response of

these blends, we carried out wide angle X-ray scattering (WAXS) studies.

Figure 5.8 WAXS of a) 2.5%, b) 5%, c) 10%, and d) 20% w/w [Pt(5)Cl](PF6) in PMMA. At 2.5% w/w loading the scattering from the platinum complexes is mostly hidden by that of the polymer matrix, however as the loading increases peaks at 2θ = 15.5 and 26.5 ° (d = 5.71 and 3.36 Å respectively) can be seen. These are consistent with alternating long and short Pt – Pt distances.

Each of the polymethacrylate controls consisting of the polymer matrix without

[Pt(5)Cl](PF6) display typical broad peaks characteristic of the amorphous polymer.

150

WAXS diffractograms were recorded of the blends at loadings ranging from 2.5 to 20% by weight of [Pt(5)Cl](PF6). Those data establish that variations in loading only impact the intensities of the peaks associated with the embedded crystallites (see Figure 5.8).

Figure 5.9 shows the WAXS patterns for the neat matrix, unexposed films, and exposed films prepared with 10% w/w [Pt(5)Cl](PF6) in the different polymer matrices. The as cast yellow films display several new peaks over the matrix alone. The dominant feature occurs at 2θ = 26.4° which corresponds to a distance of 3.38 Å. A second, slightly weaker peak appears at 2θ = 15.5° (d = 5.72 Å). These values are consistent with the presence of two different Pt – Pt distances in the yellow films which match well with previously reported long and short Pt – Pt distances found in yellow crystals of similar square-planer complexes.53,75 Thus the WAXS data are consistent with the presence of small crystallites, in which the Pt(1)Cl+ complexes pack as interacting dimers with alternating long and short Pt – Pt distances of 5.72 Å and 3.38 Å respectively.

WAXS data acquired immediately after exposing the film to a saturated acetonitrile vapor environment for 15 mins (Figure 5.9) show only one major peak at 2θ

= 26.3°, corresponding to a Pt – Pt distance of 3.39 Å, with no evidence of the longer Pt –

Pt distance. Therefore, the data are consistent with a change in the packing of the

Pt(1)Cl+ complexes upon exposure to acetonitrile vapor to yield a chain of complexes with short Pt – Pt contacts that are ultimately responsible for the red color. For both the unexposed and exposed films a peak also appears at low 2θ (5.5 and 6.4 °; d = 19.6 and

13.8 Å respectively). It is thought that these features arise from longer Pt – Pt distances, such as those between two complexes separated by 1-3 other complexes.

151

Figure 5.9 WAXS of 10% w/w [Pt(5)Cl](PF6) in a) PMMA, b) PEMA, c) PBcIBMA, and d) PBMA. The scattering for the neat matrix is shown (left) and for the as cast, yellow film (middle) which shows two dominant peaks at 2θ = 15.5 and 26.4 ° (d = 5.72 and 3.38 Å respectively) consistent of weak Pt – Pt interactions with a stepwise “long” and “short” pattern. After exposure to acetonitrile (right), only the “short” Pt – Pt interaction is observed (d = 3.39 Å) as a result of rearrangement of the crystals with increased Pt – Pt interactions. The peak at low 2θ is tentatively assigned to longer Pt – Pt distances.

152

5.2.3 Mechanochromic Behavior

Interestingly, the platinum complexes reported here, [Pt(5)Cl](PF6), were also found to show mechanochromic behavior. Further investigation into the mechanochromic behavior revealed that both the powder form of [Pt(5)Cl](PF6) as well as the polymer

blends containing [Pt(5)Cl](PF6) display such a response. An example of the

mechanochromic behavior of [Pt(5)Cl](PF6) on filter paper can be seen in Figures 5.10

where upon mechanical agitation (scraping) the yellow solid turns orange in color.

Furthermore, along with the visible color change, a more striking change in the emission

is observed where the mechanically stimulated area has a noticeable red-shift in the emission from yellow to bright orange. The normalized photoluminescence of

[Pt(5)Cl](PF6) before and after scraping is shown in Figure 5.10e. Interestingly the emission of the scraped [Pt(5)Cl](PF6) behaves in a similar fashion to the blended films which had been exposed to solvent vapor with a noticeable red-shift in λmax, a slight

increase in intensity, and an increase in the symmetry of the band. This is consistent with

stronger Pt – Pt interactions which result in emission from a lowest MMLCT state,

similar to that observed by Chen and coworkers.61

153

Figure 5.10 Images of [Pt(5)Cl](PF6) on filter paper under a) ambient light and b) UV light (λex = 365 nm). Upon scraping, a mechanochromic response is evident in both c) a visible color change and d) a change in the photoluminescence (image taken under λ = 365 nm light). e)

Emission spectra of [Pt(5)Cl](PF6) on filter paper before and after scrapping showing that the mechanophotoluminescent behavior is similar to the observed vapophotoluminescent response with a red-shift in λmax and increasing symmetry of the emission band.

The fluorescence of the [Pt(5)Cl](PF6) containing blends also display mechanochromic properties. Scratching any of the films containing [Pt(5)Cl](PF6) results in a change in the emission color from yellow to orange at the site of abrasion, as shown in Figure 5.11a for a 10% w/w [Pt(5)Cl](PF6) in PMMA film. The effect is most apparent

154

when the film is acutely deformed, such as by scratching/cutting with a razor blade,

sharply striking with the hemispherical head of a ball-peen hammer (Figure 5.11b), and

to a slightly lesser extent, applying pressure or, in the case of PBMA, stretching.

Films used to study the mechanochromic response were prepared by compressing

the films in a hydraulic press at pressures up to eight metric tons. For each of the pressed

films, little-to-no visible color change was observed, however, the emission shifted to

longer wavelengths and increased in intensity upon deformation in a manner somewhat

reminiscent of the response to acetonitrile vapor. For example, the emission spectra of a

PMMA film with 20% w/w [Pt(5)Cl](PF6) are shown in Figure 5.11c before and after compression. While the emission of the compressed film becomes more symmetrical and

λmax shifts from 575 nm to 592 nm, it is apparent that the change is not as dramatic as

that observed for similar films exposed to acetonitrile vapor. In the case of PBMA films,

the low Tg allows for the samples to be stretched, resulting in a slight change in the photoluminescence in the deformed region of the film. The resulting emission profile closely matches that of the compressed PMMA film (see Figure 5.11d). The emission

intensity slightly increases and becomes more symmetrical than the as cast film and the

λmax shifts from 574 nm to 592 nm as with the compressed film. However, it is still apparent that the response does not reach the level of change afforded in the vapochromic experiments.

155

Figure 5.11 a) Mechanoluminescent response of scratching a 10% w/w [Pt(5)Cl](PF6) in PMMA film observed under UV light (λ = 365 nm). b) Image of the mechanoluminescent response of a

10% w/w [Pt(5)Cl](PF6) in PMMA which was hit sharply with a ball-peen hammer. Normalized emission spectra of c) a pressed 20% w/w [Pt(5)Cl](PF6) in PMMA film and d) a stretched 10%

w/w [Pt(5)Cl](PF6) in PBMA film compared to the as cast and acetonitrile exposed emission. In both cases, deformation results in a red-shift of λmax and increased symmetry of the emission profile consistent with force-induced structural rearrangement of the platinum complexes.

In an attempt to probe the origin of this mechanochromic behavior the structural

properties of the platinum complexes in the compressed films were characterized using

WAXS. The observed scattering pattern for a PMMA film with 20% w/w [Pt(5)Cl](PF6)

shows four distinct peaks occurring at 2θ = 24.4, 15.46, 11.02, and 5.52 ° which equate to

distances of 3.38, 5.73, 8.03, and 16.01 Å respectively. The two shortest distances match very closely to those of the uncompressed films (3.37 and 5.72 Å) corresponding to alternating short and long Pt – Pt distances. The larger distances are consistent with longer Pt – Pt distances which may be related to complexes separated by one or multiple other complexes. The fact that the distances between platinum atoms do not appear to

156 change much after mechanical deformation suggests that possibly only a small fraction of the crystals change conformation. The mechanochromic response is consistent with force-induced structural rearrangement that results in stronger intermolecular Pt – Pt interactions. Unsurprisingly, attempts to press the films at temperatures above the polymer Tg gave no response as the applied pressure caused the films to flow which removed the driving force for aggregation of the platinum complexes.

5.3 CONCLUSIONS

The results presented in this study have shown that square-planer platinum complexes of

4-dodecyloxy-2,6-bis(N-methylbenzimidazoyl)pyridine (5) which have been blended into a series of methacrylate polymers display interesting vapochromic and vapoluminescent properties. The use of a polymer matrix allows for mechanically stable films, and the polymer properties have been shown to influence the vapochromic response recovery.

Structural studies are consistent with the previously reported mechanism of a vapor- induced rearrangement of the platinum complexes, characterized by an increase in Pt – Pt interactions. Blending the platinum complexes also results in interesting mechanochromic properties that are believed to be the result of force-induced structural rearrangement of the crystals upon stretching, scratching, or compression.

157

5.4 EXPERIMENTAL METHODS

5.4.1 Materials

All reagents and solvents were purchased from Aldrich Chemical Co. Reagents were

used without further purification. Spectrophotometric grade chloroform was used for all

76 experiments. Pt(DMSO)2Cl2 was prepared according to the literature. Poly(ethyl methacrylate) (PEMA), poly(butyl-co-isobutyl methacrylate) (PBcIBMA), and poly(butyl methacrylate) (PBMA) were purchased from Aldrich Chemical Co. with Tgs of 63, 35,

and 15 °C respectively. PMMA VS-100 was supplied by Arkema Inc. and was found to

have a Tg of 90 °C.

5.4.2 Instruments

NMR spectra were recorded using a Varian 600 NMR spectrometer. Wide-angle X-ray

scattering (WAXS) measurements were conducted using a Rigaku SA-HF3 X-ray

generator for the D/MAX2000/PC series diffractometer. UV-vis absorption spectra were

obtained using a Perkin-Elmer Lambda 800 UV-vis spectrometer. Films were solution-

cast onto glass slides and scanned in the range of 325-700 nm with an integration time of

0.24 s. Photoluminescence spectra were acquired using an Ocean Optics ACD1000-USB spectrometer (λex = 377 nm) through the use of a Y-shaped optical fiber in conjunction with a Gel Instrumente AG hotstage with a TC2 temperature controller. Molecular weights of the materials were measured by mass spectrometry on a Bruker AUTOFLEX

III MALDI TOF/TOF mass spectrometer using NALDI (Nanostructured Array Laser

Desorption/Ionization) techniques. Polarized optical microscopy (POM) studies were performed using an Olympus BX51 microscope equipped with 90° crossed polarizers and

158

a digital camera (14.2 Color Mosaic Model from Diagnostic Instruments, Inc.). Images

were acquired from the camera using Spot software (Diagnostic Instruments, Inc.).

Spatial dimensions were calibrated using a stage micrometer with 10 μm line spacing. A

50x/0.75 NA achromat long working-distance objective lens (Olympus LMPlanFI) was employed. Fluorescence microscopy studies were performed using a Leica DM2500M microscope equipped with a QImaging QiCam digital camera. Images were acquired from the camera using QCapture Pro 6.0 software. Spatial dimensions were calibrated using a stage micrometer with 10 μm line spacing. A 50x/0.75 N Plan achromat long working-distance objective lens (Leica) was employed. Samples were excited using a 100

W mercury lamp (Chiu Technical Corporation).

5.4.3 Synthesis of 5

4-hydroxy-2,6-bis(N-methylbenzimidazoyl)pyridine (HOMebip, 1.3 g, 3.7 mmol),

K2CO3 (600 mg, 4.3 mmol), and 1-bromododecane (1.37 g, 5.5 mmol) were combined in

25 mL DMSO and refluxed at 75 °C for 48 hrs. The solution was extracted with H2O and

CHCl3 after which the organic fractions were combined, dried with sodium sulphate,

filtered, and the solvent was removed in vacuo. The product was recrystallized from

hexanes to yield 1.51 g (78%). δH (600 MHz, CDCl3) 7.91 (s, 2H, Ar), 7.85 (d, 2H, Ar),

7.44 (d, 2H, Ar), 7.36 (t, 2H, Ar), 7.33 (t, 2H, Ar), 4.21 (t, 2H, -OCH2-), 4.21 (s, 6H, -

NCH3), 1.84 (m, 2H, -OCH2CH2-), 1.47 (m, 2H, -OCH2CH2CH2-), 1.39-1.21 (b, 16H, -

CH2-), 0.86 (t, 3H, -CH2CH3); δC (100 MHz, CDCl3) 166.9, 151.3, 150.7, 142.7, 137.4,

123.7, 123.0, 120.4, 112.0, 110.1, 68.9, 32.7, 32.1, 29.9, 29.84, 29.79, 29.77, 29.6, 29.5,

29.1, 26.1, 22.9, 14.3. m/z = 523.58 [M+H].

159

13 Figure 5.12 100 MHz C NMR of 5 in CDCl3.

Figure 5.13 Nanostructured Array Laser Desorption/Ionization (NALDI) spectrum of 5 [M+H].

160

5.4.4 Synthesis of [Pt(5)Cl](PF6)

A solution of 5 (400 mg, 0.76 mmoles) in chloroform (30 mL) was added to a suspension of Pt(DMSO)2Cl2 (322 mg, 0.76 mmoles) in methanol (10 mL) and stirred at 50 °C for

24 hrs under inert atmosphere. After removal of the solvent, the product was stirred in a saturated aqueous NH4PF6 solution for 24 hrs. The yellow product was washed with water and THF, dried, and recrystallized from acetone to yield 624 mg (91%) of product.

δH (600 MHz, (CD3)2SO) 7.48 (d, 2H, Ar), 7.21 (t, 2H, Ar), 7.12 (t, 2H, Ar), 7.03 (s, 2H,

Ar), 6.95 (d, 2H, Ar), 4.40 (t, 2H, -OCH2-), 3.76 (s, 6H, -NCH3), 1.87 (m, 2H, -

OCH2CH2-), 1.53 (m, 2H, -OCH2CH2CH2-), 1.60-1.20 (b, 16H, -CH2-), 0.84 (t, 3H, -

CH2CH3); δC (100 MHz, (CD3)2SO) 152.2, 147.6, 137.7, 133.5, 126.1, 125.6, 115.3,

111.9, 110.5, 70.7, 40.4, 32.2, 31.3, 29.14, 29.13, 29.11, 29.09, 28.9, 28.8, 28.2, 25.4,

22.1, 13.9. m/z = 753.54 [M+H].

13 Figure 5.14 100 MHz C NMR of [Pt(5)Cl](PF6) in (CD3)2SO.

161

Figure 5.15 a) Nanostructured Array Laser Desorption/Ionization (NALDI) spectrum of

[Pt(5)Cl](PF6) [M+H] compared with b) the theoretical isotope pattern.

5.4.5 Preparation of Films Containing [Pt(5)Cl](PF6)

Blends of [Pt(5)Cl](PF6) were prepared by dissolving 250 mg of the polymethacrylate in

4 mL of chloroform followed by addition of the appropriate amount of [Pt(5)Cl](PF6)

(ex. 25 mg for a 10% w/w film). The flask was sealed and sonicated for 16 hrs to achieve

complete dispersal of the platinum complex. The solution was cast into an aluminium walled casting dish with a Teflon sheet bottom. The solvent was allowed to evaporate at room temperature for 24 hrs then further dried in a vacuum oven at 40 °C.

5.4.6 Vapochromic Studies

Exposure to acetonitrile vapor for vapochromic studies was carried out using a sealed

vapor chamber where the solvent was placed into the bottom of the chamber and the

films were placed above the solvent on an inverted glass Petri dish. The chamber was allowed to equilibrate for 20 mins before introducing the samples. Films were exposed for a minimum of 15 mins unless otherwise stated, even though a color change was observed within 3 mins. Studies of the exposed films were performed immediately after

162

removal from the chamber.

5.4.7 Mechanochromic Studies

Films of thickness 400 ± 50 µm were used for all compression studies. Films were

pressed at room temperature at a pressure of eight metric tons for 5 mins. Attempts to

press films at temperatures above the polymer Tg resulted in no observable

mechanochromic behavior.

5.5 ACKNOWLEDGMENTS

This material is based upon work supported by the National Science Foundation under

Grants CHE-0704026, DMR-0602869, and MRI-0821515 (for the purchase of the

MALDI-TOF/TOF), the Case School of Engineering and the Kent H. Smith Charitable

Trust.

5.6 REFERENCES

1. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

2. Friese, V. A.; Kurth, D. G. Coord. Chem. Rev., 2008, 252, 199–211.

3. McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. V. M. Rotello and S. Thayumanavan, John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178.

4. Xu, D. H.; Liu, C. Y.; Craig, S. L. Macromolecules, 2011, 44, 2343-2353.

5. Piermattei, A.; Karthikeyan, S.; Sijbesma, R. P. Nat. Chem., 2009, 1, 133-137.

6. Nair, K.; Breedveld, V.; Weck, M. Macromolecules, 2011, 44, 3346-3357.

163

7. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176- 188.

8. Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565-578.

