Shuko Suzuki Thesis (PDF 6MB)

THE QUEENSLAND UNIVERSITY OF TECHNOLOGY

IN VITRO MINERALISATION OF WELL-DEFINED POLYMERS AND SURFACES

A thesis presented in fulfilment of the requirements for the degree of

Doctor of Philosophy

Shuko Suzuki M. App. Sc.

Tissue Repair and Regeneration Program

Institute of Health and Biomedical Innovation School of Physical and Chemical Sciences

November 2007

ABSTRACT

Currently, many polymeric biomaterials do not possess the most desirable surface

properties for direct bone bonding due to the lack of suitable surface functionalities.

The incorporation of negatively charged groups has been shown to enhance calcium

phosphate formation in vitro and bone bonding ability in vivo. However, there are

some conflicting literature reports that highlight the complicated nature of the

mineralisation process as well as the sometimes apparent contradictory effect of the

negatively charged groups. Surface modification using well-defined polymers offer a

more precise control of the chain structures. The aims of this study were to

synthesise well-defined polymers containing phosphate and carboxylic acid groups,

and perform various surface modification techniques. The influence of the polymer

structure on mineralisation was examined using a series of specially synthesised

phosphate-containing polymers. The mineralisation ability of the fabricated surfaces

was also tested.

Soluble poly(monoacryloxyethyl phosphate) (PMAEP) and poly(2-

(methacryloyloxy)ethyl phosphate) (PMOEP) were synthesised using reversible

addition fragmentation chain transfer (RAFT)-mediated polymerisation. The

polymerisation conversions were monitored by in situ Raman spectroscopy.

Subsequently 31P NMR investigation revealed the presence of large amounts of diene

impurities as well as free orthophosphoric acids in both the MAEP and MOEP

monomers. Elemental analyses of the polymers showed loss of phosphate groups due

to hydrolysis during the polymerisation. Both gel and soluble PMAEP polymers

were found to contain large amounts of carboxyl groups indicating hydrolysis at the

C-O-C ester linkages. Block copolymers consisting of PMAEP or PMOEP and

poly(2-(acetoacetoxy) ethyl methacrylate) PAAEMA were successfully prepared for

the purpose of immobilisation of these polymers onto aminated slides.

Well-defined fluorinated polymers, (poly(pentafluorostyrene) (PFS),

poly(tetrafluoropropyl acrylate) (TFPA) and poly(tetrafluoropropyl methacrylate)

(TFPMA)) were synthesised by RAFT-mediated polymerisation. It was found that

the Mn values of PFS at higher conversions were significantly lower than those

i

calculated from the theory, although the PDI’s were low (<1.1). One possible

explanation for this is that it may be a result of the self-initiation of FS which created

more chains than the added RAFT agents. Both TFPA and TFPMA showed well-

controlled RAFT polymerisations. Chain extension of the fluorinated polymers with

tert-butyl acrylate (tBA) followed by hydrolysis of the tBA groups produced the

amphiphilic block copolymers containing carboxylic acid groups. Block copolymers

consisting of PAAEMA segments were further reacted with glycine and L-

phenylalanyl glycine.

Three types of surface modifications were carried out: Layer-by-Layer (LbL)

assemblies of the soluble phosphate- and carboxylic acid-containing homopolymers,

coupling reactions of block copolymers consisting of phosphate and keto groups onto

aminated slides, and adsorption of fluorinated homo and block copolymers

containing carboxylic acid groups onto PTFE. For LbL assemblies XPS analyses

revealed that the thickness of the poly(acrylic acid) (PAA) layer was found to be

strongly dependent on the pH at deposition. AFM images showed that the PMAEP

LbL had a patchy morphology which was due to the carboxylate groups that were not

deprotonated at low pH. Successful coupling of the block copolymers consisting of

phosphate and keto groups onto aminated slides was evident in the XPS results. The

conformation of attached P(MOEP-b-AAEMA) was investigated by ToF-SIMS.

Adsorption of the fluorinated polymers onto the PTFE film was examined using

different solvents. PFS showed the best adsorption onto PTFE. The block

copolymers consisting of PFS and PtBA or PAA were successfully adsorbed onto

PTFE. Contact angle measurements showed that the adsorbed block copolymers

reorganised quickly to form a hydrophilic surface during the investigation.

In vitro mineralisation of various phosphate-containing polymers and the fabricated

surfaces were studied using the simulated body fluid (SBF) technique. The

SEM/EDX investigation showed that either brushite or monetite, with a tile-like

morphology, was formed on both soluble and gel PMAEP polymers after seven days

in SBF. The PMOEP gel formed a similar layer as well as a secondary growth of

hydroxyapatite (HAP) that exhibited a typical globular morphology. Fourier

transform infrared (FTIR) spectroscopy of the PMOEP film prepared from soluble

PMOEP showed large amounts of carbonated HAP formation after seven days in

SBF. Carbonated HAP is the phase that most closely resembles that found in

ii

biological systems. Both the LbL surfaces and the block copolymer-attached

aminated slides showed only patchy mineralisation even after 14 days in SBF. This

indicates that ionic interactions of the negatively charged phosphates or carboxylates

and protonated amines prevented chelation of calcium ions, which is believed to be

the first step in mineralisation. The P(FS-b-AA) adsorbed PTFE film also showed

only small amounts of mineral formation after 14 days in SBF. These results

highlight the many features controlling the mineralisation outcomes.

iii

LIST OF PUBLICATIONS

Papers: • S. Suzuki, L. Rintoul, M. J. Monteiro, E. Wentrup-Byrne, L. Grøndahl (2007),

“In vitro mineralization of phosphate-containing polymer ad-layers”, Polymer Preprints, 48 (1), 430-431

• S. Suzuki, M. R. Whittaker, L. Grøndahl, M. J. Monteiro, E. Wentrup-Byrne (2006). “Synthesis of Soluble Phosphate Polymers by RAFT and their In Vitro Mineralization”, Biomacromolecules, 7, 3178-3187

• S. Suzuki, M. Jasieniak, M. J. Monteiro, E. Wentrup-Byrne, H. Griesser, L. Grøndahl, “Probing the Orientation of a Bi-functional Di-block Copolymer Ad-layer by XPS and Static ToF-SIMS” (in preparation)

• S. Suzuki, M. R. Whittaker, L. Grøndahl, E. Wentrup-Byrne, M. J. Monteiro, “Synthesis of Well-Defined Fluorine-Containing Polymers by RAFT” (in preparation)

• E. Wentrup-Byrne, A. Chander-Temple, S. Suzuki, J.J. Suwanasilp, L. Grøndahl “Structural Characterisation of Phosphate Polymers by FTIR and XPS” (in preparation)

• E. Wentrup-Byrne, L. Grøndahl, S. Suzuki (2005). “Methacryloxyethyl phosphate-grafted expanded polytetrafluoroethylene membranes for biomedical applications”, Polymer International, 54, 1581-1588

• S. Suzuki, L. Grøndahl, D. Leavesley and E. Wentrup-Byrne (2005). “In vitro bioactivity of MOEP grafted ePTFE membranes for craniofacial applications” Biomaterials, 26 (26), 5303-5312

Oral Presentations: • International Symposium on Polymeric Materials for Regenerative Medicine

(PMRM 2007), April, 2007, Montreal Canada

• 28th Australian Polymer Symposium (APS) and 16th Australian Society for Biomaterials (ASB), February 2006, Rotorua NZ

Poster Presentations: • 233rd American Chemical Society (ACS) National Meeting, March 2007,

Chicago USA

• 27th Australian Polymer Symposium (APS), February 2005, Adelaide AU

iv

DECLARATION OF ORIGINAL AUTHORSHIP

The work submitted in this thesis has not been previously submitted for a degree or

diploma at this or any other educational institution. To the best of my knowledge and

belief, the information contained in this thesis contains no material previously

published or written by any other person except where due reference is made.

Signed ………………………..

Dated ………………………..

v

ACKNOWLEDGEMENTS This document is the result of the contributions of many people – those that have

contributed direct information that is described within, and those that have shaped

me to be able to produce this document. My work has spanned across several

Universities, with the majority of work being performed at the Queensland

University of Technology (QUT) and the University of Queensland (UQ), and thus

have collaborated with many people that have provided positive support for the

research that I was undertaking. I appreciate their assistance and thank them all.

Dr. Edeline Wentrup-Byrne, who not only took on the large role of being my

principle supervisor, but also for providing constant help and guidance through the

project and in particular for poring over my thesis for many hours. Moreover, I am

grateful for her encouraging and challenging me to get out of the comfort zone and

explore the international nature of research.

Dr. Lisbeth Grøndahl, my co-supervisor, for her guidance and dedication in ensuring

I was doing things right. Dr. Grøndahl also accommodated me with a laboratory and

office at the University of Queensland for the majority of my time performing the

research. In addition, I appreciate her spending her own time to assist with the

research and papers written.

A/Prof. Michael Monteiro, my co-supervisor, for initiating the new direction of the

research project and with guidance with RAFT polymerisation, and without whom it

would have not been possible to perform the large amount of synthesis.

Dr. Michael Whittaker, for training me with synthesis skills and his great help and

encouragements through the project, as well as running DLS samples.

The specialists, their experienced knowledge of the mechanics and intricacies of the

equipment were critical to the research process, in particular:

• Prof. Hans Griesser and Mr. Marek Jasieniak at the Ian Wark Research

Institute, University of South Australia, for performing ToF-SIMS and with

its interpretation.

• Dr. Barry Wood for his entertainment and expert assistance with the XPS

vi

• Dr. Llew Rintoul for sharing his knowledge and technical expertise in

spectroscopies

• Dr. Tri Le for his patient and sharing knowledge of NMR and in particular

helping me to understand the phosphorous-NMR

• Dr. John Bartley for assistance with the proton-NMR analysis

• Dr. Thor Bostrom, Dr. Deb Stenzel, Mr. Loc Dong and Mr. Lambert Bekessy

for their technical support with SEM and EDX characterisations

• Dr. Ihwa Tan and Mr. Yoosup Park for the technical support of the water

phase GPC

• Mr. George Blazak for performing microanalysis

Prof. Graeme George for sharing his experienced knowledge and providing kind

support.

Prof. David Hill for assistance with solving, from my perspective, difficult problems.

The members of the QUT and UQ polymer groups who afforded me a friendly and

very supportive environment.

The Queensland University of Technology for financial support. In addition, the

QUT Grant-in-Aid scheme and IHBI travel fund made possible the international

visits and presentations.

I also received travel awards from the Royal Australian Chemical Institute (RACI)

and RACI QLD polymer group which I greatly appreciate.

In addition, my colleagues, who provided enormous entertainment and motivation.

And finally, I thank my family and my friends who all have been very supportive

over the course of the research.

vii

TABLE OF CONTENTS ABSTRACT……………………………………………………………………. i

LIST OF PUBLICATIONS …………………………………………………... iv

DECLARATION OF ORIGINAL AUTHORSHIP…………………………. v

ACKNOWLEDGEMENTS…………………………………………………… vi

TABLE OF CONTENTS……………………………………………………… viii

LIST OF FIGURES ………………………………………………………….... xiv

LIST OF TABLES……………………………………………………………... xxii

LIST OF SCHEMES ………………………………………………………….. xxiv

ABBREVIATIONS…………………………………………………………….. xxvi

Chapter 1: Introduction

1.1 Biomaterials and Biocompatibility……………………………………... 1

1.2 Bone………………………………………………………………………. 2

1.3 Host Response to Polymeric Biomaterials……………………………... 5

1.3.1 Wound healing……………………………………………………. 5

1.3.2 Tissue response to implants………………………………………. 6

1.4 Material/Bone Interface………………………………………………… 8

1.5 Surface Modification: Improving the Material/Bone Interface………. 10

1.5.1 Physicochemical Surface Modification…………………………… 10

1.5.2 Morphological surface modification……………………………… 11

1.5.3 Incorporation of biological molecules…………………………….. 12

1.6 Expanded PTFE (ePTFE) in Medicine………………………………… 14

1.7 Surface Modification of PTFE and Other Fluoropolymers…………... 14

1.8 Project Outline…………………………………………………………… 18

1.9 References………………………………………………………………... 20

Chapter 2: Polymer Synthesis

2.1 Introduction……………………………………………………………… 25

viii

2.1.1 Living Radical Polymerisation (LRP)…………………………….. 25

2.1.2 Reversible Addition Fragmentation Chain Transfer (RAFT) Method …………………………………………………………...

28

2.1.3 Living Radical Polymerisation of Phosphorous-Containing Monomers………………………………………………………….

33

2.1.4 Living Radical Polymerisation of Fluorinated Monomers………... 34

2.2 Experimental……………………………………………………………... 37

2.2.1 Materials…………………………………………………………... 37

2.2.2 Methods…………………………………………………………… 38

2.2.2.1 MOEP/MAEP Homopolymer Synthesis………………….. 38

2.2.2.2 MOEP/MAEP Block Copolymer Synthesis………………. 38

2.2.2.3 Hydrolysis of PMOEP and PMAEP Polymers…………… 38

2.2.2.4 Fluorinated Homopolymer Synthesis……………………... 39

2.2.2.5 Fluorinated Block Copolymer Synthesis………………….. 40

2.2.2.6 Hydrolysis of tBA Segments……………………………… 41

2.2.2.7 Model Biomolecule Modification of Functional Fluorinated Block Polymers………………………………

42

2.2.2.8 Stability of the PFS RAFT end-groups…………………… 43

2.2.3 Analytical Techniques…………………………………………….. 43

2.2.3.1 FT-Raman spectroscopy…………………………………... 43

2.2.3.2 Gel permeation chromatography (GPC)………………….. 44

2.2.3.3 Elemental Analyses……………………………………….. 45

2.2.3.4 Nuclear Magnetic Resonance (NMR)…………………….. 45

2.2.3.5 UV/VIS Spectroscopy…………………………………….. 45

2.2.3.6 Fourier Transform Near Infrared (FT-NIR)……………..... 45

2.3 Results…………………………………………………………………….. 46

2.3.1 RAFT-Mediated Polymerisation of Phosphate-Containing Monomers………………………………………………………….

46

2.3.1.1 Monoacryloxyethyl phosphate (MAEP) polymerisation…. 46

2.3.1.2 Methacryloyloxyethyl phosphate (MOEP) polymerisation 51

2.3.1.3 NMR of monomers and polymers………………………… 54

2.3.1.4 Elemental Analyses of Monomers and Polymers………..... 61

ix

2.3.1.5 Hydrolysis of gel polymers……………………………….. 61

2.3.1.6 Synthesis of MAEP and MOEP block copolymers with AAEMA…………………………………………………...

62

2.3.2 RAFT-Mediated Polymerisations of Fluorine Containing Monomers………………………………………………………….

65

2.3.2.1 Pentafluorostyrene (FS) polymerisation…………………... 65

2.3.2.2 Tetrafluoropropyl acrylate (TFPA) polymerisation………. 72

2.3.2.3 Tetrafluoropropyl methacrylate (TFPMA) polymerisation. 74

2.3.2.4 Block copolymerisation of macrofluorinated polymers with pentafluorostyrene (FS), tert-butyl acrylate (tBA), acetoacetoxyethyl methacrylate (AAEMA) and acetoacetoxyethyl acrylate (AAEA)………………………

77

2.3.3 Biomolecule Attachment………………………………………….. 83

2.4 Discussion……………………………………………………………….... 87

2.4.1 RAFT-Mediated Polymerisation of Phosphate-Containing Monomers …………………………………………………………………

87

2.4.1.1 Polymerization with PEPDTA…………………………….. 87

2.4.1.2 Polymerization with CDB………………………………..... 87

2.4.1.3 Loss of phosphate groups………………………………….. 88

2.4.1.4 Mechanism of gel formation……………………………..... 89

2.4.1.5 Block Copolymerisation with AAEMA………………….... 91

2.4.2 RAFT-Mediated Polymerisation of Fluorine-Containing Monomers………………………………………………………….

92

2.4.2.1 FS Polymerisation…………………………………………. 92

2.4.2.2 Tetrafluoropropyl Acrylate (TFPA) and Tetrafluoropropyl Methacrylate (TFPMA) Polymerisations………………….

95

2.4.2.3 Chain Extension of Fluorinated Macro-RAFT……………. 95

2.4.3 Biomolecule Attachment…………………………………………. 95

2.5 Conclusion……………………………………………………………….. 97

2.6 References……………………………………………………………….. 98

Chapter 3: Surface Fabrication

3.1 Introduction…………………………………………………………….... 103

3.1.1 LbL Assembly…………………………………………………….. 103

x

3.1.2 Attachment of Block Copolymers onto Aminated Slides………… 107

3.1.3 Adsorption of Fluorinated Polymers…………………………….... 109

3.1.4 Surface Characterisation Techniques……………………………... 113

3.1.4.1 X-ray photoelectron spectroscopy (XPS)………………….. 113

3.1.4.2 Static time-of-flight secondary ion mass spectroscopy (ToF-SIMS)……………………………………………….

114

3.1.4.3 Infrared reflection-adsorption spectroscopy (IRRAS)…….. 115

3.1.4.4 Atomic force microscopy (AFM)………………………….. 116

3.1.4.5 Contact angle measurement……………………………....... 116

3.2 Experimental…………………………………………………………….. 118

3.2.1 Materials…………………………………………………………... 118

3.2.2 Methods…………………………………………………………… 118

3.2.2.1 Layer by Layer (LbL) assembly: PEI-PAA, PEI-PMAEP, and PEI-PMOEP…………………………………………..

118

3.2.2.2 Coupling of block copolymers to aminated slide………….. 119

3.2.2.3 Fluorinated homo and copolymer adsorption onto PTFE…. 119

3.2.3 Instrumentation……………………………………………………. 120

3.2.3.1 X-ray photoelectron spectroscopy (XPS)………………….. 120

3.2.3.2 Infrared reflection-adsorption spectroscopy (IRRAS)…….. 120

3.2.3.3 Atomic force microscopy (AFM)………………………….. 120

3.2.3.4 Static time of flight secondary ion mass spectrometry (static ToF-SIMS)…………………………………………

121

3.2.3.5 Sessile drop contact angle measurements…………………. 121

3.2.3.6 Dynamic light scattering (DLS)……………………...……. 122

3.3 Results……………………………………………………………..……… 123

3.3.1 Layer-by-Layer (LbL) Assembly………………………….……… 123

3.3.1.1 LbL of PAA…………………………………...…………… 123

3.3.1.2 LbL of phosphate-containing polymers…...………………. 127

3.3.2 Block Copolymers Coupled onto Aminated Slides…...…………... 135

3.3.2.1 Quantitative XPS investigation of attached polymers...…… 136

3.3.2.2 ToF-SIMS investigation of the conformation of attached block copolymers………………………………………….

140

xi

3.3.3 Adsorption of Fluorinated Polymers onto PTFE……………...…... 148

3.3.3.1 Effect of monomer structure and solvents for adsorption..... 148

3.3.3.2. Adsorption of PFS with varying Mn’s…………………….. 152

3.3.3.3. Adsorption of P(FS-b-tBA) and P(FS-b-AA) block copolymers onto PTFE………………………....................

153

3.4 Discussion………………………………………………………………… 159

3.4.1 Layer-by-Layer (LbL) Assembly………………………………..... 159

3.4.1.1 LbL assembly of PAA…………………………………….. 159

3.4.1.2 LbL assembly of PMAEP and PMOEP…………………… 160

3.4.2 Block Copolymer Attachment onto Aminated Slides…………….. 162

3.4.2.1 Qualitative analysis of attached polymers…………………. 162

3.4.2.2 Conformation of attached block copolymers……………… 163

3.4.3 Adsorption of Fluorinated Polymers onto PTFE………………….. 165

3.4.3.1 Homopolymer adsorption………………………………...... 165

3.4.3.2. Block copolymer adsorption onto PTFE………………….. 167

3.5 Conclusions………………………………………………………………. 170

3.6 References………………………………………………………………... 172

Chapter 4: In Vitro Mineralisation

4.1 Introduction…………………………………………………………….... 177

4.1.1 Mineralisation……………………………………………………... 177

4.1.2 Simulated Body Fluid (SBF)……………………………………… 178

4.1.3 Negatively Charged Groups………………………………………. 180

4.2 Experimental……………………………………………………………... 184

4.2.1 Materials…………………………………………………………... 184

4.2.2 Methods………………………………………………………….... 184

4.2.2.1 Synthesis of cross-linked PAA gels……………………….. 184

4.2.2.2 Simulated body fluid (SBF) experiments……………….... 185

4.2.2.3 Scanning electron microscopy with energy dispersive x-ray analysis……………………………………………………..

185

4.2.2.4 Fourier transform infrared spectroscopy – attenuated total reflectance .………………………………………………...

185

xii

4.2.2.5 XPS………………………………………………………… 186

4.3 Results…………………………………………………………………….. 187

4.3.1 SBF studies of PMAEP and PMOEP……………………………... 187

4.3.2 SBF studies of Layer-by-Layer (LbL) films……………………… 195

4.3.3 SBF studies of block copolymers coupled to aminated slides……. 199

4.3.4 SBF studies of fluorinated amphiphilic block copolymers attached onto PTFE Films…………………………………………………………..

202

4.4 Discussion…………………………………………………………………. 206

4.4.1 SBF studies of PMAEP and PMOEP……………………………… 206

4.4.2 SBF Studies of Layer-by-Layer (LbL) Films……………………... 208

4.4.3 SBF Studies of Block Copolymers Coupled to Aminated Slides…. 209

4.4.4 SBF Studies of Fluorinated Amphiphilic Block Copolymers Adsorbed onto PTFE Films………………………………………………..

210

4.5 Conclusion………………………………………………………………… 212

4.6 References………………………………………………………………… 213

Chapter 5: Overall Conclusions and Future Work

5.1 Chapter 2………………………………………………………………..... 215

5.2 Chapter 3………………………………………………………………..... 216

5.3 Chapter 4………………………………………………………………..... 217

5.4 General Discussion……………………………………………………….. 217

5.5 Future work……………………………………………………………..... 218

xiii

LIST OF FIGURES

Chapter 1

Figure 1.1: Hierarchical structure of bone(Reproduced from ref.5)............. 3

Figure 1.2: Cellular activities at wound repair (Reproduced from ref.11)…. 5

Figure 1.3: Schematic illustration of the successive events following implantation of a material (Reproduced from ref.15)…………..

9

Figure 1.4: Covalent immobilisation of RGD peptides with carboxyl groups on a modified polymer surface…………………………

12

Figure 1.5: Entrapment of biomolecules into a polymer substrate………… 13

Figure 1.6: Adsorption of functional polymers at the FEP/water interface... 17

Chapter 2

Figure 2.1: Heating block setup for in situ Raman polymerisation………... 44

Figure 2.2: Raman spectra of MAEP polymerisation solution ([PEPDTA] = 1 × 10-2 M (expt.2)) (a) initial and (b) after 7.25 h in the region of (A) 3800-360 cm-1 and (B) 1800-900 cm-1…………..

46

Figure 2.3: Conversion versus time of a MAEP polymerisation in methanol using the RAFT agent PEPDTA or CDB, and AIBN as initiator: (a) no RAFT (expt.1), (b) [PEPDTA] = 1 × 10-2 M (expt.2), and (c) [PEPDTA] = 2 × 10-2 M (expt.3) and (d) [CDB] = 1 × 10-2 M (expt.5)…………………………………...

48

Figure 2.4: 1H NMR (400 MHz) spectra of (A) linear PMAEP (expt. 2, Table 2.1) in methanol-d4 and (B) hydrolyzed PMAEP in D2O……………………………………………………………..

49

Figure 2.5: GPC traces of hydrolysed PMAEP with different conversions from expt.2, Table 2.1 (A) and expt.3, Table 2.1 (B)………….

50

Figure 2.6: Mn and PDI of PMAEP polymerized with PEPDTA after hydrolysis (a) [PEPDTA] = 1 × 10-2 M (expt.2, Table 2.1) and (b) 2 × 10-2 M (expt.3, Table 2.1). The lines show the theoretical evolution of Mn with conversion for PAA…………………………………………………………….

56

Figure 2.7: Raman spectra of MOEP polymerisation solution ([CDB] = 1 × 10-2 M (expt.9, Table 2.1)) (a) initial and (b) after 19.5 h…...

52

Figure 2.8: Conversion versus time of a MAEP polymerisation in methanol using the RAFT agent PEPDTA or CDB, and AIBN as initiator: (a) no RAFT (expt.6, Table 2.1), (b) [PEPDTA] = 1 × 10-2 M (expt.7, Table 2.1), and (c) [PEPDTA] = 2 × 10-2 M (expt.8, Table 2.1), and (d) [CDB] = 1 × 10-2 M (expt.9, Table

xiv

2.1) and (e) [CDB] = 2 × 10-2 M (expt.10, Table 2.1)…………. 52

Figure 2.9: GPC traces of hydrolysed PMOEP with different conversions from expt.9, Table 2.1 (A) and expt.10, Table 2.1 (B)………...

54

Figure 2.10: Mn and PDI of PMOEP polymerized with CDB after hydrolysis (a) [CDB] = 1 × 10-2 M (expt.9, Table 2.1) and (b) 2 × 10-2 M (expt.10, Table 2.1)…………………………………..

55

Figure 2.11: 1H NMR spectrum of MAEP in methanol-d4………………….. 55

Figure 2.12: 31P-NMR of MAEP monomer in methanol-d4 A) H-decoupled and B) H-coupled………………………………………………

56

Figure 2.13: 1H NMR spectrum of MOEP in methanol-d4………………….. 55

Figure 2.14: 31P NMR of MOEP monomer in methanol-d4 A) H-decoupled and B) H-coupled..……………………………….….................

57

Figure 2.15: 1H NMR spectrum of PMAEP (expt.4, Table 2.1) in methanol-d4………………………………………………………………..

58

Figure 2.16: 31P NMR spectrum of PMAEP (expt.4, Table 2.1) in methanol-d4………………………………………………………………..

58

Figure 2.17: 1H NMR of PMOEP (expt.10, Table 2.1) in methanol-d4………………………………………………………………..

60

Figure 2.18: 31P-NMR of PMOEP (expt.10, Table 2.1) in methanol-d4………………………………………………………………..

60

Figure 2.19: Raman spectra of PFS polymerisation solution ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM) (a) initial and (b) after 25 h..

66

Figure 2.20: Conversion based on the normalised 1620 cm-1 bands form Raman spectra versus time of a bulk PFS polymerisation using the RAFT agent PEPDTA, and Vazo-88 or AIBN as initiator: Curve a: [PEPDTA] = 29 mM and [AIBN] = 2.9 mM, 60 ºC (expt.1, Table 2.3), b: [PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM, 80 ºC (expt.3, Table 2.3), c: [PEPDTA] = 56 mM, and [Vazo-88] = 5.6 mM, 80 ºC(expt.4, Table 2.3)……………

67

Figure 2.21: Mn and PDI of PFS polymerized with PEPDTA (a) [PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM (extp.3, Table 2.3), b: [PEPDTA] = 56 mM, and [Vazo-88] = 5.6 mM (extp.4, Table 2.3). The lines show the theoretical evolution of Mn with conversion……………………………………………………...

68

Figure 2.22: FT-NIR spectra of FS polymerisation solution ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM) (a) initial and (b) after 25 h..

69

Figure 2.23: Conversion versus time of bulk FS polymerisation ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM, 80 ºC): Curve a, from in situ FT-NIR polymerisation, and b, from in situ Raman polymerisation…………….…………….…………….………..

70

xv

Figure 2.24: 1H NMR spectrum of PFS polymerisation mixture after 67h for calculation of the conversion (CDB) (this corresponds to 81% conversion). ………………………………………………

71

Figure 2.25: Raman spectra of TFPA polymerisation solution ([PEPDTA] = 26 mM and [AIBN] = 2.6 mM) (a) initial and (b) after 3 h……

72

Figure 2.26: Conversion versus time of bulk TFPA polymerisation using the RAFT agent PEPDTA, and AIBN as initiator at 60 ºC: Curve a: [PEPDTA] = 28 mM and [AIBN] = 2.8 mM (expt.11, Table 2.3), and b: [PEPDTA] = 56 mM, and [AIBN] = 5.6 mM (expt.12, Table 2.3). ………………..…………….……………

73

Figure 2.27: Mn and PDI of TFPA polymerized with PEPDTA (a) [PEPDTA] = 28 mM and [AIBN] = 2.8 mM (expt.11, Table 2.3), b: [PEPDTA] = 56 mM, and [AIBN] = 5.6 mM (expt.12, Table 2.3). The lines show the theoretical evolution of Mn with conversion. …………………………..………………………...

73

Figure 2.28: Raman spectra of TFPMA polymerisation solution ([CDB] = 25 mM and [AIBN] = 2.5 mM) (a) initial and (b) after 23 h …………………..…………….…………….………………..

74

Figure 2.29: Conversion versus time of bulk TFPMA polymerisation using the RAFT agent CDB, and AIBN as initiator at 60 ºC: Curve a: [CDB] = 25 mM and [AIBN] = 2.5 mM (expt.13, Table 2.3), and b: [CDB] = 50 mM and [AIBN] = 5.0 mM (expt.14, Table 2.3). ……………………...…………….…………….…………

75

Figure 2.30: Mn and PDI of TFPMA polymerized with CDB (a) [CDB] = 25 mM and [AIBN] = 2.5 mM (expt.13, Table 2.3), b: [CDB] = 50 mM, and [AIBN] = 5.0 mM (expt.14, Table 2.3). The lines show the theoretical evolution of Mn with conversion…...

75

Figure 2.31: Effect of CDB purity on the rate of polymerisation of TFPMA (in bulk) in the presence of 25 mM CDB and 2.5 mM AIBN at 60 oC. Using hexane as eluent, CDB was passed through: a neutral activity aluminium oxide column followed by a silica column: Curve a: once and twice, respectively (expt.13, Table 2.3), and b: twice each (expt.15, Table 2.3). …………………..

76

Figure 2.32: GPC chromatograms of the chain extension of PFS with tBA carried out in (A) tetrahydrofuran (expt.2, Table 2.4), and (B) ethyl acetate (expt.3, Table 2.4), using PDA detection at 262 nm (full line, styrenic side groups) and at 310 nm (dashed line, RAFT end group)…………........................................................

80

Figure 2.33: Degradation of the RAFT end-groups of PFS (expt.7, Table 2.3) stored in tetrahydrofuran (curve a) and ethyl acetate (curve b) monitored by UV-vis absorbance spectroscopy at 310 nm. …………………………..…………….………………

81

Figure 2.34: 1H NMR spectra of P(TFPA-b-tBA) from expt.6, Table 2.4: (A) neat polymer (in acetone-d6) and, (B) after hydrolysis with

xvi

TFA (in DMSO-d6) (* = solvent) ……………………………... 82

Figure 2.35: 1H NMR spectrum of; (A) P(TFPMA-b-AAEMA) from expt.9, Table 2.4, in DMSO-d6, (B) glycine in 1:1 = D2O:DMSO-d6 and (C) glycine modified P(TFPMA-block-AAEMA) in DMSO-d6 (* = solvent)…….. …………….…………….……..

83

Figure 2.36: FTIR-ATR of (a) P(TFPMA-b-AAEMA) and (b) after glycine attachment in the ranges of (A) 3600-547 cm-1 and (B) 1670-1340 cm-1…………….…………….…………….……………..

84

Figure 2.37: 1H NMR spectra of (A) P(TFPMA-b-AAEMA) from expt.9, Table 2.4, (B) L-phenylalanyl glycine and (C) L-phenylalanyl glycine modified P(TFPMA-b-AAEMA), all in DMSO-d6 (* = solvent) ………………..…………….…………….…………...

86

Chapter 3

Figure 3.1: Illustration of attached polymer layers………………………… 104

Figure 3.2: Chain entanglement of polymeric chains by binary-hooking structure....…………….…………….…………….……………

112

Figure 3.3: Beam geometry and polarisation of IR radiation at the interface…………….…………….…………….………………

115

Figure 3.4: Idealised electrostatically driven LbL assembly deposition…... 123

Figure 3.5: XPS survey scans of (A) blank glass slide (sample 1A), (B) PEI film (sample 1B1) and (C) PEI-PAA LbL (sample 1B2)…..

124

Figure 3.6: C 1s narrow scans of (A) PEI film (sample 1B1) and (B) PEI-PAA LbL (sample 1B2) ………………………………………..

125

Figure 3.7: N1s narrow scans of (A) PEI film (sample 1B1) and (B) PEI-PAA LbL (sample 1B2). ……………………………………….

126

Figure 3.8: XPS survey scans of (A) blank silicon wafer (sample 2A), (B) PEI film (sample 2B), (C) PEI-PMAEP LbL (sample 2C) and (D) PEI-PMOEP LbL (sample 2D) ……………………………

128

Figure 3.9: C1s narrow scans of (A) PEI film (sample 2B), (B) PEI-PMAEP LbL (sample 2C) and (C) PEI-PMOEP LbL (sample 2D). ……………………………….…………….……………...

129

Figure 3.10: N1s narrow scans of (A) PEI film (sample 2B), (B) PEI-PMAEP LbL (sample 2C) and (C) PEI-PMOEP LbL (sample 2D) ……………………………….…………….………………

130

Figure 3.11: 2D and 3D AFM images of A) silicon wafer, B) PEI film (sample 2B), C) PEI-PMAEP LbL (sample 2C) and D) PEI-PMOEP LbL (sample 2D) (analysed area 1.0×1.0 μm) ……………………………………….……………………

132

xvii

Figure 3.12: 2D and 3D AFM images of PMAEP (analysed area 10×10 μm)……… …………….…………….…………….…………..

133

Figure 3.13: FTIR spectra of (A) soluble PMOEP (ATR) and (B) PEI-PMOEP LbL (sample 2D) (IRRAS)…………………………...

134

Figure 3.14: Idealised coupling reaction of block copolymers with aminated slide……………….…………….…………….………………..

135

Figure 3.15: XPS spectra of aminated slides (A) as received (Sample 3A), (B) PAAEMA functionalized (Sample 3B), (C) PMOEP functionalized (Sample 3C) and (D) P(MOEP-b-AAEMA) functionalized (Sample 3D). …………………………………..

136

Figure 3.16: C1s narrow scans of aminated slides (A) as received (sample 3A), (B) PAAEMA functionalized (sample 3B), (C) PMOEP functionalized (sample 3C) and (D) P(MOEP-b-AAEMA) functionalized (sample 3D). …………………………………...

138

Figure 3.17: N 1s narrow scans of aminated slides (A) as received (sample 3A), (B) PAAEMA functionalized (sample 3B), (C) PMOEP functionalized (sample 3C) and (D) PMOEP-b-PAAEMA functionalized (sample 3D). …………………………………...

138

Figure 3.18: Possible conformations of block copolymers reacted with aminated slide………………………………………………….

140

Figure 3.19: Positive and negative SSIMS spectra for APS-treated amianted glass slide. Note: sodium peak (23 amu) was scaled to 25% of its original intensity; peaks above 25 amu in negative mass spectrum were magnified by 10 folds………………………….

141

Figure 3.20: Positive SSIMS spectra for: (a) Sample 3A (aminated slide), (b) Sample 3B (PAAEMA attached), (c) Sample 3C (PMOEP attached), and (d) Sample 3D (P(MOEP-b-AAEMA) attached) ………………….…………….……………………...

142

Figure 3.21: Negative SSIMS spectra for: (a) Sample 3A (aminated slide), (b) Sample 3B (PAAEMA attached), (c) Sample 3C (PMOEP attached), and (d) Sample 3D (P(MOEP-b-AAEMA) attached).

142

Figure 3.22: Score plots on PC1 and PC2 for aminated glass slide and its modifications. (a) scores derived from the positive fragments; (b) scores derived from the negative fragments………………..

144

Figure 3.23: Loadings of selected positive (a) and negative (b) fragments on PC1s.. …………….…………….…………….………………..

145

Figure 3.24: Normalised intensities of Si+ and PO3- for APS-treated

aminated glass slide and its modifications. (a) Si+ intensity reflects the APS coverage with polymers; (b) PO3

- intensity is a sign of the surface density of the PMOEP segment………….

146

Figure 3.25: Negative mass spectrum of P(MOEP-b-AAEMA) attached aminated slide and the lateral distribution of the terminating

xviii

phosphate groups (PO3-) across the sample (analysed area →

100x100 μm) …………….…………….……………………… 147

Figure 3.26: XPS Survey spectra of PTFE films (A) untreated, (B) PFS attached (sample 3B3), (C) PTFPMA attached (sample 3C1), (D) PTFPA attached (sample 3D1) ……………………………

149

Figure 3.27: C1s narrow scans of PTFE films (A) untreated (sample 4A) and (B) PFS adsorbed (sample 4B3) …………………………...