9. Dobrawa, R.; Würthner, F. J. Polym. Sci. Part A: Polym. Chem., 2005, 43, 4981-4995.

10. Fages, F. Angew. Chem., Int. Ed., 2006, 45, 1680-1682.

11. Kumpfer, J. R.; Wie, J. J.; Swanson, J. P.; Beyer, F. L.; Mackay, M. E.; Rowan, S. J. Macromolecules, 2012, 45, 473-480.

12. Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145-151.

13. Knapton, D.; Burnworth, M.; Rowan, S. J.; Weder, C. Angew. Chem., Int. Ed., 2006, 45, 5825–5829.

14. Handl, H. L.; Gillies, R. J. Life Sci., 2005, 77, 361–371.

15. Burnworth, M.; Rowan, S. J.; Weder, C. Chem. Eur. J., 2007, 13, 7828–7836.

16. Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334-337.

17. Kumpfer, J. R.; Rowan, S. J. J. Am. Chem. Soc., 2011, 133, 12866-12874.

18. Yu, S. C.; Kwok, C. C.; Chan, W. K.; Chen, C. M. Adv. Mater., 2003, 15, 1643-1647.

19. Ng, W.; Gong, X.; Chan, W. K. Chem. Mater., 1999, 11, 1165-1170.

20. Murphy, L.; Williams, J. A. G. Top. Organomet. Chem., 2010, 28, 75-111.

21. Kato, M. Bull. Chem. Soc. Jpn., 2007, 80, 287-294.

22. Anderson, B. M.; Hurst, S. K. Eur. J. Inorg. Chem., 2009, 21, 3041-3054.

164

23. Eryazici, I.; Moorefield, C. N.; Newkome, G. R. Chem. Rev., 2008, 108, 1834-1895.

24. Paw, W.; Cummings, S. D.; Mansour, M. A.; Connick, W. B.; Geiger, D. K.; Eisenberg, R. Coordin. Chem. Rev., 1998, 171, 125-150.

25. Houlding, V. H.; Miskowski, V. M. Coordin. Chem. Rev., 1991, 111, 145-152.

26. Williams, J. A. G.; Develay, S.; Rochester, D. L.; Murphy, L. Coordin. Chem. Rev., 2008, 252, 2596-2611.

27. Williams, J. A. G. Top. Curr. Chem., 2007, 281, 205-268.

28. Li, K.; Chen, Y.; Lu, W.; Zhu, N.; Che, C. M. Chem. Eur. J., 2011, 17, 4109-4112.

29. Develay, S.; Williams, J. A. G. Dalton Trans., 2008, 34, 4562-4564.

30. Ni, J.; Zhang, L. Y.; Wen, H. M.; Chen, Z. N. Chem. Commun., 2009, 3801-3803.

31. Miskowski, V. M.; Houlding, V. H. Inorg. Chem., 1989, 28, 1529-1533.

32. Chung, S. K.; Tseng, Y. R.; Chen, C. Y.; Sun, S. S. Inorg. Chem., 2011, 50, 2711- 2713.

33. Connick, W. B.; Henling, L. M.; Marsh, R. E.; Gray, H. B. Inorg. Chem., 1996, 35, 6261-6265.

34. Kato, M.; Kosuge, C.; Morii, K.; Ahn, J. S.; Kitagawa, H.; Mitani, T.; Matsushita, M.; Kato, T.; Yano, S.; Kimura, M. Inorg. Chem., 1999, 38, 1638-1641.

35. Connick, W. B.; Marsh, R. E.; Schaefer, W. P.; Gray, H. B. Inorg. Chem., 1997, 36, 913-922.

36. Hudson, Z. M.; Sun, C.; Harris, K. J.; Lucier, B. E. G.; Schurko, R. W.; Wang, S. Inorg. Chem., 2011, 50, 3447-3457.

165

37. Forniés, J.; Fuertes, S.; López, J. A.; Martín, A.; Sicilia, V. Inorg. Chem., 2008, 47, 7166-7176.

38. Lu, W.; Mi, B. X.; Chan, M. C. W.; Hui, Z.; Che, C. M.; Zhu, N.; Lee, S. T. J. Am. Chem. Soc., 2004, 126, 4958-4971.

39. Shao, P.; Li, Y.; Sun, W. Organometallics, 2008, 27, 2743-2749.

40. Ji, Z.; Li, Y.; Pritchett, T. M.; Makarov, N. S.; Haley, J. E.; Li, Z.; Drobizhev, M.; Rebane, A.; Sun, W. Chem. Eur. J., 2011, 17, 2479-2491.

41. Yam, V. W. W.; Chan, K. H. Y.; Wong, K. M. C.; Zhu, N. Chem. Eur. J., 2005, 11, 4535-4543.

42. Pritchett, T. M.; Sun, W.; Guo, F.; Zhang, B.; Ferry, M. J.; Rogers-Haley, J. E.; Shensky III, W.; Mott, A. G. Opt. Lett., 2008, 33, 1053-1055.

43. Field, J. S.; Haines, R. J.; Ledwaba, L. P.; McGuire Jr., R.; Munro, O. Q.; Low, M. R.; McMillin, D. R. Dalton Trans., 2007, 2, 192-199.

44. Shikhova, E.; Danilov, E. O.; Kinayyigit, S.; Pomestchenko, I. E.; Tregubov, A. D.; Camerel, F.; Retailleau, P.; Ziessel, R.; Castellano, F. N. Inorg. Chem., 2007, 46, 3038-3048.

45. Wang, K.; Haga, M. A.; Monjushiro, H.; Akiba, M.; Sasaki, Y. Inorg. Chem., 2000, 39, 4022-4028.

46. Rivera, E. J.; Figueroa, C.; Colón, J. L.; Grove, L.; Connick, W. B. Inorg. Chem., 2007, 46, 8569-8576.

47. Kobayashi, A.; Fukuzawa, Y.; Noro, S. I.; Nakamura, T.; Kato, M. Chem. Lett., 2009, 38, 998-999.

48. Muro, M. L.; Daws, C. A.; Castellano, F. N. Chem. Commun., 2008, 6134-6136.

166

49. Field, J. S.; Grimmer, C. D.; Munro, O. Q.; Waldron, B. P. Dalton Trans., 2010, 39, 1558-1567.

50. Du, P. Inorg. Chim. Acta, 2010, 363, 1355-1358.

51. Du, P.; Schneider, J.; Brennessel, W. W.; Eisenberg, R. Inorg. Chem., 2008, 47, 69- 77.

52. Wadas, T. J.; Wang, Q. M.; Kim, Y. J.; Flaschenreim, C.; Blanton, T. N.; Eisenberg, R. J. Am. Chem. Soc., 2004, 126, 16841-16849.

53. Grove, L. J.; Oliver, A. G.; Krause, J. A.; Connick, W. B. Inorg. Chem., 2008, 47, 1408-1410.

54. Grove, L. J.; Rennekamp, J. M.; Jude, H.; Connick, W. B. J. Am. Chem. Soc., 2004, 126, 1594-1595.

55. Mathew, I.; Sun, W. Dalton Trans., 2010, 39, 5885-5898.

56. Beiermann, B. A.; Kramer, S. L. B.; Moore, J. S.; White, S. R.; Sottos, N. R. ACS Macro Lett., 2012, 1, 163-166.

57. Pucci, A.; Ruggeri, G. J. Mater. Chem., 2011, 21, 8282-8291.

58. Lott, J.; Weder, C. Macromol. Chem. Phys., 2010, 211, 28-34.

59. Sagara, Y.; Kato, T. Nat. Chem., 2009, 1, 605-610.

60. Kunzelman, J.; Gupta, M.; Crenshaw, B. R.; Schiraldi, D. A.; Weder, C. Macromol. Mater. Eng., 2009, 294, 244-249.

61. Ni, J.; Zhang, X.; Wu, Y. H.; Zhang, L.Y.; Chen, Z. N. Chem. Eur. J., 2011, 17, 1171-1183.

62. Kozhevnikov, V. N.; Donnio, B.; Bruce, D. W. Angew. Chem. Int. Ed., 2008, 47, 6286-6289.

167

63. Abe, T.; Itakura, T.; Ikeda, N.; Shinozaki, K. Dalton Trans., 2009, 4, 711-715.

64. Zahner, J. C.; Drickamer, H. G. J. Chem. Phys., 1960, 33, 1625-1628.

65. Takagi, H. D.; Noda, K.; Itoh, S. Platinum Metals Rev., 2004, 48, 117-124.

66. Abe, T.; Takahashi, K.; Shiraishi, Y.; Toshima, N.; Kaneko, M. Macromol. Chem. Phys., 2000, 201, 102-106.

67. Nishiyama, N.; Kato, Y.; Sugiyama, Y.; Kataoka, K. Pharm. Res., 2001, 18, 1035- 1041.

68. Parrinello, G.; Stille, J. K. J. Am. Chem. Soc., 1987, 109, 7122-7127.

69. Tong, Q. X.; Li, X. H.; Wu, L. Z.; Yang, Q. Z.; Zhang, L. P.; Tung, C. H. Chinese J. Chem., 2004, 22, 1204-1207.

70. Rivera, E. J.; Barbosa, C.; Torres, R.; Grove, L.; Taylor, S.; Connick, W. B.; Clearfield, A.; Colón, J. L. J. Mater. Chem., 2011, 21, 15899-15902.

71. Li, X. H.; Wu, L. Z.; Zhang, L. P.; Tung, C. H.; Che, C. M. Chem. Commun., 2001, 2280-2281.

72. Zhang, D.; Wu, L. Z.; Yang, Q. Z.; Li, X. H.; Zhang, L. P.; Tung, C. H. Org. Lett., 2003, 5, 3221-3224.

73. Che, C. M.; Fu, W. F.; Lai, S. W.; Hou, Y. J.; Liu, Y. L. Chem. Commun., 2003, 118- 119.

74. Beck, J. B.; Rowan, S. J. Faraday Discuss., 2005, 128, 43-53.

75. Tam, A. Y. Y.; Lam, W. H.; Wong, K. M. C.; Zhu, N.; Yam, V. W. W. Chem. Eur. J., 2008, 14, 4562-4576.

76. Price, J. H.; Williamson, A. N.; Schramm, R. F.; Wayland, B. B. Inorg. Chem., 1972, 11, 1280-1284.

168

CHAPTER 6

Fabrication of Platinum Nanoparticles Utilizing

Metallosupramolecular Polymers as Templates

6.1 INTRODUCTION

The fabrication of nanoparticles embedded within a polymer matrix is of wide interest as a consequence of their potential applications in sensors,1 magnetic data storage,2 flash memory,3 and catalysis.4 Many different methods have been utilized to prepare polymer

nanocomposites with metal nanoparticles.5 One way is to mix the premade nanoparticles

with the polymer using solution or melt based processing techniques; however this

method of preparing nanocomposites generally results in heterogeneous dispersions on

account of nanoparticle agglomeration combined with high polymer viscosity.6 More

recently, in-situ formation of metal nanoparticles from metal-ions dispersed within a polymer matrix has gained attention.7 This method is performed either by adding the

metal precursor to the polymer in its processed (solution or melt) form or by processing

the two components simultaneously. The synthesis of metal nanoparticles within the

polymer matrix can then be carried out by a variety of techniques including

electrochemical8 or chemical9 reduction, photoirradiation,10,11 and thermal treatment.12,13

Metal-containing polymers14 where the metal is an intricate component of the polymer itself are interesting candidates as precursors15-17 to metal nanoparticle

containing nanocomposites as these organic/inorganic hybrid materials can combine the

processability of polymers with the functionalities of the metal and issues of macroscopic

phase separation of the metal-ion and polymer matrix are removed. A subclass of this

field is metallosupramolecular polymers,18-20 which can be prepared using a wide variety

of metal-ions and polymers.21- 24 We have previously shown that metallosupramolecular

polymers, comprised of ditopic 2,6-bis(N-methylbenzimidazoyl)pyridine (Mebip) end-

capped macromonomers self-assembled with either transition metals25,26 (Zn2+, Fe2+,

170

etc…) or lanthanides27,28 (La3+ and Eu3+), are able to form mechanically stable films.

Metallosupramolecular polymers with low Tg “soft” polymeric cores such as the poly(tetrahydrofuran)-based macromonomer 2 and the poly(ethylene-co-butylene)-based macromonomer 3 form lamellar morphologies27,28 in the solid state as a result of phase separation of the hard ionic, metal-ligand complexes and the soft polymer core . We have

also shown that the degree of long-range order and mechanical properties are strongly

dependent on the nature of the polymer used as described in Chapter 3, with the more

polar, poly(tetrahydrofuran)-based 2, showing decreased ordering and weaker films compared to the less polar poly(ethylene-co-butylene)-based 3. In these films the lamellar period was <10 nm and is related to the molecular weight of the polymer core. The presence of high concentrations of the metal-ions in one of the phases suggested to us that such materials may be interesting potential precursors to metal nanoparticle

nanocomposites.

Figure 6.1 Synthesis of the platinum metallosupramolecular polymers [Pt2(2,3 or 6)Cl2](Cl2).

171

The formation of metal nanoparticles by plasma reduction is well

documented.29,30 However general plasma reduction is hindered by the need for operating

at low pressures which results in increased processing costs and is not useful for large-

scale production. More recently, plasma sources that can be operated at atmospheric

pressure31-33 have been developed and initial work has demonstrated the reduction of

metal-cations doped into polymeric films with the use of electrons from an extracted

discharge from a microplasma at atmospheric pressure.34,35 The use of an extracted

discharge from a microplasma has the distinct advantage to microplasma alone as it is sustainable at low currents which reduces problems resulting from gas heating and sputtering which can potentially damage the samples. Therefore, we were interested in investigating the potential of using extracted discharge from a microplasma to reduce a metallosupramolecular polymer to see if we could establish a new and straightforward

strategy for nanofabrication of nanoparticles in such a polymeric matrix.

As mentioned earlier, metallosupramolecular polymers have been prepared with

many different metals, however in this study we wanted to investigate Pt-containing

polymers as platinum nanoparticles have a range of potential uses from catalysis36- 38 to

biomedical applications.39,40 There are some examples of Pt-containing supramolecular

polymers where the Pt-ligand (usually a pyridyl derivative) interaction is the reversible

motif.41-43 However, in order to take advantage of the macromonomers 2 and 3 that we have previously investigated26,27 we were interested in developing new metallosupramolecular polymers that would take advantage of weak Pt – Pt interactions.

Such interactions have been observed to occur with a number of low molecular weight

terdentate Pt complexes, including Mebip:PtX2+, and resulted in the development of

172

vapochromic and mechanochromic crystalline solids.44-46 Thus our target is the synthesis

of ditopic telechelic macromonomers which have the [Mebip:PtCl]2+ complex at the

chain-ends (Figure 6.1). The goal then would be to form films of these

metallosupramolecular polymers and study the formation of platinum nanoparticles

within these films using electrochemical reduction via extracted discharge from a

microplasma (Figure 6.2).

Figure 6.2 Schematic of platinum nanoparticle formation from the reduction of metallosupramolecular polymers via the extracted discharge from a microplasma.

6.2 RESULTS AND DISCUSSION

The platinum complexes [Pt2(2,3 or 6)Cl2](Cl2) are synthesized by reacting 2 and 6,

ditopic Mebip end-capped poly(tetrahydrofuran) macromonomers with molecular

weights of 4200 and 9600 g/mol respectively, and 3, a ditopic Mebip end-capped

poly(ethylene-co-butylene) with a molecular weight of 4000 g/mol, with Pt(DMSO)2Cl2

at 50 °C overnight as shown in Figure 6.1. Pt2+ is known to complex with 2,6-bis(N- methylbenzimidazoyl)pyridine (Mebip) ligands in a 1:1 fashion47,48 and as such does not

result in the formation of metallosupramolecular polymers via a chain extension

assembly that has previously been observed with the more common 1:2 metal:ligand

173

binding of transition metal-ions such as Zn2+ or Fe2+.49 However, it has been shown in

small molecule analogues44,47 and Chapter 5 that Pt2+:Mebip complexes can form

extended chains via Pt – Pt interactions. Thus, we wanted to see if such Pt – Pt

interactions, presumably in conjunction with phase separation of the metal complex and

polymer core,26,50 can be used to access metallosupramolecular polymers. Gratifyingly, solution-casting either via drop-casting or spin-coating [Pt2(2)Cl2](Cl2) or

[Pt2(3)Cl2](Cl2) did result in the formation of films. To our knowledge this is the first

time that this Pt – Pt motif has been used to access metallosupramolecular polymer

films.51

6.2.1 Morphological Studies

Both [Pt2(2 or 6)Cl2](Cl2) and [Pt2(3)Cl2](Cl2) are able to form metallosupramolecular

polymer films at a film thickness of ~180 µm, however the mechanical properties of each film are quite different, similar to what was shown in Chapter 3 for films prepared with

Zn2+ and Eu3+, in that metallosupramolecular polymers formed from 3 show better mechanical properties than those formed with 2. Solution-cast films of [Pt2(3)Cl2](Cl2)

are mechanically stable with a tensile storage modulus of ~70 MPa at 20 °C (Figure 6.3)

while films prepared with [Pt2(2)Cl2](Cl2) are much weaker to the extent that the

mechanical properties could not be tested using dynamic mechanical thermal analysis

(DMTA) (see Figure 6.3b and c).

174

Figure 6.3 DMTA of a mechanically stable [Pt2(3)Cl2](Cl2) film. The drop in modulus at ca. -40

°C results from the Tg of the polymer core. Films of [Pt2(2)Cl2](Cl2) were too weak to undergo

testing. Images of the mechanical properties of films prepared with b) [Pt2(3)Cl2](Cl2) and c)

[Pt2(2)Cl2](Cl2).

Small-angle (SAXS) and wide-angle (WAXS) X-ray scattering studies were

carried out to better understand the difference in the properties of these two systems. The

SAXS of films of [Pt2(2)Cl2](Cl2) and [Pt2(3)Cl2](Cl2) both show scattering which is

consistent with a lamellar-like morphology as a result of phase separation between the

metal-ligand complexes and the polymer cores, similar to what has been shown before for

other metallosupramolecular polymers with 2 and 3. For films prepared with

175

[Pt2(3)Cl](Cl2), two strong scattering peaks are observed with a primary Bragg

diffraction peak (q*) at q = 0.72 nm-1 and a secondary scattering peak (2q*) at half the distance of the primary peak as shown in Figure 6.4a. This type of scattering is consistent

with a lamellar morphology with alternating domains of the metal-ligand complex and

the polymer core as a result of phase separation. In this case the metal-ligand complexes act as a “hard” phase and the low Tg polymer core (ca. -80 °C for 2 and 6 and ca. -40 °C

for 3) acts as the “soft” phase which is consistent with previous results seen with other

metallosupramolecular films.27,28 The lamellar period, d, which is given as d = 2π/q* is

approximately 8.6 nm which correlates to the lamellar spacing between the metal-ligand

“hard” phases and matches well with previously reported metallosupramolecular polymers prepared with macromonomer 3.

176

Figure 6.4 Morphological characterization of metallosupramolecular polymers [Pt2(2)Cl2](Cl2)

and [Pt2(3)Cl2](Cl2). Films of [Pt2(3)Cl2](Cl2) display strong scattering peaks consistent with a lamellar morphology by a) small angle X-ray scattering (SAXS) and b) wide angle X-ray scattering (WAXS) shows only one short Pt – Pt distance within the metal-ligand phase. c)

[Pt2(2)Cl2](Cl2) films also show lamella in the SAXS but with reduced long-range order. d)

WAXS of [Pt2(2)Cl2](Cl2) display both “long” and “short” Pt – Pt distances in the metal-ligand hard phase.

Wide-angle X-ray scattering was used to probe the structure of the metal-ligand phase. The WAXS diffractogram for [Pt2(3)Cl2](Cl2), Figure 6.4b, displays one

dominant scattering peak at 2θ = 26.3 ° (d = 3.39 Å) as well as broad peaks at 2θ = 19.2

and 12.1 ° (d = 4.67 and 7.31 Å respectively) which result from the amorphous polymer

177

core. The peak at 3.39 Å matches closely to those previously reported from single crystal

data for low molecular weight Pt2+:Mebip complexes showing a linear chain of Pt – Pt

interactions.

In a similar manner, the SAXS of films of [Pt2(2)Cl2](Cl2), Figure 6.4c, show two

peaks with a primary diffraction peak (q*) at q = 0.75 nm-1 (d = 8.4 nm) and a secondary

peak (2q*) at q = 1.53 nm-1 (d = 4.1 nm) characteristic of a lamellar structure. However

the relative intensity of the scattering peaks is much weaker than those seen for the

[Pt2(3)Cl2](Cl2) film suggesting that the long-range ordering of the lamellae in the

[Pt2(2)Cl2](Cl2) films is inferior. The decreased intensity in the SAXS for

[Pt2(2)Cl2](Cl2) films is likely a result of poor phase separation between the metal-ligand complexes and the polymer core arising from the presence of oxygen atoms along the polymer backbone. Similarly to what was observed in Chapter 3 for the Zn2+ and Eu3+

films, the increased polarity of the poly(tetrahydrofuran) core of [Pt2(2)Cl2](Cl2) along with the potential for the polymeric oxygen atoms to weakly coordinate with the Pt2+ ions

of the metal-ligand complexes are believed to be responsible for the observed decrease in

phase separation when compared to films prepared with [Pt2(3)Cl2](Cl2).

The WAXS pattern for the [Pt2(2)Cl2](Cl2) film also shows multiple scattering

peaks rather than the one dominant peak seen in the [Pt2(3)Cl2](Cl2) film (see Figure

6.4d). The peaks in the film of [Pt2(2)Cl2](Cl2) correspond to distances of 16.19, 7.97,

5.71, 3.78, and 3.38 Å. The peaks at 3.38 and 5.71 Å match closely with previously

reported lengths from single crystal structures of small molecule Pt2+:Mebip complexes

that show a Pt – Pt chain with alternating long and short distances. The peak at 3.78 Å is

consistent with the interplaner distances between the Mebip ligands of neighboring

178

complexes52 and the scattering peaks for d = 7.97 and 16.19 Å are consistent with the

presence of longer Pt – Pt distances (e.g. the Pt(1) to Pt(3) distance in a chain of Pt-ions:

Pt(1) – Pt(2) – Pt(3) – Pt(4), etc.).

Figure 6.5 Schematic showing the partial short range structures of metallosupramolecular

polymers a) [Pt2(3)Cl2](Cl2) and b) [Pt2(2)Cl2](Cl2) formed via Pt – Pt interactions in the solid state as suggested by SAXS and WAXS data.

Based on the SAXS and WAXS data the proposed solid state structures of

metallosupramolecular polymers prepared with [Pt2(2)Cl2](Cl2) and [Pt2(3)Cl2](Cl2) are

shown in Figure 6.5. All metallosupramolecular polymers are formed via Pt – Pt interactions in the hard phase. Films prepared with [Pt2(3)Cl2](Cl2) show a well-defined, long-range lamellar morphology with one repeatable Pt – Pt spacing in the metal-ligand phase (Figure 6.5a) while films prepared with [Pt2(2)Cl2](Cl2) show decreased lamellar ordering with long and short Pt – Pt distances within the metal-ligand phase. Films of

[Pt2(3)Cl2](Cl2) represent the first known example of mechanically stable films of a

metallosupramolecular polymer formed through use of Pt – Pt interactions (presumably

aided by phase separation).

179

6.2.2 Reduction of Metallosupramolecular Polymers Using Extracted Discharge from a

Microplasma

Reduction of the platinum cations in the metallosupramolecular polymers was carried out using a low-current extracted discharge from a microplasma. The method of obtaining the extracted discharge from the microplasma is shown in Figure 6.6 where a microplasma is generated and sustained in Ar gas at atmospheric pressure between a stainless steel cathode and a metal grid anode. The extracted discharge was obtained from the microplasma using a stainless steel plate as a third electrode which was positively biased and placed near the anode.

Figure 6.6 Schematic of the experimental setup of the reduction of the metallosupramolecular polymers using extracted discharge from a microplasma.

Thin films of [Pt2(2,3 or 6)Cl2](Cl2) were prepared by spin-coating onto a

substrate and the metal-ion chain-ends were reduced by exposure to extracted electrons from a microplasma. Reduction was carried out at varying times to examine the growth of the platinum nanoparticles. Figure 6.7 shows TEM images of nanoparticle growth

180

from [Pt2(2)Cl2](Cl2) as a function of exposure time to the extracted discharge from a

microplasma. Films which were not exposed showed no features in the TEM (Figure

6.7a) however after only 30 mins of exposure nanoparticles could be observed (Figure

6.7b). Further exposure results in an increase in the average particle size as shown in

Figure 6.7c and d for exposure times of 1 hr and 3 hrs, respectively. High resolution TEM

investigations of the Pt nanoparticles formed show that the particles are crystalline and

the observed lattice spacings correspond well to the Pt(111) and Pt(200) crystalline

planes of the fcc (face centered cubic) crystalline phase platinum calculated from Bragg’s

law (see Figure 6.7d insert).

Figure 6.7 TEM images of [Pt2(2)Cl2](Cl2) for the various exposure times to the extracted discharge from the microplasma. a) Unexposed and b) 0.5 hr, c) 1 hr, and d) 3 hrs. Inset shows high resolution TEM displaying typical platinum crystalline lattice spacing. Extracted discharge from the microplasma was operated with 200 µA current.

181

Although nanoparticles were successfully formed upon exposure to the extracted discharge from a microplasma, work was carried out to make sure that it is the electrons from the extracted discharge that are responsible for the reduction of the platinum and not some other source. Initially, investigations focused on the nature of the microplasma to determine if UV and/or heat generated by the extracted discharge from the microplasma, rather than the electrons, could be responsible for the reduction. A control experiment in which one sample was directly exposed to 200 µA of extracted discharge from the microplasma (which includes electrons, UV, and heating), while another was exposed to

0 µA of extracted discharge from the microplasma (which only includes UV and heat and no electrons) was used. As shown in Figure 6.8a and b, only the sample exposed to electrons showed the formation of platinum nanoparticles, confirming that UV and heating effects caused by the extracted discharge from the microplasma are not responsible for the reduction of the metallosupramolecular polymer.

We also looked at the potential reduction by the TEM itself as TEM uses an electron beam for imaging. A [Pt2(2)Cl](Cl2) film which was spin-coated onto a TEM grid and then exposed to 300 kV TEM and images were taken at various times at the same position. Even up to 10 mins exposure to the electrons from TEM, no particle formation was observed in the [Pt2(2)Cl](Cl2) film. In addition, we checked to see if the e-beam of the TEM would induce growth of already formed Pt nanoparticles.

Gratifyingly, the particle size of the targeted Pt nanoparticles was not observed to increase upon exposure to the TEM beam for 5 mins. Thus based on these controls, it is clear that the Pt nanoparticles in the metallosupramolecular polymer are generated by electrons from the extracted discharge from the microplasma.

182

Figure 6.8 TEM images of [Pt2(2)Cl2](Cl2) studying the potential for platinum reduction by common reduction techniques. a) 1 hr exposure to extracted discharge from the microplasma (M: 4 mA, ED: 200 µA) b) 1 hr exposure with microplasma (M: 4 mA, ED: 0 µA) c) UV exposure (320-500 nm, 100 mW/cm2, 3 hrs) d) Refluxed in ethylene glycol (150 °C, 3 hrs).

Other reduction techniques were attempted to determine the importance of the extracted discharge from the microplasma. To test for photoreduction using UV light, a

[Pt2(2)Cl2](Cl2) film on a TEM grid was irradiated with 320-500 nm UV light with a surface intensity of 100 mW/cm2 for 3 hrs. As shown in Figure 6.8c, no nanoparticles were observed. Chemical reduction was also studied by refluxing [Pt2(2)Cl2](Cl2) in ethylene glycol using an adapted literature procedure of work done by Urbina and coworkers53 on the reduction of platinum nitrate to form nanoparticles by refluxing in ethylene glycol. After refluxing [Pt2(2)Cl2](Cl2) in ethylene glycol for 3 hrs, the ethylene glycol was removed, the product was redissolved in chloroform, and spin-coated onto a

183

TEM grid as done with the previous samples. Once again no nanoparticle formation could be observed as shown in Figure 6.8d. These results demonstrate the importance of using the extracted discharge from the microplasma as common reduction techniques are unable to easily reduce the metallosupramolecular polymers in this study.

6.2.3 Nanoparticle Growth and Assembly

In order to give further insight to the process of nanoparticle formation in the metallosupramolecular polymers, we investigated the effects of the polymer backbone

(poly(tetrahydrofuran) for [Pt2(2)Cl2](Cl2) vs poly(ethylene-co-butylene) for

[Pt2(3)Cl2](Cl2)) and metal-loading ([Pt2(2)Cl2](Cl2) vs [Pt2(6)Cl2](Cl2)) on the nanoparticle growth. Interestingly, the nature of the polymer core (type and size) has a

significant impact on the size and density of the nanoparticles.

In order to understand the particle formation and growth of Pt nanoparticles in the

various metallosupramolecular templates, the evolution of particle size as a function of

exposure time was monitored. As shown in Figure 6.9, Pt nanoparticle growth in all the metallosupramolecular polymer templates shows that the particle size increases with increasing exposure time to the extracted discharge from the microplasma. All polymers and exposure times yield nanoparticles with narrow size distributions with average particle diameters of < 5 nm. Examination of the TEM images showed little-to-no agglomeration, even after 3 hrs of exposure to the extracted discharge from the microplasma.

184

Figure 6.9 TEM images of nanoparticles (3 hrs exposure) and particle size distributions of a)

[Pt2(2)Cl2](Cl2), b) [Pt2(3)Cl2](Cl2), and c) [Pt2(6)Cl2](Cl2) vs exposure time to the extracted discharge from the microplasma.

Interestingly, there is an observable trend and the average size and density of the

Pt nanoparticles is strongly dependent on the polymer core of the metallosupramolecular template. Comparisons of the average particle size and densities for all metallosupramolecular polymer and exposure times are shown in Figure 6.10. For all

185

exposure times, platinum nanoparticles created from [Pt2(2)Cl2](Cl2) have the largest

average diameter, followed by [Pt2(6)Cl2](Cl2), and films prepared with [Pt2(3)Cl2](Cl2)

generated the smallest particles. We also observed a trend in the particle density with

respect to the polymer core where [Pt2(3)Cl2](Cl2) yields the highest particle density and

[Pt2(2)Cl2](Cl2) has the lowest particle density as shown in Figure 6.10b.

Figure 6.10 a) Particle size analysis vs exposure time to the extracted discharge from the

microplasma. [Pt2(2)Cl2](Cl2) yielded the largest particles while [Pt2(3)Cl2](Cl2) yielded the smallest particles. b) Particle density analysis vs exposure time to the extracted discharge from the microplasma. [Pt2(2)Cl2](Cl2) yielded the smallest particle density while [Pt2(3)Cl2](Cl2) yielded the largest particle density.

These observed trends are consistent with a mechanism of nanoparticle formation

within these metallosupramolecular polymer templates being initiated by the extracted

discharge from the microplasma reducing some of the Pt2+ complexes to form nucleation

points of Pt0 from which the nanoparticles can begin to grow. The nanoparticles grow in

size as “Pt2+” species diffuse to these nucleation points and also are reduced in

accordance with classical nucleation theory.54 Using this mechanism the observations

186

regarding the relative nanoparticle size and density of the three films containing the

different cores, [Pt2(2,3 or 6)Cl2](Cl2), can be rationalized. Films of [Pt2(3)Cl2](Cl2)

form smaller diameter particles and higher particle density compared to [Pt2(2)Cl2](Cl2)

which is consistent with a more facile diffusion of Pt2+ species through the less ordered,

more polar polymer matrix (poly(THF) vs poly(ethylene-co-butylene)). Comparison of

the compounds with the same poly(THF) core but different molecular weights

([Pt2(2)Cl2](Cl2) vs. [Pt2(6)Cl2](Cl2)) shows that larger particle sizes and lower particle densities is favored in the lower molecular weight species. This is consistent with more facile transport of the Pt2+ species which can be expected based on the higher

concentration of the polar ionic complexes in [Pt2(2)Cl2](Cl2) and the smaller diffusion distances required through the lower molecular weight core.

6.3 CONCLUSIONS

In this work, we have demonstrated that metallosupramolecular polymers formed using

Pt – Pt interactions can be used as precursors for metallic nanoparticle containing nanocomposite by exposure to electrons from an extracted discharge from a microplasma.

Very uniform, unagglomerated platinum nanoparticles were created with diameters under

5 nm, the size of which could be tailored by the exposure time to the extracted discharge from a microplasma. The nanoparticle size and densities were also found to be influenced by the polymer core of the macromonomer used and seem to relate to the ease of diffusion of the Pt2+ species. Low molecular weight, polar poly(tetrahydrofuran) cores

yielded the lowest density of larger particles while the nonpolar, poly(ethylene-co-

butylene) core yielded a higher density of smaller particles. The ability to reduce

metallosupramolecular polymers using the very mild conditions afforded by the extracted

187

discharge from a microplasma is very interesting as they can be generated at ambient

conditions and allow for the ability to easily produce them on a large scale.

6.4 EXPERIMENTAL METHODS

6.4.1 Materials

All reagents and solvents were purchased form Aldrich Chemical Co. Reagents were

used without further purification. Spectrophotometric grade chloroform was used for all

experiments. Macromonomers 2, 3, and 6 were synthesized as previously reported.26,28

55 Pt(DMSO)2Cl2 was prepared according to the literature. Molecular weights (Mn) for 2,

6, and 3 were found to be 4200 g/mol, 9600 g/mol, and 4000 g/mol respectively. TEM grids were purchased from Electron Microscopy Sciences.

6.4.2 Instruments

NMR spectra were recorded on a Varian 600 NMR spectrometer. Small-angle X-ray

scattering (SAXS) measurements were conducted using a Rigaku S-MAX 3000 SAXS

system. Cu KR X-rays from a MicroMax-002+ sealed tube source (λ = 0.154 nm) were

collimated through three pinhole slits to yield a final spot size of 0.7 mm at the sample

position. The sample-to-detector distance and the scattering vector, q, were calibrated

using a silver behenate (AgBe) standard with a characteristic (001) peak position at q =

1.076 nm-1. The calculated sample-to-detector distance was 1.5 m. Typical exposure

times were 1 hr. Wide-angle X-ray scattering (WAXS) measurements were conducted

using a Rigaku SA-HF3 X-ray generator for the D/MAX2000/PC series diffractometer.