151

Figure 3.28: C1s narrow scans of PTFE films after fluorinated homopolymer adsorption using different solvents. ……………

151

Figure 3.29: Relationship between molecular weight of PFS and adsorbed amounts of PFS onto PTFE films………………………………

153

Figure 3.30: C1s narrow scans of block copolymer attached PTFE films (A) P(FS101-b-tBA237) (sample 4F), (B) P(FS101-b-tBA141) (sample 4G), (C) P(FS101-b-AA237) (sample 4H), (D) P(FS101-b-AA141) (sample 4I1) and (E) P(FS101-b-AA141) (sample 4I2) …………..

154

Figure 3.31: Water droplet (5 μL) profiles on the surfaces of untreated PTFE (sample 4A) (A) advancing and (B) receding, P(FS101-b-tBA141) adsorbed PTFE (sample 4G) (C) advancing and (D) receding, and P(FS101-b-AA141) adsorbed PTFE (sample 4I1) (E) advancing and (F) receding ……………………………….

156

Figure 3.32: Schematic representation of PAA deposition onto PEI deposited surface at different pH ……………………………...

160

Figure 3.33: Schematic representation of PEI-PAA LbL showing free-functional groups ………………………………………………

161

Figure 3.34: Possible conformations of block copolymers reacted with aminated slide…………………………………………………..

163

Figure 3.35: Schematic representation of fluoropolymer adsorption onto PTFE by (A) fluorine adsorption and (B) chain entanglement...

166

Figure 3.36: Possible adsorption behaviour of PFS onto PTFE in different solvents.. …………….…………….…………….……………..

167

Figure 3.37: Aggregate adsorption onto PTFE surfaces in different solvents. 169

xix

Chapter 4

Figure 4.1: SEM images of sample 1A (PMAEP gel) (A) before treatment, and after SBF immersion for 7 days with (B) carbon coating and (C) gold coating, (D) sample 1B (soluble PMAEP film after 7 days in 1.5 SBF), (E) sample 1G (PAA gel) after 7 days immersion in SBF, and (F) EDX spectrum of the mineral on sample 1G………………………………………………………

188

Figure 4.2: SEM images of sample 1C (PMOEP gel) (A) before treatment, and after SBF immersion for 7 days with (B) carbon coating and (C) gold coating, sample 1D (PMOEP gel) (D) before treatment and (E) after SBF immersion for 7 days and (F) minerals dislodged from sample 1D. …………………………..

189

Figure 4.3: SEM images of sample 1E (PMOEP gel) (A) before, and (B) after SBF immersion for 7 days, (C) sample 1F (soluble PMOEP cast on glass) after SBF immersion for 7 days, and (D) EDX spectrum of the mineral on sample 1E. ……………..

190

Figure 4.4: ATR-FTIR spectra of polymer samples after 7 days immersion in SBF (solid line) and initial, untreated polymers (dotted line). (A) sample 1A (PMAEP gel); (B) sample 1B (PMAEP film) (Experiment done in 1.5×SBF.); (C) sample 1C (PMOEP gel); (D) sample 1D (PMOEP gel); (E) sample 1E (PMOEP gel); (F) sample 1F (PMOEP film on a glass surface), and (G) sample 1G (PAA gel). y-axis is absorbance…...……………….

193

Figure 4.5: ATR-FTIR spectra of Sample E (PMOEP gel) reacted with Ca(OH)2 (solid line) and untreated (dotted line). y-axis is absorbance…………..…………….…………….……………...

195

Figure 4.6: SEM images of minerals formed on the LbL surfaces after SBF immersion for 7 days: (A) sample 2A (PEI-PMAEP), (B) sample 2B (PEI-PMOEP), (C) sample 2C (PEI only), and (D) sample 2D (silicon wafer), and 14 days: (E) sample 2A, (F) sample 2B, (G) sample 2C, and (H) sample 2D………………..

197

Figure 4.7: SEM images of minerals formed on the block copolymer functionalized aminated slides after SBF immersion for 7 days: (A) sample 3A, (B) sample 3B, (C) sample 3C and (D) sample 3D, and 14 days: (E) sample 3A, (F) sample 3B, (G) sample 3C and (H) sample 3D. …………………………….…………..

200

Figure 4.8: SEM images of minerals formed on the PAAEMA functionalized and untreated aminated slides after SBF immersion for 7 days: (A) sample 3E (PAAEMA) and (B) sample 3F (untreated aminated slide), and 14 days: (C) sample 3E (PAAEMA) and (D) sample 3F. ………………...................

201

Figure 4.9: EDX spectra of the mineral (A) and the non-mineral area (B) on sample 3B (see Figure 4.7F). ………………………………

201

xx

Figure 4.10: SEM images of minerals formed on P(FSm-b-AAn) attached and untreated PTFE films after SBF immersion for 7 days: (A) sample 4A, (B) sample 4B, (C) sample 4C and (D) sample 4D (untreated PTFE), and 14 days: (E) sample 4A, (F) sample 4B, (G) sample 4C and (H) sample 4D……………………………..

204

Figure 4.11: Proposed structures of a P(AAEMA-b-MOEP) block copolymer on an aminated slide (A) in vacuum and (B) in SBF solution. ……………….…………….…………….…………...

210

xxi

LIST OF TABLES

Chapter 1

Table 1.1: Composition of human trabecular and cortical bone (vol%)4…… 2

Chapter 2

Table 2.1: Experimental conditions of MAEP and MOEP polymerisation reactions and characteristics of the polymers obtained………….

47

Table 2.2: Experimental conditions of chain extension reactions and molecular weight of the polymers obtained after hydrolysis……

64

Table 2.3: Experimental results for the RAFT polymerisation of fluorinated macromers……………………………………………………......

65

Table 2.4: Experimental Results from the Chain Extension of Fluorinated Macromers via RAFT……………………………………………

79

Table 2.5: Expected IR bands from glycine attachments64…......................... 85

Chapter 3

Table 3.1: Comparison of energies associated with intermolecular forces… 111

Table 3.2: Literature values of Ne of PTFE………………………………… 112

Table 3.3: Properties of PEI and PAA……………………………………… 124

Table 3.4: Atomic % of O, C, N and Si…………………………………….. 125

Table 3.5: Normalized atomic % of nitrogen species from the curve fitting of the N1s peak…………………………………………………..

127

Table 3.6: Properties of PMAEP/PMOEP used for this study……………… 128

Table 3.7: Atomic % of C, N, P and Si from XPS survey scans…………… 129

Table 3.8: Normalized atomic % of nitrogen species from curve fitting of the N1s peak……………………………………………………..

130

Table 3.9: Mean roughness (nm) of LbL films from the AFM images in Figure 3.7.………………………………………………………..

131

Table 3.10: Block-copolymers attached to aminated slides for SBF………… 135

Table 3.11: XPS data: Atomic % of elements concentrations from XPS survey scans……………………………………………………..

136

Table 3.12: Normalized atomic % of nitrogen species from curve fitting of the N1s peak……………………………………………………..

139

xxii

Table 3.13: Positive and negative fragments used in PCA………………….. 139

Table 3.14: Polymer characteristics………………………………………….. 148

Table 3.15: Atomic % of elements from XPS survey scans and atomic C% of different C elements from the high resolution C1s scans………..

150

Table 3.16: Properties of PFS and atomic % from C 1s scans of PFS adsorbed PTFE films……………………………………………..

152

Table 3.17: Properties of PFS block copolymers…………………………….. 153

Table 3.18: Atomic % of C-F2 and C-others…………………………………. 155

Table 3.19: Advancing and receding water contact angles of polymer-adsorbed PTFE films. ……………………………………………

157

Table 3.20: Advancing and receding contact angles of P(FS-b-AA) adsorbed surfaces after soaking in MilliQ water for 2 days……………......

158

Table 3.21: DH of P(FS-b-AA) in DMF and MEK a………………….……… 158

Chapter 4

Table 4.1: Different calcium phosphate structures8………………….……... 178

Table 4.2: Ion concentrations of human blood plasma, different SBF solutions,14 and SPF16………………….………………………....

179

Table 4.3: Properties of PMAEP and PMOEP polymers subjected to SBF studies………………….………………….……………………...

187

Table 4.4: LbL samples for SBF and their atomic % of Ca and P after 7 days in SBF (obtained from XPS survey scans) ………………...

196

Table 4.5: Block-copolymer-attached aminated slides for SBF……………. 199

Table 4.6: Properties of fluorinated block copolymer absorbed PTFE films.. 203

Table 4.7: Atomic % of Ca and P from XPS survey scans and the Ca/P ratios………………………………………………………….......

205

xxiii

LIST OF SCHEMES

Chapter 1

Scheme 1.1: Synthetic route to poly(vinylamine) with dextran lactone and N-(perfluoroundecanoyloxy)succinimide……………………...

17

Chapter 2

Scheme 2.1: Schematic representation of nitroxide-mediated polymerisation of styrene with TEMPO, monomer (e.g. styrene) and initiator (e.g BPO = benzyl peroxide). Pn• and Pm• are propagating polymer radicals……………………...…………………….......

26

Scheme 2.2: Schematic representation of the ATRP polymerisation process, Pn = polymer, X = halide, L = ligand, and M = monomer……..

27

Scheme 2.3: General structure of a RAFT agent, Z = activation group (e.g. aryl and alkyl) and R = good leaving group (e.g. cumyl and cyanoisopropyl) ……………………...…………………….......

28

Scheme 2.4: Schematic representation of proposed RAFT polymerisation process. 29

Scheme 2.5: Structures of RAFT agents……………………...……………... 32

Scheme 2.6: Structures of phosphate-containing monomers………………... 33

Scheme 2.7: Structures of fluorinated monomers……………………............ 36

Scheme 2.8: Hydrolysis of polymeric side-chains and dithiocarbonate moiety of PMAEP/PMOEP using NaOH solution; R1 = H (PMAEP) or CH3 (MOEP), R2 = H (PEPDTA) or CH3 (CDB) and R3 = CH2 (PEPDTA) or none (CDB) ……………………..

49

Scheme 2.9: Chain extension of (A) PAAEMA with MAEP/MOEP and (B) PMOEP with AAEMA, R = H: MAEP, CH3: MOEP…………………………………………………………...

63

Scheme 2.10: Schematic representations of chain extensions of PFS, PTFPA and PTFPMA, and further reactions (× = reaction did not proceed) ……………………...……………………...................

78

Scheme 2.11: Reaction scheme of PAAEMA block copolymer with glycine... 83

Scheme 2.12: Possible hydrolysis sites on the side-chain of the polymer and structure of resulting polymer (R = H: MAEP, R = CH3: MOEP) ……………………...……………………....................

89

Scheme 2.13: Structure of polymer from monomer-diene mixture(R = H: MAEP, R = CH3: MOEP) ……………………...……………...

90

Scheme 2.14: Mechanism of Diels-Alder dimer and 1,4-diradical formations of styrene. (Reproduced from Ref:86) M = styrene, AH = Diels-Alder dimer, •M2• = 1,4-diradical and DCB = 1,2

xxiv

diphenylcyclobutane…………………………………………. 94

Chapter 3

Scheme 3.1: Various reaction schemes of PAAEMA…………………… 107

Scheme 3.2: Structures of fluorinated homopolymers…………………….... 148

Chapter 4

Scheme 4.1: Structure of carboxylated polyphosphazene…………………... 181

Scheme 4.2: Structures of DEAEMA and VP……………………................. 183

Scheme 4.3: Possible hydrolysis sites on the side-chain of the polymer and structure of resulting polymer (R = H: MAEP, R = CH3: MOEP) ……………………...……………………...………….

207

xxv

ABBREVIATIONS POLYMERS

ePTFE Expanded Polytetrafluoroethylene

FEP Tetrafluoroethylene-hexafluoropropylene copolymer

PAA Poly(acrylic acid)

PAAEA Poly(2-(acetoacetoxy)ethyl acrylate)

PAAEMA Poly(2-(acetoacetoxy)ethyl methacrylate)

PAH Poly(allylamine hydrochloride)

PBT Poly(buthylene terephthalate)

PCL Poly(ε-carprolactone)

PDMS Poly(dimethyl siloxane)

PEG Poly(ethylene glycol)

PEI Polyethyleneimine

PEO Poly(ethylene oxide)

PET Poly(ethylene terephthalate)

PFS Poly(pentafluorostyrene)

PGA Poly(L-glutamic acid)

PHBV Poly(3-hydroxybutyrate-co-3-hydroxyvalerate

PHEMA Poly(2-hydroxyethyl methacrylate)

PLA Poly(lactic acid)

PMA Poly(methacylic acid)

PMAEP Poly(monoacryloxyethyl phosphate)

PMOEP Poly(methacryloyloxyethyl phosphate)

PNVF Poly(N-vinylformamide)

PS Polystyrene

PTFE Polytetrafluoroethylene

PTFPA Poly(tetrafluoropropyl acrylate)

PTFPMA Poly(tetrafluoropropyl methacrylate)

PU Polyurethane

PVAm Poly(vinylamine)

PVDF Poly(vinylidene fluoride)

PVP Poly(2-vinylpyridine)

xxvi

MONOMERS

AA Acrylic acid

AAEA 2-(acryloyloxy)ethyl acetoacetate

AAEMA 2-(methacryloyloxy)ethyl acetoacetate

AAm Acrylamide

DEAEMA (Diethylamino)ethyl methacrylate

DMAA N,N-dimethylacrylaminde

DMAEMA (N,N-dimethylamino)ethyl methacrylate

FS 2,3,4,5,6-pentafluorostyrene

HEMA 2-Hydroxyethyl methacrylate

MAA Methacrylic acid

MAEP Monoacryloxyethyl phosphate

MOEP Methacryloyloxyethyl phosphate

NaSS Sodium styrenesulfonate

tBA tert-butyl acrylate

TFPA 2,2,3,3-tetrafluoropropyl acrylate

TFPMA 2,2,3,3-tetrafluoropropyl methacrylate

VBC 4-vinylbenzyl chloride

VP 1-vinyl-2-pyrrolidinone

SOLVENTS AND OTHER CHEMICALS

APS 3-Aminopropyltrimethoxysilane

AIBN 2,2-Azobis(isobutyronitrile)

BMPs Bone morphogenic proteins

CDB Cumyl dithiobenzoate

CPDA Cumyl phenyldithioacetate

DCB 1,2-Diphenylcyclobutane

DCM Dichloromethane

DMF Dimethylformamide

DMSO Dimethyl sulfoxide

EA Ethyl acetate

EDC 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride

FB Fluorobenzene

xxvii

hTM Human Thrombomodulin

MEK Methyl ethyl ketone

PBS Phosphate-buffered saline

PEPDTA 1-Phenylethyl phenyl dithioacetate

PPDB 2-Phenyl-prop-2-yl dithiobenzoate

RGD Arginine-Glycine-Aspartate

SBF Simulated body fluid

SPF Simulated physiological fluid

TEMPO 2,2,6,6-Tetramethylpiperidinyloxy

TFA Trifluoroacetic acid

THF Tetrahydrofuran

Vazo 88 1,1’-Azobis(cyclohexanecarbonitrile)

INSTRUMENTS AND TECHNIQUES

AFM Atomic Force Microscope

ATR Attenuated Total Reflection

CVD Chemical vapour deposition

DLS Dynamic light scattering

EDX Electron Dispersive X-ray

ESCA Electron Spectroscopy for Chemical Analysis

FTIR Fourier Transform Infrared Spectroscopy

FT-NIR Fourier Transform Near-Infrared Spectroscopy

GPC Gel Permeation Chromatography

IRRAS Infrared Reflection-Adsorption Spectroscopy

NMR Nuclear Magnetic Resonance

SEM Scanning Electron Microscope

SSIMS Static Secondary Ion Mass Spectroscopy

ToF Time-of-Flight

XPS X-ray Photoelectron Spectroscopy

xxviii

OTHERS

ATRP Atom transfer radical polymerisation

CSIRO Commonwealth Scientific and Industrial Research Organization

DH Number-average hydrodynamic diameter

GBR Guided bone regeneration

ITP Iodine transfer polymerisation

LbL Layer-by-Layer

Mn Molecular weight

NMP Nitroxide-mediated polymerisation

PDI Polydispersity index

RAFT Reversible addition fragmentation chain transfer

SAMs Self Assembled Monolayers

TCPS Tissue Culture Polystyrene

xxix



1.1 Biomaterials and Biocompatibility

A biomaterial is defined as a material used to replace part of a living system or used

in close contact with a living system. Metals, ceramics, polymers, glasses, and

composites have all been widely investigated and used as biomaterials.1 Biomaterials

science is a multidisciplinary field, which encompasses not only the synthesis and

development of suitable materials but also the evaluation of their degradation,

mechanical, physical and chemical properties. The study of the complex host

response to the introduced bulk material as a whole, as well as the complex cascade

of events occurring at the material’s surface are essential parts of biomaterial-science.

Biocompatibility is the ability of a biomaterial to perform with the appropriate

response in a specific application.2 Biocompatibility, therefore, cannot be defined in

terms of the properties of a material alone; instead, it is a combination of its specific

properties and the function for which it is intended. Biocompatibility assessment is

an essential factor for determining the success of devices or materials intended for

use in contact with the body.2

Polymeric biomaterials have proven their use in a range of applications in many

areas of medicine due to their diverse chemical compositions, chemical and

mechanical properties as well as their ability to be manufactured in a wide range of

structural forms.3 There are many polymeric biomaterials on the market that meet

their bulk biocompatibility requirements. However, not many of these polymers

possess ideal surface properties. Since the first interaction of an implant with the

body is through its surface, the initial acceptance of the material in the body is highly

dependent on its surface properties.

Polymer surfaces are not rigid, which makes them very complicated as the surface

composition depends on the environment (e.g. air vs liquid). Consideration of many

parameters is required, including topography, chemical composition, surface energy,

wettability, crystallinity, surface mobility and heterogeneity. It is not yet fully

understood which parameters are the most important when considering biological

response to polymer surfaces. However, analysing as many parameters as possible

1


gives a more complete understanding of the surface and its interface reactions with

the body.

The material/tissue interface is a very complex system and is still not fully

understood. Agreement as to what constitutes, optimal surface property is also a

matter of some contention and, of course, depend on the specific application. For

example, calcification and protein adhesion onto contact lenses surfaces is not ideal

and there has been significant research focusing on how to prevent these processes

from occurring. On the other hand, bone implants require bioactive surfaces that can

support cell growth and mineralisation for direct bone bonding. Since the focus of

this PhD thesis is the creation of more reactive surfaces in clinically important

polymers in order to facilitate better bone bonding, it is necessary to consider the

nature of bone and bone fracture healing as well as the polymer/bone interface.

1.2 Bone

Bone is a unique composite material, in the sense that it consists of both an inorganic

mineral phase and organic macromolecules. It is often referred to as a “living

mineral” because it undergoes continual growth: both dissolution and remodelling

occur in response to either internal signals or external force fields, such as gravity.

The chemical composition of the two types of human bone is shown in Table 1.1.

Table 1.1: Composition of human trabecular and cortical bone (vol%).4

The inorganic mineral component of human bone is carbonated hydroxyapatite. The

organic matrix is composed of an insoluble framework of collagen fibrils (~90%),

and water soluble non-collageneous proteins (~10%) as well as trace amounts of

many other types of proteins. This special combination of inorganic and organic

2

halla

This table is not available online. Please consult the hardcopy thesis available from the QUT Library


components leads to a biological organic-inorganic composite material, with unique

mechanical properties.

The two principal types of bone are identifiable by their macroscopic structures:

cortical (or compact) bone and trabecular (or cancellous) bone. Cortical bone is a

dense tissue with a porosity of about 10%, and is primarily found in the shaft of long

bones and the outer shell around cancellous bone at the end of joints. Trabecular

bone is highly porous (50-90% porosity), and is found in the end of long bones and

in flat bones like the pelvis. Many of the unique properties of bone are a result of its

hierarchically-organized structure, which is shown in Figure 1.1.

.

Figure 1.1: Hierarchical structure of bone. (Reproduced from ref.5)

At the nanostructure level, the three main components of fibrils: mineral apatite

crystals, collagens, and non-collageneous proteins are observed. Collagen molecules

together with non-collagenous proteins constitute the organic matrix that performs

important functions including the following:

• Mechanical design control – strength and elasticity of bone. The combination

of organic and inorganic materials produces a strength greater than either the

components alone.4

3

halla

This figure is not available online. Please consult the hardcopy thesis available from the QUT Library


• Mineral stabilisation – surface stabilisation of minerals from dissolution or

phase transition.6,7

• Mineral nucleation – controls the location, as well as organization of the

nucleation sites. This in turn leads to control of the structure and orientation

of the inorganic phase.6

Although the collagen matrix, itself, does not possess either nucleation sites

or the template for mineral deposition, the non-collagenous proteins, such as

osteonectin and some phosphoproteins (which have high proportion of

anionic groups that have high affinities for calcium ions) absorbed onto the

collagen network are believed to be the biomineralisation nucleators.6,8

Four different types of bone cells have been identified: osteoblasts (bone-forming

cells), osteocytes (mature cells embedded in the bone matrix), bone-lining cells

(living in the surface of bone), and osteoclasts (bone-resorbing cells). Osteoblasts are

derived from the mesenchymal stem cells of the bone marrow stroma. They

synthesise, and lay down, the precursors of type I collagen. Osteoblasts produce

osteocalcin, as well as proteoglycans, and are rich in alkaline phosphatase (an

organic phosphate-splitting enzyme). Hence, these cells are know as bone-forming

cells and are important for biomineralisation.

Osteocytes are developed from osteoblasts and live in the bone. It has been suggested

that osteocytes can sense the mechanical load in bone from the fluid flow stresses

occurring inside the bone.8 These cells then send signals which activate the

osteoclasts and osteoblasts. Osteocytes may also participate in bone resorption to

increase the level of calcium ions whenever it is demanded.9

Bone lining cells are thought to be either inactive osteoblasts which may be activated

or a cell type of their own. They are elongated, thin cells which cover the bone

surfaces that are not under remodeling.

Osteoclasts, typically multinucleated, are derived from monocytes which are

originally derived from hematopoietic stem cells. Osteoclasts are the only cell type

that can resorb bone. They may be recruited and activated through the signals from

osteoblasts which are initially activated by osteocytes.10 It has been shown that

4


osteoclastic bone resorption does not occur in the absence of osteoblasts or added

stimuli.

1.3 Host Response to Polymeric Biomaterials

The host tissue reaction induced by an implant material is an important factor that

determines the biocompatibility of any material. Another factor is the degradation

behaviour of the material in the body. These two factors are often correlated, hence a

separate interpretation may prove difficult. Moreover, the host response to the

implant is dependent not only on material properties, such as its physical structure

and chemistry, but also the nature of the implantation site.

1.3.1 Wound healing

There are many changes in both the types and numbers of cells present at a site of

soft-tissue repair, as shown in Figure 1.2.11 Varying concentrations of chemicals and

electrolytes in body fluid are also observed during soft-tissue repair. The three

processes that occur following injury are: hemostasis, inflammation, and wound

healing.

Figure 1.2: Cellular activities at wound repair. (Reproduced from ref.11)

5

halla



The first stage (hemostasis) happens within a few minutes. Blood platelets fill the

injury site forming a temporary shield from the environment. Two important types of

leukocytes (white blood cells) that participate in the inflammation stage are

neutrophils and macrophages. These cells migrate to the wound as a result of a

number of chemicals released during injury.11 Neutrophils kill bacteria, while

macrophages remove cellular and foreign debris from the wound site. Macrophages

stay at the injury site and participate in the next stage of wound healing.

Macrophages break up the fibrous clot formed by the blood platelets so that new

blood vessels can penetrate the wound site. Capillaries bud and grow from the blood

vessels into nearby tissue, and proliferate into the injury zone and form an

interconnecting network. This network supplies oxygen and nutrients required by

other cells present at the site. Fibroblasts then migrate and synthesise extracellular

substances, such as collagen. The synthesised collagen binds into fibres, which

randomly fill the wound, forming scar tissue.

The difference between the healing mechanism of hard tissue from that of soft tissue

is the unique ability of hard tissue to regenerate without scarring.12 Where bone is

concerned, in general, there are two types of fractures: closed and open fractures.

Closed fractures have no contact with the external environment; therefore, the

healing process is only affected by local factors. Since open fractures have contact

with the outside environment, they must heal with concomitant ongoing soft tissue

regeneration, which may involve additional signals and cell types at the bone

surfaces. When a biomaterial is implanted, it is thus more complex healing scenario

that is required.

It is important to note the terms “bone bonding”, “osteointegration” and

“osseointegration” all have the same meaning of “direct bone contact of a material

with bone without interposition of non-bone tissues such as fibrous capsules”. This

provides a direct structural and functional connection between living bone and the

materials surface.

6


1.3.2 Tissue response to implants

The surgical procedures required in the introduction of any biomaterial device create

a wound and hence triggers a “wound-healing response”. Inflammation, wound

healing and foreign body response, as well as fibrous encapsulation, are generally

considered as the typical host/tissue responses to an implant.13 Inflammation is an

essential reaction occurrence at the implantation site before healing can begin. The

normal reaction of biological systems to introduced materials is to biodegrade them

or attempt digestion. Failing this, the alternative is encapsulation of the introduced

material within a fibrous collagen capsule.13 This foreign body reaction involves

foreign body giant cells and other cells such as macrophages and fibroblasts that are

already present at the wound site.

Once a fibrous capsule is formed, the material is isolated from the body.

Vascularisation cannot occur inside the capsule; therefore, no cell growth is possible.

If bacterial infections occur inside this fibrous capsule, the infection may be

prolonged as macrophages are unable to access the interior of the capsule and do the

job for which they are designed. Excessive fibrous growth can result in tissue

damage and ultimately tissue death.

As mentioned earlier, when a biomaterial is inserted into bone, a situation similar to

that of an open fracture is created. The resulting activation of the macrophages can

result in the formation of a fibrous layer or even fibrous encapsulation. This, in turn,

can lead to serious consequences such as micromotion which leads to implant

loosening and even bone resorption. In other words, good bone bonding is critical for

the successful functioning of the biomaterial implant or device.

Consequently, many implants including materials for cranio-maxillo-facial

applications require direct bone bonding without the interposition of fibrous capsules.

It is generally believed that bone bonding will take place when a layer of apatite,

similar to bone in crystallinity and composition, forms spontaneously on their surface

in contact with blood plasma.14 The mineralisation process can occur through cell-

and material-directed pathways. This depends on the material surface properties

which will be discussed in the following section.

7


1.4 Material/Bone Interface

When a material is implanted, a cascade of events occurs at the surface (Figure

1.3).15 More specifically, water molecules, which constitute 70 wt% or more of our

body, are the first to come in to contact with the implant material surface and form a

water layer (Figure 1.3 A-B) that is followed by hydrated ions (C). The behaviour of

water near the surface has been recognised to play an important role for the

following events.15,16 The various biomolecules, in particular proteins, come and

interact with the hydrated surface either through or by displacing a water layer (D-E).

When the cells arrive at the surface, they encounter the adsorbed protein layers (F). If

the proteins are denatured, the cells will not attach or may even activate a

macrophage response if they recognise the material as a foreign substance. This can

lead to fibrous encapsulation. Therefore cell and tissue interaction with material

surfaces is mediated by the adsorbed proteins.17,18

Figure 1.3: Schematic illustration of the successive events following implantation of a material. (Reproduced from ref.15)

8

halla



The proteins involved in osteoblast adhesion onto material surfaces include

extracellular matrix proteins, cytoskeletal proteins, integrins and cadherins.19 The

types, amounts, and conformations of adsorbed proteins on material surfaces

determine the osteoblast cell behaviour. Eventually cell-directed mineralisation can

occur if the adsorbed proteins signal osteoblast adhesion, spreading and

differentiation.6

Functional groups play an important role in cell behaviour via protein attachment as

well as for mineralization. The effect of functional groups on protein adsorption and

endothelial cell growth have been investigated using self-assembled monolayers

(SAMs) of alkanethiolates on gold.20 It was found that cell growth increased in the

following order -CH2OH < -C(O)OCH3 < -CH3 « -C(O)OH. However, tissue culture

polystyrene (TCPS) showed better cell growth than SAMs. XPS analysis of TCPS

revealed a surface composition of over 12% oxygen present as various polar groups

(i.e. hydroxyl, carbonyl, carboxylic acid, and esters). As well as oxygen, small

amounts of nitrogen (0.56%) were also detected by XPS analysis.21 The results of

such cell growth studies seem to indicate that multiple functionalities provide a

pronounced synergistic effect.20

The incorporation of functional groups to increase the hydrophilicity of polymer

surfaces is a common technique for improving cell adhesion on hydrophobic polymer

surfaces.22 Hydrophobic surfaces are known to induce strong irreversible protein

adsorption. This denatures their conformation and hence causes a loss in bioactivity.

Very hydrophilic surfaces, on the other hand, inhibit protein adhesion. Grafting of

water soluble poly(ethylene oxide) (PEO) has been shown to decrease the interfacial

free energy and the steric repulsion forces between the PEO chains and proteins in

polymers used for vascular grafts applications.23 It is now accepted that moderately

hydrophilic surfaces induce a positive cell response because these surfaces are able

to adsorb proteins without causing denaturing.

In addition to introducing functional groups to change the wettability of a material’s

surface, mineralisation induced by different functional groups is also important.

Apatite formation was found to be significantly enhanced on negatively charged

SAMs, compared to positively charged, or neutral surfaces.24,25 The mineral growth

9


rate decreased in the order: -PO4H2 > -COOH » -CONH2 ≃ -OH > -NH2 » -CH3 ≃

0.24

1.5 Surface Modification: Improving the Material/Bone Interface

1.5.1 Physicochemical Surface Modification

The physicochemical surface modification of polymers in order to incorporate new

functional groups has long been studied for use in a wide range of applications.

Many techniques are available, including chemical, plasma, and grafting methods.

The introduction of new functional groups gives rise to different surface properties

such as: surface energy, water-retaining capacity, and surface mobility. All of which

are important for tissue response.26

Chemical treatments, including acid or base etching that produce hydroxyl and

carboxyl groups, have been used for both non-degradable21 and biodegradable

polymeric biomaterials.27,28 Aminolysis techniques have been used for modifying

ester-containing polymers such as poly(3-hydroxybutyrate-co-3-

hydroxyvalerate)(PHBV), poly(ε-caprolactone) (PCL), poly(lactic acid) (PLA) and

polyurethane (PU).29-32 In such treatments, the polymers are treated with solutions

containing diamino compounds (e.g. diaminohexane). Despite their effectiveness,

chemical methods have several disadvantages. Degradation of the bulk properties by

the harsh chemical reagents often occurs. For biodegradable polymers, decreases in

molecular weight have been observed. It is also difficult to monitor and control the

modification depth profile. In the case of poly (lactic-co-glycolic acid) (PLGA),

chemical treatments have been optimised for minimal degradation.30

Plasma modification using glow discharges from gases such as air, O2, H2O vapour,

CO2, NH3, N2, and SO2 are frequently used to introduce functional groups such as

hydroxyl, carboxyl, amino and sulphate groups on polymeric surfaces.33 Plasma

treatment is also used for the deposition or immobilization of molecules.34 An O2-

plasma-treated PCL surface, which was subsequently alternatively dipped into

calcium and phosphate solutions, showed HAP formation after 24 hours in SBF.35 A

poly(ethylene glycol)/poly(butylene terephthalate) (PEG/PBT) segmented block

10


copolymer, which was treated with an O2 plasma exhibited improved bone marrow

stromal cell attachment and growth, which was equivalent to the result obtained for

an O2-plasma-treated polymer/HAP composite.36 However, a serious drawback of

these plasma techniques is the aging effect of the treated surface in air, which is

dependent on the environment and temperature. Although its technical operation is

simple, it is very difficult to control the actual amount of a particular functional

group formed on a surface.

Graft copolymerisation offers stable functional groups by covalently immobilising

new polymer brushes.37 Active sites (radicals or peroxides) are produced randomly

along the surface polymer chains by an electron beam, ozone oxidation, X-rays, γ-

rays or Ar plasma treatments. These sites initiate graft copolymerisation. Grafting is

applicable to many monomer/polymer combinations. The grafting of phosphate-

containing monomers has been found to induce calcium-phosphate growth38-40 with

grafted materials showing direct bone bonding in vivo.41 The incorporation of

negatively-charged functional groups onto polymeric surfaces is discussed further in

Chapter 4.

1.5.2 Morphological surface modification

Surface micro- and nano-scale patterning has shown significant influence on cell

behaviour such as: cell shape, migration, and protein synthesis.42,43 Osteoblast

proliferation has been found to be sensitive to surface topography.44 Surfaces with

grooves have been shown to induce “contact guidance”, which influences cell

spreading, alignment and migration.

It has been reported that bone cells aligned themselves parallel to the direction of the

grooves on a polystyrene (PS) surface with 5μm-deep grooves, but they did not

respond to the surface with 0.5 μm-deep grooves (both grooves were 5 μm wide).45

On the other hand, poly(L-lactic acid) (PLA) and PS surfaces with microgrooves of

0.5, 1.0, and 1.5 μm depth and 1, 2, 5, and 10 μm width showed enhanced

differentiation of osteoblast-like cells and mineral formation.46 Another study

showed that a polydimethylsiloxane (PDMS) surface with microtextures of 6 μm

11


high and 5-40 μm in diameter promoted spreading and adhesion of human bone-

marrow-derived cells.47

In general, smooth surfaces exhibit less cell adhesion than rough surfaces. In the case

of biodegradable scaffolds, surface topography and roughness have been shown to be

important in allowing the migration of cells on the scaffold surface.48 Osteoblast-like

cells seeded onto a rough PLLA surface, which had an island pattern, preferred to

elongate and showed uniform growth compared to those on a smooth PLLA

surface.49

1.5.3 Incorporation of biological molecules

The two techniques discussed in the previous sections cannot induce specific cell

behaviour due to non-specific protein adhesion. Immobilization of biomolecules onto

the surface, on the other hand, provides a way to control specific cell and tissue

responses by directly delivering molecules with a defined function to the tissue-

implant interface. Many different biologically active molecules have been either

covalently or physically immobilised onto surfaces in order to enhance biological

interaction. Some examples are proteins, peptides, polysaccharides, lipids, drugs,

ligands, and nucleic acids.50

Covalent immobilisation requires the presence of reactive groups such as –OH, –SH,

–NH2, or –COOH. Because many polymeric biomaterials do not possess these

reactive groups, surface modification, such as graft copolymerisation, is necessary in

order to introduce them. Figure 1.4 shows a schematic representation of the

introduction of peptides onto a pre-treated polymer substrate.

OO

Figure 1.4: Covalent immobilisation of RGD peptides with carboxyl groups on a modified polymer surface.

OH

O

OH

O

NOH

OO

NO

O

O

NO

O

O

NH

O

RGD

NH

O

RGDEDC

H2N–RGD

e.g poly(acrylic acid) grafting

COOH incorporation

12


RGD peptides (R: arginine, G: glycine, D: aspartic acid) have been found to promote

cell adhesion and many RGD-functionalized polymeric biomaterials have been

widely studied.51 RGD-functionalized poly(ethylene glycol-b-D,L-lactic acid),52 and

a RGD-functionalized poly(lactic acid) scaffold53 have shown significant

improvements in osteoblast attachment and spreading compared to the unmodified

materials.

Growth factors, such as bone morphogenic proteins (BMPs), are also known to

stimulate local bone regeneration. Biodegradable polymers are often used as carriers

for BMPs.54-57 BMPs have also been covalently immobilised on an allyl amine

plasma polymerised surface, where it induced significant osteoblastic activity in

vitro.58

Some examples of physical immobilisation are coating of protein layers, electrostatic

interaction, entrapment, and Layer-by-Layer techniques using negatively and

positively charged biomolecules. The entrapment technique involves immobilising

biomolecules onto the swollen polymer surface, as shown in Figure 1.5.22

Figure 1.5: Entrapment of biomolecules into a polymer substrate.