Dynamic mechanical thermal analysis experiments were performed using a TA

188

Instruments DMA Q800 under N2 with liquid N2 cooling and heated at a rate of 10 °C min-1. The UV lamp used for reduction attempts was a Bluepoint 4 Ecocure from Honle

UV America Inc. All experiments were carried out with a 320 – 500 nm filter. High

resolution transmission electron microscopy (TEM) was performed with a Philips Tecnai

F30 operated at 300 kV. X-ray photoelectron spectroscopy (XPS) measurements were

carried out with a PHI VersaProve XPS Microprobe. A monochromatic AlKα (1486.6 eV)

source was used with a spot size of 300 µm.

6.4.3 Extracted Discharge from a Microplasma

In order to extract electrons by way of an extracted discharge, atmospheric-pressure

microplasma was ignited and sustained in Ar gas (flow rate = 150 sccm) between a

stainless-steel capillary tube (Restek, Inc., I.D. = 180 lm, L = 5 cm) cathode and a metal

grid anode by a negatively-biased high voltage direct current (dc) power supply

(Keithley, Inc., Model 246). The microplasma current was controlled by adjusting the

power supply voltage. Afterwards, to extract electrons from the primary microplasma, a

third electrode (3 cm × 3 cm stainless steel plate) was positioned near the anode and

positively biased with a separate high voltage dc power supply (Gamma, Model RR30-

2P). The third electrode voltage could be varied between 0 and ~15 kV to control the

current of extracted discharge from the microplasma. During the experiment, the current

of microplasma and distance between the anode and third electrode (dee) was kept

constant at 4 mA and 0.5 cm, respectively.

189

6.4.4 Synthesis of [Pt2(2)Cl2](Cl2), [Pt2(3)Cl2](Cl2), and [Pt2(6)Cl2](Cl2)

A solution of 2 (250 mg, 0.06 mmoles) in chloroform (40 mL) was added to a suspension

of Pt(DMSO)2Cl2 (50.3 mg, 0.12 mmoles) in methanol (10 mL) under inert atmosphere.

Complexation was carried out at 50 °C for 24 hrs. After removal of the solvent, the

resulting orange solid was washed multiple times with water and THF to give the desired

product in quantitative yield. δH (600 MHz; 1:1 CDCl3:(CD3)2SO) 7.35 (b, 4H, Ar), 7.30

(b, 4H, Ar), 6.99 (b, 4H, Ar), 6.7-6.5 (b, 8H, Ar), 4.42 (t, 4H, -OCH2-), 3.76 (s, 12H, -

NCH3), 3.40, (m, 256H, -OCH2CH2-), 1.61 (m, 260H, -OCH2CH2-).

The same conditions were used for 6 with values of: 6 (50 mg, 0.005 mmoles) and

Pt(DMSO)2Cl2 (4.4 mg, 0.01 mmoles).

The same conditions were used for 3 with values of: 3 (200 mg, 0.05 mmoles) and

Pt(DMSO)2Cl2 (42.2 mg, 0.1 mmoles). δH (600 MHz; CDCl3) 7.50 (d, 4H, Ar), 7.23 (t,

4H, Ar), 7.13 (t, 4H, Ar), 7.04 (s, 4H, Ar), 6.99 (d, 4H, Ar), 4.42 (t, 4H, -OCH2-), 3.77 (s,

12H, -NCH3), 1.48-0.94 (m, 398H, -CH2-; -CH-), 0.94-0.79 (m, 125H, -CHCH3-).

190

1 Figure 6.11 Partial 600 MHz H NMR spectra of a) 2 in CDCl3 at 25 °C and b) [Pt2(2)Cl2](Cl2)

in 1:1 CDCl3:(CD3)2SO at 50 °C. No uncoordinated peaks from 2 are observed in the

spectrum of [Pt2(2)Cl2](Cl2) consistent with complete complexation.

6.4.5 Preparing Thin-Films for Reduction

0.5 mM solutions of [Pt2(2,3 or 6)Cl2](Cl2) in CHCl3 were used to create all films unless

otherwise stated. To avoid the possibility of photo-reduction, the solutions and

subsequent films were keep covered with aluminum foil at all times. Thin films were

prepared by spin-coating onto a carbon-coated TEM grid at 2000 rpm for 60 s. The films

were dried in vacuo at 40 °C overnight. Afterwards, the as-prepared sample was placed in desiccator and cooled to -10 °C overnight before and after treatment with the extracted discharge from a microplasma of the metallosupramolecular polymer films in order to maintain low humidity and low temperature. Films were exposed to the extracted discharge from a microplasma for a variety of times (30, 60, and 180 mins).

191

6.4.6 Preparing Thick Films for Structural Characterization

Thicker films were prepared by solvent-casting as described in previous chapters. 200 mg of [Pt2(2)Cl2](Cl2) or [Pt2(3)Cl2](Cl2) were dissolved in 2 mL of CHCl3 and cast into an aluminum walled casting dish with a Teflon sheet bottom. The solvent was allowed to evaporate at room temperature for 24 hrs and the films were further dried in vacuo at 40

°C overnight to yield mechanically stable films with thicknesses of 180 ± 30 µm.

6.5 ACKNOWLEGDMENTS

This material is based upon work supported by the National Science Foundation under

Grants CHE-0704026, DMR-0602869, and MRI-0821515 (for the purchase of the

MALDI-TOF/TOF), the Case School of Engineering and the Kent H. Smith Charitable

Trust. Special thanks goes to Seung Whan Lee for all of his hard work on the reduction and imaging of the samples, Dr. Mohan Sankaran for his invaluable help in interpreting the data, and Brian Michal for help with the SAXS.

6.6 REFERENCES

1. Shin, D. O.; Jeong, J. -R.; Han, T. H.; Koo, C. M.; Park, H. -J.; Lim, Y. T.; Kim, S. O. J. Mater. Chem., 2010, 20, 7241-7247.

2. Naito, K.; Hieda, H.; Sakurai, M.; Kamata, Y.; Asakawa, K. IEEE Trans. Magn., 2002, 38, 1949-1951.

3. Lee, J. -S.; Cho, J.; Lee, C.; Kim, I.; Park, J.; Kim, Y. -M.; Shin, H.; Lee, J.; Caruso, F. Nat. Nanotechnol., 2007, 2, 790-795.

192

4. Crooks, R. M.; Zhao, M.; Sun, T.; Chechik, V.; Yeung, L. K. Acc. Chem. Res., 2001, 34, 181-190.

5. Heilmann, A. Polymer Films with Embedded Metal Nanoparticles, Springer, Berlin, 2003.

6. Ghosh, K.; Maiti, S. N. J. Appl. Polym. Sci., 1996, 55, 323-331.

7. Ramesh, G. V.; Porel, S.; Radhakrishnan, T. P. Chem. Soc. Rev., 2009, 38, 2646-2656.

8. Shang, L.; Wang, Y.; Huang, L.; Dong, S. Langmuir, 2007, 23, 7738-7744.

9. Rifai, S.; Breen, C. A.; Solis, D. J.; Swager, T. M. Chem. Mater., 2006, 18, 21-25.

10. Sakamoto, M.; Tachikawa, T.; Fujitsuka, M.; Majima, T. Adv. Funct. Mater., 2007, 17, 857-862.

11. Silva, A. M. B.; de Araujo, C. B.; Santos-Silva, S.; Galembeck, A. J. Phys. Chem. Solids, 2007, 68, 729-733.

12. Zhang, J.; Gao, Y. Alvarez-Puebla, R. A.; Buriak, J. M.; Fenniri, H. Adv. Mater., 2006, 18, 3233-3237.

13. Porel, S.; Hebalker, N.; Sreedhar, B.; Radhakrishnan, T. P. Adv. Funct. Mater., 2007, 17, 2550-2556.

14. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176-188.

15. Mejia, M. L.; Agapiou, K.; Yang, X.; Holliday, B. J. J. Am. Chem. Soc., 2009, 131, 18196-18197.

16. Cheng, J. Y.; Ross, C. A.; Chan, V. Z. –H.; Thomas, E. L.; Lammertink, R. G. H.; Vancso, G. J. Adv. Mater., 2001, 13, 1174-1178.

17. Grubbs, R. B. J. Polym. Sci. Part A: Polym. Sci., 2005, 43, 4323-4336.

193

18. McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. V. M. Rotello and S. Thayumanavan, John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178.

19. Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

20. Friese, V. A.; Kurth, D. G. Coord. Chem. Rev., 2008, 252, 199–211.

21. Dobrawa, R.; Würthner, F. J. Polym. Sci. Part A Polym. Chem., 2005, 43, 4981-4995.

22. Schubert, U. S.; Eschbaumer, C. Angew. Chem. Int. Ed., 2002, 41, 2892-2926.

23. Whittell, G. R.; Manners, I. Adv. Mater., 2007, 19, 3439-3468.

24. Kurth, D. G.; Higuchi, M. Soft Matter., 2006, 2, 915-927.

25. Beck, J. B.; Ineman, J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060-5068.

26. Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145-151.

27. Kumpfer, J. R.; Wie, J. J.; Swanson, J. P.; Beyer, F. L.; Mackay, M. E.; Rowan, S. J. Macromolecules, 2012, 45, 473-480.

28. Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334-337.

29. He, J.; Ichinose, I.; Kunitake, T.; Nakao, A.; Shiraishi, Y.; Toshima, N. J. Am. Chem. Soc., 2003, 125, 11034-11040.

30. Aizawa, M.; Buriak, J. M. Chem. Mater., 2007, 19, 5090-5101.

31. Mariotti, D.; Sankaran, R. M. J. Phys. D, 2010, 43, 323001.

32. Premkumar, P. A.; Starostin, S. A.; de Vries, H.; Paffen, R. M.; Creatore, M.; Eijkemans, T. J.; Koenraad, P. M.; van de Sanden, M. C. M. Plasma Process. Polym., 2009, 6, 693-702.

194

33. Gu, M.; Kilduff, J. E.; Belfort, G. Biomaterials, 2012, 33, 1261-1270.

34. Lee, S. W.; Zmani, H.; Feng, P. X. -L.; Sankaran, R. M. J. Vac. Sci. Technol. B, 2012, 30, 010603-1-5.

35. Lee, S. W.; Liang, D.; Gao, X. P. A; Sankaran, R. M. Adv. Funct. Mater., 2011, 21, 2155-2161.

36. Cong, H.; Porco Jr., J. A. ACS Catal., 2012, 2, 65-70.

37. Serna, P.; Boronat, M.; Corma, A. Top. Catal., 2011, 54, 439-446.

38. Lee, I.; Albiter, M. A.; Zhang, Q.; Ge, J.; Yin, Y.; Zaera, F. Phys. Chem. Chem. Phys., 2011, 13, 2449-2456.

39. Liu, Y.; Li, D.; Sun, S. J. Mater. Chem., 2011, 21, 12579-12587.

40. Bhattacharya, R.; Mukherjee, P. Adv. Drug Deliver. Rev., 2008, 60, 1289-1306.

41. Xu, D.; Hawk, J. L.; Loveless, D. M.; Jeon, S. L.; Craig, S. L. Macromolecules, 2010, 43, 3556-3565.

42. Feng, K.; Zuniga, C.; Zhang, Y. –D.; Kim, D.; Barlow, S.; Marder, S. R.; Brédas, J. L.; Weck, M. Macromolecules, 2009, 42, 6855-6864.

43. Wu, P. –T.; Bull, T.; Kim, F. S.; Luscombe, C. K.; Jenekhe, S. A. Macromolecules, 2009, 43, 671-681.

44. Grove, L. J.; Oliver, A. G.; Krause, J. A.; Connick, W. B. Inorg. Chem., 2008, 47, 1408-1410.

45. Field, J. S.; Grimmer, C. D.; Munro, O. Q.; Waldron, B. P. Dalton Trans., 2010, 39, 1558-1567.

46. Du, P.; Schneider, J.; Brennessel, W. W.; Eisenberg, R. Inorg. Chem., 2008, 47, 69- 77.

195

47. Rivera, E. J.; Barbosa, C.; Torres, R.; Grove, L.; Taylor, S.; Connick, W. B.; Clearfield, A.; Colón, J. L. J. Mater. Chem., 2011, 21, 15899-15902.

48. Wang, K.; Haga, M. A.; Monjushiro, H.; Akiba, M.; Sasaki, Y. Inorg. Chem., 2000, 39, 4022-4028.

49. Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565-578.

50. Kumpfer, J. R.; Rowan, S. J. J. Am. Chem. Soc., 2011, 133, 12866-12874.

51. Yam, V.W. –W.; Chan, K. H. –Y.; Wong, K.M. –C.; Chu, B.W. –K. Angew. Chem. Int. Ed., 2006, 45, 6169–6173.

52. Tam, A. Y. Y.; Lam, W. H.; Wong, K. M. C.; Zhu, N.; Yam, V. W. W. Chem. Eur. J., 2008, 14, 4562-4576.

53. Bonet, F.; Delmas, V.; Grugeon, S.; Herrera Urbina, R.; Silvert, P. Y.; Tekaia- Elhsissen, K. Nanostruct. Mater., 1999, 11, 1277-1284.

54. Yang, C. H.; Qiu, H. J. Chem. Phys., 1986, 84, 416-423.

55. Price, J. H.; Williamson, A. N.; Schramm, R. F.; Wayland, B. B. Inorg. Chem., 1972, 11, 1280-1284.

196

CHAPTER 7

Self-Assembly of Metallosupramolecular Polymers in

the Melt

7.1 INTRODUCTION

In recent years the study of metallosupramolecular polymers and networks1,2 has received

a great amount of attention, in part, as these materials are known to display stimuli- responsive properties3,4 and have been used in many different applications ranging from

organic light-emitting diodes (OLEDs)5,6 to sensors.7,8 The ability to target the location

of the self-assembly has also become an area of interest.9,10 Specifically, investigations at

interfaces have become increasingly studied as interactions at polymer interfaces are

important for a variety of applications including layer adhesion,11-13 compatibilization of

immiscible polymer films and blends,14,15 and electronic devices.16 Examples of

interfacial assembly of metallosupramolecular polymers at liquid-liquid,17 air-liquid,18

and solid-liquid19 interfaces have been reported, however to our knowledge no examples

have yet been reported of self-assembly at a solid-solid interface. To further enhance the knowledge of metallosupramolecular polymer assemblies and develop new interfacial materials, we have investigated the ability to form metallosupramolecular polymer networks using solventless conditions at the solid-solid interface between polymer films.

Overwhelmingly, solution-based methods are used in the formation of the metallosupramolecular polymers primarily with the use of hazardous organic solvents.

While there have been systems designed to form metallosupramolecular polymer in aqueous environments, another advantage of forming metallosupramolecular polymers in the solid-state, while maintaining the desired (functional) properties of the metallosupramolecular polymer, is that it would provide a method of processing with reduced costs and decreases the use of hazardous chemicals.

198

Our approach involves the use of polymers with metal-coordinating ligands on its side-chain. Such systems are already well documented.20-22 In solution, polymers

prepared by Potts and coworkers23 with side-chain terpyridine ligands were found to become insoluble upon the addition of a transition metal-ion such as Fe2+ as metal- coordination yielded highly crosslinked metallosupramolecular networks. In a similar manner, Tew and coworkers24 used Cu2+ metal-ions to create highly inter- and intra- molecularly crosslinked polymers which were insoluble but were found to swell in DMF to form gels.

Using this knowledge, the ability to form crosslinked metallosupramolecular polymer networks at the solid-solid interface was investigated. The proposed method of

solid-state self-assembly is shown in Figure 7.1. Two different polymer matrices, one

containing a polymer with ligand side-chains, and one containing a metal salt are brought

into contact above the Tg of the matrix polymers. This should allow for diffusion of the two self-assembling components and, upon finding the complementary component, metal-coordination should occur in the two layer films. The use of a transition metal-ion,

which binds the ligand in a 2:1 fashion, should result in the formation of a crosslinked

polymer network which will consequently greatly hinder any subsequent diffusion. The

result would be a metallosupramolecular polymer formed close to the interface of the two

layers. This method also allows for the possibility of creating multilayered films where

the third layer is made of the metallosupramolecular polymer network at the interface

between the two layers.

199

Figure 7.1 Schematic representation of self-assembly of a metallosupramolecular network arising from diffusion of a ligand-functionalized polymer and metal-ions in polymer matrices when placed in contact above the matrix’s Tg.

7.2 RESULTS AND DISCUSSION

7.2.1 Polymer Synthesis and Characterization

To investigate the potential for self-assembly in a two layer film, a 2,6-bis(N- methylbenzimidazoyl)pyridine (Mebip) side-chain functionalized polystyrene was prepared using reversible addition−fragmentation chain transfer polymerization (RAFT) as shown in Figure 7.2. Random copolymers of styrene and 4-vinylbenzyl chloride were targeted and the molar ratio of the two monomers was kept constant for the two

molecular weights studied to keep the volume fraction of side-chain functionalized units consistent. The two different molecular weight polymers were synthesized by stopping the polymerization after either 6 hrs or 24 hrs to yield number average molecular weights

200

(Mn) of ~10,000 g/mol and ~50,000 g/mol with PDIs less than 1.1 (as determined by

GPC; see Figure 7.3) to give polymers 710 and 750 as shown in Figure 7.2a where the subscript 10 or 50 is used to differentiate the two polymers’ molecular weights. Synthesis

of the Mebip side-chain functionalized polymers was carried out by reacting ligand 8

with the two polymers 710 and 750 to yield polymers 910 and 950, respectively (Figure

7.2b).

Figure 7.2 a) Synthesis of random copolymer 7x by RAFT polymerization and b) functionalization with ligand 8 to yield a side-chain functionalized polymer 9x.

201

Figure 7.3 GPC traces of 710 and 750. The molecular weight (Mn) was found to be 10,000 and 50,000 g/mol respectively with PDIs under 1.1 for both.

All polymers were fully characterized by NMR and GPC. A representative 1H

NMR of 950 is shown in Figure 7.4 as the spectra for 910 and 950 are nearly identical. All

of the observed peaks are quite broad, consistent with typical NMR of polymers, and no

clear peaks from the polymer end-groups are apparent so the molecular weight is not able

to be determined by NMR. It should be noted that it is likely that the trithiocarbonate

end-group is cleaved during the reaction of 8 with 7x as the presence of strong bases are

known to result in hydrolysis.25,26 The integrations of the peaks do, however, allow for estimations of the ratio of functionalized to non-functionalized units along the polymer backbone. The ratio of the integration of peak a to peak f (chosen as it is more separated

from the other peaks) suggests ca. 12% of the units are functionalized in 910 and ca. 14% is 950 which is in good agreement with the monomer feed ratio of styrene to 4-

vinylbenzyl chloride (11%). Confirmation of the functionalization of polymers 7x with 8

was determined by investigating the 13C NMR spectra for each polymer. As shown in

202

Figure 7.5 for 950, peaks which would correspond to the CH2 protons from the free ligand, 8, and the benzylic CH2 next to chlorine from any unreacted starting polymer,

750, have disappeared. New peaks corresponding to the Mebip functionalized polymer,

950, can be clearly seen which demonstrates that a large majority if not all of the reactive side-chains in 750 have been functionalized.

1 Figure 7.4 600 MHz H NMR of a) 710 and b) 910 in CDCl3.

203

13 Figure 7.5 100 MHz C NMR of a) 8 (in 9:1 CDCl3:MeOD), b) 710, and c) 910 in CDCl3. No peaks were observed correlating to the unfunctionalized polymer (710) or ligand (8) which provides evidence for complete functionalization.

The thermal stability of the Mebip-functionalized polymers was also investigated

as the materials will be exposed to temperatures >200 °C during the preparation of the

polymer blends and melt pressing during the self-assembly step. Thermogravimetric

analysis (TGA) was used to determine at what temperatures the polymers showed degradation and the temperature limit for use in preparing blends. Figure 7.6a shows the

TGA curve for 710 (750 was found to be identical). Interestingly, two different peaks are

observed as a result of two different degradations. The first, which occurs at ~370 °C, is

204

from the degradation of the benzyl chloride group from the 4-vinylbenzyl chloride units while the rest of the polymer degrades at ~420 °C. The onset of weight loss begins at

~200 °C which is near the desired processing temperature of 210 °C. TGA of 910,

however, shows that functionalization with 8 gives a polymer with only one degradation

peak at ~405 °C with the onset occurring slightly below 300 °C. This result shows that

the functionalized polymers, 9x, are thermally stable at the desired processing

temperatures.

Figure 7.6 Thermogravimetric analysis (TGA) of a) 710 and b) 910. Polymer 710 displays two degradation peaks which have an onset temperature of ~200 °C while functionalization to give

910 displays only one thermal decomposition with an onset temperature of ~300 °C.

7.2.2 Self-Assembly in the Melt

In order to study the self-assembly in the melt between two polymer films, 9x was

blended into a polystyrene matrix and Zn(NTf2)2 was blended into a poly(methyl

methacrylate) matrix. Blends were prepared at two different concentrations (10% and

25% by weight) to determine if the loading has any effect on the size or location of the

self-assembly. For all samples, blends were prepared in a twin-screw microcompounder at 210 °C and mixed for 30 mins before extruding. The polymer and metal-containing

205

blends were then melt pressed into films 150 ± 50 µm thick to be used in the two-layer

assembly experiments.

Two-layer control films were created by pressing the 910 or 950 in PS (10% and

25% w/w) films with PMMA without any Zn2+-ions. The formation of metallosupramolecular networks was investigated by pressing films of the same loading

2+ of 9x in PS and Zn in PMMA together at 210 °C with a pressure of 2 metric tons (MTs).

2+ Diffusion of 9x and Zn -ions through the matrix above Tg should result in coordination

of the Zn2+ by the Mebip functionalized polymer. Metal coordination results in a change

in the absorption and emission which can be used to monitor the extent of self-assembly.

Figure 7.7 shows the comparison of the change in the absorption spectra from films

2+ prepared with 9x in PS with either PMMA alone or with PMMA containing Zn -ions.

The two-layer films containing Zn2+ display an increase in the absorbance from 335-375

nm (λmax = 340 nm) and a decrease in the free ligand absorbance at 315 nm. This change

in absorbance matches closely with the previously reported Mebip:Zn2+ complexes in

solution and is consistent with the formation of the metallosupramolecular polymer in the

solid-state.

206

Figure 7.7 UV-vis spectra of two-layer films prepared with 910 in PS with PMMA (black line) or with Zn2+ in PMMA (red line) at a) 10% w/w and b) 25% w/w loading. UV-vis spectra for films 2+ prepared with 950 in PS with PMMA (black line) or with Zn in PMMA (red line) at c) 10% w/w and d) 25% w/w loading.

A change in emission was also observed in the photoluminescence spectra when

compared to the control film. Figure 7.8 shows that metal-coordination results in an

increase in the emission intensity below 500 nm with a slight shift in the λmax of ca. 50

nm to ca. 450 nm. While the fluorescence of the uncomplexed polymer looks different

27 from solution (see Chapter 2), prior work and Chapter 2 have shown that λmax of the

Mebip:Zn2+ complex emission occurs between 400-450 nm. Thus the emission data are

also consistent with the formation of the metallosupramolecular polymer.

There are two possible mechanisms for the formation of the

metallosupramolecular polymers, namely the two components diffuse through the

207 matrices and either complex at the processing temperatures or the metallosupramolecular polymer is only formed upon cooling. Therefore, the emission spectra of a film that has been compression molded and contains the metallosupramolecular polymer was studied at elevated temperatures to see if the metallosupramolecular polymer is present at the processing temperatures. Figure 7.9a shows the unnormalized emission profiles for a

2+ 910:PS / Zn :PMMA (10% w/w) two-layer film. Upon heating above the matrix polymers’ Tgs, the emission intensity decreases as expected from increased molecular collisions, non-radiative transmission, and quenching which occur at elevated

28 temperatures; however, more importantly, the PL spectra and λmax (ca. 450 nm) of the films at elevated temperatures match those of the compression molded films at room temperature (Figure 7.8). Also, no significant change in the peak intensity ratios at ca.

450 nm and 500 nm is observed even at 200 °C (Figure 7.9b), which would be expected if some of the metal complex was converted to free ligand. This suggests that the metallosupramolecular polymers are formed during the processing conditions.

208

Figure 7.8 Photoluminescence spectra of two-layer films prepared with 910 in PS with PMMA (black line) or with Zn2+ in PMMA (red line) at a) 10% w/w and b) 25% w/w loading.

Photoluminescence spectra for films prepared with 950 in PS with PMMA (black line) or with Zn2+ in PMMA (red line) at c) 10% w/w and d) 25% w/w loading.

Figure 7.9 a) Unnormalized and b) normalized variable-temperature PL spectra of a two-layer 2+ film of 910:PS / Zn :PMMA (10% w/w) showing evidence of metal-coordination at 200 °C.

209

Figure 7.10a shows an example of the distinctive change in emission upon

formation of metallosupramolecular polymer. A film of the 10% w/w 910 in PS blend

(left side of the picture) was partially overlapped with a film of 10% w/w Zn2+ in PMMA

(right side of the picture) and, upon pressing the two films together, self-assembly of the

metallosupramolecular polymer occurred in the center, overlapped region. The change in

emission after self-assembly when viewed under UV light (λex = 365 nm) is easily

discernible as shown in Figure 7.10a.

Interestingly, a green colored emission was observed after removing the UV light source which gradually fades after a ~2-3 seconds. It was found that exciting the two-

layer film with a handheld UV lamp (λex = 365 nm) for a short time (~1-2 mins) and then

removing the light source produced this effect. This is thought to be phosphorescence

originating from the metallosupramolecular polymer and, as shown in Figure 7.10b, the

green emission is only located at the overlap in the two films and corresponds to the exact

same area where the increased emission seen in Figure 7.10a is found. Interestingly, this

behavior is not seen in the control films or with any other of the metallosupramolecular

polymer films studied in the previous chapters. Others have demonstrated

phosphorescence of Zn2+ complexes with nitrogen-containing ligands29 in glassy

solutions at low temperatures (ca. 77 K) and have attributed it a 3(π – π*) transition in the

metal-ligand complex.30,31 Others have also observed phosphorescence in Ru3+, Pt2+, and

Ir3+ complexes.30,32,33 The mechanism behind the phosphorescence in our two-layered

system is currently unclear; however based on these prior studies it is consistent with the

metal-ligand complexes being able to be excited into the triplet state. The matrix

polymers seem to play a role as this behavior is not seen in the neat

210

metallosupramolecular polymers discussed previously and may be related to the ability of

34 the matrices to hinder quenching (ex. O2) of the phosphorescent emission.

Figure 7.10 a) Image of the change in emission color and intensity upon formation of a 2+ metallosupramolecular network from 910:PS / Zn :PMMA (10%) (λex = 365 nm; the edge of the film is outlined for clarification). Sample was prepared by only overlapping the films in the center 2+ 2+ to highlight the emission change (left: 910 in PS, middle: 910:PS / Zn :PMMA, right: Zn in PMMA). b) Image of the phosphorescence observed in the same film which is only present at the overlapped area. The phosphorescence disappears after ~3 s (λex = 365 nm).

Information on the rate of the metallosupramolecular polymer formation during

processing of films that have been compression molded for set periods of time was

monitored by UV-vis and PL. It was found that for all polymers and loadings the

absorption and emission showed no change after 5 mins as shown in Figure 7.11. Films pressed for only 1 min displayed similar absorbance and emission to the control films, however films pressed for a total of 5 mins displayed increased absorption from 335-375

nm and emission below 500 nm. Pressing for additional time (10 or 15 mins total)

211

showed no significant changes in the spectra other than slight differences in intensity. All

spectra were corrected for the film thickness, however it is likely that the small changes

in intensity result from the change in the film’s thickness after pressing. Although a small

amount of additional self-assembly after 5 mins may take place, the majority of the

complexation takes place between 1-5 mins. Also, there appears to be no significant difference in the rate of formation between two-layer films prepared with 910 vs 950 or

with differences in loading on the time scales measured, however more detailed

investigation in the 1-5 mins time scale may elucidate subtle differences.

Figure 7.11 Influence of contact time on the self-assembly. a) UV-vis spectra of 910:PS / 2+ 2+ Zn :PMMA (10%), b) PL spectra of 910:PS / Zn :PMMA (10%), c) UV-vis spectra of 950:PS / 2+ 2+ Zn :PMMA (10%), and PL spectra of 950:PS / Zn :PMMA (10%) after different total pressing times which show that the absorbance and emission profile are unchanged after only 5 mins.

212

The data above suggests that the metallosupramolecular polymer is formed during

compression molding. It would be expected that once incorporated into the metallosupramolecular polymer network diffusion of the metal-ion and 9x components

would be greatly reduced. As such, one would expect that the majority of the

metallosupramolecular polymer would be formed close to the PS/PMMA interface. To

study this, confocal microscopy was used as it allows for three-dimensional fluorescence

imaging. The samples were excited at 405 nm and the green colored emission (ca. 450-

600 nm) was monitored during the experiment which should show both the free and

complexed 9x although the complexed species should exhibit higher emission intensity

(Figure 7.8). Scans through the depth of the two layer films gave direct evidence of the

location of the self-assembled polymers. Both an edge-on view and a surface view of all

2+ of the two layer films (9x:PS / Zn :PMMA; 10 and 25% w/w) can be seen in Figure 7.12

with the PMMA layer on top and the PS layer on the bottom. No fluorescence was

observed in the PMMA layer, which confirms that 9x does not diffuse into this layer. The

PS layer does show a faint green emission, likely resulting from a small amount of free ligands still present. However the greatest intensity is observed to be located primarily in the PS layer at the interface of the two different polymer layers. These results are consistent with the hypothesis that the Mebip-functionalized polymer and Zn2+ diffuse to

the interface and, upon coordination, result in a metallosupramolecular network, which

greatly hinders further diffusion.

The three-dimensional scans also allow for observation of the extent of assembly,

i.e. continuous layer vs agglomeration. As shown in Figure 7.12, a large area of the

interfacial surface is covered with the metallosupramolecular polymer network and, as

213

2+ would be expected, at higher starting concentrations of 9x and Zn there is evidence of

more metallosupramolecular polymer formed at the interface. All samples appeared to display a random dispersion of agglomerated self-assembled polymer along the interface.

There are many reasons why we see agglomeration instead of a continuous layer such as too little loading, intramolecular coordination, or even phase separation of the metallosupramolecular polymer from the matrices. It is difficult to examine increased loadings (> 25% w/w) as there already appears to be phase separation between the PS

35 matrix and 9x. It is also likely that intramolecular metal-ion coordination is favorable as

nearby Mebip ligands on the same polymer chain would have less distance to diffuse to

find a metal-ion. It is also possible that the self-assembly behaves analogous to nucleation as self-assembly greatly hinders diffusion of the metallosupramolecular polymer network.

Nonetheless, the formation of a metallosupramolecular polymer at the interface between

two films in the melt has been verified using spectroscopy and confocal microscopy.

214

Figure 7.12 3D images using confocal microscopy to look at the green colored fluorescence (ca.

450-600 nm; λex = 405 nm) which shows intense fluorescence from metallosupramolecular 2+ 2+ polymers a) 910:PS / Zn :PMMA (10% w/w), b) 910:PS / Zn :PMMA (25% w/w), c) 950:PS / 2+ 2+ Zn :PMMA (10% w/w), and d) 950:PS / Zn :PMMA (25% w/w) located at/near the interface between the layers from an edge-on view (top) and surface (bottom). Scanning range is marked for clarity with dimensions 365 x 365 x 100 µm.

7.3 CONCLUSIONS

Compression molding of two polymer films, one a blend of a Mebip side-chain functionalized polystyrene in a PS matrix and the other a blend of Zn(NTf2)2 in PMMA,

above the Tg of the matrix polymers resulted in two-layer films which hava a self- assembled metallosupramolecular polymer network at the polymer-polymer interface.

The degree and rates of self-assembly were monitored spectroscopically, and complexation was found to occur within 5 mins and that higher loading of the assembling components into the polymer matrices yielded higher concentrations of the

215

metallosupramolecular polymer at the interface. The ability to design polymers which

will self-assemble in the solid-state at known locations provides a method of targeted

delivery of substances to the interface and the inherent optoelectronic properties of

metallosupramolecular films could make them interesting in devices.

7.4 EXPERIMENTAL METHODS

7.4.1 Materials

All reagents and solvents were purchased form Aldrich Chemical Co. Styrene and 4-

vinylbenzyl chloride was purified by passing them through neutral alumina. All other reagents were used without further purification. The chain-transfer agent (CTA) used for the RAFT polymerizations was prepared according to a slightly modified literature

36,37 procedure. PMMA VS-100 was supplied by Arkema Inc. Zn(NTf2)2 was prepared according to the literature.38

7.4.2 Instruments

NMR spectra were recorded using a Varian 600 NMR spectrometer. UV-vis absorption

spectra were obtained using a Perkin-Elmer Lambda 800 UV-vis spectrometer scanning in the range of 300-400 nm with an integration time of 0.24 s. Photoluminescence spectra were acquired using an Ocean Optics ACD1000-USB spectrometer (λex = 377 nm)

through the use of a Y-shaped optical fiber in conjunction with a Gel Instrumente AG

hotstage with a TC2 temperature controller with a temperature range of 25 to 200 °C.

Thermogravimetric analysis (TGA) experiments were done using a TA Instruments TGA

Q500 under N2 with a heating rate of 10 °C/min. Size exclusion chromatography (SEC)

216

was performed on a Varian ProStar 210/215 equipped with a Varian ProStar 350/352

Refractive Index Detector, a Viscotek Model 270 Dual Detector, and Varian OligoPore

and ResiPore SEC/GPC columns running tetrahydrofuran (THF) at a flow rate of 1.0 mL/min. Confocal microscopy was done on an Olympus FV1000 filter-based confocal

scan head coupled to an Olympus BX62 upright motorized microscope (Olympus Corp.,

Tokyo, Japan). Four lasers provide a total of six different wavelengths (405, 458, 488,

514.5, 543 and 633 nm), which can be combined into a single beam prior to imaging.