Klee et al.59 investigated human osteoblast behaviour on surfaces with both

physically and covalently attached fibronectin. Poly(vinylydenefluoride) (PVDF)

was first pre-treated by either the plasma graft copolymerisation of acrylic acid, or

the chemical vapour deposition (CVD) of an amine-containing monomer, followed

by immobilisation of fibronectin using different techniques. Although both

physically adsorbed and covalently attached fibronectin surfaces showed enhanced

cell attachment, proliferation was only enhanced on the surfaces with the covalently

Surface swelling Entrapment

e.g. PLA In a solvent and

non-solvent mixture with biomolecules

In pure non-solvent

biomolecules

13


bound protein. This was interpreted as supporting the view that the long term

presence of biomolecules is necessary for cell adhesion and proliferation.59

Despite showing many promising results, there are also some concerns with covalent

immobilization. It has been shown that the covalent immobilization of enzymes and

proteins resulted in a substantial loss of activity due to the loss of active sites by

bonding or reorientation of the biomolecules after bonding.60 Sterilization becomes

more difficult, fouling by other biomolecules occurs, and adverse biological

responses of enzyme-supported surfaces have all been reported.50

1.6 Expanded PTFE (ePTFE) in Medicine

In 1969, expanded PTFE (ePTFE) was developed by W.L. Gore and associates.61

ePTFE has a highly fibrillated structure with an improved resistance to creep.

Although ePTFE is soft, and very flexible, it is also very strong and extremely

resistant to stretching or tearing. Because of its bioinertness, ePTFE has found many

applications in medicine, including vascular grafts,62-65 cranio-maxillo-facial

surgery,62 and guided bone regeneration (GBR).66-68 It is also used for catheter

coatings, sutures, aneurism clips, and oxygenation membranes. In general, ePTFE

does not support soft-tissue growth and ingrowth in vivo, which is a benefit for use in

vascular grafts and GBR. However, for cranio-maxillo-facial applications, where

implants are in contact with both soft and hard tissues, a more bioactive surface is

desirable.

1.7 Surface Modification of PTFE and Other Fluoropolymers

Fluoropolymers have a wide range of applications in industry due to their

outstanding properties such as high-temperature stability, excellent chemical

resistance, low water sorption, and low dielectric constant. Depending on the

application, the low surface energy of fluoropolymers, which results in poor adhesion

to other materials, is not desirable. Surface modification of these materials has been

widely studied, including modification for medical applications.69,70

Chemical modification of fluoropolymers has been successfully used for

defluorination and refunctionalisation.69 However, as mentioned before harsh

14


chemical treatments alter PTFE bulk properties. The chemical treatment of ePTFE

has been reported to cause destruction of the structural integrity even under relatively

mild conditions.71 Although fluoropolymers are sensitive to irradiation treatment,

radiation treatments, such as: X-ray, γ-ray, UV, laser, electron beam and ion beam

methods have all been investigated for the surface modification of

fluoropolymers.69,72 Most of the irradiation treatments result in the formation of

alkyl-type radicals and lead to the formation of carbon-carbon double bonds. The

surface modification of ePTFE (Gore-Tex) by N2+, Ar+, and Ca+ ion implantation has

also been reported.73

Plasma treatments of fluoropolymers have been carried out with glow discharges

generated from various gases including: H2,74 CH4,74 He,74,75 Ne,75 O2,76 H2,77 N2,76,78

Ar,74,76,78,79 a H2O/Ar mixture,80 SO281, NH3

74,76,82-84, and an NH3/O2 mixture85. The

aging process of O2, Ar, N2, and NH3 plasma-treated PTFE surfaces in air, and

phosphate-buffered saline (PBS), have been investigated for up to 1 month.86,87

Lappan et al.88 reported the surface modification of PTFE using either O2 or NH3

plasma-treatment followed by adsorption of polyelectrolytes.

Radiation-induced grafting onto fluoropolymers is a well-established method89 and

wide varieties of monomers have been grafted onto PTFE (and ePTFE) this way.

Such monomers include: styrene,90-94 MMA,94-98 2-hydroxyethyl methacrylate

(HEMA),94,97 acrylic acid (AA),99-102 methacrylic acid,97,98 acrylamide (AAm),103

N,N-dimethylacrylamide (DMAA)104-106, PEO107,108, MAEP102 and MOEP109.

Another common grafting technique of PTFE is plasma activation. Kang and

coworkers have reported UV-induced graft copolymerisation of hydrophilic

monomers onto Ar-plasma-pretreated PTFE.110 The hydrophilic monomers used in

Kang’s study included: AAm, AA, the Na salt of styrenesulfonic acid (NaSS),

DMAA, and (N,N-dimethylamino)ethyl methacrylate (DMAEMA).110 Tu et al.111

used ozone treatment on H2-plasma-pre-treated PTFE to create hydroperoxides, and

peroxides, which were then used to graft AAm and NaSS. AA grafting onto plasma-

pre-treated PTFE has also been extensively studied.112,113

More recently, the surface-initiated living radical polymerisation of fluoropolymers

has been reported using atom transfer radical polymerisation (ATRP),114 and

reversible addition-fragmentation chain transfer (RAFT)115 polymerisation. The

15


PTFE surfaces were pre-treated with plasma, in combination with ozone114, to create

active sites. In the case of ATRP, the initiators were immobilised on the plasma-

treated PTFE surface. For the RAFT techniques, graft copolymerisation of HEMA

onto plasma-treated tetrafluoroethylene-hexafluoropropylene copolymer (FEP) was

conducted in the presence of a RAFT agent and N,N-dimethylaniline.115 The addition

of N,N-dimethylaniline was to accelerate the decomposition of the peroxides formed

during the surface pre-treatment. Yu et al.116 grafted comb-copolymer brushes onto

PTFE using both RAFT, and ATRP techniques. Glycidyl methacrylate was first

grafted onto an Ar-plasma-treated PTFE surface using a RAFT technique. ATRP

initiators were then attached onto the epoxide side-chains of the grafted polymer.

Hydrophilic monomers were finally grafted from the ATRP initiator sites to form

comb structures.

Localised grafting onto a PTFE surface has also been reported. The grafting of 2-

dimethylaminoethyl methacrylate (DMAEMA) through a redox catalysis process

onto n-doped PTFE surfaces which had been obtained by local scanning

electrochemical microscopy has been achieved.117 This concept developed from

work on electro-grafting of polymers onto conducting, and semi-conducting

electrodes. The electro-grafting process is made possible by the electro-reduction of

vinylic monomers.

Grafted PTFE surfaces have also been used to immobilise biomolecules. Covalent

immobilisation of fibronectin with the carboxylic acid groups of poly(methacrylic

acid)-grafted PTFE has shown improved healing for vascular grafts.118,119 Human

thrombomodulin (hTM) has been immobilised onto poly(acrylic acid)-grafted PTFE.

The modified surface exhibited the expected improved anticoagulation activity.120,121

Fluorosurfactants can be used for the surface modification of fluoropolymers. This

modification method involves noncovalent interactions between the surfactant and

the polymer. Marchant and coworkers investigated the surface modification of PTFE

by the physisorption of fluorosurfactant polymers.122 Poly(vinylamine) (PVAm),

which was obtained from the hydrolysis of poly(N-vinylformamide) (PNVF), was

reacted with dextran lactone, followed by reaction with perfluorocarbon succinimide

(Scheme 1.1).

16


NH

Scheme 1.1: Synthetic route to poly(vinylamine) with dextran lactone and N-(perfluoroundecanoyloxy)succinimide.

XPS was used to characterise the fluorosurfactant-adsorbed PTFE surfaces. It was

found that adsorbed polymer, which contained 45 mol% fluorocarbon chains (y =

366 in Scheme 1.3), was stable under shear stress (1-20 dyn/cm2), whereas some

delamination of the polymers containing 15 and 21 mol% fluorocarbon chains was

observed. Marchant also created fluorosurfactants containing an RGD peptide, or

endothelial cell (EC)-selective peptide, for ePTFE surface modification.123,124 These

surfactants were used as PTFE and ePTFE surface modifiers for vascular graft

applications. The resulting surfaces facilitated endothelial cell attachment, growth,

and function.

Although PTFE is thought to be “non-stick”, due to its low surface energy,

spontaneous adsorption of biopolymers125,126 as well as poly(L-lysine) (PLL)127 from

aqueous solution, due to hydrophobic interactions, have been reported. Coupe and

coworkers have investigated the adsorption of functional polymers onto

poly(tetrafluoroethylene-co-hexafluoropropylene) (FEP) from aqueous solution.128

Since the interfacial energy of FEP/water is high, the polymers are thought to adsorb

with the functional groups extending to water (see Figure 1.6).

Figure 1.6: Adsorption of functional polymers at the FEP/water interface.

CH

O

** n

NH2

** n

NH* x

NH2

*m

Dextran

NH* x

NHy

Dextran(CF2)9

O

CF3

NH2

*n

Dextran Lactone NaOH

DMSO DCC, NHSH2O

C10F21COOH

x = 179 y = 122-366 n = 269-512 Total = 814

X X

X X

X X X

X = COOH or NH2

H2O Polymer Adsorption

FEP

17


Poly(acrylic acid) (PAA), poly (allylamine hydrochloride) (PAH), polyethyleneimine

(PEI) and PLL all showed adhesion to FEP. The highest adsorption of these

polymers to FEP was observed at pH values where the polymers were not charged.

This is due to the low solubility of these polymers in this state. Salt addition

increased adsorption of these polymers to FEP due to screening of the repulsive

interactions between charged groups.

1.8 Project Outline

It is widely accepted that polymeric materials used in orthopaedic applications

generally require some kind of surface modification to improve their bioactivity.

Surface modification has been shown to enhance osseointegration and bone ingrowth.

In previous studies, it has been shown that it is possible to graft phosphate-containing

monomers onto expanded PTFE (ePTFE) using a radiation-induced grafting

technique.102,109 The modified surfaces showed enhanced mineralisation in vitro.38,39

However, it was difficult to control the structure of the grafted polymers since the

phosphate-containing monomers used in the early studies are known to form

insoluble gels by conventional polymerisation.129,130

The approach in this study represents a new direction in the surface modification of

biomaterials in particular PTFE. It involves synthesis of well-defined phosphate- and

carboxylate-containing polymers by RAFT-mediated polymerisation, followed by

the surface fabrication using these synthesised polymers. The fluorinated segment of

the block copolymers allows adsorption onto PTFE surfaces, while not requiring pre-

treatment of PTFE. This technique allows control of the molecular weights of

polymers, as well as control over the absorbed mass, and spatial distribution of

adsorbed polymers.

There are three equally important experimental sections within this thesis: synthesis

(Chapter 2), surface fabrication (Chapter 3) and in vitro mineralisation (Chapter 4).

These are followed by the overall conclusions and proposed future work (Chapter 5).

The outlines for chapters 2-4 are as follows:

18


Chapter 2: Synthesis

1. Synthesis of well-defined phosphate-containing polymers from

monoacryloxyethyl phosphate (MAEP) and methacryloyloxyethyl phosphate

(MOEP) using RAFT-mediated polymerisation

2. Chain extension of phosphate polymers with an 2-(methacryloyloxy)ethyl

acetoacetate (AAEMA) keto group

3. Synthesis of well-defined fluorine-containing polymers from 2,3,4,5,6-

pentafluorostyrene (FS), 2,2,3,3-tetrafluoropropyl acrylate (TFPA), and

2,2,3,3-tetrafluoropropyl methacrylate (TFPMA) using RAFT-mediated

polymerisation

4. Chain extension of fluorinated polymers with FS, tert-butyl acrylate tBA, and

AAEMA/2-(acryloyloxy)ethyl acetoacetate (AAEA)

5. Amino acid attachment onto PAAEMA block copolymers


1. Layer-by-Layer (LbL) assembly of phosphate- and carboxylate-containing

homopolymers

2. Coupling of PAAEMA block copolymers onto aminated slides

3. Adsorption of fluorinated homo- and copolymers onto PTFE films

1 and 2 are model systems and the substrates used were silicon wafer/glass slides and

aminated glass slides, respectively. These functional groups can be introduced onto

PTFE surfaces, or other biomaterials’ surfaces, by pre-treatments such as plasma, as

discussed in section 1.3. Adsorption of fluorinated polymers onto PTFE does not

require any pre-treatment.

Chapter 4: In vitro mineralisation

Mineralisation of phosphate-containing homopolymers, and fabricated surfaces was

studied using SBF for 1-2 weeks. Samples used for this study are as follows:

1. Phosphate-containing gel and soluble polymers

2. Phosphate- and carboxylate-containing LbL surfaces

3. Phosphate-containing block copolymers coupled to aminated slides

4. Carboxylic acid-containing fluorinated copolymers adsorbed onto PTFE films

19


1.9 References

(1) Ratner, B. D. In Biomaterials science : an introduction to materials in medicine; Ratner, B.D., Hoffman, A.S., Schoen, F.S., and Lemons, J.E., Ed.; Academic Press: San Diego, 1996.

(2) Jones, A. J., and Denning, N.T. Polymeric Biomaterials (Bio- and Eco-compatible polymers): A perspective for Australia; Dept. of Industry, Technology and Commerce: Canberra, 1988.

(3) Visser, S. A., Hergenrother, R.W., and Cooper, S.L. In Biomaterials science : an introduction to materials in medicine; Ratner, B.D., Hoffman, A.S., Schoen, F.S., and Lemons, J.E., Ed.; Academic Press: San Diego, 1996.

(4) Ontanon, M., Aparicio, C., Ginebra, M.P., and Planell, J.A. In Structural biological materials; Elices, M., Ed.; Pergamon: UK, 2000.

(5) Rho, J. Y., Kuhn-Soearing, L., and Zioupos, P. Medical Engineering & Physics 1998, 20, 92-102.

(6) Mann, S. Biomineralization : principles and concepts in bioinorganic materials chemistry; Oxford: N.Y., 2001.

(7) Weiner, S., and Traub, W. FEBS lett 1986, 206, 262-266. (8) Sikavitsas, V. I., Temenoff, J.S., and Mikos, A.G. Biomaterials 2001, 22,

2581-2593. (9) Shimizu, H., Sakamoto, M., Sakamoto, S. J Bone Miner Res 1990, 5, 411-418. (10) Lerner, U. H. Matrix Biology 2000, 19, 107-120. (11) Hench, L. L., and Ethridge, E.C. Biomaterials: An interfacial approach;

Academic Press: New York, 1982. (12) Rosen, V. a. T., R. S. The cellular and molecular basis of bone formation and

repair; R.G. Landes: Austin, 1995. (13) Anderson, J. M. In Biomaterials science : an introduction to materials in

medicine; Ratner, B.D., Hoffman, A.S., Schoen, F.S., and Lemons, J.E., Ed.; Academic Press: San Diego, 1996.

(14) Jagur-Grodzinski, J. Reactive & Functional Polymers 1999, 39, 99-138. (15) Kasemo, B., and Gold, J. Adv Dent Res 1999, 13, 8-20. (16) Vogler, E. A. Adv Colloid Interf Sci 1998, 74, 69-117. (17) Kasemo, B. Surf Sci 2002, 500, 656-677. (18) Wilson, C. J., Clegg, R.E., Leavesley, D.I., and Pearcy, M.J. Tissue Eng 2005,

11, 1-18. (19) Anselme, K. Biomaterials 2000, 21, 667-681. (20) Tidwell, C. D., Ertel, S.I., and Ratner, B.D. Langmuir 1997, 13, 3404-3413. (21) Dalton, B. A., McFarland, C., Gengenbach, T.R., Griesser, H.J., and Steele,

J.G. J Biomater Sci Polymer Edn 1998, 9, 781-799. (22) Ma, Z., Mao, Z., and Gao, C. Colloids and Surfaces B: Biointerfaces 2007,

doi:10.1016/j.colsurfb.2007.06.019. (23) Mao, C., Qiu, Y., Sang, H., Mei, H., Zhu, A., Shen, J., and Lin, S. Adv

Colloid Interf Sci 2004, 110, 5-17. (24) Tanahashi, M., and Matsuda, T. J Biomed Mater Res 1997, 34, 305-315. (25) Liu, Q., Ding, J., Mante, F.K., Wunder S.L., and Baran, J.R. Biomaterials

2002, 23, 3103-3111. (26) Tengvall, P. In Bio-implant interface: Improving biomaterials and tissue

reactions; Ellingsen, J. E., and Lyngstadaas, S.P., Ed.; CRC Press: Florida, USA, 2003.

(27) Murphy, W. L., and Mooney, D.J. J Am Chem Soc 2002, 124, 1910-1917.

20


(28) Park, G. E., Pattison, M.A., Park, K., and Webster, T.J. Biomaterials 2005, 26, 3075-3082.

(29) Keen, I., Broota, P., Rintoul, L., Fredericks, P., Trau, M., and Grøndahl, L. Biomacromolecules 2006, 7, 427-434.

(30) Croll, T. I., O'Connor, A.J., Stevens, G.W., and Cooper-White, J.J. Biomacromolecules 2004, 5.

(31) Zhu, Y. B., Gao, C.Y., Liu, X.Y., and Shen, J.C. J Biomed Mater Res 2004, 69A, 436-443.

(32) Zhu, Y. B., Gao, C.Y., Liu, X.Y., and Shen, J.C. Biomacromolecules 2002, 3, 1312-1319.

(33) Chan, C.-M. Polymer surface modification and characterization; Hanser: Munich, 1993.

(34) Tseng, D. Y., and Edelman, E.R. J Biomed Mater Res 1998, 42, 188-198. (35) Oyane, A., Uchida, M., Choong, C., Triffitt, J., Jones, J. and Ito, A.

Biomaterials 2005, 26, 2407–2413. (36) Claase, M. B., Riekerink, M.B.O., de Bruijn, J.D., Grijpma, D.W., Engbers,

G.H.M., and Feijen, J. Biomacromolecules 2003, 4, 57-63. (37) Kato, K., Uchida, E., Kang, E.T., Uyama, Y., and Ikada, Y. Prog. Polym. Sic

2003, 28, 209-259. (38) Grøndahl, L., Cardona, F., Chiem, K., Wentrup-Byrne, E., and Bostrom, T.

J.Mater Sci Mater Med 2003, 14, 503-510. (39) Suzuki, S., Grøndahl, L., Leavesley, D., and Wentrup-Byrne, E. Biomaterials

2005, 26, 5303–5312. (40) Tretinnikov, O. N., Kato, K., and Ikada, Y. J Biomed Mater Res 1994, 28,

1365-1373. (41) Kamei, S., Tomita, N., Tamai, S., Kato, K., and Ikada. J Biomed Mater Res

1997, 37, 384-393. (42) Curtis, A., and Wilkinson, C. Biomaterials 1997, 18, 1573-1583. (43) Flemming, R. G., Murphy, C.J., Abrams, G.A., Goodman, S.L., and Nealey,

P.F. Biomaterials 1999, 20, 573-588. (44) Boyan, B. D., Hummert, T.W., Dean, D.D., and Schwartz, Z. Biomaterials

1996, 17, 1068-1077. (45) Chesmel, K. D., Clark, C.C., Brighton, C.T., and Black, J. J Biomed Mater

Res 1995, 29, 1101-1110. (46) Matsuzaka, K., Walboomers, X.F., de Ruijter, J.E., and Jansen, J.A.

Biomaterials 1999, 20, 1293-1301. (47) Mata, A., Boehm, C., Fleischman, A.J., Muschler, J., and Roy, S. J Biomed

Mater Res 2002, 62, 499-506. (48) Cheung, H.-Y., Lau, K.-T., Lu, T.-P., and Hui, D. Composites:Part B 2007,

38, 291-300. (49) Wan, Y. Q., Wang, Y., Liu, Z.M., Qu, X., Han, B.X., Bei, J.Z., and Wang, S.

Biomaterials 2005, 26, 4453–4459. (50) Hoffman, A. S. In Biomaterials science : an introduction to materials in

medicine; Ratner, B.D., Hoffman, A.S., Schoen, F.S., and Lemons, J.E., Ed.; Academic Press: San Diego, 1996.

(51) Hersel, U., Dahmen, C., and Kessler, H. Biomaterials 2003, 24, 4385–4415. (52) Lieb, E., Hacker, M., Tessmar, J., Kunz-Schughart, L.A., Fiedler, J., Dahmen,

C., Hersel, U., Kessler, H., Schulz, M.B., Gopferich, A. Biomaterials 2005, 26, 2333-2341.

21


(53) Hu, Y., Winn, S.R., Krajbich, I., and Hollinger, J.O. J Biomed Mater Res 2003, 64A, 583-590.

(54) Saitoa, N., and Takaoka, K. Biomaterials 2003, 24, 2287-2293. (55) Seeherman, H., and Wozney, J.M. Cytokine Growth Ractor Reviews 2005, 16,

329-345. (56) Kato, M., Toyoda, H., Namikawa, T., Hoshino, M., Terai, H., Miyamoto, S.,

and Takaoka, K. Biomaterials 2006, 27, 2035-2041. (57) Lee, S.-H., and Shin, H. Adv Drug Delivery Rev 2007, 59, 339-359. (58) Puleo, D. A., Kissling, R.A., and Sheu, M.-S. Biomaterials 2002, 23, 2079-

2087. (59) Klee, D., Ademovic, A., Bosserhoff, A., Hoecker,; H., M., G., and Erli, H.-J.

Biomaterials 2003, 24, 3663-3670. (60) Kang, E. T., Tan, K.L., Kato, K., Uyama, U., and Ikada, Y. Macromolecules

1996, 29, 6872-6879. (61) http://www.devicelink.com/company98/w/w00057.html. 31/1/2003. (62) http://www.emedicine.com/ent/topic376.htm. 01/8/2007. (63) Pu, F. R., Williams, R.L., Markkula, T.K., and Hunt, J.A. Biomaterials 2002,

23, 2411-2428. (64) Jarrell, B. E., Williams, S.K., Rose, D., Garibaldi, D., Talbot, C., and Kapelan,

B. J Biomed Eng 1991, 113, 120-122. (65) Zhang, Q., Wang, C., Babukutty, Y., Ohyama, T., Kogoma, M., and Kodama,

M. J Biomed Mater Res 2002, 60, 502-509. (66) Kellomaki, M., Niiranen, H., Puumanen, K., Ashammakhi, N., Waris, T., and

Tormala, P. Biomaterials 2000, 21, 2495-2505. (67) Nyman, R., Magnusson, M., Sennerby, L., Nyman, S., and Lundgren, D. Acta

Orthop Scand 1995, 66, 169-173. (68) Salata, L. A., Hatton, P.V., Devlin, A.J., Craig, G.T., and Brook, I.M. Clin

Oral Impl 2001, 12, 62-68. (69) Kang, E. T., Zhang, Y. Adv Mater 2000, 12, 1481-1495. (70) Sacher, E. Prog Surf Sci 1994, 47, 273. (71) Graziani, E. I., McKeown, N.B., Kalman, P.G., and Thompson, M. Int Biodet

Biodeg 1992, 30, 217-231. (72) Choi, Y. J., Kim, M.S., and Noh, I. Surf Coating Tech 2007, 201, 5724-5728. (73) Colwell, J. M., Wentrup-Byrne, E., Bell, J.M., and Wielunski, L.S. Surf Coat

Technol 2003, 168, 216-222. (74) Chen, J.-R., and Wakida, T. J Appl Poly Sci 1997, 63, 1733-1739. (75) Ryan, M. E., and Badyal, J.P.S. Macromolecules 1995, 28, 1377-1382. (76) Wilson, D. J., Eccles, A.J., Steel, T.A., Williams, R.L., and Pond, R.C.

Surface and interface Analysis 2000, 30, 36-39. (77) Yamada, Y., Yamada, T., Tasaka, S., and Inagaki, N. Macromolecules 1996,

29, 4331-4339. (78) Biederman, H., Zeuner, M., Zalman, J., Bilkova, P., Slavinska, D., Stelmasuk,

V., and Boldyreva, A. Thin Solid Films 2001, 392, 208-213. (79) Chandy, T., Das, G.S., Wilson, R.F., Rao, G.H.R. Biomaterials 2000, 21,

699-712. (80) Xu, H., Hu, Z., Wu, S., and Chen, Y. Mater Chem Phys 2003, 80, 278–282. (81) Caro, J. C., Lappan, U., and Lunkwitz, K. Surf Coating Tech 1999, 116-119,

792–795. (82) Gauvreau, V., Chevallier, P., Vallie`res, P., Petitclerc, E., C.-Gaudreault, R.,

and Laroche, G. Bioconjugate Chem 2004, 15, 1146-1156.

22

http://www.devicelink.com/company98/w/w00057.html

http://www.emedicine.com/ent/topic376.htm


(83) Chevallier, P., Castonguay, M., Turgeon, S., Dubrulle, M., Mantovani, D., McBreen, P.H., Wittmann, J.-C., and Laroche, G. J Phys Chem B 2001, 105, 12490-12497.

(84) Pringle, S. D., Joss, V.S., and Jones, C. Surf Interf Anal 1996, 24, 821-829. (85) Chen, M., Zamora, P.O., Som, P., Pena, L.A., and Osaki, S. J Biomater Sci

Polym Edn 2003, 14, 917-935. (86) Wilson, D. J., Wiliams, R.L., and Pond, R.C. Surf Interf Anal 2001, 31, 385-

396. (87) Wilson, D. J., Wiliams, R.L., and Pond, R.C. Surf Interf Anal 2001, 31, 397-

408. (88) Lappan, U., Buchhammer, H.-M., and Lunkwitz, K. Polymer 1999, 40, 4087-

4091. (89) Dargaville, T. R., George, G.A., Hill, D.J.T., and Whittaker, A.K. Prog

Polym Sci 2003, 28, 1355-1376. (90) Charlesby, A. Atomic radiation and polymers; Pergamon Press: Great Britain,

1960. (91) Nasef, M. M., Saidi, H., Dessouki, A.M., and El-Nesr, E.M. Polym Int 2000,

49, 399-406. (92) Nasef, M. M. Polym Int 2001, 50, 338-346. (93) Dobo, J., Somogji, A., and Czvikovszky, J. J Polym Sci Part C 1964, 4,

1173-1193. (94) Takamatsu, T., Wada, J., Fukada, E., and Matsumoto, H. Kobunshi

Ronbunshu 1980, 37, 782-787. (95) Ali-Miraftab, H., Chapiro, A., and Le Doan, T. Eur Polym J 1978, 14, 639-

646. (96) Yamakawa, S. Macromolecules 1979, 12, 1222-1227. (97) Mohy, E. M. S., Portaccio, M., Diano, N., Rossi, S., Bencivenge, U., and

D'Uva, A. J Molecular Catalysis B: Enzymatic 1999, 7, 251-261. (98) Hegazy, E. A., Taher, N.H., Ebaid, A.R., Rabie, A-G., and Kamal, H. J Appl

Poly Sci 1990, 39, 1029-1043. (99) Ishigaki, I., Kamiya, N., Sugo, Y., Machi, S. Polym J 1978, 10, 513-519. (100) Chapiro, A. Radiat Phys Chem 1977, 9, 55-67. (101) Hegazy, E. A., Ishigaki, I., and Okamoto, J. J Appl Polym Sci 1981, 26, 3117-

3124. (102) Grøndahl, L., Cardona, F., Chiem, K., Wentrup-Byrne, E. J Appl Poly Sci

2002, 86, 2550-2556. (103) Hegazy, E. A., El-Assy, N.B., Taher, N.H., and Dessouki, A.M. Radiat Phys

Chem 1989, 33, 539-543. (104) Otsuhata, K., Razzak, M.T., Castanares, R.L., Tabata, Y., Ohashi, F., and

Takeuchi, A. Radiat Phys Chem 1985, 25, 537-548. (105) Razzak, M. T., and Otsuhata, K. and Otsuhata, K. J Appl Polym Sci 1987, 33,

2345-2351. (106) Razzak, M. T., Otsuhata, K., and Tabata, Y. J Appl Poly Sci 2003, 33, 2345-

2351. (107) Kidane, A., Lantz, G.C., Jo, S., and Park, K. J Biomater Sci Polymer Edn

1999, 10, 1089-1105. (108) Park, K., Shim, H.S., Dewanjee, M.K., and Eigler, N.L. J Biomater Sci

Polymer Edn 2000, 11, 1121-1134. (109) Wentrup-Byrne, E., Grøndahl, L., and Suzuki, S. Polym Int 2005, 54, 1581–

1588.

23


(110) Tan, K. L., Woon, L.L., Wong, H.K., Kang, E.T., and Neoh, K.G. Macromolecules 1993, 26, 2832-2836.

(111) Tu, C.-Y., Liu, Y.-L., Lee, K.-R., and Lai, J.-Y. J Membrane Sci 2006, 274, 47-55.

(112) Wang, C., and Chen, J.-R. Appl Surf Sci 2007, 253, 4599–4606. (113) Sun, H.-X., Zhang, L., Chai, H., and Chen, H.-L. Desalination 2006, 192,

271–279. (114) Liu, Y.-L., Luo, M.-T., and Lai, J.-Y. Macromol Rapid Commun 2007, 28,

329–333. (115) Yoshikawa, C., Goto, A., Tsujii, Y., Fukuda, T., Yamamoto, K. and Kishida,

A. Macromolecules 2005, 38, 4604-4610. (116) Yu, W. H., Kang, E.T., and Neoh, K.G. Langmuir 2005, 21, 450-456. (117) Combellas, C., Kanoufi, F., and Nunige, S. Chem Mater 2007, 19, 3830-3839. (118) Nishibe, T., O'Donnel, S., Pikoulis, E., and Rich, N. J Cardiovascular Surg

2001, 42, 667. (119) Nishibe, T., Okuda, Y., Kumada, T., Tanabe, T., and Yasuda, K. Surg Today

2000, 30, 426-431. (120) Sperling, C., Konig, U., Hermel, G., Werner, C., Muller, M., Simon, F.,

Grundke, K., and Jacobasch, J. J Mater Sci: Mater Med 1997, 8, 789-791. (121) Vasilets, V. N., Hermel, G., Konh, U., Werner, G., Muller, M., Simon, F.,

Gtundke, K., Ikada, Y., and Jacobasch, H.J. Biomaterials 1997, 18, 1139-1145.

(122) Wang, S., and Marchant, R.E. Macromolecules 2004, 37, 3353-3359. (123) Larsen, C. C., Kligman, F., Kottke-Marchan, K., and Marchant, R.E.

Biomaterials 2006, 27, 4846–4855. (124) Larsen, C. C., Kligman, F., Tanga, C., Kottke-Marchant, K., and Marchant,

R.E. Biomaterials 2007, 28, 3537–3548. (125) Zoungrana, T., Findenegg, G., and Norde, W. J Colloid and Interface Sci

1997, 190, 437. (126) Vermeer, A. W. P., Bremer, M.G.E.G., and Norde, W. Biocheim Biophy Acta

Gen Subjects 1998, 1425, 1. (127) Shoichet, M. S., and McCartyy, T.J. Macromolecules 1991, 24, 1441. (128) Coupe, B., Evangelista, M.E., Yeung, R.M., and Chen, W. Langmuir 2001,

17, 1956-1960. (129) Reghunadhan Nair, C. P., Clouet, G., and Brossas, J. J Polym Sci A: Polym

Chem 1988, 26, 1791-1807. (130) Pike, R. M., and Cohen, R.A. J Polym Sci 1960, 44, 531-538.

24



2.1 Introduction

The aim of this synthesis section of the work was to create well-defined polymers for

the purpose of surface modification of polymers to be used in biomedical

applications. Living radical polymerisation (LRP) was the technique of choice since

it provides good control over both polymer structure and composition, and due to the

fact that it requires moderately simple reaction conditions.

2.1.1 Living Radical Polymerisation (LRP)

LRP enables us to engineer well-defined polymers with tuneable structures,

compositions and properties.1 In an ideal LRP the molecular weight (Mn) increases

linearly with conversion and narrow molecular weight distributions can be obtained

(Mw/Mn<1.1). The three main techniques that have found widespread applications are:

nitroxide-mediated polymerisation (NMP), atom transfer radical polymerisation

(ATRP) and reversible addition fragmentation chain transfer (RAFT). Advantages of

these methods over anionic/cationic living polymerisation are that they can be carried

out under moderate reaction conditions and in the presence of small amounts of

impurities that are difficult to eliminate.

In NMP, stable nitroxide free radicals (e.g. 2,2,6,6-tetramethylpiperidinyloxy;

TEMPO) act as reversible terminating agents.2 NMP-mediated polymerisations is

carried out by two methods: one involves the thermal decomposition of an

alkoxyamine into a reactive radical and a stable radical, and the other uses a mixture

of a conventional radical initiator and the nitroxide radical. Scheme 2.1 represents

the mechanism of styrene polymerisation using TEMPO.

25


O

Pn + Pm Pn+mTermination

Pn + Pm= H

(Combination)

(Disproportionation)

Scheme 2.1: Schematic representation of nitroxide-mediated polymerisation of styrene with TEMPO, monomer (e.g. styrene) and initiator (e.g BPO = benzyl

peroxide). Pn• and Pm• are propagating polymer radicals.

NMP process involves an equilibrium between dormant species (i.e. reversibly

terminated with the stable free radical) and active chains (i.e. propagating polymer

radicals). Hence, the amount of free nitroxide in the polymerization solution is a key

because an excess of free nitroxide shifts the equilibrium to the dormant species

which results in very little monomer addition.3

NMP requires elevated temperatures (>120oC) and long reaction times (1-3 days) to

activate highly stable TEMPO-capped dormant chains. It has been successfully

applied to limited numbers of monomers such as acrylate and styrenic systems. It has

been traditionally difficult to obtain low polydispersity indexes (PDI’s) for other

monomers.4 However, Hawker and coworkers developed alkoxyamine-based

initiators which can be applied to a wide range of monomers resulting in good

molecular weight control and lower PDI’s (1.05-1.15).3,5

ATRP reactions involve an organic halide that undergoes a reversible redox process

catalyzed by a transition metal compound such as cuprous halide.6 The mechanism

by which ATRP proceeds is described below:

NNO

NO

NO

NO

Monomer

BPO

TEMPO

26


Pn – X + Cu(I)/L2 Pn• + Cu(II)X/L2

M

Pn+m / Pn= + Pm

HTermination:

Scheme 2.2: Schematic representation of the ATRP polymerisation process, Pn = polymer, X = halide, L = ligand, and M = monomer.

The copper (I) complex accepts a halogen atom (X; usually Cl, Br) and undergoes a

one-electron oxidation to form a Cu(II) complex. At the same time, an organic

radical is generated that either reacts with a monomer, or it can reabstract the halogen

atom from the copper (II) complex. This technique has been successfully applied to

obtain well-defined polymers with various structures (e.g. linear, block, star and

comb) from a variety of monomers.6,7 However, there are some disadvantages:8

• Specialised initiators are required, although they are currently commercially

available

• Stringent purification of the resulting polymers is required to remove the

metal complex (This is particularly important for biomedical applications

where there is concern over the release of even trace amounts of metals when

polymer materials or devices are implanted.)

• Reactions are sensitive to traces of oxygen since Cu(I) will oxidise to form

Cu(II) deactivator

• Generally acidic monomers need to be protected

In comparison to NMP and ATRP reactions which control chain growth by

reversible termination, the RAFT process involves reversible chain transfer reactions.

This is a more versatile technique since a much wider range of monomers can be

successfully polymerised, including many functional monomers. RAFT

27


polymerisations have been carried out in both aqueous and organic solvents and

generally use a lower polymerisation temperature than either ATRP or NMP

systems.9

2.1.2 Reversible Addition Fragmentation Chain Transfer (RAFT) Method

The RAFT process was invented by Rizzardo and coworkers at the Commonwealth

Scientific and Industrial Research Organization (CSIRO) (Australia) in 1998.9 It is a

radical polymerisation which uses a thiocarbonylthio compound added which acts as

a highly efficient RAFT agent (Scheme 2.3). This transfer of the S=C(Z)S- moiety

between the active and dormant chains maintains the living characteristics of the

polymerisation.

S S

z

R

Scheme 2.3: General structure of a RAFT agent, Z = activation group (e.g. aryl and alkyl) and R = good leaving group (e.g. cumyl and cyanoisopropyl).