7.4.3 Synthesis of 710

4-cyano-4-(dodecylsulfanylthiocarbonyl)sulfanyl pentanoic acid (CTA) (641.6 mg, 1.6 mmol) and azobisisobutyronitrile (AIBN) (25.4 mg, 0.15 mmol) were placed in a 50 mL round bottom flask and purged with Ar. Styrene (18.8 mL, 164 mmol) and 4-vinylbenzyl choloride (2.8 mL, 20 mmol) were added by syringe and the flask was again purged with

Ar. The reaction was put through three freeze, pump, thaw cycles and purged once more with Ar before placing in a preheated 70 °C oil bath. The reaction was carried out for 6 hrs before cooling to room temperature. CHCl3 (30 mL) was added to dissolve the

product which was then precipitated out of stirring methanol (2 L). The product was filtered and dried to yield 5.02 g of 710 (25% conversion). Mn = 10,200 g/ml by GPC.

PDI = 1.09.

217

7.4.4 Synthesis of 750

750 was prepared in the same manner with 646.1 mg CTA, 24.3 mg AIBN, 18.8 mL

styrene, 2.8 mL 4-vinylbenzyl chloride, and allowed to polymerize for 24 hrs to yield

11.3 g (56% conversion). Mn = 50,700 g/ml by GPC. PDI = 1.10.

7.4.5 Synthesis of 8

4-hydroxy-2,6-bis(N-methylbenzimidazoyl)pyridine39 (HOMebip, 10.5 g, 30 mmol) and

K2CO3 (16 g, 116 mmol) were suspended in 150 mL dry THF under Ar. 2-bromoethanol

(18.5 g, 148 mmol) was then added while stirring and refluxed at 85 °C for 24 hrs. The reaction was cooled and the solvent was removed in vacuo. The solids were suspended in boiling toluene with stirring and filtered directly from the boiling suspension. The product was recrystallized from toluene, filtered, washed with cold toluene, and dried to

yield 10.35 g of 8 (72%). δH (600 MHz, CDCl3:CD3OD 9:1) 7.42 (s, 2H, Ar), 7.85 (d,

2H, Ar), 7.44 (d, 2H, Ar), 7.36 (t, 2H, Ar), 7.33 (t, 2H, Ar), 4.24 (t, 2H, -OCH2-), 4.21 (s,

6H, -NCH3), 4.02 (t, 2H, -OCH2CH2OH); δC (100 MHz, CDCl3:CD3OD 90:10) 166.5,

150.5, 149.9, 141.7, 123.8, 125.2, 123.1, 119.5, 111.8, 110.1, 70.2, 60.2, 32.4.

218

1 Figure 7.13 600 MHz H NMR of 8 in CDCl3:CD3OD 9:1.

13 Figure 7.14 100 MHz C NMR of 8 in CDCl3:CD3OD 9:1.

219

Figure 7.15 Nanostructured Array Laser Desorption/Ionization (NALDI) spectrum of 8 [M+H].

7.4.6 Synthesis of 910

8 (522.6 mg, mmol), 18-crown-6 (687.6 mg, mmol), and sodium hydride (60 mg) were suspended in 125 mL dry THF and purged with Ar. 710 (4.93 g, mmol) was added and the

reaction was stirred at room temperature for 24 hrs. The THF was removed in vacuo, the

product dissolved in 20 mL of CHCl3 and precipitated out of stirring methanol (2 L). The solid was filtered and dried to give 4.45 g of 910 (90%).

7.4.7 Synthesis of 950

950 was prepared in the same manner with 1.00 g 8, 1.42 g 18-crown-6, 120 mg sodium

hydride, and 10.13 g 750 to yield 9.9 g of product (93%).

220

7.4.8 Preparation of Blends

Blends were prepared using a DSM Research Micro5 microextruder by mixing 10 or 25

% by weight of 910 and 950 with PS or Zn(NTf2)2 with PMMA. Samples were mixed at

210 °C and 100 rpm for 30 mins before collecting.

7.4.9 Example Two-Layer Assembly

Films of the extruded material were pressed at 210 °C using a Carver hydraulic

laboratory press at 5 metric tons. A PS film containing 10% w/w 910 was placed on top of

a PMMA containing 10% w/w Zn2+ and pressed together using the hydraulic press. Films

were compressed at 210 °C and a pressure of 2 metric tons as it produced the best looking films, however it should be noted that self-assembly occurs even with no pressure.

7.5 ACKNOWLEDGMENTS

This material is based upon work supported by the National Science Foundation under

Grants CHE-0704026, and DMR-0602869, the Case School of Engineering and the Kent

H. Smith Charitable Trust. Special thanks to David Hovis and the Swagelok Center for

Surface Analysis of Materials for imaging help with the confocal microscopy.

7.6 REFERENCES

1. Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565-578.

2. Whittell, G. R.; Manners, I. Adv. Mater., 2007, 19, 3439-3468.

221

3. McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. Rotello, V. M.; Thayumanavan, S. John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178.

4. Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176- 188.

5. Wu, F. –I.; Yang, X. –H.; Neher, D.; Dodda, R.; Tseng, Y. –H.; Shu, C. –F. Adv. Funct. Mater., 2007, 17, 1786-1798.

6. Burnworth, M. Mendez, J. D.; Schroeter, M.; Rowan, S. J.; Weder, C. Macromolecules, 2008, 41, 2157-2163.

7. Holliday, B. J.; Stanford, T. B.; Swager, T. M. Chem. Mater., 2006, 18, 5649-5651.

8. Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145–151.

9. Ariga, K.; Hill, J. P.; Lee, M. V.; Vinu, A.; Charvet, R.; Acharya, S. Sci. Technol. Adv. Mater., 2008, 9, 014109.

10. Krishnaswamy, R.; Sood, A. K. J. Mater. Chem., 2010, 20, 3539-3552.

11. Laurila, T.; Vuorinen, V.; Mattila, T. T.; Turunen, M.; Paulasto-Kröckel, M.; Kivilahti, J. K. Interfacial Compatibility in Microelectronics, Springer-Verlag, London, 2012, Chap. 5, 101-133.

12. Davey, R. J.; Williams-Seton, L.; Lieberman, H. F.; Blagden, N. Nature, 1999, 402, 797-799.

13. Zhang, J.; Lodge, T. P.; Macosko, C. W. J. Polym. Sci. Part B: Polym. Phys., 2009, 47, 2313-2319.

14. Wang, D.; Fujinami, S.; Liu, H.; Nakajima, K.; Nishi, T. Macromolecules, 2010, 43, 5521-5523.

15. Jiang, G.; Wu, H.; Guo, S. Polym. Eng. Sci., 2010, 50, 2273-2286.

222

16. Peet, J.; Senatore, M. L.; Heeger, A. J.; Baan, G. C. Adv. Mater., 2009, 21, 1521- 1527.

17. Puigmarti-Luis, J.; Rubio-Martínez, M.; Hartfelder, U.; Imaz, I.; Maspoch, D.; Dittrich, P. S. J. Am. Chem. Soc., 2011, 133, 4216-4219.

18. Lehmann, P.; Symietz, C.; Brezesinski, G.; Krass, H.; Kurth, D. G. Langmuir, 2005, 21, 5901-5906.

19. Krass, H.; Papastavrou, G.; Kurth, D. G. Chem. Mater., 2003, 15, 196-203.

20. Pollino, J. M.; Weck, M. Chem. Soc. Rev., 2005, 34, 193-207.

21. Aamer, K. A.; Tew, G. N. Macromolecules, 2007, 40, 2737-2744.

22. Carlise, J. R.; Weck, M. J. Polym. Sci. Part A: Pol. Chem., 2004, 42, 2973-2984.

23. Potts, K. T.; Usifer, D. A. Macromolecules, 1988, 21, 1985-1991.

24. Tew, G. N.; Aamer, K. A.; Shunmugam, R. Polymer, 2005, 46, 8440-8447.

25. Lima, V.; Jiang, X.; Brokken-Zijp, J.; Schoenmakers, P. J.; Klumperman, B.; van der Linde, R. J. Polym. Sci. Part A: Polym. Chem., 2005, 43, 959-973.

26. Barner, L.; Perrier, S. Handbook of RAFT Polymerization. Ed. Barner-Kowollik, C., Wiley-VCH Verlag GmbH & Co., KGaA, Weinheim, 2008.

27. Beck, J. B.; Ineman, J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060-5068.

28. Kumar Rai, V. Appl. Phys. B., 2007, 88, 297-303.

29. Liu, Q. –D.; Wang, R.; Wang, S. Dalton Trans., 2004, 2073-2079.

30. Bose, D.; Rahaman, S. H.; Mostafa, G.; Bailey Walsh, R. D.; Zaworotko, M. J.; Ghosh, B. K. Polyhedron, 2004, 23, 545-552.

223

31. Ho, K. –Y.; Yu, W. –Y.; Cheung, K. –K.; Che, C. –M. Dalton Trans., 1999, 1581- 1586.

32. Diring, S.; Ziessel, R.; Barigelletti, F.; Barbieri, A.; Ventura, B. Chem. Eur. J., 2010, 16, 9226-9236.

33. Zhao, Q.; Liu, S. –J.; Huang, W. Macromol. Rapid Commun., 2010, 31, 794-807.

34. Xie, Z.; Ma, L.; deKrafft, K. E.; Jin, A.; Lin, W. J. Am. Chem. Soc., 2010, 132, 922- 923.

35. Hanabusa, K.; Nakano, K.; Koyama, T.; Shirai, H.; Hojo, N.; Kurose, A. Makromol. Chem., 1990, 191, 391-396.

36. Weber, W. G.; McLeary, J. B.; Sanderson, R. D. Tetrahedron Lett., 2006, 47, 4771- 4774.

37. Moad, G.; Chong, Y. K.; Postma, A.; Rizzardo, E.; Thang, S. H. Polymer, 2005, 46, 8458-8468.

38. Nie, J.; Kobayashi, H.; Sonoda, T. Catal. Today, 1997, 36, 81–84.

39. Beck, J. B.; Rowan, S. J. Faraday Discuss., 2005, 128, 43-53.

224

Bibliography

Aamer, K. A.; Tew, G. N. Macromolecules, 2007, 40, 2737-2744.

Abe, T.; Itakura, T.; Ikeda, N.; Shinozaki, K. Dalton Trans., 2009, 4, 711-715.

Adarsh, N.N.; Dastidar, P. Cryst. Growth. Des., 2011, 11, 328–336.

Ahn, S.; Deshmukh, P.; Kasi, R. M. Macromolecules, 2010, 43, 7330–7340.

Aida, T.; Meijer, E. W.; Stupp, S. I. Science, 2012, 335, 813-817.

Aizawa, M.; Buriak, J. M. Chem. Mater., 2007, 19, 5090-5101.

Anderson, B. M.; Hurst, S. K. Eur. J. Inorg. Chem., 2009, 21, 3041-3054.

Appel, W. P. J.; Nieuwenhuizen, M. M. L.; Meijer, E. W. Multiple Hydrogen-Bonded Supramolecular Polymers, in Supramolecular Polymer Chemistry, ed. Harada, A.; Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2011.

Ariga, K.; Hill, J. P.; Lee, M. V.; Vinu, A.; Charvet, R.; Acharya, S. Sci. Technol. Adv. Mater., 2008, 9, 014109.

Armstrong, G.; Buggy, M. J. Mater. Sci., 2005, 40, 547–559.

Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev., 2010, 110, 1857-1959.

Astruc, D.; Chardac, F. Chem. Rev., 2001, 101, 2991-3024.

Barner, L.; Perrier, S. Handbook of RAFT Polymerization. Ed. Barner-Kowollik, C., Wiley-VCH Verlag GmbH & Co., KGaA, Weinheim, 2008.

Batabyal, S. K.; Leong, W. L.; Vittal, J. J. Langmuir, 2010, 26, 7464–7468.

Beck, J. B.; Ineman, J. M.; Rowan, S. J. Macromolecules, 2005, 38, 5060-5068.

226

Beck, J. B.; Rowan, S. J. Faraday Discuss., 2005, 128, 43-53.

Beck, J. B.; Rowan, S. J. J. Am. Chem. Soc., 2003, 125, 13922– 13923.

Behl, M.; Razzaq, M. Y.; Lendlein, A. Adv. Mater., 2010, 22, 3388–3410.

Beiermann, B. A.; Kramer, S. L. B.; Moore, J. S.; White, S. R.; Sottos, N. R. ACS Macro Lett., 2012, 1, 163-166.

Bender, J. L.; Corbin, P. S.; Fraser, C. L.; Metcalf, D. H.; Richardson, F. S.; Thomas, E. L.; Urbas, A. M. J. Am. Chem. Soc., 2002, 124, 8526–8527.

Bhattacharya, R.; Mukherjee, P. Adv. Drug Deliver. Rev., 2008, 60, 1289-1306.

Binder, W. H.; Zirbs, R. Adv. Polym. Sci., 2007, 207, 1–78.

Binnemans, K. Liquid-crystalline lanthanide complexes in Inorganic Chemistry in Focus II, ed. Meyer, G.; Naumann, D.; Wesemann, L. Wiley-VCH, Weinheim, 2005, ch. 15, pp. 267–291.

Binnemans, K.; Goerller-Walrand, C. Chem. Rev., 2002, 102, 2303–2345.

Bonet, F.; Delmas, V.; Grugeon, S.; Herrera Urbina, R.; Silvert, P. Y.; Tekaia-Elhsissen, K. Nanostruct. Mater., 1999, 11, 1277-1284.

Bonnet, S.; Collin, J. –P.; Gruber, N.; Sauvage, J. –P.; Schofield, E. R. Dalton Trans., 2003, 4654-4662.

Bose, D.; Rahaman, S. H.; Mostafa, G.; Bailey Walsh, R. D.; Zaworotko, M. J.; Ghosh, B. K. Polyhedron, 2004, 23, 545-552.

Bouteiller, L. Adv. Polym. Sci., 2007, 207, 79–112.

227

Buckley, P. R.; McKinley, G. H.; Wilson, T. S.; Small, W., IV; Benett, W. J.; Bearinger, J. P.; McElfresh, M. W.; Maitland, D. J. IEEE Trans. Bio-Med. Electron., 2006, 53, 2075–2083.

Bünzli, J. -C. G.; Piguet, C. Chem. Soc. Rev., 2005, 34, 1048–1077.

Burattini, S.; Colquhoun, H. M.; Fox, J. D.; Friedmann, D.; Greenland, B. W.; Harris, P. J. F.; Hayes, W.; Mackay, M. E.; Rowan, S. J. Chem. Commun., 2009, 6717-6719.

Burnworth, M. Mendez, J. D.; Schroeter, M.; Rowan, S. J.; Weder, C. Macromolecules, 2008, 41, 2157-2163.

Burnworth, M.; Rowan, S. J.; Weder, C. Chem. Eur. J., 2007, 13, 7828–7836.

Burnworth, M.; Tang, L.; Kumpfer, J. R.; Duncan, A. J.; Beyer, F. L.; Fiore, G. L.; Rowan, S. J.; Weder, C. Nature, 2011, 472, 334-337.

Carlise, J. R.; Weck, M. J. Polym. Sci. Part A: Polym. Chem., 2004, 42, 2973-2984.

Che, C. M.; Fu, W. F.; Lai, S. W.; Hou, Y. J.; Liu, Y. L. Chem. Commun., 2003, 118-119.

Chen, C. -C.; Dormidontova, E. E. J. Am. Chem. Soc., 2004, 126, 14972–14978.

Chen, M.; Xing, X.; Liu, Z.; Zhu, Y.; Liu, H.; Yu, Y.; Cheng, F. Appl. Phys. A: Mater. Sci. Process., 2010, 100, 39–43.

Chen, S.; Cao, Q.; Jing, B.; Cai, Y.; Liu, P.; Hu, J. J. Appl. Polym. Sci., 2006, 102, 5224– 5231.

Chen, S.; Hu, J.; Liu, Y.; Liem, H.; Zhu, Y.; Liu, Y. J. Polym. Sci., Part B: Polym. Phys., 2007, 45, 444–454.

Chen, S.; Hu, J.; Zhuo, H.; Yuen, C.; Chan, L. Polymer, 2010, 51, 240–248.

228

Cheng, J. Y.; Ross, C. A.; Chan, V. Z. –H.; Thomas, E. L.; Lammertink, R. G. H.; Vancso, G. J. Adv. Mater., 2001, 13, 1174-1178.

Cheong, S.; Watt, J. D.; Tilley, R. D. Nanoscale, 2010, 2, 2045-2053.

Chiper, M.; Hoogenboom, R.; Schubert, U. S. Macromol. Rapid Commun., 2009, 30, 565-578.

Chow, C. F.; Fujii, S.; Lehn, J. –M. Angew. Chem. Int. Ed., 2007, 46, 5007-5010.

Chow, C. F.; Fujii, S.; Lehn, J. –M. Chem. Asian J., 2008, 3, 1324-1335.

Chung, S. K.; Tseng, Y. R.; Chen, C. Y.; Sun, S. S. Inorg. Chem., 2011, 50, 2711-2713.

Ciferri, A. Macromol. Rapid Commun., 2002, 23, 511–529.