There are four groups of RAFT agents categorised according to the nature of the Z

group: (1) dithioesters (Z = aryl or alkyl), (2) trithiocarbonates (Z = substituted

sulfur), (3) dithiocarbonates (xanthates: Z = substituted oxygen) and (4)

dithiocarbamates (Z = substituted nitrogen).10 A wide range of RAFT agents have

been synthesized and utilized for different types of monomers.11

The choice of RAFT agent is important since different monomers may require

different RAFT agents depending on the reactivity of the propagating polymer

radical.12 For example, methacrylate monomers form radicals that are very good

leaving groups. To obtain an ideal living polymerisation of this monomer, the RAFT

agent has to have an equally good or superior leaving group, such as a cumyl or

cyanoisopropyl group. Several reviews have summarized RAFT agents and the

polymerisations in which they have been applied.11,13,14

28


S

z

S R S

z

RS S

z

S

S

z

SS

z

S S

z

S

Pm

Initiator 2I

I Pm

+ Pm Pm + R

Monomer

R PnMonomer

Pn + Pm + Pm

Pm Pn Pn

Pn + Pm Pn+m

(I) Initiation

(II) Chain Transfer

(III) Reinitiation

(IV) Addition/Fragmentation

(V) Termination

Pn + Pm= H

(Combination)

(Disproportionation)

kad, 1

kad, 2

kad

kad

kβ, 1

kβ, 2

kβ

kβ

Pre-equilibrium

Main equilibrium

Scheme 2.4: Schematic representation of proposed RAFT polymerisation process.

The mechanism of the RAFT process is illustrated in Scheme 2.4. In the early stage

of the polymerisation, the propagating polymeric radical (Pm•, mainly oligomeric

nature) reacts with the initial RAFT agent, forming the intermediate radical, which

then fragments into a polymeric thiocarbonylthio compound and a new radical (R•)

(pre-equilibrium, II). The reaction of the radical with a monomer forms a new

propagating radical Pn• (reinitiation, III). Subsequent addition fragmentation steps

form an equilibrium between the propagation polymeric radicals (Pn• and Pm•) and

the dormant polymeric RAFT agents (main equilibrium, IV). This step results in an

equal probability for all chains to grow, which results in the formation of polymers

with low PDI’s.11 Polymerisation is unavoidably terminated by either combination or

disproportionation (V), which is dictated by the amount of initiator that is

decomposed. The most of the polymer chains contain thiocarbonylthio groups at one

end and re-initiating groups (R) at the other.

Kinetically, the individual reactions of pre-equilibrium (II) and main equilibrium (II)

are called “addition rate coefficients” (kad) and “fragmentation rate coefficients”

(kβ).15 It is important to note that, in pre-equilibrium, the reactions are asymmetrical

since different radical species are attacking and leaving. Hence as mentioned before,

29


the choice of R group is crucial. Contrarily, these reactions are symmetrical in the

main equilibrium.

The rate of polymerisation (Rp) in RAFT polymerisations using dithiobenzoates has

been shown to be retarded, and decreases with increasing the initial RAFT agent

concentrations. There are two rate retardation effects: “induction period” and “rate

retardation”.15 An induction period is at the beginning of the polymerisation where

there is no polymerisation occurring. Rate retardation is when the polymerisation rate

is slower than that of conventional polymerisation without a RAFT agent. There are

two models which have been proposed to describe this behaviour: intermediate

radical termination or slow fragmentation.15-18 The former model explains the severe

retardation through the termination of the intermediate radicals with other radicals in

the system that results in the formation of three- and four-arm stars.17,18 The latter

model involves the slow fragmentation of the stable intermediate radicals.16 These

models are currently being widely debated in the literature. It has also been found

that impurities such as dithiobenzoic acid, which is the starting reagent for CDB,

induces inhibition in the polymerisations of 2-hydroxyethyl methacrylate (HEMA),

styrene and methyl acrylate.19 Despite this, the polymerisation can proceed in a

controlled manner.

The molecular weight increases linearly with conversion in the RAFT process, and

the equation (1) has been traditionally used to predict the molecular weight:

Mn = [M]0 x

[RAFT]0 Mw(M) + Mw(RAFT)×

where [M]0 is the initial monomer concentration, x is the fractional monomer

conversion to the polymer, [RAFT]0 is the initial RAFT agent concentration, and

Mw(M) and Mw(RAFT) are the molar masses of the monomer unit and RAFT agent,

respectively. More precisely, taking into account the initiator-derived chains, as well

as the chain transfer constant (Ctr) of RAFT agent, the equation becomes as

follows:18

(1)

Mn = [M]0 x

([RAFT]0 – [RAFT]x) + af([I]0 – [I]x) Mw(M) + Mw(RAFT) (2) ×

30


where [RAFT]x is the RAFT agent concentration at x, a is the mode of termination (a

equals 1 for termination by combination and 2 for disproportionation), f is the

initiator efficiency, [I]0 is the initial initiator concentration and [I]x is the initiator

concentration at x. The concentration of the RAFT agent at x can be calculated from

the equation (3).

Many types of monomers have been successfully polymerized by the RAFT process.

These include functional (meth)acrylates, (meth)acrylamide, styrenic, and vinyl

monomers. However, there are certain types of monomers where RAFT-mediated

polymerisation can not be used. For example, olefins such as ethylene and propylene

are difficult to homopolymerise by the RAFT process.11 Other examples are

monomers containing primary or secondary amines that are known to undergo facile

reaction with the thiocarbonylthio compounds.

Depending on the intended application such as in a medical use, the residual RAFT

end-group can sometimes be problematic. Stenzel et al.20 investigated the

cytotoxicity of two RAFT agents: benzyl dithiobenzoate and 3-

benzylsulfanylthiocarbonylsulfanyl propanoic acid. The assay they used was based

on the Australian Standards Protocol for Biological Evaluation of Medical Devices:

test for in vitro cytoxicity AS/ISO10993.5-2002. The leached liquid from the RAFT

agent (supernatant of 0.125g of RAFT agent in 2.5mL cell culture medium) was

diluted 1:6 in the medium. The 0.8mL of this solution was placed onto the cultured

mouse fibroblast L929 cell line, and after 48 hours cell numbers were counted and

compared to a control culture without RAFT agent. Benzyl dithiobenzoate, a

commonly used RAFT agent inhibited cell growth by 72% when compared to the

controls, whereas only 9.7% inhibition for 3-benzylsulfanylthiocarbonylsulfanyl

propinoic acid was observed.20 Hence less toxic RAFT agents should be

preferentially used for polymers used in medical applications, or alternatively RAFT

end-group removal is required.

The thiocarbonylthio groups can be cleaved by several methods which includes

radical induced reduction (to provide a hydrocarbon end-group), thermal elimination

(to provide an unsaturated end-group) and reaction with a nucleophile e.g. amine,

[RAFT]x = [RAFT]0(1-x)Ctr (3)

31


hydroxide, borohydride (to provide a thiol end-group).21 When forming thiol groups,

it is important to eliminate air completely from the reaction, since thiols undergo

oxidative coupling to give the disulfide and lead to a doubling of the molecular

weight of the polymers.22

In this study, RAFT-mediated polymerisation was chosen since acidic monomers

were being used. Both ATRP and NMP are not suitable for acidic monomers.

Initially the intention was to synthesise phosphate- and fluorine-containing block

copolymers using RAFT but this approach proved unfeasible because of the fact that

the solvents for these monomer-polymer combinations proved to be non-common.

Two RAFT agents, 1-phenylethyl phenyldithioacetate (PEPDTA) and cumyl

dithiobenzoate (CDB) were synthesised and used in this research (Scheme 2.5).

SS

SS

1-Phenylethyl phenyldithioacetate (PEPDTA) Cumyl dithiobenzoate (CDB)

Scheme 2.5: Structures of RAFT agents.

PEPDTA has shown good control of acrylate and styrene polymerisations, as well as

faster rates of polymerisation compared to that of CDB.23-25 CDB is one of the most

versatile RAFT agent, and used for methacrylate monomers. In the literature, 2-

cyanoprop-2-yl dithiomethacrylate (CPDB) has been shown to induce less rate

retardation of methyl methacrylate (MMA) polymerisation than CDB while still

providing good control.26-28 However, the synthesis of this RAFT agent requires

large amounts of AIBN currently not commercially available in Australia. Due to

the limited supply of AIBN and the necessity of avoiding the costly and time-

consuming AIBN synthesis experiments, this RAFT agent was not used.

32


2.1.3 Living Radical Polymerisation of Phosphorous-Containing Monomers

Phosphorous-containing polymers have attracted significant attention over several

decades due to their wide range of applications including biomedical use.

Incorporation of these groups in polymers provides properties such as enhanced

mineralisation, imprinting of biological molecules and improved blood and other

biocompatibilities.29-34

Several studies have reported the successful LRP of phosphorous-containing

monomers. ATRP of phosphate-containing monomers in their non-acidic forms such

as dimethyl(1-ethoxycarbonyl)vinyl phosphate and deprotonated 2-

methacryloyloxyethyl phosphate (MOEP) have been successfully carried out.35,36

However, limited monomer conversion was observed suggesting that complexation

between the copper ions and the phosphoryl oxygen of the phosphonate groups was

occurring.35,36

2-Methacryloyloxyethyl phosphorylcholine was polymerised using ATRP in both

aqueous or alcoholic solutions37 as well as RAFT20,38 processes. Incorporation of

phosphorylcholine groups onto biomaterial surfaces has shown extremely high

biocompatibility and antithrombogenicity due to the reduction of protein

adsorption.33,34

Phosphonated methacrylates have been successfully copolymerised with vinyldene

chloride and methyl acrylate using the RAFT technique to obtain statistical, gradient

and diblock terpolymers.39

Two commercially available phosphate-containing monomers monoacryloxyethyl

phosphate (MAEP) and MOEP were polymerised using the RAFT technique in this

study (Scheme 2.6). These monomers were chosen since previous studies have

shown that their graft copolymers are capable of inducing calcium phosphate

nucleation.29,31,40

OO

PO

OHOH

O

OO

PO

OHOH

O Monoacryloxyethyl phosphate (MAEP) Methacryloyloxyethyl phosphate (MOEP)

Scheme 2.6: Structures of phosphate-containing monomers.

33


2.1.4 Living Radical Polymerisation of Fluorinated Monomers

Fluorine-containing polymers are of industrial interest due to their excellent physical

and chemical properties.41,42 However, there are only a few fluoropolymers that

possess functional groups. Since LRP offers the possibility of introducing

functionalisation and unique architectures in polymers, there have been several

studies which have investigated its potential in the synthesis of fluoropolymers.43

2,3,4,5,6-Pentafluorostyrene (FS) has been polymerised mainly by the ATRP process,

possibly due to the fast polymerisation rate (e.g. 90% conversion in 90 min). PFS and

its block copolymers with styrene have been prepared by ATRP and extensively

characterized.44 Borkar et al.45 investigated the synthesis of highly fluorinated

styrene monomers by nucleophilic substitution of FS with 1H,1H-

pentafluoropropane-1-ol and 1H,1H-pentadecafluorooctane-1-ol. These monomers

were subsequently homo and copolymerised with fluorinated styrene and styrene by

copper-mediated ATRP.45 Triblock copolymers based on central poly(ethylene

glycol) (PEG) or poly(ethylene glycol-co-propylene glycol) (PEGPG) blocks with FS

outer blocks have also been prepared by ATRP and showed low PDI’s (1.2-1.3).46

These triblock copolymers were further functionalized by complexing lithium

bis(trifluoromethylsulfonyl)imide salt and liquid PEGPG precursor to form ion

conducting polymer electrolytes.

Fu et al.47 synthesised well-defined block copolymers of PFS and poly(tert-butyl

acrylate) (PtBA) via ATRP. Amphiphilic block copolymers of PFS and poly(acrylic

acid) (PAA) were prepared by the hydrolysis of the tBA segments of the

corresponding P(FS-b-tBA) copolymers. These block copolymers formed

membranes with well-defined pores in sizes in the micrometer range due to inverse

micelle formation in aqueous media.

Fluorinated methacrylate, 2-[(perfluorononenyl)oxy]ethyl methacrylate, has been

used to prepare a series of di- and triblock copolymers by ATRP.48,49 Perrier et al.

have successfully synthesised poly(methacrylate)s and polystyrenes incorporating

fluorinated moieties as either the initiator or monomer by copper-mediated ATRP

with pyridine imine ligands.50,51 Shemper and Mathias52 reported the synthesis of a

fluorinated macroinitiator from a fluorinated surfactant for ATRP polymerisations.

34


Statistical and block copolymers with linear and star-like architectures were prepared

using this initiator.

Only a few studies of fluorinated polymer syntheses have been carried out using the

RAFT technique. Pai et al.53 modified poly(dimethyl siloxane) (PDMS) to form a

di(trithiocarbonate) functional molecule to use in the RAFT process. A-B-A triblock

copolymers were synthesized from this macro-RAFT and polymerized with N,N-

dimethyl acrylamide (DMA) and 2-(N-butyl perfluorooctanefluoro-sulfonamido)

ethyl acrylate (BFA). PDI’s were under 1.25. An A-B block copolymer of

poly(ethylene oxide)-b-poly(1,1,2,2-tetrahydroperfluorodecyl acrylate) was

synthesized via RAFT polymerisation, iodine transfer polymerisation (ITP), and

ATRP, in the presence of either degenerative transfer agents or a macroinitiator

based on poly(ethylene oxide).54 Both RAFT and ATRP gave better control of the

expected copolymer preparation than ITP.

Yu et al.55 have investigated the surface-initiated RAFT graft polymerisation of 4-

vinylbenzyl chloride (VBC) hydrogen-terminated silicon surfaces using cumyl

dithobenzoate (CDB) or cumyl phenyldithioacetate (CPDA) as the RAFT agent.

Tethered poly(4-vinylbenzyl chloride) (PVBC) brushes were further copolymerized

with FS to prove the livingness of the chain ends. The formation of block copolymer

brushes was confirmed by ellipsometry and XPS. Surface-grafted styrene brush-

based homopolymers and diblock copolymers bearing semifluorinated alkyl side

groups of different length, i.e., -(CH2)2(-CF2)n (n=6 or 8), have been synthesized by

NMP on silicon oxide surfaces.56

Three types of commercially available fluorine-containing monomers have been bulk

polymerised by the RAFT technique in this study (Scheme 2.7). In the literature,

only one study showing the chain extension of surface-initiated polymers with RAFT

end-groups using FS.55 As far as I can assertain, there are no reports on RAFT

polymerisation of 2,2,3,3-tetrafluoropropyl acrylate (TFPA) and 2,2,3,3-

tetrafluoropropyl methacrylate (TFPMA).

35


O

O

HF F

F F

2,2,3,3-tetrafluoropropyl

acrylate (TFPA)

O

O

HF F

F F

2,2,3,3-tetrafluoropropyl methacrylate (TFPMA)

F

FF

F

F

2,3,4,5,6-pentafluorostyrene

(FS)

Scheme 2.7: Structures of fluorinated monomers.

36


2.2 Experimental

2.2.1 Materials

MOEP was purchased from Aldrich (USA, no stated-purity) and the inhibitor

removed by toluene extraction.57 Inhibitor-free MAEP (Polysciences, USA, stated

purity of 97%) was used as received. 2-(methacryloyloxy)ethyl acetoacetate

(AAEMA, Aldrich, 95%) was distilled under vacuum to remove the inhibitor and

polymer. 2,3,4,5,6-pentafluorostyrene (PFS, Aldrich, USA, 99%), 2,2,3,3-

tetrafluoropropyl methacrylate (TFPMA, Aldrich, 99%), 2,2,3,3-tetrafluoropropyl

acrylate (TFPA, Matrix Scientific, USA, 97%), 2-(acryloyloxy)ethyl acetoacetate

(AAEA, Aldrich, 95%), tert-butyl acrylate (tBA, Aldrich, 98%) were all passed

through a column of basic alumina to remove the inhibitor. 2,2-azobis

(isobutyronitrile) (AIBN, Fluka, 98%) and 1,1’-Azobis(cyclohexanecarbonitrile)

(Vazo 88, DuPont, 98%) were recrystallised twice from methanol prior to use. AR

grade reagents ethyl acetate (99.5%), methanol (99.8%), n-hexane (95%), DMSO

(99.9%), and the ACS reagent acetone (99.5%) were used without further

purification. HPLC grade tetrahydrofuran (THF, 99.8%) was used in most cases,

except for the polymerisation when inhibitor-free anhydrous THF (Aldrich, 99.9%)

was used. Trifluoroacetic acid (TFA, 98%, Aldrich) and sodium cyanoborohydride

(95%, Aldrich), glycine (Ajax Finechem, AU) and L-phenylalanyl glycine (Aldrich)

were used as received. Benzoylated dialysis tubing (Mw cut off of 1200, Aldrich)

was used for the purification of PMOEP and PMAEP. Snake skin dialysis tubing

(Mw cut off of 3000, Pierce) was used for other polymers.

The RAFT agents, 1-phenylethyl phenyl dithioacetate (PEPDTA) and cumyl

dithiobenzoate (CDB) were synthesized according to the literature.58,59 PEPDTA was

purified by passing through a neutral aluminium oxide column using petroleum spirit

(40-60 ºC) as an eluent. The solid product was further purified by recrystallisation

from methanol. CDB was purified by passing through silica gel and neutral

aluminium oxide columns using n-hexane as an eluent.

37


2.2.2 Methods

2.2.2.1 MOEP/MAEP Homopolymer Synthesis

Solutions containing monomer, RAFT agent and AIBN in methanol were prepared in

the concentrations given in Table 2.1. A typical homopolymerisation included the

following; 2 g of MAEP (1.01 M), 27.2 mg of PEPDTA (1 × 10-2 M), and 3.3 mg of

AIBN (2 × 10-3 M) were dissolved in 6.7 g of methanol. Aliquots of 1.5 mL were

transferred to five individual ampoules which were degassed by four freeze-

evacuate-thaw cycles and sealed. These samples were placed in an oil bath at 60 ºC

and removed after the required time such that five different time points were reached

for each experiment. Conversion was measured by Raman spectroscopy. The

reaction was stopped by cooling the solution in liquid N2. Soluble polymer samples

were purified by dialysing against methanol for 3 days. Gels were extensively

washed with methanol followed by acetone. The samples were dried in a vacuum

oven at 40 ºC for 3 days.

PAAEMA homopolymer was prepared with CDB as the RAFT agent (EXPT 10 in

Table 2.2, p.63). Solutions containing monomer (2.92 M), CDB (0.125 M), and

AIBN (0.017 M) in ethyl acetate were prepared. Polymerisation was carried out as

described previously. Polymerisation was stopped after 16.7 h. The polymer was

precipitated twice in methanol and dried under vacuum at 25 ºC for 1 day.

2.2.2.2 MOEP/MAEP Block Copolymer Synthesis

The conditions of block copolymer synthesis (macro-RAFT agent, monomer, and

concentrations) are given in Table 2.2 (in p.63). The same procedure as that in the

polymerisation of the homopolymer was used except ethyl acetate was used as the

solvent. The polymer samples were purified by dialysing against methanol for 3

days.

2.2.2.3 Hydrolysis of PMOEP and PMAEP Polymers

The soluble PMOEP and PMAEP polymers were hydrolysed to poly(acrylic acid)

(PAA) and poly(methacrylic acid) (PMA) respectively, by stirring in excess 5M

38


NaOH at 80 ºC for 24 hours (see Scheme 2.6). The resulting polymers were purified

by dialysing against water in benzoylated dialysis tubing. Gel samples with high

conversions were hydrolysed with 5 or 10M NaOH at 80 ºC for 7 days. Full

conversion to the acid analogues was confirmed by 1H NMR.

2.2.2.4 Fluorinated Homopolymer Synthesis

The concentrations of monomers, RAFT agents and AIBN are given in Table 2.3 (p.

64). A typical polymerisation procedure is given below.

Typical RAFT Polymerisation of FS

21.8 mg of PEPDTA (8.00×10-5 mol, 2.81×10-2 M), and 1.95 mg of Vazo 88

(8.00×10-6 mol, 2.81×10-3 M) were dissolved in 4.00 g of FS (2.06×10-2 mol, 7.25

M). Aliquots of 0.5 mL were transferred to six individual ampoules which were

degassed by four freeze-evacuate-thaw cycles and sealed. These samples were placed

in an oil bath at 80 ºC and removed after the required time such that six different

time points (i.e. conversions) were obtained for each experiment. Conversion was

measured by Raman spectroscopy. The polymerisation was stopped by quenching

with liquid nitrogen, exposure to air and dilution with ethyl acetate. The polymer

(PFS) was purified by precipitating into methanol, vacuum filtered and dried

exhaustively under vacuum at room temperature.

Typical RAFT Polymerisation of TFPMA

19.6 mg of CDB (7.20×10-5 mol, 2.50×10-2 M), and 1.18 mg of AIBN (7.20×10-6 mol,

2.50×10-3 M) were dissolved in 3.60 g of TFPMA (1.80×10-2 mol, 6.25 M). Aliquots

of 0.5 mL were transferred to six individual ampoules which were degassed by four

freeze-evacuate-thaw cycles and sealed. These samples were placed in an oil bath at

60 ºC and removed after the required time such that six different time points were

reached for each experiment. Conversion was measured by Raman spectroscopy. The

polymerisation was stopped by quenching with liquid nitrogen, exposure to air and

dilution with ethyl acetate. The polymer (PTFPMA) was purified by precipitating

into n-hexane, vacuum filtered and dried exhaustively under vacuum at room

temperature.

39


Typical RAFT Polymerisation of TFPA

21.8 mg of PEPDTA (8.00×10-5 mol, 2.63×10-2 M), and 1.31 mg of AIBN (8.00×10-6

mol, 2.63×10-3 M) were dissolved in 4.00 g of TFPA (2.15×10-2 mol, 7.08 M).

Aliquots of 0.5 mL were transferred to six individual ampoules that were degassed

by four freeze-evacuate-thaw cycles and then sealed. These samples were placed in

an oil bath at 60 ºC and removed after the required time such that six different

conversions were reached for each experiment. Conversion was measured by Raman

spectroscopy. The polymerisation was stopped by quenching with liquid nitrogen,

exposure to air and dilution with ethyl acetate. The polymer (PTFPA) was purified

by precipitating into n-hexane, and dried exhaustively under vacuum at room

temperature.

2.2.2.5 Fluorinated Block Copolymer Synthesis

The conditions of block copolymer synthesis (macro-RAFT agent, monomer, solvent

and concentrations) are summarised in Table 2.4. Examples of each macro-RAFT

agent are given below:

Typical Chain Extension of PFS: Synthesis of Poly(FS-b-tBA),

expts.1 and 2, Table 2.4:

0.18 g of PFS (Exp. 7, Table 2.3:, Mn = 29209, PDI = 1.06, 6.03×10-6 mol, 6.86×10-3

M) was dissolved into a solution of tBA (0.406g, 3.17×10-3 mol, 3.45M), AIBN (0.29

mg, 1.77 ×10-6 mol, 1.92×10-3 M) and THF (0.403 g, 5.59×10-3 mol, 6.09M). The

mixture was placed in a glass ampoule, deoxygenated by five freeze-thaw-pump

cycles and sealed. The sample was polymerized at 60 ºC for 1.7 hr (expt.1, Table 2.4)

and 2.8 hr (expt.2, Table 2.4) and conversion was monitored by Raman spectroscopy.

The polymerisation was stopped by quenching with liquid nitrogen, exposure to air

and dilution with THF. The polymer was purified by precipitating into n-hexane,

vacuum filtered and dried exhaustively under vacuum at room temperature. In the

case of P(FS-b-AAEA), polymer was purified by precipitating into methanol.

40


Typical Chain Extension of PTFPA: Synthesis of Poly(TFPA-b-AAEA),

expt.5, Table 2.4:

0.16 g of PTFPA (expt.11, Table 2.3: Mn = 39959, PDI = 1.06, 3.91×10-6 mol,

9.57×10-3 M) was dissolved into a solution of AAEA (0.20 g, 9.95×10-4 mol, 2.44 M),

AIBN (0.07 mg, 4.26×10-7 mol, 1.04×10-3 M) and ethyl acetate (0.20 g, 2.27×10-3

mol, 5.56 M). The mixture was placed in a glass ampoule, deoxygenated by five

freeze-thaw-pump cycles and sealed. The sample was polymerized at 60 ºC and

conversion was monitored by Raman spectroscopy. The polymerisation was stopped

by quenching with liquid nitrogen, exposure to air and dilution with ethyl acetate.

The gel did not dissolve in either ethyl acetate or THF. The extract obtained from the

gel by ethyl acetate was precipitated in n-hexane, centrifuged and dried exhaustively

under vacuum at room temperature for GPC. In the case of P(TFPA-b-tBA), the

polymer was purified by precipitating into methanol/water (1:1), and dried

exhaustively under vacuum at room temperature.

Typical Chain Extension of PTFPMA: Synthesis of Poly(TFPMA-b-tBA),

expt.8, Table 2.4:

0.19 g of PTFPMA (expt.15, Table 2.3: Mn = 38025, PDI = 1.11, 5.01×10-6 mol,

8.46×10-3 M) was dissolved into a solution of tBA (0.26 g, 2.03×10-3 mol, 3.43 M),

AIBN (0.16 mg, 9.74×10-7 mol, 1.65×10-3 M) and THF (0.26 g, 3.63×10-3 mol, 6.13

M). The mixture was placed in a glass ampoule, deoxygenated by five freeze-thaw-

pump cycles and sealed. The sample was polymerized at 60 ºC and conversion was

monitored by Raman spectroscopy. The polymerisation was stopped by quenching

with liquid nitrogen, exposure to air and dilution with THF. The polymer was

purified by precipitating into n-hexane, vacuum filtered and dried exhaustively under

vacuum at room temperature.

2.2.2.6 Hydrolysis of tBA Segments

The hydrolysis of tBA side groups on the block copolymers to acrylic acid was

carried out according to the literature procedure.60

41


Synthesis of Poly(TFPMA-b-AA):

A 5-fold molar excess of TFA (18.1 mg, 1.58×10-4 mol, 3.00×10-1 M) with respect to

the tBA groups of the block copolymer was added dropwise to a solution of 20 mg

P(TFPMA-b-tBA) (expt.8, Table 2.4) dissolved in 0.5 mL of DCM. The reaction was

carried out at room temperature with stirring for 16 hours. DCM was evaporated by

blowing N2 on the surface of the solution, and the polymer was dried exhaustively at

room temperature under high vacuum overnight.

2.2.2.7 Model Biomolecule Modification of Functional Fluorinated Block

Polymers

Glycine attachment

A 2.5-fold molar excess of glycine (14.0 mg, 1.86×10-4 mol) with respect to the

AAEMA groups was added to the solution of P(TFPMA-b-AAEMA) (expt.9, Table

2.4, 50mg, 7.3×10-5 mol AAEMA units) in acetone (5 mL). The mixture was stirred

at room temperature for 7 days in order to react the amines with the ketones to form

an imine which was then stabilized by mild reduction with addition of NaBH3CN (10

mg, 1.59×10-4 mol, 3.18×10-8 M) to form a stable secondary amine bond by stirring

for a further 2 days. After evaporating the acetone to half the volume, the polymer

was precipitated into n-hexane, vacuum filtered and washed extensively with MilliQ

water. It was dried exhaustively under vacuum at 40 ºC overnight. The modified

polymer was characterized by 1H NMR.

L-Phenylalanyl glycine attachment

A 2-fold molar excess of L-phenylalanyl glycine (19.5 mg, 8.76×10-5 mol) with

respect to the AAEMA groups and equal mole of triethylamine (TEA, 8.9 mg,

8.76×10-5 mol, 2.50×10-8 M) were added to the solution of P(TFPMA-b-AAEMA),

(expt.9, Table 2.4, 30.0 mg, 4.38×10-5 mol AAEMA units) in anhydrous

dimethylformamide (DMF) (3.5 mL). The mixture was stirred at 60 ºC for 1 day to

react the amines with the ketones to form an imine which was then stabilized by mild

reduction with addition of NaBH3CN (29.0 mg, 4.16×10-4 mol, 1.32×10-7 M) to form

a stable secondary amine bond by stirring further 2 days at 60 ºC. The DMF was

evaporated to half the volume and the polymer was precipitated into MilliQ water,

42


and washed extensively with water. It was dried exhaustively under vacuum for 2

days at 40 ºC. The modified polymer was characterized by 1H NMR.

2.2.2.8 Stability of the PFS RAFT end-groups

A 1 mg/mL solution of fluorinated polystyrene prepared from the RAFT

polymerisation using PEPDTA (expt.7, Table 2.3) in THF or ethyl acetate was left at

room temperature without stirring. Changes in the UV-vis absorption at 310nm,

indicating hydrolysis of the dithioester RAFT end-groups were monitored over an 18

day period.

2.2.3 Analytical Techniques

2.2.3.1 FT-Raman spectroscopy

To obtain the degree of conversion, samples were prepared in ampoules and FT-

Raman spectra (PE Spectrum 2000 NIR FTIR, 64 scans, 8 cm-1 resolution, wave

number range 4000 – 360 cm-1) were recorded at various time points. Spectral

information was extracted by means of spectral analysis software (GRAMS/32,

Galactic Inductries Corp., Salem, NH). The area under the C=C double bond

stretching band at 1640 cm-1 (1620 cm-1 for PFS) was normalised to the non-

changing signal (for methanol at 1000 cm-1) and used for the conversion calculation.

Although Raman intensity is linear to the concentration of the species, normalisation

using an internal signal was performed to compensate any changes due to the

instrumentation (e.g. laser power and instrument arraignments). In addition, a control

reaction involving one sample for each reaction condition was polymerised in situ in

the FT-Raman spectrometer (8 cm-1 resolution, wavenumber range 4000 – 360 cm-1)

at 60 ºC to obtain a conversion/time curve. Figure 2.1 shows the set up of heating

block inside the FT-Raman spectrometer.

43


Heating Block

Figure 2.1: Heating block setup for in situ Raman polymerisation.

Spectra were collected as follows: every 3 minutes, 16 scans for MOEP/MAEP

polymerisations, every 30 minutes, 64 scans for FS polymerisations, and every 10

minutes, 32 scans for TFPMA and TFPA polymerisations.

2.2.3.2 Gel permeation chromatography (GPC)

Average molar mass and molar mass distributions of the hydrolysed MOEP and

MAEP homo and block copolymers were measured by GPC using a Waters system

(Alliance GPCV 2000) equipped with three Ultrahydrogel columns (7.8 × 300 mm),

a UV detector, and a refractive index (RI) detector. The sample was prepared by

diluting the polymer solution with 5 mM NaOH to obtain a 10 mg/mL solution. The

analyses were carried out in a 5 mM NaOH solution at 60 ºC, with a flow rate of 0.5

mL/min. Calibration was relative to 10 PAA standards (Mn range 830 – 888,900 )

(Polymer Standards Service, Mainz, Germany).

GPC measurements for fluorinated polymers were performed using a Waters

Alliance 2690 Separations Module equipped with three 7.8 × 300mm Waters

Styragel GPC columns (2 linear Ultrastyragel and one Styragel HR3 columns), an

autosampler, column heater, differential refractive index detector and a Photo Diode

Array (PDA) connected in series. HPLC grade tetrahydrofuran was used as eluent at

a flow rate of 1 mL min-1. Polystyrene standards Mn ranging from 517 – 2000000

44


were used for calibration. Molecular weights of all polymers are reported relative to

polystyrene standards.

2.2.3.3 Elemental Analyses

Elemental carbon, hydrogen and sulphur were determined using a Carlo Erba

Elemental analyzer model 1106. Elemental phosphorous was determined using an

ICPAES Spectro spectroflame P instrument using a forward power of 1200W, a flow

rate of 1.0 mL/min and a Meinhard concentric nebuliser. Soluble polymers,

monomers, and standards were prepared in methanol. The insoluble gels were acid

digested prior to characterization.

2.2.3.4 Nuclear Magnetic Resonance (NMR)

1H-NMR spectra were recorded using either a 300, 400, or 500 MHz spectrometer

(Avance Bruker). Software used was TOPSPIN 1.3. Chemical shifts are given in

ppm relative to the residual solvent peak.

31P-NMR spectra were recorded using a 400 MHz spectrometer (Avance Bruker).

Orthophosphoric acid, H3PO4, in D2O was used as an external reference. The

polymers were analysed in methanol-d4.

2.2.3.5 UV/VIS Spectroscopy

UV spectra were recorded on a Hitachi U-3000 UV-VIS spectrometer in the

wavelength range from 190 to 700 nm.

2.2.3.6 Fourier Transform Near Infrared (FT-NIR)

The in situ FT-NIR measurements of FS polymerisation were performed using a

Nicolet 5700 FTIR spectrometer with a heating block set to 80 °C. Each spectrum in

the NIR region ranging from 7000-5000 cm-1 was recorded every 30 min, 32 scans,

with a resolution of 8 cm-1. The decrease in area of the vinyl C-H stretching overtone

band of the monomer was calculated to give percent conversion.

45

Inte

nsity

(a.u

.)


46

(b)

2.3 Results

2.3.1 RAFT-Mediated Polymerisation of Phosphate-Containing Monomers

Two molecular weights of 10 and 20k were the targeted for both MAEP and MOEP

polymerisations involving the changing RAFT concentrations only. Therefore the

concentrations of initiator, monomer and solvent were kept constant (e.g. for 10k,

[RAFT]:[Initiator] = 10:1, and for 20k, [RAFT]:[Initiator] = 5:1) (See Table 2.1 for

polymerisation conditions). All polymerisation reactions were monitored by Raman

spectroscopy.

2.3.1.1 Monoacryloxyethyl phosphate (MAEP) polymerisation

The homopolymerisation of MAEP was carried out in the presence of either

PEPDTA or CDB with AIBN as initiator in methanol at 60ºC. The experimental

conditions and characteristics of the polymer structure (i.e. either cross-linked or

soluble in methanol) are given in Table 2.1. Figure 2.2 shows the Raman spectra of a

typical MAEP polymerisation mixture both before and after polymerisation.

Conversion was calculated from the C=C band at 1638 cm-1 which was normalized

against the methanol solvent band at 1032 cm-1. The local linear baselines used for

these peaks were 1653–1607 cm-1 and 1079–961 cm-1, respectively.

A B

1638

1032 (a) (a)

Figure 2.2: Raman spectra of MAEP polymerisation solution ([PEPDTA] = 1 × 10-2 M (expt.2)) (a) initial and (b) after 7.25 h in the region of (A) 3800-360 cm-1 and (B) 1800-900 cm-1.

Raman Shift (cm-1)

(b)

Raman Shift (cm-1)


47

Table 2.1: Experimental conditions of MAEP 1 and MOEP 2 polymerisation reactions 3 and characteristics of the polymers obtained.

1: [MAEP] = 1.01 mol/L

Monomer Experimental 5 Elemental Analyses ICP Expt.

RAFT [RAFT] (mol/L)

Time (h)

Conv (%)

Characteristics

Theoretical Mn

4

Mn PDI %C %H %S %P 1 MAEP ─ ─ 3 90 Gel 125000 6 2.98 6 35.6 5.5 0 9.5 2 MAEP PEPDTA 1 × 10-2 7 75 Soluble 7200 9000 1.18 3 MAEP PEPDTA 2 × 10-2 10 84 Soluble 4113 5500 1.23 4 MAEP PEPDTA 1 × 10-2 9 83 Soluble 7325 5500 1.46 39.3 5.9 0.2 7.0 5 MAEP CDB 1 × 10-2 40 35 Soluble 3532 1300 1.22 6 MOEP ─ ─ 3 96 Gel 215000 6 2.07 6 37.7 5.9 0 11.2 7 MOEP PEPDTA 1 × 10-2 0.7

2 7

23 44 81

Gel Gel Gel

2499 4545 8502

263000 100000

1.82 3.63

37.6

5.9

0

11.5 8 MOEP PEPDTA 2 × 10-2 7 95 Gel 9 MOEP CDB 1 × 10-2 20 74 Soluble 7780 10500 1.94 38.3 6.0 0.1 9.9

10 MOEP CDB 2 × 10-2 36 44 Soluble 2438 2800 1.84 MAEP 29.3 4.9 15.9

MOEP 34.5 5.5 14.7 Theoretical MAEP 7 30.6 4.6 15.8 Theoretical MAEP based on the MAEP-Diene-H3PO4 mixture 8 30.1 4.6 16.1 Theoretical PMAEP based on the MAEP-Diene mixture 9 34.7 4.8 13.7 Theoretical MOEP 7 34.3 5.3 14.8 Theoretical MOEP based on the MOEP-Diene-H3PO4 mixture 8 34.7 5.2 14.8 Theoretical PMOEP based on the MOEP-Diene mixture 9 38.9 5.6 12.5

2: [MOEP] = 0.96 mol/L 3: Solvent = methanol, [AIBN] = 2 × 10-3, Reaction temperature = 60 ºC 4: Theoretical Mn was calculated as conversion × a × [Monomer]/[RAFT] + b. a = 94 or 108 (molecular weight of sodium acrylate for MAEP or sodium methacrylate for MOEP), and b = 106 or 174 (molecular weight of polymer end-groups after hydrolysis for PEPDTA or CDB) 5: Polymers were hydrolysed for GPC analysis with 5M NaOH at 80 ºC for 24 hours. 6: After hydrolyses with 10M NaOH 7: Theoretical values calculated from the monomer structure. 8: Theoretical values calculated from the monomer-diene-H3PO4 mixture: ratios obtained from NMR. Monomer: diene: H3PO4 = 50.9: 22.8: 26.3(%, MAEP) and 50.8: 24.9: 24.2 (%, MOEP). 9: Theoretical values calculated from the monomer-diene mixture: ratios obtained from NMR. Monomer: diene = 69.0: 31.0 (%, PMAEP), and 65.8: 34.2 (%, PMOEP).


Figure 2.3 shows conversion versus time plots of MAEP polymerisations. A control

MAEP polymerisation was carried out under the same experimental conditions as above

but in the absence of a RAFT agent. This experiment resulted in rapid polymerisation,

reaching 100% conversion in under 1 h (curve a, Figure 2.3), and resulted in the

formation of an insoluble cross-linked gel. Polymerisation of MAEP in the presence of

PEPDTA at two different concentrations gave polymers that were soluble in methanol

with no evidence of gel formation. The rate of polymerisation was retarded in the

PEPDTA-mediated polymerisation and there is a marked increase in the inhibition time

with increased PEPDTA (curves b and c, Figure 2.3) from approximately 50 min (at

[PEPDTA]= 1 × 10-2 M) to 120 min (at [PEPDTA]= 2 × 10-2 M). When MAEP was

polymerized with CDB, the rate of polymerisation was much slower than that of

PEPDTA but with no inhibition period (curve d, Figure 2.3).

Time (min)0 500 1000 1500 2000

Con

vers

ion

(%)

0

20

40

60

80

100(a)

(b)

(d)

(c)

Figure 2.3: Conversion versus time of a MAEP polymerisation in methanol using the RAFT agent PEPDTA or CDB, and AIBN as initiator: (a) no RAFT (expt.1), (b) [PEPDTA] = 1 × 10-2 M (expt.2), and (c) [PEPDTA] = 2 × 10-2 M (expt.3) and (d) [CDB] = 1 × 10-2 M (expt.5).

Although PMAEP was soluble in water, the eluent used in the GPC analysis, it was

hydrolyzed with 5 M NaOH to give poly(acrylic acid) (PAA), and its molecular weight

48


distribution was determined with high accuracy using PAA calibration standards. The

proposed hydrolysis reaction is shown in Scheme 2.8.

O

OP O

OHOH

OS n

S

R

RR3 2

1

+ excess NaOH 80 oC

ONaO

NaS n

R

R

1

2

ONaS

R3 NaOONa + Na3PO4+ +

Scheme 2.8: Hydrolysis of polymeric side-chains and dithiocarbonate moiety of PMAEP/PMOEP using NaOH solution; R1 = H (PMAEP) or CH3 (MOEP), R2 = H (PEPDTA) or CH3 (CDB) and R3 = CH2 (PEPDTA) or none (CDB).

7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm

a

b

water CH2

CH

* *

OHO

n

b

a

Water

-(CH2)2-

methanol

b

a

CH2

CH

* *

OCH2

CH2

O

OPO OHOH

n

b

a

A B

Figure 2.4: 1H NMR (400 MHz) spectra of (A) linear PMAEP (expt. 2, Table 2.1) in methanol-d4 and (B) hydrolyzed PMAEP in D2O.

49


This method was used in the analysis of all the PMAEP and PMOEP as well as their

block copolymer samples. The disappearance of the methylene resonances (δ = 3.6-4.6)

from the side-chain in the 1H NMR spectra (Figure 2.4) showed that the polymer was

quantitatively converted to the acid form.

Figure 2.5 shows GPC chromatograms of hydrolysed PMAEP with different conversions.

Evolution of Mn is observed with increasing conversion. The values of Mn and PDI

obtained from GPC are shown in Figure 2.6. The Mn increased linearly with conversion

for the two PEPDTA concentrations, and the PDI’s decreased from 1.3 to approximately

1.2 over the conversion range. Although the Mn profile for the PEPDTA at 1 x 10-2 M

was close to theory (solid line, Figure 2.6), at the higher concentration of PEPDTA the

experimental Mns were greater than theory by a factor of approximately two (dotted line,

Figure 2.6) and converged towards the theoretical line above conversions of 70% due to

the increased amount of dead polymer arising from initiator-derived radicals. The

experimental Mn of hydrolysed polymer obtained from the CDB-mediated

polymerisation was lower (1300, expt.5 in Table 2.1) than the theoretical value (3400)

and the PDI was close to 1.22.

75% 63% 38%

16% 37% 69% A B

20% 84%

Figure 2.5: GPC traces of hydrolysed PMAEP with different conversions from expt.2, Table 2.1 (A) and expt.3, Table 2.1 (B).

50


Conversion (%)0 20 40 60 80

Mn

0

2000

4000

6000

8000

PDI

1.01.11.21.3

(a)

(b)

Figure 2.6: Mn and PDI of PMAEP polymerized with PEPDTA after hydrolysis (a) [PEPDTA] = 1 × 10-2 M (expt.2, Table 2.1 ) and (b) 2 × 10-2 M (expt.3, Table 2.1 ). The lines show the theoretical evolution of Mn with conversion for PAA. Theoretical Mn was calculated as: conversion × 94 × [MAEP]/[PEPDTA] + 106. The molecular weight of sodium acrylate is 94 and 106 is the molecular weight of polymer end-groups after hydrolysis.

2.3.1.2 Methacryloyloxyethyl phosphate (MOEP) polymerisation

The MOEP homopolymerisation was carried out in the presence of either PEPDTA or

CDB with AIBN as initiator in methanol at 60ºC. The experimental conditions and

characteristics of the polymer structure are given in Table 2.1. The polymerisations were

monitored by Raman spectroscopy and the conversion was calculated from the C=C

band at 1638 cm-1 which was normalized against the methanol solvent band at 1032 cm-1

(Figure 2.7). The local linear baselines used for these peaks were 1656–1620 cm-1 and

1069–979 cm-1, respectively. The conversion profiles are shown in Figure 2.8.

51


A B

(a) (a)

1032

1638

Inte

nsity

(a.u

.)

(b) (b)

Raman Shift (cm-1) Raman Shift (cm-1)

Figure 2.7: Raman spectra of MOEP polymerisation solution ([CDB] = 1 × 10-2 M (expt.9, Table 2.1)) (a) initial and (b) after 19.5 h in the region of (A) 3800-360 cm-1 and (B) 1800-900 cm-1.

Time (min)0 500 1000 1500 2000

Con

vers

ion

(%)

0

20

40

60

80

100 (c) (a) (b)

(d)

(e)

Figure 2.8: Conversion versus time of a MOEP polymerisation in methanol using the RAFT agent PEPDTA or CDB, and AIBN as initiator: (a) no RAFT (expt.6, Table 2.1), (b) [PEPDTA] = 1 × 10-2 M (expt.7, Table 2.1), and (c) [PEPDTA] = 2 × 10-2 M (expt.8, Table 2.1), and (d) [CDB] = 1 × 10-2 M (expt.9, Table 2.1) and (e) [CDB] = 2 × 10-2 M (expt.10, Table 2.1).

52


In the absence of a RAFT agent, the polymerisation was fast and reached high

conversion (over 85%) after 200 min (curve a, Figure 2.8). This resulted in the formation

of an insoluble gel. Polymerisation in the presence of PEPDTA showed only a slight

decrease in rate with increased RAFT agent concentration (curves b and c, Figure 2.8),

suggesting that re-initiation is not the rate determining step. The polymers formed for

the two PEPDTA concentrations were also cross-linked gels, even at low (23%)

conversion (Table 2.1). However, when CDB is used as the RAFT agent at the same

concentrations both severe retardation and inhibition of the polymerisation rates (curves

d and e, Figure 2.8) were observed. The resulting polymer was found to be soluble.

Figure 2.9 shows GPC chromatograms of hydrolysed PMOEP obtained from the CDB

mediated polymerisations (expts. 9 and 10 in Table 2.1) with different conversions. The

Mn and PDI values are plotted in Figure 2.10.

58% 43%

25%

30%

27% 20%

A B

71% 44%

Figure 2.9: GPC traces of hydrolysed PMOEP with different conversions from expt.9, Table 2.1 (A) and expt.10, Table 2.1 (B).

53


Conversion (%)0 20 40 60 80

Mn

0

2000

4000

6000

8000

10000PD

I1.01.21.41.61.82.02.2

(a)

Figure 2.10: Mn and PDI of PMOEP polymerized with CDB after hydrolysis (a) [CDB] = 1 × 10-2 M (expt.9, Table 2.1 ) and (b) 2 × 10-2 M (expt.10, Table 2.1 ). The solid lines show the theoretical evolution of Mn with conversion calculated as: conversion × 108 × [MOEP]/[CDB] + 174. The molecular weight of sodium acrylate is 108 and 174 is the molecular weight of polymer end-groups after hydrolysis.

(b)

The Mn values increased linearly with conversion and were close to theory for both CDB

concentrations (Figure 2.10), suggesting that the RAFT agent had been consumed early

in the polymerisation. However, the PDI profiles deviate from ideal ‘living’ behaviour.

The PDI profiles (Figure 2.10) are similar for the two CDB concentrations, and increase

from 1.6 at low conversion to over 2 at high conversion.

2.3.1.3 NMR of monomers and polymers

When intended for medical application, it is important to characterise the polymer

thoroughly as in this case, factors such as the amount and distribution of phosphate

groups as well as their states are predicted to affect mineralisation in SBF (details in

Chapter 4).

54


7 6 5 4 3 2 1 ppm

0.

68

9

0.

68

5

2.

00

41

.9

65

0.

96

5

1.

00

0

0.

94

5

CH2 C

H

C

O

O CH2

CH2

O P

O

OH

OHa,a’

b

c d a

a’

b c

d

I I

Methanol Water

I = Impurity

I

Figure 2.11: 1H NMR spectrum of MAEP in methanol-d4.

OHPO

OHOH

22.8%

26.3% 50.9%

A

OO

PO

OHOH

O

OO

PO

OHO

OO

O

B

Figure 2.12: 31P-NMR of MAEP monomer in methanol-d4 A) H-decoupled and B) H-coupled.

55


The 1H-NMR of MAEP monomer as received is shown in Figure 2.11. Characteristic

signals for MAEP were observed with the correct integrations. However, some

impurities were also evident at around 2.5-2.8, 3.6-3.8 and 3.9-4.1 ppm.

The 31P-NMR spectrum of MAEP monomer is shown in Figure 2.12. The proton

decoupled 31P-NMR spectrum (Figure 2.12 A) exhibits three peaks at 1.09, 0.13 and -

0.87 ppm. When coupled with protons, the patterns of these peaks become singlet, triplet

and pentet, respectively. Thus, they were assigned to free-orthophosphate (i.e.

orthophosphoric acid), MAEP monomer and diacrylate (or diene), respectively. It is

important to note that the phosphorus external standard (i.e. orthophosphoric acid) was

run in D2O, whereas all samples were in methanol-d4, which caused small shifts. The

integration of these peaks gives the following molar ratio: free-phosphate, 26.3%;

MAEP, 50.9%; diene, 22.8%.

7 6 5 4 3 2 1 ppm

3.

04

8

0.

48

1

2.

07

12

.1

44

1.

12

3

1.

00

0

0.

44

8

a,a’

b

c d

a

a’

b

c

d

I

Water

Methanol

I = Impurity Benzene

CH2 C

CH3

C

O

O CH2

CH2

O P

O

OH

OH

Figure 2.13: 1H NMR spectrum of MOEP in methanol-d4.

56


Figure 2.14: 31P NMR of MOEP monomer in methanol-d4 A) H-decoupled and B) H-coupled.

The 1H NMR of MOEP as received is shown in Figure 2.13. Besides the characteristic

signals for MOEP with the correct integrations, the presence of less impurities (at 3.7-

3.8, 4.0-4.1 ppm) compared to that of MAEP was observed. A benzene resonance at 7.3

ppm was also found to be present and is possibly due to residual benzene used for the

purification of the monomer.

The 31P NMR of MOEP also showed three peaks at 0.91, 0.07, -0.74 ppm, with the same

H-coupling pattern as MAEP (Figure 2.14). The molar ratios of these peaks are: free-

phosphate, 24.2%; MOEP, 50.8%; diene, 24.9%.

The 31P NMR spectra above unexpectedly revealed that large amounts of diene are

present in both the MAEP and MOEP monomers. However, this was discovered late in

the project and therefore, purification of the monomers was not performed. It is

generally accepted that diene is formed during the purification of monomer by

distillation due to the esterification process, hence further distillation will probably not

remove this completely. Other purification techniques such as solvent extraction need to

be established.

OO

PO

OHOH

O

O

50.8%O

OPO

OHO

OO

O24.9%

OHP

OHOH

24.2%

A

B

57


8 7 6 5 4 3 2 1 ppm

19

7.

16

6

12

2.

84

1

11

1.

62

4

20

4.

65

6

10

.0

00

Water

Methanol

b a

OCH2

CH2O

P OOH

OH

O

CH2

Sn

HS

c

d

d

d

a

b

c

Figure 2.15: 1H NMR spectrum of PMAEP (expt.4, Table 2.1) in methanol-d4.

-3-2-13 2 1 0 ppm

6.

66

9

31

.4

84

61

.8

47

PMAEP

Polydiene I

I

I = Impurity

Figure 2.16: 31P NMR spectrum of PMAEP (expt.4, Table 2.1) in methanol-d4.

58


Figure 2.15 shows the 1H NMR of PMAEP (expt.4, Table 2.1). The methylene

resonance of the side-chains shows a broad peak at 4.0 – 4.5 ppm. There is another peak

at 3.5 – 3.8 ppm which is unidentified. Since after hydrolysis this peak also disappeared,

it is thought to be another methylene with an ester linkage. The peaks around 7.1-7.4

ppm are from the phenyl protons of the RAFT end-groups. Using these peaks, the

number of repeating units on PMAEP was calculated to be 107, which is higher than that

obtained from GPC (n=76).

The 31P NMR spectrum of this polymer is shown in Figure 2.16. Broadening of peaks is

often observed for polymers. The peaks at 0.1 and -0.5– -1.15 ppm are assigned to the

phosphorus from MAEP and diene in PMAEP. The broad peak at -1.9 ppm is

unassigned. There are also unidentified sharp peaks at 0.5 and -0.2 ppm. The

orthophosphate peak at 1.09 ppm is no longer present, since these polymers were

dialysed after polymerisation and thus the orthophosphates have been removed.

Figure 2.17 shows the 1H NMR spectrum of PMOEP. There is a broad peak at 4.0 – 4.7

ppm assigned to the methylene groups in the side-chain. The unidentified peak at 3.5 –

3.9 ppm is much smaller compared to that of PMAEP. The number of repeating unit in

PMOEP calculated using the RAFT end-group peaks was 26, which is close to that

obtained from GPC (n=24). The 31P-NMR spectrum of PMOEP exhibited two broad

peaks of monomer and diene units of PMOEP at 0.0 and -0.7 – -1.43 ppm, respectively

(Figure 2.18). Again, the orthophosphate peak at 0.9 ppm was no longer present.

59


OCH2

CH2O

P OOH

OH

O

CH2

Sn

CH3S

8 7 6 5 4 3 2 1 ppm

68

.8

49

34

.2

34

11

.0

15

73

.6

76

10

.0

00

Water Methanol

b a

c

d

d

b

a

c

d

Figure 2.17: 1H NMR of PMOEP (expt.10, Table 2.1) in methanol-d4.

-3-2-13 2 1 0 ppm

29

.7

57

70

.2

43

PMOEP

Polydiene

Figure 2.18: 31P-NMR of PMOEP (expt.10, Table 2.1) in methanol-d4.

60


2.3.1.4 Elemental Analyses of Monomers and Polymers

Elemental analyses were used to establish the total amounts of C, H, and P in the

polymer. Table 2.1 shows the theoretical C, H, and P% and experimental values of

monomers and polymers. Two monomer theoretical values from the monomer structures

and the monomer-diene-H3PO4 mixture with the concentrations obtained from 31P NMR

are listed. Their values are very close. Expected theoretical values for polymers are

calculated from the monomer-diene mixture from 31P NMR. Both experimental values

of MAEP and MOEP monomers showed good agreements of C, H, and P% with the

theoretical value calculated from the monomer structures.

Free-orthophosphates are removed during the purification of the polymers by dialysis as

is evident from the lack of a resonance ~1.0 ppm in the 31P NMR spectra. Therefore, the

expected polymer elemental composition can be estimated from the monomer-diene

mixture, assuming that the reactivity of monomer and diene are similar. The soluble

(expt.2) and gel (expt.1) PMAEP showed phosphorous contents of 7.0 and 9.5%,

respectively, which are much smaller than the expected value for the MAEP-diene

mixture (13.7%). Their carbon% (35.6 and 39.3% for gel and soluble PMAEP,

respectively) was also higher than the expected value (34.7%).

The PMOEP gels (expts. 6 and 7) showed P% that is slightly lower (~11%) than the

expected value of the MOEP-diene mixture (12.5%). In contrast, the soluble PMOEP

(expt.9) displayed a much lower phosphorous content (9.9%). However, the carbon% for

this sample is 38.3%, which is close to the theoretical value of the MOEP-diene mixture

(38.9%). Diene impurities and hydrolysis of phosphates are discussed in Section 2.4.1.

2.3.1.5 Hydrolysis of gel polymers

In order to gain some insight into the mechanism of gel formation, the gel polymers

obtained from the non-living systems were hydrolysed with excess NaOH for 7 days at

80ºC. This quite harsh treatment of the polymer should result in the conversion of all

ester groups in the polymer side-chains to carboxylic acid groups. Therefore, if as a

result of this treatment the polymers become soluble in good solvents, we can

61


confidently infer that the crosslinks are formed through the polymer side chains and not

the backbone carbons.

PMOEP gels from the PEPDTA-mediated polymerisation (expt.7) were hydrolysed with

5 M NaOH for 7 days. The two samples, 23 and 44% conversions were rendered soluble

and successfully characterised by GPC. In contrast the high conversion sample remained

a gel. The results showed very high molecular weights and high PDI’s. For the 23%

conversion, the value of Mn obtained was 263,000 (theoretical Mn =1,800) and the PDI

was 1.82. The experimental Mn of the PMA from the 44% PMOEP conversion was

100,000 (theoretical Mn = 3,500) and the PDI was 3.63. These results strongly indicate

that the polymerisation must have occurred in a non-controlled way.

Hydrolysis with 10 M NaOH solution was necessary to dissolve the PMAEP (expt.1)

and PMOEP (expt.6) synthesised without RAFT agent since no reaction was observed

even after 7 days using 5M NaOH. As expected for conventional polymerisation these

polymers showed very high Mn’s and high PDI’s (expt.1 and expt.6 in Table 2.1). These

results suggest that most if not all cross-linking of PMOEP and PMAEP gels is through

the side-chain and not through the polymer backbone. (See scheme 2.13 in the

discussion)

2.3.1.6 Synthesis of MAEP and MOEP block copolymers with AAEMA

The aim of this series of reactions was to synthesise block copolymers where one block

contained functional groups enabling immobilisation onto a surface for the

mineralization study. AAEMA has a ketone side group that can react under mild

conditions with amines to yield an imine, which can then be reduced to the more stable

secondary amines.61 In this study, a selection of block copolymers containing PAAEMA

were synthesised in order to immobilise the soluble PMOEP and PMAEP on to aminated

surfaces through the AAEMA units.

The homopolymerisation of AAEMA with CDB (expt.10, Table 2.2) gave a Mn of 5000

which is double that of the theoretical value (i.e. 2200). It had a low PDI (1.13). One

explanation could be that the efficiency of CDB for AAEMA polymerisations is only

62


~50%. It is also possible that the use of PS standards is inappropriate in this system. It

was noted earlier that CDB is an effective RAFT agent for MOEP polymerisation since

the theoretical and the experimental Mn values agreed. In that case, hydrolysis before

GPC and the use of PAA standards produced an appropriate calibration set. Krasia et al.

found similar results when AAEMA was polymerised with either 2-cyano- or 2-phenyl-

prop-2-yl dithiobenzoate (CPDB or PPDB, respectively) as the RAFT agent.62 Their

results showed a Mn value ~70% higher than the theoretical value.

OO

OO

O

n SS

OO

PO

OHOH

O

R

OO

OO

O

O

OP O

OHOH

Om n

SSR

O

OP O

OHOH

O

Sn S

OO

OO

O

m

O

OP O

OHOH

On

SSO

OO

O O

(A)

(B)

Scheme 2.9: Chain extension of (A) PAAEMA with MAEP/MOEP and (B) PMOEP with AAEMA, R = H: MAEP, CH3: MOEP.

PAAEMA was further chain extended with either MAEP (expts.11 and 12, Table 2.2)

and MOEP (expt.14, Table 2.2) (See Scheme 2.9A). The block copolymers were

hydrolysed before GPC measurements. The Mn’s for these polymerisations were double

that of theory and the PDI’s close to 1.38 which are satisfactory for our purpose since

the majority of chains contain both MOEP/MAEP and AAEMA units. It was also

necessary to couple a P(AAEMA) block of much greater than 22 units. Therefore,

PMOEP made with CDB (expt.13, Table 2.2) was chain extended with AAEMA (See

63


64

Scheme 2.9B) to give a Mn of 22500 (109 units of AAEMA) and a PDI of 1.38, again an

acceptable PDI for the purpose of this study.


65

Table 2.2: Experimental conditions of chain extension reactions and molecular weight of the polymers obtained after hydrolysis.

GPC Units Expt. Macro RAFT Mn PDI Monomer [M]

(mol/L)

[RAFT]

(mol/L)

[AIBN]

(mol/L)

Polym

Time

(min)

Conv

(%)

Theoretical

Mn Mn PDI PAAEMA PMOEP/

PMAEP

10 CDB AAEMA 3.4 0.15 0.019 980 44 2200 5000 2 1.13 22

11 PAAEMA-X 1 5000 2 1.13 MAEP 0.53 0.005 0.001 270 40 6711 12000 1.38 22 99

12 PAAEMA-X 1 5000 2 1.13 MAEP 0.53 0.005 0.001 545 66 9302 17500 1.38 22 159

13 PMOEP-X 1 10500 1.94 AAEMA 0.22 0.005 0.0006 160 20 6664 22500 1.38 109 97

14 PAAEMA-X 1 5000 2 1.13 MOEP 0.49 0.005 0.001 180 39 6857 19500 1.41 22 155

1: X ≡ SC(Ph)=S 2: Mn obtained without hydrolysis using THF GPC


2.3.2 RAFT-Mediated Polymerisations of Fluorine Containing Monomers

In this series of experiments, two targeted molecular weights of 25 and 50k were set for

all monomers. The RAFT to the initiator concentrations were set to 10:1 in all cases.

Table 2.3: Experimental results for the RAFT polymerisation of fluorinated macromers. expt. Monomer [RAFT] [I] Temp Time Conv Theory GPC

(mmol L-1) (mmol L-1) (oC) (h) (%) 1 Mn Mn PDI

1 FS 29 (PEPDTA) 2.9 (AIBN) 60 60.7 27 12000 13000 1.09

2 FS 28 (PEPDTA) ― 100 36 2

3 FS 28 (PEPDTA) 2.8 (V- 88) 80 25 92 46500 29000 1.05

4 FS 56 (PEPDTA) 5.6 (V- 88) 80 18.5 96 24000 16500 1.06

5 FS 28 (PEPDTA) 2.8 (V- 88) 80 22.5 53 27000 20000 1.06

6 FS 56 (PEPDTA) 5.6 (V- 88) 80 21.5 95 23000 14000 1.03

7 FS 28(PEPDTA) 2.8 (V-88) 80 14 87 44000 29000 1.06

8 FS 28(PEPDTA) 2.8 (V-88) 80 25 96 2 48500 31500 1.37

9 FS 28(PEPDTA) 2.8 (V-88) 80 24.5 72 78 3

36000 39000 4

28500 1.08

10 FS 28(CDB 5) 2.8 (V-88) 80 67 73 81 3

36500 40500 4

29000 1.08

11 TFPA 26 (PEPDTA) 2.6 (AIBN) 60 3 90 44500 40000 1.06

12 TFPA 53 (PEPDTA) 5.3 (AIBN) 60 2.6 88 22000 18500 1.07

13 TFPMA 25 (CDB 5) 2.5 (AIBN) 60 23 88 41000 39000 1.13

14 TFPMA 50 (CDB 5) 5.0 (AIBN) 60 18 91 21500 22500 1.14

15 TFPMA 25 (CDB 6) 2.5 (AIBN) 60 15 76 36500 38000 1.11

1: conversion obtained from FT-Raman, 2: conversion obtained from FT-NIR, 3: conversion obtained from 1H NMR, 4: Theoretical Mn calculated based on the conversion from 1H NMR, 5 and 6: CDB was purified by passing it through a neutral activity aluminium oxide column followed by a silica column in one of two protocols: (5) once and twice, respectively, or (6) twice each, using hexane as eluent.

2.3.2.1 Pentafluorostyrene (FS) polymerisation

The bulk hompolymerisation of pentafluorostyrene (FS) was carried out in the presence

of 1-phenylethyl phenyldithioactetate (PEPDTA) and 1,1’-

azobis(cyclohexanecarbonitrile) (Vazo 88) or AIBN at 80 or 60 oC, respectively. Raman

spectra of a PFS polymerisation solution are shown in Figure 2.19. The olefinic band at

1620 cm-1 was normalized against the 385 cm-1 band for conversion calculations. The

66


local linear baselines used for these peaks were 1629–1603 cm-1 (see Figure 2.19C) and

416–360 cm-1, respectively. The overlapping peak at 1645 cm-1 is assigned to the C=C

ring band. This band comes at 1600 cm-1 for styrene. With fluorine substitution, the band

shifts to a higher wavenumber.

Raman Shift (cm-1)

(b)

B A (a)

Figure 2.19: Raman spectra of PFS polymerisation solution ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM) (a) initial and (b) after 25 h, in the region of (A) 3800-200 cm-1, (B) 1800-260 cm-1 and (C) 1700-1560 cm-1, showing the local linear baseline for the area analysis under the 1620 cm-1 peak.

(b)

Inte

nsity

(a.u

.)

Raman Shift (cm-1)

(a) 1620

Raman Shift (cm-1)

C

385

(a)

67


0

20

40

60

80

100

Figure 2.20: Conversion based on the normalised 1620 cm-1 bands from Raman spectra versus time of a bulk PFS polymerisation using the RAFT agent PEPDTA, and Vazo-88 or AIBN as initiator: Curve a: [PEPDTA] = 29 mM and [AIBN] = 2.9 mM, 60 ºC (expt.1, Table 2.3), b: [PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM, 80 ºC (expt.3, Table 2.3), c: [PEPDTA] = 56 mM, and [Vazo-88] = 5.6 mM, 80 ºC(expt.4, Table 2.3).

Initially, AIBN was used as the initiator at a polymerisation temperature of 60 ºC

(expt.1, Table 2.3). However the rate of polymerisation was very slow (Figure 2.20,

curve a), and it took 60.7 h to obtain a conversion of only 26.8%. Since styrene and

substituted styrenes are known to thermally initiate, PFS and PEPDTA without an

initiator in a degassed ampule were heated at 100 ºC for 36 hours. However, only a 2%

conversion was achieved (expt.2, Table 2.3). Hence a higher temperature initiator (i.e.

1,1’-Azobis(cyclohexanecarbonitrile) (Vazo-88)) was ultimately selected for this

experiment.

The conversion profiles for two concentrations of PEPDTA (where [PEPDTA]/[Vazo 88]

= 10), targeting 25 and 50 K polymers at full conversion, are shown in Figure 2.20,

curves b and c, respectively. The rates of polymerisation for both reactions were

relatively linear until approximately 50% conversion, after which a gel effect is observed

resulting in a significantly increased rate. High conversions (> 90%) were obtained for

the 25 and 50 K polymerisations after 1500 and 1000 min, respectively.

Time (min)

(b) (c)

(a)

Con

vers

ion

(%)

0 1000 2000 3000

68


Conversion (%) 0 20 40 60 80 100

Mn

0

10000

20000

30000

40000

PDI

1.001.021.041.061.081.10

(a)

(b)

Figure 2.21: Mn and PDI of PFS polymerized with PEPDTA (a) [PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM (extp.3, Table 2.3 ), b: [PEPDTA] = 56 mM, and [Vazo-88] = 5.6 mM (extp.4, Table 2.3 ). The lines show the theoretical evolution of Mn with conversion.

Figure 2.21 shows the Mn and PDI of PFS obtained from the GPC. All the GPC

chromatograms of PFS showed narrow and unimodal distributions. There was no

evidence of termination by recombination since no high molecular weight shoulders

were observed. As can be see in Figure 2.21, the Mn for both polymerisations increased

linearly with conversion, and was in good agreement with theory up to 50% conversion.

At higher conversions, the Mn values deviated significantly from the theory. The PDI

values started at below 1.1 and at full conversion were below 1.06, suggesting that the

polymer chains were of near uniform length.

Conversion measurements using FT-NIR and NMR

In order to confirm the accuracy of the conversion obtained from FT-Raman, FT-NIR

and 1H NMR were used. In the case of FT-NIR, in situ polymerisation was performed.

69


The PFS polymerisation solution was placed in the heating block in the FT-NIR

instrument and conversion was monitored using the change in the absolute area under

the vinyl C-H stretching overtone at 6200 cm-1 (Figure 2.22).

6200 cm-1

(a)

(b)

6600 6200 5800 5400Wavenumber

Figure 2.22: FT-NIR spectra of FS polymerisation solution ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM) (a) initial and (b) after 25 h.

The conversion plot is shown in Figure 2.23(a). Compared to that of the FT-Raman

(Figure 2.23(b), 92% conversion in 25h), slightly higher rate of polymerisation was

observed (93% conversion in 20h). This can be attributed to the different heating block

setups. The FS polymerisation temperature of 80 ºC is higher than the boiling point of

PFS under high pressure (62-63 ºC at 50mmHg), and the experiments showed that

depending on whether a different heating block or oil bath was used, some deviation in

the rate of polymerisation was observed. This was confirmed by measuring the

conversion by the same technique (i.e. FT-Raman). Nevertheless, the polymer obtained

with 96% conversion from FT-NIR showed a lower Mn than the theoretical (expt.8,

Table 2.3). The high PDI (1.37) of this sample can be explained by the fact that extended

polymerisation after reaching high conversion (as shown in Figure 2.23a) and some

termination by combination has occurred.

(cm-1)

Abs

orba

nce

(a.u

.)

70


0

20

40

60

80

100

0 200 400 600 800 1000 1200 1400Time (min)

Con

vers

ion

(%)

(a)

(b)

Figure 2.23: Conversion versus time of bulk FS polymerisation ([PEPDTA] = 28 mM and [Vazo-88] = 2.8 mM, 80 ºC): Curve a, from in situ FT-NIR polymerisation, and b, from in situ Raman polymerisation.

The 1H-NMR was also used to calculate the conversion rates. After the polymerisation

was performed by in situ Raman, the ampoules were opened to take samples for NMR

analysis for direct comparison of conversions from these two techniques. Figure 2.24

shows the 1H NRM of the PFS polymerisation mixture in CDCl3. The normalised area

for the olefinic –CH=CH2 (6.65 ppm) and –CH=CH2 (5.85 and 6.19 ppm) protons for

FS compared to the sum of the aliphatic –CH-CH2 protons (1.7-3.0 ppm) for PFS were

used. Table 2.3 shows the conversions of polymerisation samples obtained from Raman

and 1H NMR (expts.9 and 10). In order to establish whether or not the nature of the

RAFT agent would influence the reaction outcome, CDB was used in a control

experiment (expt.10).

71


H HH

Figure 2.24: 1H NMR spectrum of PFS polymerisation mixture after 67h for calculation of the conversion (CDB) (this corresponds to 81% conversion).

The conversions from 1H NMR and FT-Raman were found to be in good agreement,

although the 1H NMR gave a slightly higher conversion than that from the Raman. It is

possible that during the sample preparation for 1H NMR, some monomer evaporated.

Also the polymerisation solution at this conversion is very viscous and since the sample

for NMR was taken from the top, this may have let to some discrepancy. There may also

be some variations in conversion in the polymerisation solution due to the gel effect.

PFS obtained from CDB-mediated polymerisation also showed a lower Mn compared to

the theoretical value. Both FT-NIR and 1H NMR measurements agree with the

conversion obtained from Raman.

F

FF

F

F

HH

F

FF

F

F

n

Hp b

c a

CHCl3

Water

Hc

HaHb

3 Hp

p p

72


2.3.2.2 Tetrafluoropropyl acrylate (TFPA) polymerisation

The bulk homopolymerisation of the fluorinated acrylate monomer (TFPA) was carried

out in the presence of PEPDTA and 2,2-azobis(isobutyronitrile) (AIBN) at 60 oC. Figure

2.25 shows the Raman spectra of a bulk TFPA polymerisation. The C=C band at 1636

cm-1 was normalized against the 360 cm-1 band in order to measure the conversion. The

local linear baselines used for these peaks were 1672–1588 cm-1 and 382–332 cm-1,

respectively.

(a)

(b)

A

(a)

1636

B

(b)

360


Inte

nsity

(a.u

.)

Figure 2.25: Raman spectra of TFPA polymerisation solution ([PEPDTA] = 26 mM and [AIBN] = 2.6 mM) (a) initial and (b) after 3 h, in the region of (A) 3800-200 cm-1 and (B) 1900-200 cm-1.

The conversion profiles at two concentrations of PEPDTA, targeting 25 and 50 K at full

conversion, are shown in Figure 2.26 (a) and (b), respectively. With increased PEPDTA

concentration the inhibition time increased from 20 to 40 min despite the increased

initiator concentration. After this initial inhibition period the rates were similar and rapid,

reaching high conversions (> 80%) in under 160 min.

73


0

20

40

60

80

100

Figure 2.26: Conversion versus time of bulk TFPA polymerisation using the RAFT agent PEPDTA, and AIBN as initiator at 60 ºC: Curve a: [PEPDTA] = 28 mM and [AIBN] = 2.8 mM (expt.11, Table 2.3), and b: [PEPDTA] = 56 mM, and [AIBN] = 5.6 mM (expt.12, Table 2.3).

Conversion (%)0 20 40 60 80 100

Mn

0

10000

20000

30000

40000

PDI

1.001.041.081.121.16

Figure 2.27: Mn and PDI of TFPA polymerized with PEPDTA (a) [PEPDTA] = 28 mM and [AIBN] = 2.8 mM (expt.11, Table 2.3 ), b: [PEPDTA] = 56 mM, and [AIBN] = 5.6 mM (expt.12, Table 2.3 ). The lines show the theoretical evolution of Mn with conversion.

Time (min)

Con

vers

ion

(%)

(a)

(b)

0 50 100 150 200

(a)

(b)

74


The Mn and PDI obtained from GPC are shown in Figure 2.27. The Mn profiles were in

excellent agreement with the theoretical values for both PEPDTA concentrations and the

PDI’s for all conversions were low (between 1.05 and 1.14).

2.3.2.3 Tetrafluoropropyl methacrylate (TFPMA) polymerisation

CDB was the RAFT agent of choice to mediate the fluorinated methacrylate monomer,

as PEPDTA has been shown to be a poor agent for most methacrylates including MOEP.

The bulk homopolymerisation of TFPMA was carried out in the presence of CDB and

AIBN at 60 oC. Raman spectra of TFPMA polymerisation solutions are shown in Figure

2.28 The C=C peak at 1630 cm-1 was normalized against the 602 cm-1 peak to measure

the conversion. The local linear baselines used for these peaks were 1675–1601 cm-1 and

629–568 cm-1, respectively.

A B

1630 (a) (a)

(b) (b)

602

Inte

nsity

(a.u

.)


Figure 2.28: Raman spectra of TFPMA polymerisation solution ([CDB] = 25 mM and [AIBN] = 2.5 mM) (a) initial and (b) after 23 h, in the region of (A) 3800-200 cm-1 and (B) 1820-550 cm-1.