Cong, H.; Porco Jr., J. A. ACS Catal., 2012, 2, 65-70.

Connick, W. B.; Marsh, R. E.; Schaefer, W. P.; Gray, H. B. Inorg. Chem., 1997, 36, 913- 922.

Courtois, J.; Baroudi, I.; Nouvel, N.; Degrandi, E.; Pensen, S.; Ducouret, G.; Chanéac, C.; Bouteiller, L.; Creton, C. Adv. Funct. Mater., 2010, 20, 1803–1811.

Crooks, R. M.; Zhao, M.; Sun, T.; Chechik, V.; Yeung, L. K. Acc. Chem. Res., 2001, 34, 181-190.

Curry, R. J.; Gillin, W. P. Curr. Opin. Solid State Mater. Sci., 2001, 5, 481–486.

Dai, J.; Bruening, M. L. Nano Lett. 2002, 2, 497-501.

Davey, R. J.; Williams-Seton, L.; Lieberman, H. F.; Blagden, N. Nature, 1999, 402, 797- 799.

229

De Greef, T. F. A.; Smulders, M. M. J.; Wolffs, M.; Schenning, A. P. H. J.; Sijbesma, R. P.; Meijer, E. W. Chem. Rev., 2009, 109, 5687–5754.

Develay, S.; Williams, J. A. G. Dalton Trans., 2008, 34, 4562-4564.

Diring, S.; Ziessel, R.; Barigelletti, F.; Barbieri, A.; Ventura, B. Chem. Eur. J., 2010, 16, 9226-9236.

Dobrawa, R.; Würthner, F. J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4981–4995.

Du, P. Inorg. Chim. Acta, 2010, 363, 1355-1358.

Du, P.; Schneider, J.; Brennessel, W. W.; Eisenberg, R. Inorg. Chem., 2008, 47, 69-77.

Eisenbach, C. D.; Schubert, U. S. Macromolecules, 1993, 26, 7372-7374.

Elkins, C. L.; Park, T.; Mckee, M. G.; Long, T. E. J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4618–4631.

Eryazici, I.; Moorefield, C. N.; Newkome, G. R. Chem. Rev., 2008, 108, 1834-1895.

Escande, A.; Guénée, L.; Buchwalder, K. -L.; Piguet, C. Inorg. Chem., 2009, 48, 1132– 1147.

Fages, F. Angew. Chem., Int. Ed., 2006, 45, 1680-1682.

Feng, K.; Zuniga, C.; Zhang, Y. –D.; Kim, D.; Barlow, S.; Marder, S. R.; Brédas, J. L.; Weck, M. Macromolecules, 2009, 42, 6855-6864.

Ferry, J. D. Viscoelastic Properties of Polymers, 3rd ed.; John Wiley & Sons, Inc.: New York, 1980.

230

Field, J. S.; Grimmer, C. D.; Munro, O. Q.; Waldron, B. P. Dalton Trans., 2010, 39, 1558-1567.

Field, J. S.; Haines, R. J.; Ledwaba, L. P.; McGuire Jr., R.; Munro, O. Q.; Low, M. R.; McMillin, D. R. Dalton Trans., 2007, 2, 192-199.

Fluegel, A.;Varshneya,A.K.; Earl,D.A.; Seward,T. P.; Oksoy, D. Ceram. Trans., 2005, 170, 129–143.

Forniés, J.; Fuertes, S.; López, J. A.; Martín, A.; Sicilia, V. Inorg. Chem., 2008, 47, 7166- 7176.

Fox, J. D.; Rowan, S. J. Macromolecules, 2009, 42, 6823–6835.

Friese, V. A.; Kurth, D. G. Coordin. Chem. Rev., 2008, 252, 199-211.

Friese, V. A.; Kurth, D. G. Curr. Opin. Colloid Interface Sci., 2009, 14, 81–93.

Fustin, C. –A.; Guillet, P.; Schubert, U. S.; Gohy, J. –F. Adv. Mater., 2007, 19, 1665- 1673.

Gasnier, A.; Royal, G.; Terech, P. Langmuir, 2009, 25, 8751-8762.

Ghosh, K.; Maiti, S. N. J. Appl. Polym. Sci., 1996, 55, 323-331.

Godlewska, P.; Macalik, L.; Hanuza, J. J. Alloys Compd., 2008, 451, 236–239.

Grondin, P.; Roubeau, O.; Castro, M.; Saadaoui, H.; Colin, A.; Clérac, R. Langmuir, 2010, 26, 5184–5195.

Grove, L. J.; Oliver, A. G.; Krause, J. A.; Connick, W. B. Inorg. Chem., 2008, 47, 1408- 1410.

231

Grove, L. J.; Rennekamp, J. M.; Jude, H.; Connick, W. B. J. Am. Chem. Soc., 2004, 126, 1594-1595.

Grubbs, R. B. J. Polym. Sci. Part A: Polym. Sci., 2005, 43, 4323-4336.

Gu, M.; Kilduff, J. E.; Belfort, G. Biomaterials, 2012, 33, 1261-1270.

Guan, Y.; Cao, Y.; Peng, Y.; Xu, J.; Chen, A. S. C. Chem. Commun., 2001, 1694–1695.

Guillet, P.; Fustin, C. –A.; Lohmeijer, B. G. G.; Schubert, U. S.; Gohy, J. –F. Macromolecules, 2006, 39, 5484-5488.

Gunnlaugsson, T.; Leonard, J. P. Chem. Commun., 2005, 3114–3131.

Hammond, M. R.; Andreopoulou, A. K.; Pefkianakis, E.; Kallitsis, J. K.; Mezzenga, R. Chem. Mater., 2009, 21, 2169-2172.

Han, S.-I.; Gu, B. H.; Nam, K. H.; Im, S. J.; Kim, S. C.; Im, S. S. Polymer, 2007, 48, 1830–1834.

Hanabusa, K.; Nakano, K.; Koyama, T.; Shirai, H.; Hojo, N.; Kurose, A. Makromol. Chem., 1990, 191, 391-396.

Handl, H. L.; Gillies, R. J. Life Sci., 2005, 77, 361–371.

Harada, A.; Takashima, Y.; Yamaguchi, H. Chem. Soc. Rev., 2009, 38, 875–882.

He, J.; Ichinose, I.; Kunitake, T.; Nakao, A.; Shiraishi, Y.; Toshima, N. J. Am. Chem. Soc., 2003, 125, 11034-11040.

Heilmann, A. Polymer Films with Embedded Metal Nanoparticles, Springer, Berlin, 2003.

232

Herbst, F.; Schröter, K.;Gunkel, I.;Gröger, S.; Thurn-Albrecht,T.; Balbach, J.; Binder, W. H. Macromolecules, 2010, 43, 10006–10016.

Ho, K. –Y.; Yu, W. –Y.; Cheung, K. –K.; Che, C. –M. Dalton Trans., 1999, 1581-1586.

Hofmeier, H.; Schubert, U. S. Chem. Soc. Rev., 2004, 33, 373–399.

Holliday, B. J.; Stanford, T. B.; Swager, T. M. Chem. Mater., 2006, 18, 5649-5651.

Holyer, R. H.; Hubbard, C. D.; Kettle, S. F. A.; Wilkins, R. G. Inorg. Chem., 1966, 5, 622-625.

Hoogenboom, R.; Schubert, U. S. Chem. Soc. Rev., 2006, 35, 622-629.

Houlding, V. H.; Miskowski, V. M. Coordin. Chem. Rev., 1991, 111, 145-152.

Hsu, L.; Weder, C.; Rowan, S. J. J. Mater. Chem., 2011, 21, 2812–2822.

Huang, W. M. Open Med. Devices J., 2010, 2, 11–19.

Hudson, Z. M.; Sun, C.; Harris, K. J.; Lucier, B. E. G.; Schurko, R. W.; Wang, S. Inorg. Chem., 2011, 50, 3447-3457.

Ilavsky, J.; Jemian, P. R. J. Appl. Cyrstallogr., 2009, 42, 347.

Ishida, K.; Yoshie, N. Macromolecules, 2008, 41, 4753–4757.

Iyer, P.; Beck, J. B.; Rowan, S. J.; Weder, C. Chem. Commun., 2005, 3, 319–321.

Ji, Z.; Li, Y.; Pritchett, T. M.; Makarov, N. S.; Haley, J. E.; Li, Z.; Drobizhev, M.; Rebane, A.; Sun, W. Chem. Eur. J., 2011, 17, 2479-2491.

Jiang, G.; Wu, H.; Guo, S. Polym. Eng. Sci., 2010, 50, 2273-2286.

233

Jiang, H.; Kelch, S.; Lendlein, A. Adv. Mater., 2006, 18, 1471–1475.

Kade, M. J.; Burke, D. J.; Hawker, C. J. J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 743–750.

Kato, M. Bull. Chem. Soc. Jpn., 2007, 80, 287-294.

Kawano, S.; Fujita, N.; Shinkai, S. J. Am. Chem. Soc., 2004, 126, 8592–8593.

Kelch, S.; Rehahn, M. Macromolecules, 1999, 32, 5818-5828.

Kido, J.; Okamoto, Y. Chem. Rev., 2002, 102, 2357–2368.

Kim, H. -J.; Lee, J. -H.; Lee, M. Angew. Chem., Int. Ed., 2005, 44, 5810–5814.

Kim, H. -J.; Lim, Y. -B.; Lee, M. J. Polym. Sci., Part A: Polym. Chem., 2008, 46, 1925– 1935.

Kim, H. -J.; Zin W. -C.; Lee, M. J. Am. Chem. Soc., 2004, 126, 7009–7014.

Kishimura, A.; Yamashita, T.; Yamaguchi, K.; Aida, T. Nature Mater., 2005, 4, 546-549.

Kishimura, A.; Yamishita, T.; Aida, T. J. Am. Chem. Soc., 2005, 126, 179–183.

Knapton, D.; Burnworth, M.; Rowan, S. J.; Weder, C. Angew. Chem., Int. Ed., 2006, 45, 5825–5829.

Knapton, D.; Rowan, S. J.; Weder, C. Macromolecules, 2006, 39, 651–657.

Kobayashi, A.; Fukuzawa, Y.; Noro, S. I.; Nakamura, T.; Kato, M. Chem. Lett., 2009, 38, 998-999.

234

Koerner, H.; Price, G.; Pearce, N. A.; Alexander, M.; Vaia, R. A. Nat. Mater., 2004, 3, 115–120.

Kozhevnikov, V. N.; Donnio, B.; Bruce, D. W. Angew. Chem. Int. Ed., 2008, 47, 6286- 6289.

Krass, H.; Papastavrou, G.; Kurth, D. G. Chem. Mater., 2003, 15, 196-203.

Krishnaswamy, R.; Sood, A. K. J. Mater. Chem., 2010, 20, 3539-3552.

Kumar Rai, V. Appl. Phys. B., 2007, 88, 297-303.

Kumpfer, J. R.; Jin, J.; Rowan, S. J. J. Mater. Chem., 2010, 20, 145-151.

Kumpfer, J. R.; Rowan, S. J. J. Am. Chem. Soc., 2011, 133, 12866-12874.

Kumpfer, J. R.; Wie, J. J.; Swanson, J. P.; Beyer, F. L.; Mackay, M. M.; Rowan, S. J. Macromolecules, 2012, 45, 473-480.

Kunzelman, J.; Chung, T.; Mather, P. T.; Weder, C. J. Mater. Chem., 2008, 18, 1082– 1086.

Kunzelman, J.; Gupta, M.; Crenshaw, B. R.; Schiraldi, D. A.; Weder, C. Macromol. Mater. Eng., 2009, 294, 244-249.

Kuriki, K.; Koike Y.; Okamoto, Y. Chem. Rev., 2002, 102, 2347–2356.

Kuroiwa, K.; Shibata, T.; Takada, A.; Nemoto, N.; Kimizuka, N. J. Am. Chem. Soc., 2004, 126, 2016–2021.

Kurth, D. G.; Higuchi, M. Soft Matter., 2006, 2, 915-927.

235

Kushner, A. M.; Vossler, J. D.; Williams, G. A.; Guan, Z. J. Am. Chem. Soc., 2009, 131, 8766–8768.

Lahn, B.; Rehahn, M. Macromol. Symp., 2001, 163, 157-176.

Larson, R. G. Constitutive Equations for Polymer Melts and Solutions; Butterworths: Boston, 1988.

Laurila, T.; Vuorinen, V.; Mattila, T. T.; Turunen, M.; Paulasto-Kröckel, M.; Kivilahti, J. K. Interfacial Compatibility in Microelectronics, Springer-Verlag, London, 2012, Chap. 5, 101-133.

Lee, I.; Albiter, M. A.; Zhang, Q.; Ge, J.; Yin, Y.; Zaera, F. Phys. Chem. Chem. Phys., 2011, 13, 2449-2456.

Lee, J. H.; Lee, H.; Seo, S.; Jaworski, J.; Seo, M. L.; Kang, S.; Lee, J. Y.; Jung, J. H. New J. Chem., 2011, 35, 1054–1059.

Lee, J. -S.; Cho, J.; Lee, C.; Kim, I.; Park, J.; Kim, Y. -M.; Shin, H.; Lee, J.; Caruso, F. Nat. Nanotechnol., 2007, 2, 790-795.

Lee, S. W.; Liang, D.; Gao, X. P. A; Sankaran, R. M. Adv. Funct. Mater., 2011, 21, 2155- 2161.

Lee, S. W.; Zmani, H.; Feng, P. X. -L.; Sankaran, R. M. J. Vac. Sci. Technol. B, 2012, 30, 010603-1-5.

Lehmann, P.; Symietz, C.; Brezesinski, G.; Krass, H.; Kurth, D. G. Langmuir, 2005, 21, 5901-5906.

Lehn, J. -M. Chem. Soc. Rev., 2007, 36, 151–160.

Lehn, J. -M. Prog. Polym. Sci., 2005, 30, 814–831.

236

Lendlein, A.; Jiang, H.; Jünger, O.; Langer, R. Nature, 2005, 434, 879–882.

Lendlein, A.; Kelch, S. Angew. Chem., Int. Ed., 2002, 41, 2034–2057.

Lendlein, A.; Shastri, V. P. Adv. Mater., 2010, 22, 3344–3347.

Leng, J.; Lan, X.; Liu, Y.; Du, S. Prog. Mater. Sci., 2011, 56, 1077–1135.

Leng, J.; Wu, X.; Liu, Y. J. Appl. Polym. Sci., 2009, 114, 2455–2460.

Leng, J.; Zhang, D.; Liu, Y.; Yu, K.; Lan, X. Appl. Phys. Lett., 2010, 96, 111905/1– 111905/3.

Li, J.; Viveros, A.; Wrue, M. H.; Anthamatten, M. Adv. Mater., 2007, 19, 2851–2855.

Li, K.; Chen, Y.; Lu, W.; Zhu, N.; Che, C. M. Chem. Eur. J., 2011, 17, 4109-4112.

Li, X. H.; Wu, L. Z.; Zhang, L. P.; Tung, C. H.; Che, C. M. Chem. Commun., 2001, 2280-2281.

Li, Z.; Djohari, H.; Dormidontova, E. E. J. Chem. Phys., 2010, 133, 184904-1-9.

Lima, V.; Jiang, X.; Brokken-Zijp, J.; Schoenmakers, P. J.; Klumperman, B.; van der Linde, R. J. Polym. Sci. Part A: Polym. Chem., 2005, 43, 959-973.

Lis, S. J. Alloys Compd., 2002, 341, 45–50.

Liu, C.; Qin, H.; Mather, P. T. J. Mater. Chem., 2007, 17, 1543–1558.

Liu, G.; Ding, X.; Cao, Y.; Zheng, Z; Peng, Y. Macromol. Rapid Commun., 2005, 26, 649-652.

Liu, J.; Chen, G.; Goo, M.; Jiang, M. Macromolecules, 2010, 43, 8086–8093.

237

Liu, Q. –D.; Wang, R.; Wang, S. Dalton Trans., 2004, 2073-2079.

Liu, Y.; Li, D.; Sun, S. J. Mater. Chem., 2011, 21, 12579-12587.

Lott, J.; Weder, C. Macromol. Chem. Phys., 2010, 211, 28-34.

Loveless, D. M.; Jeon, S. L.; Craig, S. L. J. Mater. Chem., 2007, 17, 56-61.

Lu, H.; Liu, Y.; Leng, J.; Du, S. Smart Mater. Struct., 2009, 18, 1–5.

Lu, W.; Mi, B. X.; Chan, M. C. W.; Hui, Z.; Che, C. M.; Zhu, N.; Lee, S. T. J. Am. Chem. Soc., 2004, 126, 4958-4971.

Mariotti, D.; Sankaran, R. M. J. Phys. D, 2010, 43, 323001.

Mather, P. T.; Luo, X.; Rousseau, I. A. Annu. Rev. Mater. Res., 2009, 39, 445–471.

Mathew, I.; Sun, W. Dalton Trans., 2010, 39, 5885-5898.

Matsumoto, S.; Yamaguchi, S.; Ueno, S.; Komatsu, H.; Ikeda, M.; Ishizuka, K.; Iko, Y.; Tabata, K. V.; Aoki, H.; Ito, S.; Noji, H.; Hamachi, I. Chem. Eur. J., 2008, 14, 3977– 3986.