75


Figure 2.29 shows the conversion profiles for the TFPMA polymerisation. High

conversions (~90%) were obtained after 1380 and 1080 min for the target 25 and 50 K

polymers, respectively.

Time (min)0 200 400 600 800 1000 1200 1400

Con

vers

ion

(%)

0

20

40

60

80

100

(b) (a)

Figure 2.29: Conversion versus time of bulk TFPMA polymerisation using the RAFT agent CDB, and AIBN as initiator at 60 ºC: Curve a: [CDB] = 25 mM and [AIBN] = 2.5 mM (expt.13, Table 2.3), and b: [CDB] = 50 mM and [AIBN] = 5.0 mM (expt.14, Table 2.3).

Conversion (%)0 20 40 60 80 100

Mn

0

10000

20000

30000

40000

PDI

1.001.101.201.301.40

(a)

Figure 2.30: Mn and PDI of TFPMA polymerized with CDB (a) [CDB] = 25 mM and [AIBN] = 2.5 mM (expt.13, Table 2.3 ), b: [CDB] = 50 mM, and [AIBN] = 5.0 mM (expt.14, Table 2.3 ). The lines show the theoretical evolution of Mn with conversion.

(b)

76


Figure 2.30 shows the Mn and the PDI values obtained from GPC. The Mn profiles

showed excellent agreement between experiment and theory, although the PDI’s were

slightly higher than that observed for the other two monomers, decreasing from 1.3 at

low conversion to 1.13 at high conversion.

Figure 2.31 shows the conversion profiles for the TFPMA polymerisations in the

presence of CDB with different purities. In the reaction shown as curve (a), the CDB had

been purified by passing through an aluminium oxide column once and then twice

through a silica column. The rate was significantly increased (curve (b)) when the CDB

was more intensively purified by passing it first through an aluminium oxide column

twice and followed by a further twice through a silica column. This significant change in

the rate of polymerisation by increased purification of the RAFT agent highlights both

the sensitivity of this type of reaction and the importance of reagent purity. However, the

rate of TFPMA polymerisation did not affect the control over Mn or PDI.

Time (min)0 200 400 600 800 1000 1200 1400

Con

vers

ion

(%)

0

20

40

60

80

100

(b)

(a)

Figure 2.31: Effect of CDB purity on the rate of polymerisation of TFPMA (in bulk) in the presence of 25 mM CDB and 2.5 mM AIBN at 60 oC. Using hexane as eluent, CDB was passed through: a neutral activity aluminium oxide column followed by a silica column: Curve a: once and twice, respectively (expt.13, Table 2.3), and b: twice each (expt.15, Table 2.3).

77


2.3.2.4 Block copolymerisation of macrofluorinated polymers with

pentafluorostyrene (FS), tert-butyl acrylate (tBA), acetoacetoxyethyl methacrylate

(AAEMA) and acetoacetoxyethyl acrylate (AAEA)

Scheme 2.10 summarises the chain extension reaction of fluorinated homopolymers

(PFS, PTFPA and PTFPMA) with with FS, tBA, AAEMA or acetoacetoxyethyl acrylate

(AAEA). Amphiphilic block copolymers can be prepared from the blocks consisting of

tert-butyl groups (from PtBA) through hydrolysis to form carboxylic acid groups

(Scheme 2.10D). AAEMA and AAEA have a reactive keto functionality which allows

the coupling of glycine or L-phenylalanyl glycine (Scheme 2.10H). Blocks consisting of

methacrylate and acrylate fluorinated polymers with FS were also synthesised. The

molecular weight data for all these polymerisations are given in Table 2.4.

78


79

Scheme 2.10: Schematic representations of chain extensions of PFS, PTFPA and PTFPMA, and further reactions (× = reaction did not proceed).

O

O

O

OO

O O

F

FF

F

F

n S S

F

FF

F

Fn

O

Om S S

F

FF

F

Fn

OO

m

OO

O

S S

(A)

(B)AAEA

tBA

TFA

(C)

(D)

(E)

OO

n

F

F

FF

H

S S O

O

O

OO

O O

O

n

F

F

FF

H

OO

m S S

O

F

FF

F

F

O

n

F

F

FF

H

m S S

O

F

FF

F

F

O

O

n

F

F

FF

H

OHO

m S S

O

O

n

F

F

FF

H

OO

m

OO

O

S S

(F)

(G)

(H)AAEMA

OO

n

F

F

FF

H

S S O

O

O

OO

O O

O

n

F

F

FF

H

OO

m

OO

O

S SO

NH2 OH

O

F

FF

F

F

O

n

F

F

FF

H

m S S

O

F

FF

F

F

OO

n

F

F

FF

H

OO

m S S

O

O

n

F

F

FF

H

OO

m

OO

S S

NH

OH

O

1.

NaBH3CN2.


80

Table 2.4: Experimental Results from the Chain Extension of Fluorinated Macromers via RAFT.

Expt Macro-RAFT Solvent [Monomer] [MacroRAFT] [AIBN] Polym time Conv. Theory GPC

Polymer Mn PDI (mol L-1) (mol L-1) (mol L-1) (mol L-1) (h) (%) Mn Mn PDI

1 PFS 29000 1.06 6.1 (THF) tBA (3.4) 0.00686 0.00192 1.7 39.92 54500 50000 1 1.23

2 PFS 29000 1.06 6.1 (THF) tBA (3.4) 0.00686 0.00192 2.8 71.86 74000 80000 1 1.31

3 PFS 20000 1.06 4.9 (EA) tBA (3.4) 0.00860 0.00092 1.8 46.24 43500 62500 1.15

4 PFS 14000 1.03 7.7 (EA) AAEA (1.1) 0.00456 0.00049 5.0 23.10 25000 19500 1.09

5 PTFPA 40000 1.06 5.6 (EA) AAEA (2.4) 0.00957 0.00104 3.0 50.47 65500 37500 1.10

6 PTFPA 40000 1.06 4.9 (EA) tBA (3.4) 0.00845 0.00093 1.9 59.09 70000 81500 1.11

7 PTFPA 40000 1.06 6.1 (EA) FS (2.8) 0.01038 0.00114 15.0 19.49 49500 56000 1.09

8 PTFPMA 38000 1.11 6.1 (THF) tBA (3.4) 0.00846 0.00165 20.1 56.71 64000 69000 1.18

9 PTFPMA 38000 1.11 6.5 (THF) AAEMA (2.5) 0.00903 0.00165 9.0 52.84 67000 55500 1.22

10 PTFPMA 38000 1.11 4.7 (EA) FS (2.3) 0.00840 0.00089 16.8 19.95 48000 56500 1.24

1: GPC chromatogram was bimodal


Initially chain extension of PFS with tBA was carried in tetrahydrofuran (THF).

However, the GPC chromatogram of this polymer showed a bimodal distribution

(Figure 2.32A, solid line, using photodiode array (PDA) detection at 262 nm

(styrenic side groups)) (expts.1 and 2, Table 2.4). When PDA detection at 310 nm

was used to measure the RAFT end-group (Figure 2.32A, dashed line), it showed

that the polymer in the low Mn peak had lost this group. As a consequence, the

polymerisation solvent was then changed to ethyl acetate. GPC chromatograms of

P(PFS-b-tBA) obtained from polymers synthesised in ethyl acetate (expt.3, Table 2.4)

showed a unimodal distribution as well as the presence of RAFT end-groups over the

whole PFS distribution (Figure 2.32B). The chain extension of PFS with AAEA was

also successful using ethyl acetate (expt.4, Table 2.4).

Retension Time (min)20 22 24 26

UV

Res

pons

e

0.00

0.02

0.04

0.06

Retension Time (min)20 22 24 26

UV

Res

pons

e

0.00

0.02

0.04

0.06

0.08

(A) (B)

Retention Time (min) Retention Time (min)

Figure 2.32: GPC chromatograms of the chain extension of PFS with tBA carried out in (A) tetrahydrofuran (expt.2, Table 2.4), and (B) ethyl acetate (expt.3, Table 2.4), using PDA detection at 262 nm (full line, styrenic side groups) and at 310 nm (dashed line, RAFT end group).

The stability of the RAFT end-group of PFS in THF and ethyl acetate at room

temperature was investigated over an 18 day period. UV-vis spectroscopy was used

to measure the loss of dithioester absorbance at 310 nm and a plot of the absorbance

of this peak versus time is shown in Figure 2.33.

81


0.08

0.12

0.16

0.20

0.24

0.28

0.32

0.36

0 2 4 6 8 10 12 14 16 18 20Day

Abs

(a)

(b)

Figure 2.33: Degradation of the RAFT end-groups of PFS (expt.7, Table 2.3) stored in tetrahydrofuran (curve a) and ethyl acetate (curve b) monitored by UV-vis absorbance spectroscopy at 310 nm.

These results clearly show that PFS stored in THF resulted in the significant loss of

RAFT end-groups, presumably due to oxidation of these end-groups with residual

amounts of peroxides in THF, whereas there is little or no loss when stored in ethyl

acetate.63 Electrospray ionization mass spectrometry (ESI-MS) has been used for

end-functional structural characterization.63 In future work, analysis of polymers

before and after storage in THF using this technique would make it possible to

confirm the UV-vis spectroscopy results as well as to identify the nature of the end

groups.

Chain extension of PTFPA with either tBA or FS was successfully carried out in

ethyl acetate solution. The Mn and PDI’s of the block copolymers are shown in Table

2.4 (expts.6 and 7). GPC traces of these polymers show a unimodal distribution

indicating the involvement of essentially all the macro-RAFT chains in the

polymerisation. In both monomers, Mn was well-controlled and the PDI was low. In

the case of chain extension with AAEA in ethyl acetate, some formation of gel was

observed already in the early stages of polymerisation (within 10 min). The gel

obtained was insoluble in both excess ethyl acetate and THF. The polymer that could

be extracted with THF was purified and analysed by GPC (expt.5, Table 2.4). The

Mn was close to that of the macro-RAFT, indicating no chain extension had occurred.

82


The result was the same when the solvent was changed to dimethyl sulfoxide

(DMSO).

PTFPMA was successfully chain extended with tBA, AAEMA or PFS and the Mn

and PDI’s of the block copolymers are shown in Table 2.4 (expts.8-10). In all cases,

the PDI was below 1.25. For chain extension with tBA or AAEMA, THF was used

as a solvent (expts.8 and 9). This prevented the loss of RAFT end-groups, although

oxidation was observed for the PFS-PEPDTA macro-RAFT. This indicates that

methacrylate-dithiobenzoate is much more stable to peroxides compared to PFS-

phenylditioacetate.

Amphiphilic block copolymers consisting of carboxylic acid groups were prepared

by the hydrolysis of the t-butyl groups by stirring with TFA (5-fold molar excess) in

DCM solution at room temperature overnight.60 Figure 2.34A shows the 1H NMR

spectrum of P(TFPA-b-tBA) (expt.6, Table 2.4) before hydrolysis. The chemical

shifts between 1.5-2.6 ppm are attributed to the protons of the aliphatic –CHCH2– of

both PTFPA and PtBA polymer backbones (peaks 1, 2, 5, and 6). The triplets at 4.6

and 6.4 ppm (peaks 3 and 4) correspond to the protons of –CH2-CF2– and –CHF2 of

PTFPA block, respectively. The large peak at 1.4 ppm (peak 7) is attributed to the

protons of –C(CH3)3 of the PtBA block. After hydrolysis, this large peak at 1.4 ppm

disappeared which is consistent with the hydrolysed structure (Figure 2.34B). The

broad peak at 3.4 ppm is from the water present in the system.

H2O

Figure 2.34: 1H NMR spectra of P(TFPA-b-tBA) from expt.6, Table 2.4: (A) neat polymer (in acetone-d6) and, (B) after hydrolysis with TFA (in DMSO-d6) (* = solvent).

83


2.3.3 Biomolecule Attachment

Glycine and L-phenylalanyl glycine were coupled to P(TFPMA-b-AAEMA) (expt.9,

Table 2.4) by reacting the primary amine groups with the keto group to form imines,

followed by reductive amination to form secondary amines (Scheme 2.11).

Scheme 2.11: Reaction scheme of PAAEMA block copolymer with glycine.

Figure 2.35: 1H NMR spectrum of; (A) P(TFPMA-b-AAEMA) from expt.9, Table 2.4, in DMSO-d6, (B) glycine in 1:1 = D2O:DMSO-d6 and (C) glycine modified P(TFPMA-block-AAEMA) in DMSO-d6 (* = solvent).

O

n

F

F

FF

H

OO

m

OO

O

S S

O

n

F

F

FF

H

OO

m

OO

S S

N OH

O

NH2 OH

O

O

n

F

F

FF

H

OO

m

OO

S S

NH

OH

O

NaBH3CN

Acetone

+ -

2

84


Glycine was attached to P(TFPMA-b-AAEMA) by stirring the reagents in acetone

for 7 days at room temperature, followed by mild reduction using NaBH3CN and

further stirring for 2 days. Glycine is sparingly soluble in acetone and the reaction

mixture was slightly cloudy. The resulting polymer was extensively washed with

MilliQ water. Figure 2.35 shows the 1H NMR spectra of P(TFPMA-b-AAEMA) (A),

glycine (B) and polymer after coupling with glycine (C).

The glycine peak at 3.2 ppm (peak 12) is overlapped with the water peak in Figure

2.35C. There is a peak at 8.8 ppm (peak 11), possibly from the protonated secondary

amine. Since proton exchange between amines and residual water can occur,

quantification of attached amino acid using this peak integration will not be accurate.

The shape of the peak 9 has also changed after attachment, as well as peak 9’

disappeared.

Figure 2.36 shows FTIR-ATR of P(TFPMA-b-AAEMA) (a) before and (b) after

glycine attachment in the ranges of (A) 3600-547 cm-1 and (B) 1670-1340 cm-1.

(b)

(a)

(b)

(a)

1579

1394

(A) (B)

Figure 2.36: FTIR-ATR of (a) P(TFPMA-b-AAEMA) and (b) after glycine attachment in the ranges of (A) 3600-547 cm-1 and (B) 1670-1340 cm-1.

85


Table 2.5: Expected IR bands from glycine attachments64.

Table 2.5 shows positions of expected IR bands from glycine attachments. Peaks at

1579 and 1394 cm-1 (Figure 2.36 B) indicate the presence of carboxylic acid salts

(antisymmetric and symmetric CO2 stretching bands at 1650-1540 cm-1 and 1450-

1360 cm-1, respectively). Carboxylic acid groups of glycine were possibly

deprotonated by the presence of trace amounts of cations in MilliQ water (confirmed

by ICP analysis). N-H stretching from secondary amine at 3320-3280 cm-1 is weak

and overlapped with OH stretching of bound water. There was also no evidence of

protonated amines which can be identified from the Fermi resonance at 2700-2300

cm-1.

The evidence of glycine attachment is not strong probably because the solubility of

the glycine in acetone limited the process, so it was then changed to L-phenylalanyl

glycine which is soluble in DMF or DMSO at elevated temperatures. This dipeptide

has the additional advantage that it has a phenyl group which is easy to detect by 1H

NMR.

86

halla



Figure 2.37: 1H NMR spectra of (A) P(TFPMA-b-AAEMA) from expt.9, Table 2.4, (B) L-phenylalanyl glycine and (C) L-phenylalanyl glycine modified P(TFPMA-b-AAEMA), all in DMSO-d6 (* = solvent).

The DMF solution containing L-phenylalanyl glycine, P(TFPMA-b-AAEMA) and

triethylamine was stirred at 60 ºC for 1 day followed by the addition of NaBH3CN

and then further stirred for 2 days. The polymer was extensively purified by washing

with MilliQ water before NMR. Figure 2.37 shows the 1H NMR spectra of

P(TFPMA-b-AAEMA) (A), L-phenylalanyl glycine (B) and polymer after coupling

with L-phenylalanyl glycine (C). From Figure 2.37(C), peaks from the dipeptide,

especially the phenyl group (peak 13), are clearly observed indicating successful

attachment. From the integrations of peaks 13 and 10 (from PAAEMA), 1 out of 8

AAEMA units was found to contain the dipeptide.

87


2.4 Discussion

2.4.1 RAFT-Mediated Polymerisation of Phosphate-Containing Monomers

In this study, the homo and block copolymerisation of MAEP and MOEP by RAFT

techniques using two different RAFT agents; PEPDTA and CDB were investigated.

The polymerisations without a RAFT agent were fast (high conversions at 100 and

200 min for MAEP and MOEP, respectively), but gave insoluble cross-linked gels.

However, judicious choice of RAFT agent gave soluble PMAEP and PMOEP.

2.4.1.1 Polymerization with PEPDTA

MAEP polymerisation with PEPDTA as the RAFT agent gave soluble polymers with

low PDI’s. Although there was inhibition as well as retardation, the PEPDTA-

mediated polymerisation proceeded faster compared to that using CDB. It has been

proposed that the intermediate radical becomes less stable when the Z activation

group is a benzyl group rather than a phenyl group, resulting in faster fragmentation

and hence reducing retardation.58,65 The inhibition periods increased with the

increase in the concentration of the RAFT agent, from approximately 50 min (at

[PEPDTA]= 1 x 10-2 M) to 120 min (at [PEPDTA]= 2 x 10-2 M). This suggests that

the leaving radical R (Ph(CH3)CH.) is slow to re-initiate polymerisation.12,66,67 (see

Scheme 1 (III) Reinitiation step)

The rate of MOEP polymerisation with PEPDTA was slightly retarded but no

inhibition period was observed. However, the resulting polymer was an insoluble gel

even at low conversion (23%). For methacrylate monomers, the phenylethyl group

(R group) of PEPDTA is a poor leaving group with respect to the methacrylyl

propagating radical, therefore, polymerisation is not controlled.68

2.4.1.2 Polymerization with CDB

Although the PMAEP obtained from the CDB-mediated polymerisation was soluble,

the rate of polymerisation was extremely slow and conversion only reached 40%

88


after 40 hours. Significant retardation in the polymerization of acrylates, as well as

other monomers such as styrenes and methacrylates in the presence of CDB is well

known, and has been discussed in the introduction (section 2.1.2).

The molecular weight of PMAEP-CDB obtained from GPC was 2.6 times smaller

than that of the theoretical value. Loiseau et al.69 found that the polymerization of

acrylic acid (AA) with trithiocarbonic acid dibenzyl ester as the RAFT agent resulted

in a lower Mn than the theoretical value when the [AA]/[RAFT] ratio was more than

100. This was ascribed to chain transfer to the solvent. This was observed even in

methanol which is known for its low chain transfer constant. It is possible that a

similar chain transfer reaction is occurring in the MAEP polymerisation with CDB in

methanol, although this was not tested in this study. Another possibility is that

although acrylates do not undergo self-initiation, the other initiation sources such as

impurities in the monomer and the long polymerisation time caused creation of more

chains than the RAFT agents and therefore shorter chains were obtained.

MOEP polymerisation was found to be well controlled with CDB, although the rate

of polymerisation was retarded to a lesser degree than that in the MAEP

polymerisation. However, the PDI was quite high (1.6–2.1) which increased with

conversion followed by a slight decrease at the high conversion. The reason for this

is possibly due to the high diene content and this will be further discussed in Section

2.4.1.4.

2.4.1.3 Loss of phosphate groups

From the elemental analysis of the polymers, it becomes apparent that phosphorous

groups were being lost in all systems, and that this phenomenon was most

pronounced for the soluble polymers produced by RAFT-mediated synthesis. This

loss could be attributed to hydrolysis during the polymerisation reaction by the

orthophosphoric acid present in the monomer at elevated temperatures as well as the

additional presence of acidic protons since the solvents used may contain small

amounts of water that would catalyse the reaction.

Polymerisation conducted in the presence of a RAFT agent generally took longer

(~20h) compared to without (~2h), hence hydrolysis could be expected to occur to a

89


larger extent. In addition, it was noted that the PMAEP systems were more prone to

such hydrolysis than the PMOEP systems. This is also in agreement with the general

observation that methacrylate polymers are much more stable than acrylate polymers

towards spontaneous hydrolysis in aqueous media.38

There are two possible cleavage sites on the side chains: the C-O-P phosphate ester

bond (1, Scheme 2.12) and the C-O-C ester bond (2). Cleavage 1 forms a hydroxyl

group whereas cleavage 2 leads to a carboxylic acid group.

O

OP O

OHOH

O*

R

n

O

OH

O

R

m

OHO

*

R

l

O

OP O

OHOH

O**

R

n

1

2

Scheme 2.12: Possible hydrolysis sites on the side-chain of the polymer and structure of resulting polymer (R = H: MAEP, R = CH3: MOEP).

The most likely cleavage site is the C-O-P bond (1) which is known to be unstable.

From the FTIR investigation (Section 4.3.1 in Chapter 4), it was evident that both

soluble and gel PMAEP contained a high proportion of carboxylic acid groups and

thus must have undergone hydrolysis at the C-O-C ester bond (2). This was not

observed in the FTIR spectra of PMOEP polymers.

2.4.1.4 Mechanism of gel formation

It has been proposed that phosphate-containing monomers and other monomers (e.g.

HEMA70) crosslink in free-radical polymerisations due to the presence of either

small amounts of residual diacrylate or dimethacrylate contaminants or as a result of

chain transfer to monomer or polymer.71,72 From the 31P NMR spectra, it was evident

that both MAEP and MOEP contain unexpectedly large amounts of dienes (22.8 and

24.9%, respectively), although suppliers do not mention this. Although this is

90


expected to cause extensive cross-linking, even with these high diene concentrations

with the right choice of RAFT agent, it was possible to obtain non-crosslinked

soluble polymers.

O

O

O

R

OP OOH

O

OP O

OHOH

O*

R

n

OO

R

*m

Scheme 2.13: Structure of polymer from monomer-diene mixture(R = H: MAEP, R = CH3: MOEP).

Scheme 2.13 shows the expected structure of polymer obtained from what we now

know to be more of a monomer-diene starting reagent mixture. In a system where all

the double bonds (i.e. monomer and both bonds in diene) have the same reactivity,

the critical extent of reaction at the gel point pc is given by:

[Diene ] Xw

[Monomer] + [Diene] = pc (4)

where Xw is the weight-average degree of polymerisation that would be observed in

the polymerisation of monomer in the absence of diene.73 In other words, cross-

linking in this case is inversely dependent upon the chain length of the polymer: the

greater the chain length the greater the conditional probability for crosslinking.74,75

Based on the Mn values obtained from the hydrolysis of the gels and RAFT-mediated

polymerisation (Table 2.1), it was clear that cross-linked polymer forms when high

91


molecular weight polymer is obtained and soluble polymer when molecular weights

below 20 K are produced. Therefore, by targeting low molecular weights (in this case

20K) using the RAFT technique, non-cross-linked polymers were obtained.

Another possible cross-linking mechanism is hydrogen abstraction from anywhere on

the polymer chain: more likely from the methylene groups on the side chain. This has

been observed for PMOEP synthesised by radiation.76 However, chain transfer

reactions to polymer (or even monomer) is conversion dependent (or more precisely

dependent upon the weight fraction of polymer in the reaction mixture) and not

molecular weight dependent. Therefore we can confidently exclude this mechanism

from this system.

The fact that monomers contain large amounts of dienes means that polymers formed

from these monomers by the RAFT technique are most probably highly branched

instead of having a liner structure. This would explain the high PDI’s obtained from

the CDB-mediated MOEP polymerisation (1.6–2.1). In addition, PMOEP had less

hydrolysis compared to that of PMAEP. The PDI’s of PMAEP were low (1.2–1.3),

and PMAEP contained high amounts of carboxylic acids. This indicates that since

PMAEP (and MAEP) are prone to hydrolysis, early in the polymerisation, the

hydrolysis took place resulting in less dienes present in the system.

Sherrington et al.77 have shown that a soluble branched polymer can be formed when

a monomer is polymerised with a bifunctional monomer (i.e. diene) in the presence

of a chain transfer agent. This concept can be applied to living radical polymerisation.

ATRP78-80 and RAFT81 have both been used to create branched polymers. In general,

PDI’s of branched polymers are very high. Wang78 and Li79 have used dienes that

contain cleavable linkages. The GPC data of their hydrolysed polymers showed

living character as well as narrow PDI’s. In this study the concentration of diene was

found to be ~30% and the hydrolysed PMOEPs still showed high PDI’s, whereas in

some literature reports, they are as low as 1-2%.78,79

2.4.1.5 Block Copolymerisation with AAEMA

The block copolymers containing PMAEP or PMOEP and PAAEMA were prepared

in order to immobilise the linear PMOEP and PMAEP on a surface suitable for SBF

92


studies. RAFT polymerization is known as one of the most versatile methods for the

synthesis of block copolymers since the homopolymers obtained by the RAFT

technique have RAFT end-groups which can be further chain extended with suitable

monomers. The rule of block copolymer synthesis in RAFT is that the propagating

radical for the first formed block must be a good homolytic leaving group with

respect to that of the second block.11

In the case of PMAEP blocks, since the acrylate propagating radical is not a good

leaving group with respect to the methacrylate propagating radical, PAAEMA was

synthesised first and then used as a macro-RAFT to chain extend using MAEP. The

living nature of the soluble PMOEP prepared with CDB was observed when it was

used as a macro-RAFT agent to chain extend with AAEMA to form the block

copolymer. Chain extension of PAAEMA with MOEP was also successful. In all

cases, the experimental Mns after hydrolysis of the side chains were much higher

than the corresponding theoretically calculated values. However, the PDI was

reasonable and in the range from 1.38-1.41.

2.4.2 RAFT-Mediated Polymerisation of Fluorine-Containing Monomers

2.4.2.1 FS Polymerisation

The bulk hompolymerisation of FS was carried out in the presence of PEPDTA and

Vazo 88 at 80 oC. High conversions (>90%) were obtained after 1500 and 1000 min

for polymerisations targeting 25 and 50 K at full conversions, respectively. In

comparison, the ATRP polymerisation of FS in the bulk at 110 ºC was very rapid.

When a [M]0:[Initiator]0 ratio of 55:1 was employed, around 90% conversion was

obtained in 90 min.44 In this study, the [M]0:[Initiator]0 ratios are 2579:1 and 1290:1

for targeting 25 and 50 K at full conversion, respectively.

The Mn for both polymerisation conditions showed good agreement with theoretical

values up to 50% conversion, whereas large deviations were observed at higher

conversion. Since the PDI values decreased from 1.1 to below 1.06 at full conversion,

as well as the fact that the polymerisation rate at the high conversion was not

93


retarded, this indicates that the RAFT process was dominating. Three possible

explanations were considered for the observed lower Mn compared to the theory:

• The conversion obtained from FT-Raman was incorrect

• Mn obtained from GPC was inaccurate

• The creation of more chains during the polymerisation

The conversion of the FS polymerisation was also monitored by FT-NIR and 1H

NMR which are often the techniques of choice in the literature.25,44 Although, there

were slight variations, possibly due to the experimental errors, the results correlated

well with the initial FT-Raman analysis.

For highly fluorinated polymers, it was found that a deviation occurred between the

calculated Mn and the Mn obtained from GPC. This was explained by the smaller

hydrodynamic volume of the fluorinated polymers compared to the polystyrene

standards. This is also commonly observed for other types of monomers when

standard and sample are different. However in the case of PFS synthesised by the

ATRP technique, this was not observed.44 Good agreement was found between the

calculated Mn and the experimental data obtained from the GPC with the same GPC

conditions used here (i.e. PS standards, THF as an eluent, room temperature).

Therefore it can be concluded that GPC values in this study should be reasonably

accurate. However, in future work, these polymers should be characterised with more

advanced GPC, such as with multiple angle laser light scattering (MALLS) detectors,

to obtain the absolute molecular weights. Although 1H NMR has been used to

calculate the molecular weights in the literature, the chains derived from the other

initiation sources do not contain the RAFT end groups. Therefore the Mn vs

conversion based on 1H NMR is not an accurate measure of the RAFT efficiency.

According to the RAFT mechanism, the lower Mn than the theoretical value may be

an indication of the existence of other sources of radicals than the RAFT-derived

ones, such as initiator-derived chains. This should not be the case, since the initiator

concentration was kept small to minimise initiator-derived chains (i.e. [RAFT]:[I] =

10:1).18

94


Styrene and substituted styrenes are known to undergo thermal polymerisation by

two proposed mechanisms. Mayo proposed the formation of a Diels-Alder dimer

(AH) followed by transfer of a hydrogen atom from the dimer to a monomer

(mechanism shown in Scheme 2.14).73,82,83 Flory’s mechanism involves the

formation of 1,4-diradicals (•M2•, in Scheme 2.14).84 This diradical can transfer a

radical to a monomer and then start monoradial polymerisation, or start diradical

polymerisation. It has also been proposed that a diradical transform to a Diels-Alder

dimer or 1,2-diphenylcyclobutane (DCB) that is inactive in polymerisation.84 In the

case of FS, it is suggested that 1,4-diradicals form which then initiate the

polymerisation.85 This study also showed thermal initiation of FS, although it was

very slow (2% conversion after 36 hours).

Scheme 2.14: Mechanism of Diels-Alder dimer and 1,4-diradical formations of styrene. (Reproduced from Ref: 86) M = styrene, AH = Diels-Alder dimer, •M2• =

1,4-diradical and DCB = 1,2-diphenylcyclobutane.

ATRP polymerisations of FS did not show a deviation of GPC determined Mn

compared to theory, possibly due to the fast polymerisation time (~90 min). For

RAFT-mediated FS polymerisations with PEPDTA, thermal initiation effects are

possibly more pronounced due to the long polymerisation times (1000 and 1500 min).

95

halla



The use of another RAFT agent, CDB, which showed even longer polymerisation

time also showed a lower Mn than the theoretical value.

2.4.2.2 Tetrafluoropropyl Acrylate (TFPA) and Tetrafluoropropyl Methacrylate

(TFPMA) Polymerisations

The PTFPAs obtained from PEPDA-mediated polymerisation showed the excellent

agreement of Mn and the theory as well as the low PDI (1.05-1.14), indicating that

this is a well controlled RAFT polymerisation. TFPMA polymerisation was carried

out in the presence of CDB, and the Mn obtained from the GPC correlated well with

the theoretical values. The PDI was in the range 1.13-1.3, which is slightly higher

than those of the other two monomers. It was also observed that the purity of CDB

was an important consideration in controlling the rates of polymerisation for TFPMA,

in agreement with previous findings.19

2.4.2.3 Chain Extension of Fluorinated Macro-RAFT

The fluorinated homopolymers were chain extended with FS, tBA, AAEMA or

AAEA. The residual peroxides in THF were found to cleave the RAFT end-groups

of PFS-PEPDTA by oxidation. Whereas, PTFPMA-CDB was found to be stable in

THF resulting in successful chain extension in this solvent. Chain extension of

PTFPA-PEPDTA with AAEA did not proceed successfully. This suggests that

although TFPA and AAEA are both acrylates, the PTFPA macro-RAFT is less stable

and sterically hindered the propagating radical compared to the AAEA radical. One

way to overcome this may be by polymerizing AAEA first and then chain extending

with TFPA. Amphiphilic block copolymers were successfully prepared from the

blocks consisting of t-butyl groups (from t-butyl acrylate) through hydrolysis to form

acids.

2.4.3 Biomolecule Attachment

Attachment of biomolecules to functional polymers has been widely studied due to

many applications in medicine such as drug delivery, gene therapy and improving the

96


biocompatibility. For example, functionalisation of polymer surfaces with peptides

containing the cell recognition motif RGD (R: arginine, G: glycine, D: aspartic acid)

is one approach to instigating cell-material interaction.87-90 The functional groups of

polymers most often used for coupling are: carboxylic acid, amine, and hydroxyl

groups. These groups are first preactivated (e.g. carboxyl group with N-

hydroxysuccinimide (NHS) to form the active ester), followed by the coupling with

the amines of biomolecules.

Polymers containing reactive AAEMA or AAEA allow attachment of biomolecles,

by simple coupling of the keto group with the amines. Attachment of glycine and L-

phenylalanyl glycine, onto P(TFPMA-b-AAEMA) was investigated in this study.

Glycine attachment onto P(TFPMA-b-AAEMA) was not clearly observed from 1H

NMR. This is possibly due to the low solubility of glycine in acetone, which was the

reaction solvent. L-phenylalanyl glycine is soluble in DMF at elevated temperatures,

and the coupling reaction onto P(TFPMA-b-AAEMA) was successful in this solvent.

In literature, amine compounds have been successfully coupled onto PAAEMA in

the common solvent.91 (This will be further discussed in Chapter 3, Section 3.2.1) To

overcome solubility problem, protected amino acids can be used.87 However, this

strategy also has a drawback of the need of harsh conditions to remove the protecting

groups after attachment.

97


2.5 Conclusion

To the best of my knowledge, this is the first successful report of soluble PMAEP

and PMOEP synthesised by RAFT-mediated polymerization. It is also the first time

that such large amounts of diene impurities were identified from the 31P NMR of

both MAEP and MOEP. Such dienes are known to cause cross-linking, and this

study was also able to show that the cross-linking is molecular weight dependent.

Therefore, by limiting the molecular weight below 20K using the RAFT technique, it

was possible to prevent cross-linking. Block copolymers consisting of PMAEP or

PMOEP with PAAEMA were successfully synthesised for the purpose of

immobilising them onto aminated slides.

Well-defined fluorinated homopolymers were synthesised using the RAFT

technique. PFS homopolymers showed much lower Mns compared to the theoretical

values. This is proposed to be due to the thermal initiation of PFS and the longer

polymerisation times compared to the ATRP technique. The purity of the RAFT

agent, CDB, has also been shown to be a factor important in maintaining fast rates of

polymerization, in agreement with literature. It was also found that storing PFS in

THF resulted in the loss of the RAFT end-groups, presumably due to oxidation by

peroxides. Block copolymers consisting of fluorinated polymers as the first block

and either the protecting (t-butyl) or reactive groups (ketones) as the second were

successfully synthesized. It was found that the formation of amphiphilic block

copolymers from the hydrolysis of the t-butyl groups was facile, and the attachment

of amino acids (glycine or L-phenylalanyl glycine) to the ketones of PAAEMA

provided a simple methodology for the attachment of important model biomolecules.

The resulting well-defined fluorinated homo and copolymers are highly suitable for

investigation of surface modification of fluoropolymers by simple adsorption.

98


2.6 References

(1) Matyjaszewski, K., Ed. Controlled/living radical polymerization : progress in ATRP, NMP, and RAFT; ACS: Washington, D.C., 2000.

(2) Matyjaszewski, K., and Davis, T.P., Ed. Handbook of radical polymerization; John Wiley and Sons, Inc.: USA, 2002.

(3) Hawker, C. J., Bosman, A W., and Harth, E. Chem Rev 2001, 101, 3661-3688. (4) Listigovers, N. A., Georges, M. K., Odell, P. G., and Keoshkerian, B.

Macromolecules 1996, 29, 8992-8993. (5) Benoit, D., Chaplinski, V., Braslau, R., and Hawker, C.J. J Am Chem Soc

1999, 121, 3904-3920. (6) Matyjaszewski, K. a. X., J. Chem Rev 2001, 101, 2921-2990. (7) Kamigaito, M., Ando, T. and Sawamoto, M. Chem Rev 2001, 101, 3689-3746. (8) Matyjaszewski, K. Macromol Symp 2003, 195, 25-31. (9) Chiefari, J., Chong, Y. K., Ercole, F., Krstina, J., Jeffery, J., Le, T. P. T.,

Mayadunne, R. T. A., Meijs, G. F., Moad, C. L., Moad, G., Rizzardo, E. and Thang, S. H. Macromolecules 1998, 31, 5559-5562.

(10) Chiefari, J., and Rizzardo, E. In Handbook of radical polymerization; Matyjaszewski, K., and Davis, T.P., Ed.; Wiley-Interscience: Hoboken, 2002.

(11) Moad, G., Rizzardo, E., and Thang, S.H. Aust J Chem 2005, 58, 379-410. (12) Moad, G., Chiefari, J., Chong, Y.K., Krstina, J., Mayadunne, R.T.A., Postma,

A., Rizzardo, E. and Thang, S.H. Polym Int 2000, 49, 993-1001. (13) Favier, A., and Charreyre, M.-T. Macromol Rapid Commun 2006, 27, 653-

692. (14) Perrier, S., and Takolpuckdee, P. J Polym Sci: Part A: Polym Chem 2005, 43,

5347-5393. (15) Barner-Kowollik, C., Buback, M., Charleux, B., Coote, M.L., Drache, M.,

Fukuda, T., Goto, A., Klumperman, B., Lowe, A.B., Mcleary, J.B., Moad, G., Monteiro, M.J., Sanderson, R.D., Tonge, M.P., and Vana, P. J Polym Sci A: Polym Chem 2006, 44, 5809-5831.

(16) Barner-Kowollik, C., Quinn, J.F., Morsley, D.R., and Davis, T.P. J Polym Sci Part A: Polym Chem 2001, 39, 1353-1365.

(17) Monteiro, M. J., and de Brouwer, H. Macromolecules 2001, 34, 349-352. (18) Monteiro, M. J. J Polym Sci A: Polym Chem 2005, 43, 3189-3204. (19) Plummer, R., Goh, Y-K., Whittaker, A.K., and Monteiro, M.J.

Macromolecules 2005, 38, 5352-5355. (20) Stenzel, M. H., Barner-Kowollik, C., Davis, T.P., and Dalton, H.M.

Macromol Biosci 2004, 4, 445-453. (21) Moad, G., Chong, Y.K., Postma, A., Rizzardo, E. and Thang, S.H. Polymer

2005, 46, 8458-8468. (22) Whittaker, M. R., Goh, Y.-K., Gemici, H., Legge, T.M., Perrier, S., and

Monteiro, M.J. Macromolecules 2006, 39, 9028-9034. (23) Lansalot, M., Davis, T.P., and Heuts, J.P.A. Macromolecules 2002, 35, 7582-

7591. (24) Barner-Kowollik, C., Quinn, J.F., Nguyen, T.L.U., Heuts, J.P.A., and Davis,

T.P. Macromolecules 2001, 34, 7849-7857. (25) Goh, Y.-K., Whittaker, M.R., and Monteiro, M.J. J Polym Sci A: Polym

Chem 2005, 43, 5232-5245.

99


(26) Johnston-Hall, G., Theis, A., Monteiro, M.J., Davis, T.P., Stenzel, M.H., and Barner-Kowollik, C. Macromol Chem Phys 2005, 206.

(27) Johnston-Hall, G., Stenzel, M.H., Davis, T.P., Barner-Kowollik, C., and Monteiro, M.J. Macromolecules 2007, 40, 2730-2736.

(28) Chong, Y. K., Krstina, J., Le, T.P.T., Moad, G., Postma, A., Rizzardo, E., and Thang, S.H. Macromolecules 2003, 36, 2256-2272.

(29) Grøndahl, L., Cardona, F., Chiem, K., Wentrup-Byrne, E., and Bostrom, T. J.Mater Sci Mater Med 2003, 14, 503-510.

(30) Nuttelmana, C. R., Benoita, D.S.W., Tripodia, M.C., Anseth, K.S. Biomaterials 2006, 27, 1377-1386.

(31) Tretinnikov, O. N., Kato, K., and Ikada, Y. J Biomed Mater Res 1994, 28, 1365-1373.

(32) Kugimiya, A., Kuwada, Y., and Takeuchi, T. J Chromatography A 2001, 938, 131-135.

(33) Ishihara, K., Ziats, N.P., Tierney, B.P., Nakabayashi, N. and Anderson, J. J Biomed Mater Res 1991, 25, 1397-1407.

(34) Ishihara, K., Nomura, H., Mihara, T., Kurita, K., Iwasaki, Y. and Nakabayashi, N. J Biomed Mater Res 1998, 39, 323-330.

(35) Huang, J., and Matyjaszewski, K. Macromolecules 2005, 38, 3577-3583. (36) Zhou, F., and Huck, W.T. Chem Commun 2005, 5999-6001. (37) Ma, I. Y., Lobb, E. J., Billingham, N. C., Armes, S. P., Lewis, A. L., Lloyd,

A. W., and Salvage, J. Macromolecules 2002, 35, 9306-9314. (38) Yusa, S., Fukuda, K., Yamamoto, T., Ishihara, K., and Morishima, Y.

Biomacromolecules 2005, 6, 663-670. (39) Rixens, B., Severac, R., Boutevin, B., and Lacroix-Desmazes, P. J Polym Sci

A: Polym Chem 2006, 44, 13-24. (40) Suzuki, S., Grøndahl, L., Leavesley, D., and Wentrup-Byrne, E. Biomaterials

2005, 26, 5303–5312. (41) Koo, G. P. In Fluoropolymers; Wall, L. A., Ed.; Wiley-Interscience: N.Y.,

1972. (42) Sperati, C. A. In Polymer handbook; Brandrup, J., and Immergut, E.H., Ed.;

John Wiley & Sons: USA, 1989. (43) Hansen, N. M. L., Jankova, K., and Hvilsted, S. Eur Polym J 2007, 43, 255–

293. (44) Jankova, K., and Hvilsted, S. Macromolecules 2003, 36, 1753-1758. (45) Borkar, S., Jankova, K., Siesler, H.W., Hvilsted, S. Macromolecules 2004, 37,

788-794. (46) Jankova, K., Jannasch, P. and Hvilsted, S. J Mater Chem 2004, 14, 2902-

2908. (47) Fu, G. D., Kang, E.T., and Neoh, K.G. Langmuir 2005, 21, 3619-3624. (48) Zhang, Z.-B., Ying, S.-K. and Shi, Z.-Q. Polymer 1999, 40, 5439-5444. (49) Li, H., Zhang, Z.B., Hu, C.P., Ying, S.K., Wu, S.S. and Xu, X.D. Reactive &

Functional Polymers 2003, 56, 189-197. (50) Perrier, S., Jackson, S.G., Haddleton, D.M., Améduri, B. and Boutevin, B.

Tetrahedron 2002, 58, 4053-4059. (51) Perrier, S., Jackson, S.G., Haddleton, D.M., Améduri, B. and Boutevin, B.

Macromolecules 2003, 36, 9042-9049. (52) Shemper, B. S., and Mathias, L.J. Euro Polym J 2004, 40, 651-665. (53) Pai, T. S. C., Barner-Kowollik, C., Davis, T.P. and Stenzel, M.H. Polymer

2004, 45, 4383-4389.

100


(54) Ma, Z., and Lacroix-Desmazes, P. J Polym Sci: Part A: Polym Chem 2004, 42, 2405-2415.

(55) Yu, W. H., Kang, E.T. and Neoh, K.G. Ind. Eng. Chem. Res. 2004, 43, 5194-5202.

(56) Andruzzi, L., Hexemer, A., Li, X., Ober, C.K., Kramer, E.J., Galli, G., Chiellini, E. and Fischer, D.A. Langmuir 2004, 20, 10498-10506.

(57) Stancu, I. C., Filmon, R., Cincu, C., Marculescu, B., Zaharia, C., Tourmen, Y., Basle, M.F., and Chappard, D. Biomaterials 2004, 25, 205-213.

(58) Quinn, J. F., Rizzardo, E., and Davis, T.P. Chem. Commun. 2001, 1044-1045. (59) Oae, S. Y., T. and Okabe, T. Tetrahedron 1972, 28, 3203. (60) Ma, Q., and Wooley, K.L. J Polym Sci A: Polym Chem 2000, 38, 4805-4820. (61) Park, J. P., Monteiro, M.J., van Es, S., and German, A.L. Euro Polym J 2001,

37, 965-973. (62) Krasia, T., Soula, R., Borner, H.G., and Schlaad, H. Chem Commun 2003,

538-539. (63) Vana, P., Albertin, L., Barner, L., Davis, T.P., and Barner-kowollik, C. J

Polym Sci A: Polym Chem 2002, 40, 4032-4037. (64) Lin-Vien, D., Colthup, N.B., Fateley, W.G., and Grasselli, J.G. The handbook

of Infrared and Raman characteristic frequencies of organic molecules; Academic Press: London, 1991.

(65) Quinn, J. F., Barner, L., Rizzardo, E., and Davis, T.P. J Polym Sci A: Polym Chem 2002, 40, 19-25.

(66) Theis, A., Feldermann, A., Charton, N., Stenzel, M.H., Davis, T.P., and Barner-Kowollik, C. Macromolecules 2005, 38, 2595-2605.

(67) Monteiro, M. J. J Polym Sci Part A: Polym Chem 2005, 43, 3189-3204. (68) Quinn, J. F., Barner, L., Barner-Kowollik, C., Rizzardo, E., and Davis, T.P.

Macromolecules 2002, 35, 7620-7627. (69) Loiseau, J., Doerr, N., Suau, J.M., Egraz, J.B., Llauro, M.F., Ladaviere, C.,

and Claverie, J. Macromolecules 2003, 36, 3066-3077. (70) Robinson, K. L.; Khan, M. A.; de Paz Banez, M. V.; Wang, X. S.; Armes, S.

P. Macromolecules 2001, 34, 3155-3158. (71) Reghunadhan Nair, C. P., Clouet, G., and Brossas, J. J Polym Sci A: Polym

Chem 1988, 26, 1791-1807. (72) Pike, R. M., and Cohen, R.A. J Polym Sci 1960, 44, 531-538. (73) Odian, G. Principles of polymerization; John Wiley & Sons, Inc.: N.Y., 2004. (74) Landin, D. T.; Macosko, C. W. Macromolecules 1988, 21, 846-851. (75) Li, W. H.-.; Hamielec, A. E.; Crowe, C. M. Polymer 1989, 30, 1518-1523. (76) Chandler-Temple, A. In School of Molecular and Microbial Sciences; The

University of Queensland: Brisbane, 2007. (77) Costello, P. A., Martin, I.K., Slark, A.T., Sherrington, D.C., and Titterton, A.

Polymer 2002, 43, 245-254. (78) Wang, A. R., and Zhu, S. J Polym Sci A: Polym Chem 2005, 43, 5710-5714. (79) Li, Y., and Armes, S.P. Macromolecules 2005, 38, 8155-8162. (80) Bannister, I., Billingham, N.C., Armes, S.P., Rannard, S.P., and Findlay, P.

Macromolecules 2006, 39, 7483-7492. (81) Liu, B., Kazlauciunas, A., Guthrie, J.T., and Perrier, S. Macromolecules 2005,

38, 2131-2136. (82) Mayo, F. R. J Am Chem Soc 1953, 75, 6133-6141. (83) Mayo, F. R. J Am Chem Soc 1968, 90, 1289-1295. (84) Flory, P. J. J Am Chem Soc 1937, 59, 241-253.

101


(85) Pryor, W. A., Iino, M., and Newkome, G.R. J Am Chem Soc 1977, 99, 6003-6007.

(86) Khuong, K. S., Jones, W.H., Pryor, W.A., and Houk, K.N. J Am Chem Soc 2005, 127, 1265-1277.

(87) Hersel, U., Dahmen, C., and Kessler, H. Biomaterials 2003, 24, 4385–4415. (88) Hern, D. L., and Hubbell, J.A. J Biomed Mater Res 1998, 39, 266-276. (89) Quirk, R. A., Chan, W.C., Davies, M.C., Tendler, S.J.B., and Shakeshe, K.M.

Biomaterials 2001, 22, 865-872. (90) Jo, S., Engel, P.S., and Mikos, A.G. Polymer 2000, 41, 7595-7604. (91) Moszner, N., Salz, U., and Rheinberger, V. Polymer Bulletin 1994, 32, 419-

426.

102



3.1 Introduction A wide range of surface modification techniques has been used to introduce new

surface properties onto biomaterials. However, in many cases it has been difficult to

modify only the outermost layer of the materials using conventional techniques

without affecting more than the desired depth.

The greatest challenge is to alter the surface of materials in a controlled way. The

attachment of well-defined homo and copolymers produced by living radical

polymerisation is one approach to modifying the outermost surface of a polymeric

material with the ability to restrict the penetration of the change. Three relatively

simple yet robust techniques that fall into this category of surface fabrication are

listed below:

1. Layer-by-Layer (LbL) assemblies of soluble phosphate- and carboxylate-

containing homopolymers

2. Coupling reactions of keto-containing block copolymers onto aminated

surfaces

3. Adsorption of fluorinated homo and block copolymers containing carboxylic

acid groups onto PTFE

3.1.1 LbL Assembly

The Layer-by-Layer (LbL) ultrathin film fabrication was pioneered by Decher and

coworkers in 1991.1-4 This technique employs the alternating immersion of a solid

material into oppositely charged polyelectrolyte solutions to produce a

polyelectrolyte multilayer as an insoluble polymeric complex film on the material

surface. The surface charge is reversed upon each exposure. LbL films can be

prepared with up to a hundred layers.

There are many advantages of this innovative technique including the following:

103


• A much greater thermal and mechanical stability compared to the Langmuir-

Blodgett film technique

• It can be applied to materials of any shape, including colloids5 and porous

scaffolds6

• A large choice of polyelectrolytes can be employed

• The incorporation of other molecules, such as inorganic particles,5,7,8 DNA,9-

11 and proteins12-16 is possible

When polymers are attached onto substrate surfaces, this is conventionally illustrated

as involving conformational orientation of molecular “loops”, “tails” and “trains” as

shown in Figure 3.1.

Loop

Tail

Train

Figure 3.1: Illustration of attached polymer layers.

In LbL assemblies, it is possible to control the amounts of trains and loops through

altering the physicochemical conditions of the dipping solutions, such as the ionic

strength, and controlling the multilayer structure. The nature of the polyelectrolytes

and the concentration of the polyelectrolyte solution as well as charge density are

also important factors for tuning surface functionality and thickness.17,18 It is also

known that the pH of the solution strongly affects the deposition of weak

polyelectrolytes such as poly(acrylic acid) (PAA). Small changes in pH can cause

dramatic differences in film properties such as film thickness19,20 and morphology,21

hence this aspect has been highly studied.

104


Shiratori et al. 20 showed that the pH of the deposition solution has a large effect on

the thickness of poly(allylamine hydrochloride) (PAH)/PAA films. The PAH and

PAA bilayer thickness can be altered from 5 to 80 Å by controlling the pH. Kim and

Bruening19 investigated the deposition of hyperbranced poly(amidoamine)

dendrimers/PAA films in the pH range 2.0-8.0. The film thickness increased with

decreasing the deposition pH for PAA, or by increasing the deposition pH for the

dendrimers. Thicker films are formed because the lower charge density of the

polyelectrolytes requires more adsorption to compensate for the surface charge.

Although most of the polyelectrolytes used in LbL studies are not well defined,

Morgan et al.22 used well-defined strong and weak homo and block

copolyelectrolytes, synthesised using the RAFT technique, to investigate the pH

response on LbL film production. Their weak polyanion/strong polycation film

showed exponential film thickness growth when deposited in a pH 5.5 solution

without salt. This behaviour is indicative of strong hydrogen bonds between multiple

partially ionized weak polyanions via carboxylates. The addition of an 0.1 M NaCl

solution disrupted this hydrogen bonding and the layer thickness displayed linear

growth. Multilayer adsorption caused or enhanced by hydrogen bonding has also

been reported by other groups.23-25 The LbL films fabricated from polyanion

copolymers, either block or random structure, showed dramatic differences in film

dimensions and morphology even though these copolymers possessed equivalent

degrees of ionization, Mn’s, PDI’s, and compositions.22

Due to the ease of assembly and the high structural versatility, the LbL technique has

found a wide range of applications in many fields of science and technology.

Examples include sensors, optical materials, filtration membranes, corrosion

prevention, conducting polymers, electrochromics, and light emitting systems.

Another area where it has found interesting use is in biomedical applications. Caruso

et al.26,27 used the LbL technique to form hollow polymeric microspheres (or

capsules) by removing template particles (e.g. silica) chemically after LbL

assemblies. This technique can generate hollow polyelectrolyte spheres of a

controlled diameter, wall thickness and composition. The same technique has been

used to create hollow capsules consisting solely of DNA multilayers by hydrogen

bonding of the base pairs,28 as well as nanoporous PAH/PAA films.29 Caruso and

coworkers also were able to create PAA-only capsules using click chemistry of PAA

105


copolymers with alkyne or azide alternatively assembled onto a silica template.27

These capsules showed a pH response as well as further functionalisation. Some of

the applications for these capsules are in drug delivery, sensing and biocatalysis.

Michel et al.30 developed a technique to load a high quantity of an active substance

into LbL films in one step using large unilamellar liposome vesicles. In one study,

they loaded calcium ions, spermin and alkaline phosphatase into liposomes and

coated them with a layer of poly(D-lysine). This was then embedded into poly(L-

glutamic acid)/poly(allylamine hydrochloride) (PGA/PAH) LbL films.12 The films

were then put in contact with a solution of paranitrophenyl phosphate (PNP) which

diffused through the layer as well as liposomes and the hydrolysed phosphate esters

of alkaline phosphatase to free phosphates. After 18 hours of reaction, calcium

phosphate (CaP) mineralisation had occurred inside the closed space. This was

characterised using ATR-FTIR and AFM.

Although not within the scope of this particular project, the stability of LbL films can

be further increased by post modification to form covalent bonds. Cross-linking of

these films has been found to greatly improve film stability. For example, layers

constructed from a photoreactive diazo resin as the polycation with polyanions

containing sulphonate or carboxylic groups were post UV treated to change ionic

interactions into covalent bonds.31 Another example is that of PAH/PAA multilayer

films which were further treated by heating32 or EDC coupling29 to form amide

bonds. Lawrie et al.33 used gamma radiation to cross-link PEI layers on silica

particles.

In this study, PEI/PAA LbL films were prepared from polymer solutions with

different pHs. The thickness of the PAA film was found to be strongly dependent on

the deposition pH. Soluble PMAEP and PMOEP synthesised by RAFT

polymerisation were also used to create phosphate-containing surfaces by the LbL

technique using PEI. Although these polymers are strong polyelectrolytes due to the

phosphate groups, since PMAEP was also shown to contain large amounts of

carboxylic acid groups (discussed in Chapter 4, Section 4.3.1) this ultimately

affected the film morphology.

106


3.1.2 Attachment of Block Copolymers onto Aminated Slides

Functional polymers have attracted much attention over the years.34 According to the

IUPAC Compendium of Chemical Terminology, a functional polymer is defined as:

• A polymer that bears specific chemical groups; or

• A polymer that has specified physical, chemical, biological, pharmacological,

or other uses which depend on specific chemical groups

Functional polymers are ubiquitous even in daily life. Some examples which are

widely studied are: polymeric catalysts, polymers that are photoresponsive,

conductive, magnetic and oxygen carrying-polymers, polymers used in drug delivery,

biomaterials and as biosensors.34

2-(acetoacetoxy)ethyl methacrylate (AAEMA) is a functional monomer containing

β-ketoester moieties. The potential of this monomer has been increasingly explored

over the last few decades mainly due to two types of reactions it undergoes. Some

example reactions are shown in Scheme 3.1. The most important of these involve the

methylene and ketone carbonyl groups of these monomers participating in numerous

crosslinking reactions such as enamine formation, Michael addition, reductive

amination, as well as conventional methods involving melamines and isocyanates.35

AAEMA has attracted a wide range of applications in thermoset coatings, adhesives

and paints.

Enamine formation

Reductive amination

Michael addition

Metal Chelation

*O

OO

O O

*

n

NHR1

R2

NH2 R1

R1

*O

OO

O

*

n O

R1

*O

O O

*

n

O

O

*O

OO

O

*

n NR1 R2

NR1

*O

OO

O

*

n *O

OO

O

*

n NHR1

*

O

O

O

*n O

OM

M

Imine

Scheme 3.1: Various reaction schemes of PAAEMA.

107


A secondary, but important reaction is metal chelation.36-39 Acetoacetoxy groups act

as strong bidentate ligands capable of coordinating a wide range of metal ions with

different geometries.

The aim of this study is to immobilise block copolymers consisting of PAAEMA and

PMAEP or PMOEP onto aminated slides by reaction of the keto groups of PAAEMA

with the primary amines. The reaction of β-keto groups and amines is well

established.35 Moszner et al.40 synthesised monomers containing enamines by

reacting AAEMA with various aliphatic mono- and diamines in THF. These

monomers were successfully polymerised with AIBN as an initiator at 60 °C. They

were also able to modify PAAEMA with n-butyl amines at room temperature. Park

et al.41 synthesised composite latexes based on polystyrene and poly(n-butyl

acrylate-co-AAEMA) followed by post-crosslinking using 1,6-hexanediamine at

room temperature. The resulting film showed a significant increase in tensile strength.

Yu and coworkers42 reported that AAEMA-based resins can selectively remove

primary amines in the presence of secondary amines.

AAEMA has been homo and copolymerised with various monomers using

conventional polymerisation techniques.43 A hyperbranched AAEMA polymer has

also been produced by the Michael addition of this monomer.44 More recently,

Schlaad and coworkers synthesised well-defined homo and copolymers based on

AAEMA via RAFT-mediated polymerisation.38,45 The resulting polymers showed

strong coordination to metals and metal ions. Interestingly, it was found that

PAAEMA homopolymers self-assembled into hollow, double-stranded hierarchical

superstructures driven by the hydrogen-bridging interactions between adjacent

acetoacetoxy groups and compensation of dipole moments.45 The researchers also

produced block copolymers containing acetoacetoxy groups from

transacetoacetoxylation of PHEMA copolymer and the Claisen acylation of esters of

poly(2-hydroxyethyl ethylene) copolymer with sodium acetate.39

In this study the coupling reactions of block copolymers consisting of PAAEMA and

either PMAEP or PMOEP onto aminated glass slides was successfully accomplished.

The unstable imines formed through reaction of the amines and ketones were then

reduced to form stable secondary amines by the addition of NaCNBH3. XPS was

used to characterise the successful immobilisation. Since the PMOEP homopolymer

108


was found to be capable of attaching onto the aminated slides through electrostatic

interactions, the conformation of attached block copolymers was also investigated

using ToF-SIMS.

3.1.3 Adsorption of Fluorinated Polymers

Surfactants are known for their ability to adsorb at an interface (i.e. air/liquid,

liquid/liquid and solid/liquid) in order to reduce surface tension. The term surfactant

is a blend of the phrase “surface active agent”. They contain hydrophobic (tail) and

hydrophilic (head) groups which make them soluble in both organic solvents and

water.

Fluorinated surfactants are an important class of surfactants with a wide range of

applications, such as in coatings, textiles, moldings, fire fighting foams and

lubrication.46 Fluorocarbons are substantially greater than hydrocarbons in terms of

thermal, chemical and biological inertness, surface activity, gas dissolving capacity,

hydrophobicity and lipophobicity.47,48

Block copolymers are known to exhibit similar behaviour as surfactants. The

adsorption of block copolymers onto substrates, including polymers, has been widely

investigated.49 The driving force for adsorption is dependent on the nature of the

system. Surface micellisation and micelle adsorption of neutral block copolymers

onto hydrophilic and hydrophobic surfaces have been well studied.46 Greater rates of

adsorption for micellar solutions over non-micellar solution have been observed. For

block copolymers containing polyelectrolyte segments, adsorption is strongly

influenced by the charge on the copolymer relative to the surface.

Segregation of A-B block copolymers can be used to enhance the interfacial strength

between homopolymers A and B.46 According to Leibler’s theory, if the degree of

polymerization of the A block of the copolymer is greater than the degree of

polymerization of the A homopolymer, the homopolymer chains will penetrate the

copolymer brush.50,51

Several important applications of the adsorption of block copolymers are in the

detergency industry, in oil recovery and lubrication uses.49 However, the surface

109


modification of polymers by this technique is not well suited for industrial

application due to limitations in the thin film fabrication technology.52

Koberstein reported that if the block copolymer consist of functional groups with low

surface tension, the surface segregation of these groups occurs.53 However, this is not

applicable if the block copolymer contains high-energy reactive functional groups.

Adsorption of poly(styrene-b-tBA) (P(S-b-tBA)) onto a polystyrene substrate has

been carried out by spin coating followed by annealing.54 The block copolymers

were segregated on the surface of the polystyrene substrate due to the low surface

tension of the polyacrylate blocks. The t-butyl ester groups were subsequently

hydrolysed to form surface carboxylic acid groups.

In another study, Koberstein and coworkers formed poly(styrene-b-dimethylsiloxane)

films onto polystyrene by adsorption of these block copolymers in super critical

carbon dioxide which swelled the polystyrene.52 The driving force for adsorption in

this case is the reduction of interfacial tension between the supercritical fluid and the

polymeric substrate. The polystyrene segment of the block copolymer provides a

mechanical anchor through the formation of entanglements with the polymer chains

of the substrate.

As mentioned in the Chapter 1 (Section1.8), Marchant et al.55-57 used fluoropolymers

for the surface modification of PTFE films and expanded PTFE (ePTFE). In their

studies, random copolymers containing side-chains of perfluroundecanoyloxy groups

were adsorbed onto PTFE substrates from water. These modified polymers were

found to be stable under shear stress and supported endothelial cell attachment,

growth and function.

Some of the relevant intermolecular forces are shown in Table 3.1. Weak-electron

sharing bonds (or Lewis acid-Lewis base bonds) occur between electron donor (e.g.

I2) and electron deficient (e.g. aromatic compound) groups.58 Hydrophobic

interactions occur between non-polar groups in an aqueous solution when the

hydrogen-bonding structure of water is altered.

110


Table 3.1: Comparison of energies associated with intermolecular forces.

Intermolecular force Energy (kJ/mol)

Hydrogen bond58,59 10-50

Weak-electron sharing bond58 50

Dipole-dipole59 5-25

London dispersion59 0.05-40

Dipole-induced-dipole59 2-10

Hydrophobic60 4-21

In the case of fluorinated polymers, fluorine-fluorine (F-F) interaction, which is one

type of van der Waals force, can be expected between the fluorinated substrate and

the polymers. Halogen-halogen interactions have long been known in X-ray

crystallographic structures and are well studied. Alkorta and Elguero61 studied F-F

interaction using NMR and Atoms in Molecules (AIM) analysis. Boyd and

coworkers62 investigated F-F bonding in aromatic compounds based on electron

density. They found that the presence of F-F bonds can impart as much as 14

kcal/mol (or 59 kJ/mol) of local stabilisation in the molecules.62

Poly(pentafluorostyrene) (PFS) has previously been used for the surface modification

of PTFE, however, not by adsorption. Minko and coworkers63 fabricated amine-

functionalised PTFE surfaces and reacted this with carboxyl terminated PFS (PFS-

COOH) by coating substrates with this polymer and heating for 6 h at 150 °C. The

same procedure was used to immobilise poly(2-vinylpyridine) (PVP-COOH) onto

the PFS-functionalised surface. These grafted polymers were reversibly tuneable

from highly hydrophobic to highly hydrophilic surfaces by soaking in a solvent

selective to one of the components of the grafted brush.

One characteristic feature of polymer is known as “chain entanglement”.64 The

entanglement structure (or junction) involves binary-hooking geometry as shown in

Figure 3.2.

111


Figure 3.2: Chain entanglement of polymeric chains by binary-hooking structure.

Chain entanglement creates physical cross-linking and is a major factor controlling

the melt rheological, solid mechanical and adhesive properties of polymers.65 The

molecular weight between entanglements (Me) is often determined from the flow

properties of polymer melts. The literature values of the number of chain backbone

atoms between entanglements (Ne) of PTFE are listed in Table 3.2.

Table 3.2: Literature values of Ne of PTFE.

Ne Published Year Reference

93 1988 66

110 1988 67

112 1989 65

119 1983 68

132 1963 69

If the chain length of fluorinated segments of block copolymers is longer than this

value, the entanglement of these segments and the PTFE chain may be possible as

long as the right solvent is used to swell PTFE outer layer. This will further improve

the adhesion forces.

This study showed the adsorption of three types of fluorinated homopolymers (PFS,

PTFPMA, and PTFPA) as well as block copolymers consisting of PFS and PtBA or

PAA onto the PTFE substrate was performed. Of particular interest is the question

whether or not the polymer chains can entangle with the swollen PTFE outermost

chains. The effect of the chain length of PFS was also investigated.

112


3.1.4 Surface Characterisation Techniques

Surface analysis techniques have become vital since the recognition of the

importance of the surface properties of materials, and surface modification has

become commonplace in such a wide variety of applications. Since each technique

has inherent advantages and disadvantages, typically a multi-technique approach is

used. The choice of which depends on the information required. An important point

to note is that polymer surfaces or even polymer films that are attached are not in fact

rigid and reorientation of the chains can occur depending on the environment. This

makes it difficult to fully characterise polymer surfaces under one set of conditions

(e.g. wet or dry). In this study, the polymer on the surface could consist of even a

monolayer in thickness, meaning that the possibility that the techniques used covered

a wide range and included XPS, ToF-SIMS, AFM, IRRAS and contact angle

measurements.

3.1.4.1 X-ray photoelectron spectroscopy (XPS)

X-ray photoelectron spectroscopy (XPS) is also known as electron spectroscopy for

chemical analysis (ESCA). XPS is a sensitive technique used to determine the

chemical composition of a surface. Typically it measures the first 5-10 nm for all

elements except hydrogen. A detection limit of about 0.1% atomic concentration can

be measured.70 Although most polymers can be analysed by this technique,

specimens must be vacuum compatible, and the degradation of some radiation-

sensitive polymer samples may occur due to the extended measurement times

required. However, an advantage is that liquid samples can also be analysed in

conjunction with cryogenic sampling.

Although X-rays penetrate deeply into the sample, the photoelectrons only eject from

the surface region. The equation expressing the sampling depth (d) of XPS is given

by:

d = 3λ sin θ (1)

λ is the attenuation length of the photoelectron and θ is the angle between the sample

surface and the analyser. λ for PTFE at 1000 eV is reported to be 2.4 nm.71

113


3.1.4.2 Static time-of-flight secondary ion mass spectroscopy (ToF-SIMS)

The underlying principle of static secondary ion mass spectroscopy (SIMS)

resembles that of XPS, but the source of bombardment is an accelerated ion beam.72

The atomic and molecular fragments emitted from the material surface on

bombardment of the high-energy ion beam are subsequently detected. Static time-of-

flight (ToF) SIMS has greatly enhanced both the reliability and surface sensitivity

(sampling depth of 1-2 nm). Trace amounts (pico or even femto mole) of organic

molecules can be detected and identified.72

ToF-SIMS is widely used for polymer surface characterization. Polymer

identification is possible from the finger print spectrum (m/z ≤ 200). The polymer

surface tacticities of both poly(methyl methacrylate) (PMMA) and polystyrene (PS)

have been studied using ToF-SIMS.73

Over the past decade, ToF-SIMS has become a widely used technique for the

characterisation of adsorbed protein films. Since the sampling depth is less than the

typical dimension of most proteins (4-10 nm), this technique has proven invaluable

for studying not only the amount and conformation of the adsorbed proteins but also

their orientation.74-76 ToF-SIMS has also been used to identify the molecular

orientation of phospholipids on alkane thiol SAMs on gold.77

ToF-SIMS data are generally generating complex spectra which contain hundreds of

peaks. Therefore, multivariate analysis technique (MVA), such as principal

component analysis (PCA), have increasingly been used to aid in the interpretation of

ToF-SIMS spectra.78,79 Before PCA, data sets are commonly pre-treated with, for

example, normalization to the total sum of intensities of the selected peaks and

mean-centering. The input to PCA is a matrix where the rows are samples (i.e.

spectra) and the columns are variables (i.e. peak intensities). Mathematically, PCA

consists of the singular value decomposition of the variance-covariance matrixes.

Thus the new variables (PC1, PC2, etc) formed by this transform are linear

combinations of the original variables (the ToF-SIMS peak intensities). The process

of data decomposition using PCA results in the formation of two new matrices: a

score matrix and a loading matrix. The scores show the relationship between the

samples in the new coordinate (PC defined) space, whilst the loadings illustrate the

relationship between the original variables and the principal components.

114


3.1.4.3 Infrared reflection-adsorption spectroscopy (IRRAS)

IRRAS is a well-established technique used for the analysis of thin films, monolayers

and molecules adsorbed on flat surfaces.80 The substrate of choice is often a metal for

several reasons including its 100% reflectivity regardless of the incident angle and

polarisation. The technique utilizes a polarized IR beam: either p- or s-. That means

it either parallel or perpendicular polarized radiation with respect to the plane of

incidence, respectively (Figure 3.3).

p

s Reflected

Transmitted

Incident Beam θ θ1 1

θ2

Figure 3.3: Beam geometry and polarisation of IR radiation at the interface.

The signal-to-noise ratio (S/N) is calculated from the following equation80:

S/N = = ( – )R R

where R and R are the reflectivities of the clean and film-covered substrates,

respectively, at the frequency of the absorption maximum, R is the intrinsic noise

level of the single-beam sample spectrum, and ρ =1/R . Hence, the S/N is

proportional to the reflectivity difference between sample and reference (R –R ). For

dielectric substrates, the R –R strongly depends on polarisation, incidence angle and

the optical properties of the particular substrate. For silicon, optimum incidence

angles (θ ) are 86º and 0-50º for p- and s-, respectively.

0 F

N

N

0 F

0 F

opt80 When a beam is nearly

parallel to the surface, it is called grazing angle. Grazing angle IR has been widely

used to characterise LbL films,81 self-assembled monolayers82-84 and adsorbed

proteins.85

ρ(R0 – RF)0 F

(2) RN

115


3.1.4.4 Atomic force microscopy (AFM)

Atomic force microscopy (AFM) investigates the surface morphology on a

nanometer scale. Most types of samples can be examined and they are not required to

be conductive. AFM measures the forces of interaction between a probe and a

surface.

While a cantilever probes the sample surface, the motion of the cantilever, which is

deflected depending on the surface features, is accurately measured by a laser beam

which is reflected from the top of the cantilever. Subsequently three-dimensional

images are reconstructed. In addition to the morphology, the electrical charge and

hydrophilicity can also be assessed using AFM, leading to its increasing popularity

as a tool for characterising biomaterial surfaces as well as interfaces. Soft biological

samples such as cell surfaces and proteins on biomaterials have been successfully

characterised by non-contact AFM techniques.86,87

3.1.4.5 Contact angle measurement

Contact angle measurement techniques are one of the most sensitive measurements

for obtaining true surface information from the outermost few Angstroms of solid

surfaces.88 Although widely used for the investigation of the surface free energy of

many materials, the data can be difficult to interpret since artefacts often occur.

According to Young’s equation, the surface free energy of the solid can be derived

from the cosine of the contact angle of the liquid to the surface and the surface

tension of this liquid.86

LV

SLSV

γγγ

θ−

=cos (3)

where θ = contact angle, γ = interfacial tensions of SV (solid –vapour), SL (solid–

liquid), and LV (solid–vapour).

The assumptions which need to be valid for the straightforward interpretation of

contact angles are: be thermodynamically equilibrated, smooth, homogeneous, rigid,

immobile, non-deformable surfaces which do not swell or dissolve in the test

liquid.88 Many polymer surfaces do not satisfy all of these conditions. However, in

116


spite of this limitation, raw data from wetting experiments often contain useful

information such as the degree of hydrophilicity and molecular reorientation.

The hysteresis of contact angles can be measured by increasing (advancing) and

decreasing (receding) the test droplet volume. There are two types of hysteresis; true

or thermodynamic hysteresis and kinetic hysteresis.88 True or thermodynamic

hysteresis is generally based on the microscopic domains of non-equilibrium

transitions. Slow curve changes with time or frequency are due to the kinetic

hysteresis. In many polymer surface systems, reorientation and mobility of the

functional groups and macromolecules depending on the environment can be

observed due to thermodynamic hysteresis.

117


3.2 Experimental

3.2.1 Materials

Silicon wafers (100) and microscope glass cover slips were used as the substrates for

the LbL fabrications. Branched polyethyleneimine (PEI, Mw = 70k, 30 % aqueous

solution) was obtained from Polysciences (USA), poly(acrylic acid) (PAA, Mw = 2k)

from Aldrich. The syntheses of soluble PMAEP and PMOEP are described in

Chapter 2. HEPES (2-[4-(2-Hydroxyethyl)-1-piperazinyl]ethanesulfonic acid, 99.9%)

and MES were used as buffers to make 0.1M buffer solutions with pH 7 and pH 5.5,

respectively. 1M NaOH and 1M HCl were used to adjust the pH.

3-Aminopropyltrimethoxysilane (APS) treated glass slides (or aminated slides) were

kindly donated by Asper Biotech (ES). The syntheses of PAAEMA homopolymer

and block copolymers are described in Chapter 2. Anhydrous dimethylformamide

(DMF) and NaCNBH3 were bought from Aldrich and used as received.