McKenzie, B. M.; Rowan, S. J. Molecular Recognition and Polymers, ed. Rotello, V. M.; Thayumanavan, S. John Wiley and Sons, Hoboken, NJ, 2008, ch. 7, pp. 157–178.

Mejia, M. L.; Agapiou, K.; Yang, X.; Holliday, B. J. J. Am. Chem. Soc., 2009, 131, 18196-18197.

Mendes, P. M. Chem. Soc. Rev., 2008, 37, 2512–2529.

Meng, Q. H.; Hu, J. L. Composites, Part A, 2009, 40, 1661–1672.

238

Meng, Q. H.; Hu, J. L.; Zhu, Y.; Lu, J.; Liu, Y. Smart Mater. Struct., 2007, 16, 1192– 1197.

Merline, J. D.; Nair, C. P. R.; Gouri, C.; Shrisudha, T.; Ninan, K. N. J. Mater. Sci., 2007, 42, 5897–5902.

Mes, T.; Smulders, M. M. J.; Palmans, A. R. A.; Meijer, E. W. Macromolecules, 2010, 43, 1981–1991.

Meudtner, R. M.; Hecht, S. Macromol. Rapid Commun., 2008, 29, 347-351.

Meudtner, R. M.; Ostermeier, M.; Goddard, R.; Limberg, C.; Hecht, S. Chem. Eur. J., 2007, 13, 9834- 9840.

Miskowski, V. M.; Houlding, V. H. Inorg. Chem., 1989, 28, 1529-1533.

Moad, G.; Chong, Y. K.; Postma, A.; Rizzardo, E.; Thang, S. H. Polymer, 2005, 46, 8458-8468.

Moughton, A. O.; O’Reilly, R. K. Macromol. Rapid Commun., 2010, 31, 37–52.

Muro, M. L.; Daws, C. A.; Castellano, F. N. Chem. Commun., 2008, 6134-6136.

Murphy, L.; Williams, J. A. G. Top. Organomet. Chem., 2010, 28, 75-111.

Nair, K. P.; Breedveld, V.; Weck, M. Macromolecules, 2011, 44, 3346–3357.

Naito, K.; Hieda, H.; Sakurai, M.; Kamata, Y.; Asakawa, K. IEEE Trans. Magn., 2002, 38, 1949-1951.

Ng, W.; Gong, X.; Chan, W. K. Chem. Mater., 1999, 11, 1165-1170.

Ni, J.; Zhang, L. Y.; Wen, H. M.; Chen, Z. N. Chem. Commun., 2009, 3801-3803.

239

Ni, J.; Zhang, X.; Wu, Y. H.; Zhang, L.Y.; Chen, Z. N. Chem. Eur. J., 2011, 17, 1171- 1183.

Nie, J.; Kobayashi, H.; Sonoda, T. Catal. Today, 1997, 36, 81–84.

Nishiyama, N.; Kato, Y.; Sugiyama, Y.; Kataoka, K. Pharm. Res., 2001, 18, 1035-1041.

Parrinello, G.; Stille, J. K. J. Am. Chem. Soc., 1987, 109, 7122-7127.

Paulusse, J. M. J.; Huijbers, J. P. J.; Sijbesma, R. P. Chem. Eur. J., 2006, 12, 4928-4934.

Paulusse, J. M. J.; van Beek, D. J. M.; Sijbesma, R. P. J. Am. Chem. Soc., 2004, 126, 15802–15808.

Paulusse, J. M. J.; van Beek, D. J. M.; Sijbesma, R. P. J. Am. Chem. Soc., 2007, 129, 2392-2397.

Paw, W.; Cummings, S. D.; Mansour, M. A.; Connick, W. B.; Geiger, D. K.; Eisenberg, R. Coordin. Chem. Rev., 1998, 171, 125-150.

Payne, S. J.; Fiore, G. L.; Fraser, C. L.; Demas, J. N. Anal. Chem., 2010, 82, 917-921.

Peet, J.; Senatore, M. L.; Heeger, A. J.; Baan, G. C. Adv. Mater., 2009, 21, 1521-1527.

Petoud, S.; Bünzli, J. -C. G.; Glanzman, T.; Piguet, C.; Xiang, Q.; Thummel, P. J. Lumin., 1999, 82, 69–79.

Piermattei, A.; Karthikeyan, S.; Sijbesma, R. P. Nat. Chem., 2009, 1, 133-137.

Piguet, C.; Bünzli, J.- C. G.; Bernardinelli, G.; Hopfgartner, G.; Williams, A. F. J. Alloys Compd., 1995, 225, 324–330.

240

Piguet, C.; Williams, A. F.; Bernardinelli, G.; Bünzli, J. -C. G. Inorg. Chem., 1993, 32, 4139–4149.

Pollino, J. M.; Weck, M. Chem. Soc. Rev., 2005, 34, 193-207.

Porel, S.; Hebalker, N.; Sreedhar, B.; Radhakrishnan, T. P. Adv. Funct. Mater., 2007, 17, 2550-2556.

Potts, K. T.; Usifer, D. A. Macromolecules, 1988, 21, 1985-1991.

Premkumar, P. A.; Starostin, S. A.; de Vries, H.; Paffen, R. M.; Creatore, M.; Eijkemans, T. J.; Koenraad, P. M.; van de Sanden, M. C. M. Plasma Process. Polym., 2009, 6, 693- 702.

Price, J. H.; Williamson, A. N.; Schramm, R. F.; Wayland, B. B. Inorg. Chem., 1972, 11, 1280-1284.

Pritchett, T. M.; Sun, W.; Guo, F.; Zhang, B.; Ferry, M. J.; Rogers-Haley, J. E.; Shensky III, W.; Mott, A. G. Opt. Lett., 2008, 33, 1053-1055.

Pucci, A.; Ruggeri, G. J. Mater. Chem., 2011, 21, 8282-8291.

Puigmarti-Luis, J.; Rubio-Martínez, M.; Hartfelder, U.; Imaz, I.; Maspoch, D.; Dittrich, P. S. J. Am. Chem. Soc., 2011, 133, 4216-4219.

Ramesh, G. V.; Porel, S.; Radhakrishnan, T. P. Chem. Soc. Rev., 2009, 38, 2646-2656.

Ratna, D.; Kocsis, J. K. J. Mater. Sci., 2008, 43, 254–269.

Rieth, S.; Baddeley, C.; Badjić, J. D. Soft Matter, 2007, 3, 137–154.

Rifai, S.; Breen, C. A.; Solis, D. J.; Swager, T. M. Chem. Mater., 2006, 18, 21-25.

241

Rivera, E. J.; Barbosa, C.; Torres, R.; Grove, L.; Taylor, S.; Connick, W. B.; Clearfield, A.; Colón, J. L. J. Mater. Chem., 2011, 21, 15899-15902.

Rivera, E. J.; Figueroa, C.; Colón, J. L.; Grove, L.; Connick, W. B. Inorg. Chem., 2007, 46, 8569-8576.

Sabbatini, N.; Guardigli, M.; Lehn, J.-M. Coordin. Chem. Rev., 1993, 123, 201–228.

Sagara, Y.; Kato, T. Nat. Chem., 2009, 1, 605-610.

Sakamoto, M.; Tachikawa, T.; Fujitsuka, M.; Majima, T. Adv. Funct. Mater., 2007, 17, 857-862.

Schmatloch, S.; van den Berg, A. M. J.; Alexeev, A. S.; Hofmeier, H.; Schubert, U. S. Macromolecules, 2003, 36, 9943–9949.

Schubert, U. S.; Eschbaumer, C. Angew. Chem. Int. Ed., 2002, 41, 2892-2926.

Serna, P.; Boronat, M.; Corma, A. Top. Catal., 2011, 54, 439-446.

Serpe, M. J.; Craig, S. L. Langmuir, 2007, 23, 1626–1634.

Shang, L.; Wang, Y.; Huang, L.; Dong, S. Langmuir, 2007, 23, 7738-7744.

Shao, P.; Li, Y.; Sun, W. Organometallics, 2008, 27, 2743-2749.

Shikhova, E.; Danilov, E. O.; Kinayyigit, S.; Pomestchenko, I. E.; Tregubov, A. D.; Camerel, F.; Retailleau, P.; Ziessel, R.; Castellano, F. N. Inorg. Chem., 2007, 46, 3038- 3048.

Shimizu, L. S. Polym. Int., 2007, 56, 444–452.

242

Shin, D. O.; Jeong, J. -R.; Han, T. H.; Koo, C. M.; Park, H. -J.; Lim, Y. T.; Kim, S. O. J. Mater. Chem., 2010, 20, 7241-7247.

Sholrollahi, P.; Mirzadeh, H.; Scherman, O. A.; Huck, W. T. S. J. Biomed. Mater. Res., Part A, 2010, 95A, 209–221.

Shunmugam, R.; Tew, G. N. J. Am. Chem. Soc., 2005, 127, 13567-13572.

Shunmugam, R.; Tew, G.N. Macromol. Rapid Commun., 2008, 29, 1355-1362.

Sievers, T. K.; Vergin, A.; Möhwald, H.; Kurth, D. G. Langmuir, 2007, 23, 12179-12184.

Silva, A. M. B.; de Araujo, C. B.; Santos-Silva, S.; Galembeck, A. J. Phys. Chem. Solids, 2007, 68, 729-733.

Sim, H. G.; Ahn, K. H.; Lee, S. J. J. Non-Newtonian Fluid Mech., 2003, 112, 237–250.

Sivakova, S.; Bohnsack, D. A.; Mackay, M. E.; Suwanmala, P.; Rowan, S. J. J. Am. Chem. Soc., 2005, 127, 18202–18211.

Small, W., IV; Metzger, M. F.; Wilson, T. S.; Maitland, D. J. IEEE J. Sel. Top. Quantum Electron., 2005, 11, 892–901.

Small, W., IV; Wilson, T. S.; Benett, W. J.; Loge, J. M.; Maitland, D. J. Opt. Express, 2005, 13, 8204–8213.

Sokolowski, W.; Metcalfe, A.; Hayashi, S.; Yahia, L.; Raymond, J. Biomed. Mater., 2007, 2, S23–S27.

South, C. R.; Pinon, V.; Weck, M. Angew. Chem., Int. Ed., 2008, 47, 1425–1428.

Tam, A. Y. Y.; Lam, W. H.; Wong, K. M. C.; Zhu, N.; Yam, V. W. W. Chem. Eur. J., 2008, 14, 4562-4576.

243 ten Brinke, G.; Ruokolainen, J.; Ikkala, O. Adv. Polym. Sci., 2007, 207, 113–177.

Tew, G. N.; Aamer, K. A.; Shunmugam, R. Polymer, 2005, 46, 8440-8447.

Tong, Q. X.; Li, X. H.; Wu, L. Z.; Yang, Q. Z.; Zhang, L. P.; Tung, C. H. Chinese J. Chem., 2004, 22, 1204-1207. van Beek, D. J. M.; Spiering, A. J. H.; Peters, G. W. M.; te Nijenhuis, K.; Sijbesma, R. P. Macromolecules, 2007, 40, 8464–8475.

Vermonden, T.; van der Gucht, J.; de Waard, P.; Marcelis, A. T. M.; Besseling, N. A. M.; Sudhölter, E. J. R.; Fleer, G. J.; Cohen Stuart, M. A. Macromolecules, 2003, 36, 7035– 7044.

Vermonden, T.; van Steenbergen, M. J.; Besseling, N. A. M.; Marcelis, A. T. M.; Hennink, W. E.; Sudhölter; E. J. R.; Cohen Stuart, M. A. J. Am. Chem. Soc., 2004, 126, 15802–15808.

Wadas, T. J.; Wang, Q. M.; Kim, Y. J.; Flaschenreim, C.; Blanton, T. N.; Eisenberg, R. J. Am. Chem. Soc., 2004, 126, 16841-16849.

Wang, D.; Fujinami, S.; Liu, H.; Nakajima, K.; Nishi, T. Macromolecules, 2010, 43, 5521-5523.

Wang, K.; Haga, M. A.; Monjushiro, H.; Akiba, M.; Sasaki, Y. Inorg. Chem., 2000, 39, 4022-4028.

Wang, S. H.; Chen, C.- C.; Dormidontova, E. E. Soft Matter, 2008, 4, 2039–2053.

Weber, W. G.; McLeary, J. B.; Sanderson, R. D. Tetrahedron Lett., 2006, 47, 4771-4774.

Weck, M. Polym. Int., 2007, 56, 453–460.

244

Weigel, T.; Mohr, R.; Lendlein, A. Smart Mater. Struct., 2009, 18, 1–9.

Weiss, R. A.; Izzo, E.; Mandelbaum, S. Macromolecules, 2008, 41, 2978–2980.

Weng, W.; Beck, J. B.; Jamieson, A. M.; Rowan, S. J. J. Am. Chem. Soc., 2006, 128, 11663–11672.

Weng, W.; Li, Z.; Jamieson, A. M.; Rowan, S. J. Macromolecules, 2009, 42, 236–246.

Weng, W.; Li, Z.; Jamieson, A. M.; Rowan, S. J. Soft Matter, 2009, 5, 4647–4657.

Whittell, G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater., 2011, 10, 176- 188.

Whittell, G. R.; Manners, I. Adv. Mater., 2007, 19, 3439-3468.

Wietor, J.-L.; van Beek, D. J. M.; Peters, G. W.; Mendes, E.; Sijbesma, R. P. Macromolecules, 2011, 44, 1211–1219.

Williams, J. A. G. Top. Curr. Chem., 2007, 281, 205-268.

Wojtecki, R. J.; Meador, M. A.; Rowan, S. J. Nat. Mater., 2011, 10, 14–27.

Woodward, P. J.;Merino, D. H.;Greenland, B.W.;Hamley, I.W.; Light, Z.; Slark, A. T.; Hayes, W. Macromolecules, 2010, 43, 2512–2517.

Wu, F. –I.; Yang, X. –H.; Neher, D.; Dodda, R.; Tseng, Y. –H.; Shu, C. –F. Adv. Funct. Mater., 2007, 17, 1786-1798.

Wu, P. –T.; Bull, T.; Kim, F. S.; Luscombe, C. K.; Jenekhe, S. A. Macromolecules, 2009, 43, 671-681.

Xie, Z.; Ma, L.; deKrafft, K. E.; Jin, A.; Lin, W. J. Am. Chem. Soc., 2010, 132, 922-923.

245

Xu, D. H.; Liu, C. Y.; Craig, S. L. Macromolecules, 2011, 44, 2343–2353.

Xu, D.; Hawk, J. L.; Loveless, D. M.; Jeon, S. L.; Craig, S. L. Macromolecules, 2010, 43, 3556-3565.

Xu, D.; Craig, S. L. J. Phys. Chem. Lett., 2010, 1, 1683–1686.

Xu, D.; Craig, S. L. Macromolecules, 2011, 44, 5465–5472.

Yakacki, C. M.; Satarkar, N. S.; Gall, K.; Likos, R.; Hilt, J. Z. J. Appl. Polym. Sci., 2009, 112, 3166–3176.

Yam, V. W. W.; Chan, K. H. Y.; Wong, K. M. C.; Zhu, N. Chem. Eur. J., 2005, 11, 4535-4543.

Yan, Y.; Huang, J. Coordin. Chem. Rev., 2010, 254, 1072-1080.

Yang, C. H.; Qiu, H. J. Chem. Phys., 1986, 84, 416-423.

Yang, S. K.; Ambade, A. V.; Weck, M. Chem. Soc. Rev., 2011, 40, 129-137.

Yang, S. K.; Ambade, A. V.; Weck, M. J. Am. Chem. Soc., 2010, 132, 1637-1645.

Yount, W. C.; Loveless, D. M.; Craig, S. L. J. Am. Chem. Soc., 2005, 127, 14488-14498.

Yu, S. C.; Kwok, C. C.; Chan, W. K.; Chen, C. M. Adv. Mater., 2003, 15, 1643-1647.

Yu, Y.; Ikeda, T. Macromol. Chem. Phys., 2005, 206, 1705–1708.

Yuen, F.; Tam, K. C. Soft Matter, 2010, 6, 4613–4630.

Zhang, D.; Wu, L. Z.; Yang, Q. Z.; Li, X. H.; Zhang, L. P.; Tung, C. H. Org. Lett., 2003, 5, 3221-3224.

246

Zhang, J.; Gao, Y. Alvarez-Puebla, R. A.; Buriak, J. M.; Fenniri, H. Adv. Mater., 2006, 18, 3233-3237.

Zhang, J.; Lodge, T. P.; Macosko, C. W. J. Polym. Sci. Part B: Polym. Phys., 2009, 47, 2313-2319.

Zhang, S.; Yu, Z.; Govender, T.; Luo, H.; Li, B. Polymer, 2008, 49, 3205–3210.

Zhao, Q.; Liu, S. –J.; Huang, W. Macromol. Rapid Commun., 2010, 31, 794-807.

Zhao, Y.; Beck, J. B.; Rowan, S. J.; Jamieson, A. M. Macromolecules, 2004, 37, 3529– 3531.

Zhou, J.; Schmidt, A. M.; Ritter, H. Macromolecules, 2010, 43, 939–942.

247