Polytetrafluoroethylene (PTFE) virgin tape (1.5 mm thickness) was obtained from E-

Plas, Victoria, Australia. The thickness was measured to be 1.54-1.58 mm. Syntheses

of fluorinated homopolymers and block copolymers are described in Chapter 2.

Dichloromethane (DCM), fluorobenzene (FB), methyl ethyl ketone (MEK) and DMF

were analytical grade and used as received.

3.2.2 Methods

3.2.2.1 Layer by Layer (LbL) assembly: PEI-PAA, PEI-PMAEP, and PEI-

PMOEP

Glass cover slips (1.0 × 1.0 cm2) and silicon wafers (approx. 1.0 × 0.5 cm2) were

acid-piranha treated (3:1 (v:v) conc H2SO4/H2O2) for at least 6 hours at 80 ºC to

ensure clean and uniformly oxidised surfaces. They were then washed extensively

with MilliQ water and dried under a gentle stream of N2. The clean substrates were

used immediately. LbL film formation was carried out by immersing the substrate in

an aqueous solution of branched polyethyleneimine (PEI) (2 mg/mL) for 20 min

followed by washing with MilliQ water. Then they were immersed in an aqueous

polyanion (either PAA, PMAEP or PMOEP) solution (2mg/mL) for 20 min and

118


washed again before drying under stream of N2. Only one layer of each polycation

and polyanion was deposited on all samples.

In the case of the PEI-PAA films, for deposition solutions with pH 5.5 and 7.0, 0.1M

MES and HEPES buffers were prepared, and after polyelectrolyte addition (2mg/mL),

the pH was adjusted using 1M HCl and NaOH.

3.2.2.2 Coupling of block copolymers to aminated slide

The aminated slide (approx. 0.5 × 0.5 cm2) was immersed in either 0.5 mL of the

PAAEMA homopolymer or block copolymer solution in dry DMF (2 mg/mL). After

reacting these individually in a glass tube with a septa seal for around 16 hours to

react the primary amine on the glass surface with the ketone group on the PAAEMA

to form the unstable imine. This linkage was then stabilized by mild reduction to

form the stable secondary amine bonds by the addition of excess NaCNBH3 (~1 mg).

The solution was left to react for over 4 hours. The PMOEP homopolymer was

reacted with the aminated slide following the same procedure except for the

reduction step. The slides were then washed thoroughly with DMF, rinsed with

acetone, and dried under vacuum.

3.2.2.3 Fluorinated homo and copolymer adsorption onto PTFE

PTFE films (approx. 1.0 × 0.5 cm2) were washed in a series of solvents (2h each in

chloroform and n-hexane and overnight in methanol with stirring at room

temperature) and dried before use. The film was then soaked in the relevant polymer

solution (1 mg/mL) in different solvents (e.g. DCM, FB, MEK or DMF) in a glass

tube overnight. The sample was then washed more than three times (~10 seconds

each) with the soaking solvent and then dried.

119


3.2.3 Instrumentation

3.2.3.1 X-ray photoelectron spectroscopy (XPS)

XPS spectra were recorded using a Kratos Axis Ultra X-ray photoelectron

spectrometer with monochromated Al Kα X-ray source (1486.6 eV) running at 150

W (15 kV, 10 mA emission current). The survey scans were collected at 1200-0 eV

with 1.0 eV steps at a pass energy of 160 eV; the narrow scans at 0.1 eV steps at a

pass energy of 20 eV. Vision 2 software was used for data acquisition and processing.

The binding energies were charge-corrected using a saturated hydrocarbon C1s peak

(285.0 eV) for LbL and aminated slide samples.89 For adsorption of fluoropolymers

onto PTFE, the binding energies were charge corrected to 292.5 eV for the C-F2 of

PTFE.90

High resolution spectra were resolved into individual Gaussian-Lorentzian peaks

using a least squares fitting program (CasaXPS, Casa Software Ltd.). Component

energies, number of peaks and peak widths (FWHM of 1.0 and 1.3 for all Cs and Ns,

respectively) were fixed initially and refinement was done only for peak heights. In a

final refinement cycle, component energies and peak widths were also refined and

these changed by less than 1.0 %. Peak fit results were imported into Excel for final

illustrations.

3.2.3.2 Infrared reflection-adsorption spectroscopy (IRRAS)

IRRAS was carried out using a 19650 Series grazing angle accessory (SPECAC) in a

Nicolet FTIR spectrometer, under continuous purging with CO2 free dry air. The s-

polarized light was incident at 40º relative to the substrate surface and an MCT

detector was used. To increase the signal-to-noise ratio, four spectra recorded at 8

cm-1 resolution and 512 scans were averaged.

3.2.3.3 Atomic force microscopy (AFM)

A NanoScope IVa Mutimode AFM instrument (Veeco, USA) was used in tapping

mode to obtain information on surface roughness. Measurements were performed

using a phosphorous-doped silicon cantilever (Veeco, USA) with a nominal tip

120


radius of less than 10 nm and a scan speed of approximately 3.05 Hz. The cantilever

had a spring constant of 20-80 N/m. A nominal area of either 1.0 μm × 1.0 μm or

10.0 μm × 10.0 μm was scanned. The Ra value is obtained from analyzing the total

area of 1.0 μm × 1.0 μm of the AFM image (Figure 3.7).

3.2.3.4 Static time of flight secondary ion mass spectrometry (static ToF-SIMS)

The ToF-SIMS analyses were performed using a PHI TRIFT II (model 2100)

spectrometer (PHI Electronics Ltd, USA) equipped with a 69Ga liquid metal ion gun

(LMIG). A 15 keV pulsed primary ion beam was used to desorb and ionise species

from the sample surface. Pulsed, low energy electrons were used for charge

compensation. Stainless steel grids were additionally used to minimise charging

effects. Mass axis calibration was done with CH3+, C2H5

+ and C3H7+

in positive mode

and with CH-, C2H- and Cl- in negative mode of operation. A mass resolution m/Δm

of ~ 4500 at nominal m/z = 27 amu (C2H3+) was typically achieved. Each sample was

characterised by 8 positive and 5 negative spectra: 49 positive and 49 negative peaks

were selected for sample characterisation. Principal Component Analysis was

accomplished using PLS_Toolbox Version 3.0 (Eigenvector Research, Inc., Manson,

WA) working in the MATLAB Platform (MATLAB Version 6.5, the MathWorks

Inc., Natic, MA).

3.2.3.5 Sessile drop contact angle measurements

A custom build apparatus fitted with a Kodak Digital Science DC120 camera linked

to a Kodak Digital Science Picture Postcard Software imaging program was used to

measure the contact angles. The measurements were performed manually at room

temperature.

For advancing contact angle measurements, a drop of MilliQ water (5 μL) was

placed into contact with the flat surface using a microsyringe. An image was

recorded immediately. The syringe tip was then placed in contact with the drop and

another 5 μL was added to advance the drop edge slowly. This addition was repeated

twice to create a total of 20 μL, with images recorded each time. The images of 5, 10

121


and 15 μL drops were used for determining the advancing angle measurements.

Receding contact angles were measured following the same procedure after

withdrawing water from the drop. Only the image of the last 5μL drop was used.

The following equation was used to calculate the average contact angle (θ) using

height (h) and distance (d):58

The reported values are the average values of measurements for two samples from

three different locations for each sample. The errors in the angle values are the

standard deviations.

3.2.3.6 Dynamic light scattering (DLS)

Dynamic light scattering measurements were performed using a Malvern Zetasizer

Nano Series instrument running DTS software and operating a 4 mW He-Ne laser at

633 nm. Analysis was performed at an angle of 90° and a constant temperature of 25

°C. All samples were run 10 times. The number average particle size is reported.

h θd

(3) h = d/2 tan(× θ / 2)

122


3.3 Results 3.3.1 Layer-by-Layer (LbL) Assembly

In this study, LbL assembly of PEI and PAA, PMAEP or PMOEP was carried out to

create carboxylate- or phosphate-containing surfaces (Figure 3.4). Only one layer of

each of the polycation and polyanions is adsorbed on either glass or silicon surfaces.

XPS was used to analyse the adsorbed layers. The topographies of PMAEP and

PMOEP LbLs were imaged by AFM. IRRAS was also used to investigate the

polyelectrolytes complex.

pHopt =

Figure 3.4: Idealised electrostatically driven LbL assembly deposition.

3.3.1.1 LbL of PAA

Table 3.3 shows properties of polymers used for this study. Since PAA is a weak

polyanion, its deposition is expected to be strongly affected by the pH of the solution.

The literature values of pKa of PEI (Mw = 25k) and PAA are 8.591and 5.5,

respectively. The optimum pH (pHopt) for deposition is calculated using the

following equation:

For PEI and PAA, the pHopt is 7. In this study, two pH values 7 and 5.5 as well as a

control, unadjusted MilliQ water were used. The pHs of the PEI and PAA solutions

(2 mg/mL) in unadjusted water were found to be 9.9 and 3.3, respectively (Table 3.3).

pKa(polyanion) + pKa(polycation)

2(4)

OH

OH

OH

OH

OH

OH

O-

O-

O-

O-

O-

O-

+ +

+ +

+ +

+ +

+

O-

O-

O-

+

O-

O-

O-

+ +

+ +

+ +

+ +

_ _

_ _

_

Piranha Solution Polycation Polyanion

(PEI) (PAA/ PMAEP/ PMOEP)

_

Carbon contaminants

123


Table 3.3: Properties of PEI and PAA.

Mw pKa pH of 2mg/mL solution

PEI 70,000 ~8.5 9.9

PAA 2,000 5.5 3.3

020040060080010001200Binding Energy (eV)

(C) (B) (A)

C1sSi2s Si2p N1s

O1s

O KLL

O2s

Figure 3.5: XPS survey scans of (A) blank glass slide (sample 1A), (B) PEI film (sample 1B1) and (C) PEI-PAA LbL (sample 1B2).

After preparation of the PEI-PAA LbL, the XPS survey spectrum was compared to

those of the PEI and the untreated slides as shown in Figure 3.5. The peak labelled

KLL arises from the Auger process. Glass slides contain O, C, Si, as well as trace

amounts of Na (0.2%), Ti (0.4%), N (1.2%), K (0.2%), B (2.15%) and S (0.8%)

(Figure 3.5A). PEI deposition was clearly observed from the increase in the N 1s and

C 1s peaks (Figure 3.5B). The PAA film also showed an increase in C 1s (Figure

3.5C). Table 3.4 summarises the atomic % of O, C, N and Si obtained from the XPS

survey scans of all samples.

The C/N ratio is also shown in Table 3.4. After PEI deposition, the ratios of C/N

were in the range of 3.3-4.0, which was slightly higher than the expected value from

the PEI structure (C/N=2). This is most probably due to the co-contribution of

hydrocarbons from the contaminants on the glass slide. After PAA deposition, the

124


C/N ratios increased to 5.9-9.6, indicating successful film formation at all pH values

studied.

Table 3.4: Atomic % of O, C, N and Si.

Sample Solution Polymer O C N Si C(=O)O C/N

1A a ― ― 65.0 10.3 1.2 19.8 ― 8.6

1B1 Water PEI 58.8 16.3 4.9 20.0 ― 3.3

1B2 Water PEI-PAA 46.3 35.0 4.1 14.6 7.1 8.5

1C1 pH 5.5 PEI 58.3 15.0 4.5 22.3 ― 3.3

1C2 pH 5.5 PEI-PAA 55.0 22.0 3.7 19.3 2.5 5.9

1D1 pH 7 PEI 61.2 12.3 3.1 23.3 ― 4.0

1D2 pH 7 PEI-PAA 58.5 15.3 1.6 24.6 4.0 9.6 a: Untreated glass slide

The deposition of PEI was not dramatically affected by pH, since the atomic C% was

found to lie in a reasonably close range (12.3-16.3%). On the other hand the atomic

C % of the PAA deposited surface showed large differences depending on the pH. In

pH 7, the PAA deposition only increased its atomic C% from 12 (PEI deposited

surface) to 15%, whereas at pH 3.3 (unadjusted water), it increased from 16 to 35%.

This indicates that the PAA film fabricated at pH 3.3 was much thicker than that

obtained from the pH 7 solution.

282285288291294

Binding Energy (eV)

282285288291294

Binding Energy (eV)(A) (B)

K1s 1 2

3

4

1

2

3 5

4 5

6

Figure 3.6: C 1s narrow scans of (A) PEI film (sample 1B1) and (B) PEI-PAA LbL (sample 1B2).

125


Figure 3.6 shows the XPS C1s high resolution spectra of PEI and PAA deposited

surfaces. The C1s peak from the PEI deposited surface (Figure 3.6A) required five

peaks to fit, and these peaks were assigned to C*-C (peak 1, 285.0 eV), C*-C-O

(peak 2, 285.8 eV), C*-N (amino, peak 3, 286.1 eV), C-O, (peak 4, 286.9 eV) and N-

C*=O (amide, peak 5, 288.1 eV).89,92,93 Peak 1, 2 and 4 possibly arise from the

carbon contaminants. The small amide peak indicates some possible oxidation of

amines. The C1s spectrum of the PAA deposited surface (Figure 3.6B) showed one

additional peak at 289.1 eV (peak 6) due to C*(=O)-O. The atomic % of this

carboxylic acid carbon against the total elements is also shown in Table 3.3. Peak 2

at 285.5 eV is also assigned to C*-C(=O)O from PAA.


396398400402404406

Binding Energy (eV) (B) (A)

1

2

3 1

2

3

Figure 3.7: N1s narrow scans of (A) PEI film (sample 1B1) and (B) PEI-PAA LbL (sample 1B2).

Figure 3.7 shows the N1s high resolution spectra of the PEI film and PEI-PAA LbL

on glass surface. Three peaks were required for fitting and assigned to the amine

(399.5±0.2 eV),33,94 the amide (400.7±0.2 eV),33 and the protonated amine

(401.9±0.2 eV).17,33,94-98 The presence of amide peaks correlates well with C1s peak

(peak 5 in Figure 3.6). The XPS spectrum of the thicker PEI film obtained from

drying the original PEI solution did not show evidence of any amide peak. This result

is in agreement with the fact that the oxidation of amines is more likely to occur

when it is very thin or monolayer film.33,99 Figure 3.7B shows a large increase in the

protonated amine peak when PAA was deposited (sample 1B2). Table 3.5

summarises the relative ratios of the fitted N1s peaks.

126


Table 3.5: Normalized atomic % of nitrogen species from the curve fitting of the N1s peak.

Sample Solution Polymer N1 a N2 b N3 c

1B1 Water PEI 59.1 10.6 30.4

1B2 Water PEI-PAA 33.9 8.8 57.3

1C1 pH 5.5 PEI 36.8 15.8 47.4

1C2 pH 5.5 PEI-PAA 40.0 14.3 45.7

1D1 pH 7 PEI 43.3 16.7 40.0

1D2 pH 7 PEI-PAA 31.8 27.3 40.9

a: amine (399.5±0.2eV), b: amide (400.7±0.2eV) and c: protonated amine (401.9±0.2 eV)

Although the atomic C% values are similar for all the PEI films at different pHs, the

high resolution N1s peaks show different patterns. For samples 1D1 and 1C1 (pH 7

and 5.5), there were more protonated amines (40 and 47%, respectively) compared to

the 30% found for sample 1B1 deposited in unadjusted water (pH 9.9). At pH 7 and

5.5, the ionization of PEI is 100% and thus very probably leads to stronger

interaction with the glass slides. This also indicated that fewer amine groups are

available for interaction with PAA and therefore less PAA chains were attached to

these surfaces compared to PEI deposited in unadjusted water (pH 9.9).

When PAA was attached in unadjusted water (sample 1B2), a large increase in the

protonated amine peak was observed. This indicates successful PAA deposition. At

pHs 5.5 and 7, only slight increases in the protonated amine peaks were observed,

which correlates well with small increase in the total C% that was observed in the

survey scans (samples 1C2 and 1D2).

3.3.1.2 LbL of phosphate-containing polymers

The LbL deposition of PEI and PMAEP or PMOEP onto silicon wafers was carried

out in unadjusted water. The pH of the PMAEP and PMOEP solutions (2 mg/mL)

were found to be 2.7 and 2.8, respectively (Table 3.6). The same branched PEI was

used.

127


Table 3.6: Properties of PMAEP/PMOEP used for this study.

n Mn a PDI pH of 2mg/mL solution

PMAEP 57 11500 1.46 2.7

PMOEP 97 20500 1.94 2.8 a: calculated from Mn of hydrolysed polymers obtained from GPC, assuming no hydrolysis of side-chains. The actual Mn is expected to be lower than stated here due to hydrolysis of some side-chains.


(C)

(B)

(A)

(D)

O1s

N1s

C1s

Si2p

Si2s P2p O KLL

Plasmon

Figure 3.8: XPS survey scans of (A) blank silicon wafer (sample 2A), (B) PEI film (sample 2B), (C) PEI-PMAEP LbL (sample 2C) and (D) PEI-PMOEP LbL (sample 2D).

XPS survey scans (Figure 3.8) were used to analyse LbL formation. Table 3.7

summarises the elemental compositions of the surfaces. After acid Piranha treatment,

clean silicon wafers showed only 5.5% carbon, which corresponds to a clean

surface.100 As layers of polymers were deposited, increases in the C% and decreases

in the Si% were observed, indicating both successful film and LbL formations. The

appearance of the P peak gave further evidence of successful PMAEP and PMOEP

film formations. The lower P% found for sample 2C (PMAEP layer) compared to

that of sample 2D (PMOEP layer) can be explained by the lower phosphorous

content of the PMAEP polymer as characterised and discussed in Chapter 2.

128


Table 3.7: Atomic % of C, N, P and Si from XPS survey scans.

Sample Polymer O C N P Si C(=O)O C/N

2A a ― 46.1 5.5 ― ― 48.4 ― ―

2B PEI 31.9 21.4 5.9 ― 40.8 ― 3.6

2C PEI-PMAEP 33.7 44.1 2.8 2.2 17.3 7.2 15.8

2D PEI-PMOEP 34.9 48.5 4.2 3.5 9.0 6.5 11.5

a: Untreated silicon wafer

Table 3.7 also shows the C/N ratios. The PEI deposited surface showed a C/N ratio

of 3.6, which is similar to that obtained for the PEI layer on the glass surface (sample

1B1, Table 3.4). LbL results for the phosphate polymers showed higher atomic C%

and C/N ratios than for the PAA samples.

282284286288290292294

Binding Energy (eV)

282284286288290292294

Binding Energy (eV)282284286288290292294

Binding Energy (eV)

(A)

(B) (C)

1 2 3

4

5

1 2

3 4 4’ 6

1

2 4 4’ 6

3

Figure 3.9: C1s narrow scans of (A) PEI film (sample 2B), (B) PEI-PMAEP LbL (sample 2C) and (C) PEI-PMOEP LbL (sample 2D).

The high-resolution C1s spectra of these samples are shown in Figure 3.9. The

spectrum of the PEI film (sample 2B) was fitted with five peaks: C*-C (peak 1, 285.0

129


eV), C*-C-O (peak 2, 285.6 eV), C*-N (amino, peak 3, 286.1 eV), C-O, (peak 4,

286.7 eV), and N-C*=O (amide, peak 5, 288.2 eV). Peaks 1, 2 and 4 are from the

carbon contaminants.

For the C1s spectra of the PMAEP and PMOEP LbL’s, peaks 2, 4 and 4’ were

assigned to C*-C=O (285.6 eV) and C*-O-C (286.1 eV), C*-O-P (286.5 eV),

respectively. Appearance of the C(=O)-O peak (peak 6) was observed. Atomic % are

shown in Table 3.7. No amide peaks were observed in these spectra.

396398400402404406408

Binding Energy (eV)396398400402404406408

Binding Energy (eV)

396398400402404406408

Binding Energy (eV)

(C)

(A)

(B)

1

2 3

1

3 3 1

2 2

Figure 3.10: N1s narrow scans of (A) PEI film (sample 2B), (B) PEI-PMAEP LbL (sample 2C) and (C) PEI-PMOEP LbL (sample 2D).

Table 3.8: Normalized atomic % of nitrogen species from curve fitting of the N1s peak.

Sample Polymer N1 a N2 b N3 c

2B PEI 68.1 19.3 24.5

2C PEI-PMAEP 29.3 2.9 67.8

2D PEI-PMOEP 31.2 2.2 66.6 a: amine (399.5±0.1eV), b: amide (400.9eV) and c: protonated amine (401.9eV)

130


Curve fitting of the N1s peak spectra required three peaks: the same as for LbL on

glass slides (Figure 3.10). The normalised atomic % of these peaks are shown in

Table 3.8. Sample 2B (PEI film) showed amine, amide and protonated amine

concentrations of 68, 19 and 25%, respectively. In the case of samples 2C and 2D

(PMAEP and PMOEP LbL’s), only small amide peaks were observed (2.9 and 2.2%,

respectively). This could be attributed to the low resolution of the spectra. As

expected, large increases in the protonated amine peaks were observed for these

samples (~67%).

To investigate the topography of these samples, tapping mode AFM was employed

(Figure 3.11). The mean roughness (Ra) from these images is shown in Table 3.9.

The clean silicon surface showed smooth topography (Figure 3.11A). The PEI film

(sample 2B) did not show increased roughness compared to the silicon wafer (Figure

3.11B). When PMAEP was deposited (sample 2C), the surface became much

rougher compared to the PEI film surface (Ra = 0.69 nm). Some “aggregates” (or

“patches”) of polymer molecules were clearly visible (Figure 3.11C). The PEI-

PMOEP LbL surface (sample 2D, Figure 3.11D) was much smoother than that of the

PEI-PMAEP surface, with the Ra value of 0.27 nm.

Table 3.9: Mean roughness (nm) of LbL films from the AFM images in Figure 3.7.

Sample Attached Polymer Ra (nm)

2A ― 0.10

2B PEI 0.10

2C PEI-PMAEP 0.69

2D PEI-PMOEP 0.27

131


Data Scale = 5.0 μm




A

B

C

D

Figure 3.11: 2D and 3D AFM images of A) silicon wafer, B) PEI film (sample 2B), C) PEI-PMAEP LbL (sample 2C) and D) PEI-PMOEP LbL (sample 2D) (analysed area 1.0×1.0 μm).

132


Large area scans (10×10 μm, Figure 3.12) were also carried out because surface

roughness and layer homogeneity could be expected to play an important role in

subsequent mineralisation studies. Although patchy adsorption was observed, the

PMAEP was found to uniformly cover the surface especially at the scale of SEM

images (area of 25×25 μm and 50×50 μm) which were subsequently used in the SBF

study (Chapter 4, section 4.3.2).

Data Scale = 50.0 μm Figure 3.12: 2D and 3D AFM images of PMAEP (analysed area 10×10 μm).

FTIR is a useful technique commonly used to characterise functional groups. In more

specific applications, IRRAS has been used to analyse thin films including

monolayers on silicon wafers and flat metal surfaces. Attenuated total reflectance

FTIR (FTIR-ATR) has been used throughout this study for other samples, since it

does not require any special sampling (experimental in Chapter 4, Section 4.2.2). The

FTIR-ATR spectrum of soluble PMOEP showed a large band at 974 cm-1,

corresponding to the P-O(H) stretching (Figure 3.13A). The IRRAS spectrum of the

LbL surface showed the two large bands at 1065 and 967 cm-1 corresponding to the

out-of-phase and in-phase P-O stretches of phosphate salts, respectively (Figure

3.13B).6 This indicates that large amounts of phosphate groups are interacting with

protonated amines.

133


Wavenumber (cm-1)

967 1065

(B)

974 (A)

Figure 3.13: FTIR spectra of (A) soluble PMOEP (ATR) and (B) PEI-PMOEP LbL (sample 2D) (IRRAS).

134


3.3.2 Block Copolymers Coupled onto Aminated Slides

Coupling of the block copolymers to the aminated slides was carried out in dry DMF

at room temperature followed by mild reduction with NaCNBH3 to convert the

unstable imines to the more stable secondary amines (Figure 3.14).

Figure 3.14: Idealised coupling reaction of block copolymers with aminated slide.

Table 3.10: Block-copolymers attached to aminated slides for SBF.

Coupled Polymer Sample

m a n b PDI

3B PAAEMAm 22 ― 1.13

3C PMOEPn ― 19 1.98

3D P(MOEPn-b-AAEMAm) 109 97 1.38

3E P(AAEMAm-b-MOEPn) 22 132 1.41

3F P(AAEMAm-b-MAEPn) 22 99 1.38

3G P(AAEMAm-b-MAEPn) 22 160 1.38 a: m = unit of PAAEMA segment, b: n = unit of PMOEP/PMAEP segment

Table 3.10 summarizes the properties of the polymers used for this study. PAAEMA

and PMOPE homopolymers were also reacted with aminated slides as controls. The

successful immobilization of the polymers onto the aminated slides was verified by

analysing the XPS spectral changes. ToF-SIMS was also used to identify the

conformation of one of the coupled block copolymers (sample 3D).

NH2

NH2

NH2

NH2

N

N

N

N

NH

NH

NH

NH

PA

AE

MA

PMAEP/PMOEP

Reduction

NaCNBH3

Coupling

Polymer in dry DMF

135


3.3.2.1 Quantitative XPS investigation of attached polymers

Figure 3.15 shows the XPS survey spectra of the control slides (untreated aminated,

PAAEMA and PMOEP), as well as the P(MOEP-b-AAEMA) functionalised slides.

Table 3.11 summarises the elemental compositions of the polymer attached surfaces

obtained from the XPS survey spectra.


(A) (B)

(C) (D)

O1s

N1s C1s

Si2p Si2s

P2p Na1s

Na KLL Ca2pK1sP2s

O KLL

Figure 3.15: XPS spectra of aminated slides (A) as received (Sample 3A), (B) PAAEMA functionalized (Sample 3B), (C) PMOEP functionalized (Sample 3C) and (D) P(MOEP-b-AAEMA) functionalized (Sample 3D).

Table 3.11: XPS data: Atomic % of elements concentrations from XPS survey scans.

Sample Polymer Attached Na O N Ca C P Si C(=O)O

3A – 4.0 54.4 2.6 0.9 22.3 – 15.9 0.9

3B PAAEMA 1.6 49.7 1.8 1.2 26.3 – 19.4 2.0

3C PMOEP 1.8 49.9 1.7 1.0 26.1 1.3 18.1 2.2

3D P(MOEP-b-AAEMA) 3.5 44.3 1.5 0.5 36.0 1.5 12.8 4.3

3E P(AAEMA-b-MOEP) 3.3 38.6 1.5 0.3 46.1 1.8 8.5 6.6

3F P(AAEMA-b-MAEP) 2.3 47.5 1.5 0.8 30.7 1.0 16.2 4.3

3G P(AAEMA-b-MAEP) 1.9 46.5 1.5 0.8 32.2 0.9 16.1 4.4

136


The XPS spectrum of the 3-aminopropyltrimethoxysilane (APS) treated glass slide

(aminated slide), showed the presence of N and C from the APS, as well as O, Si and

small amounts of Na and Ca from the glass slide.

Immobilisation of the homopolymers PAAEMA (sample 3B) and PMOEP (sample

3C) onto aminated slides resulted in a reduction in the amount of N and an increase

in the amount of C (to 26 atom% in both systems) indicating successful attachment

of the homopolymers. In addition, the PMOEP-modified slide showed the

appearance of P as expected (1.3%).

The amounts of C increased to 30.7-46.1% when block copolymers were attached.

This suggests that a higher amount was attached for the copolymer than for the

homopolymers. For both PMOEP and PMAEP copolymers, those with the longer

PMOEP or PMAEP segments (samples 3E and 3G) showed higher attachments

compared to those with shorter chains (samples 3D and 3F). In all cases, the amount

of N was reduced to 1.5% and presence of P was observed (0.9-1.8%).

A decrease in Si% was only observed for samples 3D and 3E (PMOEP copolymers)

compared to the aminated slide. Others showed an increase in Si% after polymer

attachment. Most probably this can be attributed to the carbon contaminants on the

aminated slides which were either partially replaced after polymer adsorption or

removed by the washing steps.

Figure 3.16 shows the high resolution C 1s peaks for samples 3A-3D. The untreated

aminated slide (Figure 3.11A) shows peaks related to APS (C*-C (peak 1, 285.0 eV)

and C*-N (amino, peak 3, 286.1 eV)) as well as from carbon contaminants (C*-C-O

(peak 2, 285.5 eV), C-O, (peak 4, 286.9 eV), and C*(=O)-O (peak 6, 288.8 eV)).

Amide peaks (N-C*=O (peak 5, 287.8 eV)) was also observed. When polymers were

attached, changes in the peak shapes were observed. The increase in the ester peak

(peak 6) after polymer attachment was clearly observed and the atomic % of this

peak is shown in Table 3.11.

137


282284286288290292294

Binding Energy (eV)

282284286288290292294

Binding Energy (eV)

282284286288290292294

Binding Energy (eV)

282284286288290292294

Binding Energy (eV)

(A) (B)

(C) (D)

1

2 3 4

5 6

1

2

4 4’ 6

5 3

1

2

4 4’ 6

5 3

1

2

4 4’ 6

5 3

Figure 3.16: C1s narrow scans of aminated slides (A) as received (sample 3A), (B) PAAEMA functionalized (sample 3B), (C) PMOEP functionalized (sample 3C) and (D) P(MOEP-b-AAEMA) functionalized (sample 3D).

395397399401403405

Binding Energy (eV)

395397399401403405

Binding Energy (eV)

395397399401403405

Binding Energy (eV)

395397399401403405

Binding Energy (eV)

(A)

(C) (D)

(B)

Figure 3.17: N 1s narrow scans of aminated slides (A) as received (sample 3A), (B) PAAEMA functionalized (sample 3B), (C) PMOEP functionalized (sample 3C) and (D) PMOEP-b-PAAEMA functionalized (sample 3D).

138


The high-resolution spectra of the N1s peak for all the samples (Figure 3.17)

required three peaks for the fit. These peaks were assigned to amine (399.6±0.2 eV),

amide (400.7±0.2 eV) and the protonated amine (401.8±0.2 eV), as for the LbL films.

No oxidised nitrogen species (i.e. NO3-) were detected for any of the samples by XPS.

The shape of the N1s peak changed depending on which polymer was attached to the

slide. The relative ratios of the fitted nitrogen peaks are given in Table 3.12.

Table 3.12: Normalized atomic % of nitrogen species from curve fitting of the N1s peak.

Sample Polymer Attached N1 a N2 b N3 c

3A – 46.2 27.3 26.4

3B PAAEMA 55.2 22.7 22.1

3C PMOEP 30.6 26.9 42.6

3D P(MOEP-b-AAEMA) 41.7 26.7 31.7

3E P(AAEMA-b-MOEP) 38.2 26.8 35.0

3F P(AAEMA-b-MAEP) 40.5 26.2 33.3

3G P(AAEMA-b-MAEP) 37.2 27.2 35.6 a: amine (399.6±0.1eV), b: amide (400.7±0.2eV) and c: protonated amine (401.8±0.2eV)

The APS-treated aminated slide (sample 3A) showed 46.2% amines and 26.4%

protonated amines. The presence of protonated amines on this sample indicate that

some APS amines have reacted with the hydroxyl groups on the glass surface during

silanation possibly by as a result of residual water or proton exchange.95 Polymer

attachment did not alter the amount of the amide peak (in the range of 22.7–27.3 %),

suggesting that no further oxidation occurred as a result of the procedures used to

attach the polymers to the slides. When PAAEMA was attached (sample 3B), there

was an increase in the amine peak (55.2%) and a decrease in the protonated amine

(22.1%) peak. Attachment of the PMOEP homopolymer caused an increase in

protonated amine (42.6%) and a decrease in amine (30.6%) peaks (sample 3C).

Compared to the aminated slide, the block copolymer attachment resulted in a slight

decrease in amine (37.2-41.7%) and an increase in protonated amine peaks (31.7-

35.6%). These values are in fact between those of the PAAEMA and PMOEP-

139


functionalised slides. For both PMOEP and PMAEP copolymers, there is a trend that

those with longer PMOEP/PMAEP segments (samples 3E and 3G) showed larger

protonated amine peaks compared to the shorter ones (samples 3D and 3F). However,

differences between them are only ~3%, and this may not be significant.

3.3.2.2 ToF-SIMS investigation of the conformation of attached block

copolymers

ToF-SIMS was used to assess the conformation of one of the attached block

copolymers on the aminated surfaces as well as control samples: samples 3A

(aminated slide), 3B (PAAEMA-attached), 3C (PMOEP-attached), and 3D

(P(MOEP-b-AAEMA)-attached). Figure 3.18 represents the possible conformations

of the block copolymers through different chemical interactions: (A) coupling

reaction of keto groups of PAAEMA segment and amines, (B) ionic interaction of

PMAEP/PMOEP segment of phosphates and protonated amines, and (C)

combination of both types of interaction.

Figure 3.18: Possible conformations of block copolymers reacted with aminated slide.

Since XPS showed some variations of APS-treatments between the slides, it was

determined to use the same slide for the samples and the untreated one for ToF-SIMS.

Figure 3.19 shows the positive and negative SSIMS spectra of the APS-treated

aminated slide surface (sample 3A). The positive spectrum showed distinct

hydrocarbon peaks such as C+, CH3+, C2H3

+, C3H3+ and C3H5

+. The signal at m/z 28

contains 2 peaks corresponding to Si+ and C2H4+. The spectrum also reveals the

presence of sodium confirming its identification by XPS. The negative spectrum is

O

NH

O

PAAEMA

PMAEP/PMOEP

(A) (B) (C)

O

OP O

O-O

NH3+

140


dominated by signals at m/z = 16 (O-) and at m/z = 17 (OH-). Signals at m/z = 46 and

62 amu indicate oxidized nitrogen species (NO2- and NO3

- respectively). XPS

investigation did not reveal the presence of these species. Hence it appears that these

groups are only present on the outer surface in very low concentration and thus

appear over-represented in the negative mass spectrum.

0 20 40 60 80

Inte

nsity

20050610A.TDC + Ions 100µm 1496964 cts

11 67 53 15 71 69 39 55

57 43 41

28 23

27 29

x0.25

m/z

+ SIMS

0

20050615A.TDC - Ions 100µm 1372525 cts

25

13

16

26

62 46

x10

0 40 60

Inte

nsity

m/z

- SIMS

80 20

35

Figure 3.19: Positive and negative SSIMS spectra for APS-treated amianted glass slide (samples 3A). Note: sodium peak (23 amu) was scaled to 25% of its original intensity; peaks above 25 amu in negative mass spectrum were magnified by 10 folds.

141


30 40 50 60 70 80 90In

tens

ity

20050610A.TDC + Ions 100µm 14969

53 67 716939 575543 27 41 29

28 + SIMS (a)

m/z

20050625A.TDC + Ions 100µm 12791

53 716939 5755

43 27 41

29 28 + SIMS (b)

30 40 50 60 70 80 90

Inte

nsity

m/z

20050640A.TDC + Ions 100µm 13868

53 67 71 69

39 57 55

43 27 41 29

28 + SIMS (c)

30 40 50 60 70 80 90

Inte

nsity

m/z

20050655A.TDC + Ions 100µm 123955

53 67 71 69

39 57 55

43 27 41 29 28

+ SIMS (d)

30 40 50 60 70 80 90

Inte

nsity

m/z

Figure 3.20: Positive SSIMS spectra for: (a) sample 3A (aminated slide), (b) sample 3B (PAAEMA attached), (c) sample 3C (PMOEP attached), and (d) sample 3D (P(MOEP-b-AAEMA) attached).

30 40 50 70 80 90

20050615A.TDC - Ions 100µm 137252

37 42 32 29 46 35 62

26

Inte

nsity

m/z

(a) - SIMS

20050630A.TDC - Ions 100µm 171669

30 40 50 70 80 90

37 41 32

29 46 35 62

26

Inte

nsity

m/z59

(b) - SIMS

20050645A.TDC - Ions 100µm 166643

30 40 50 70 80 90

37 41

32

29

46

35 62

26

Inte

nsity

m/z59

63

79

(c) - SIMS

20050660A.TDC - Ions 100µm 104055

30 40 50 70 80 90

37 41

32

29

46 35

62

26

Inte

nsity

m/z

59

63

79

- SIMS (d)

Figure 3.21: Negative SSIMS spectra for: (a) sample 3A (aminated slide), (b) sample 3B (PAAEMA attached), (c) sample 3C (PMOEP attached), and (d) sample 3D (P(MOEP-b-AAEMA) attached).

142


Figure 3.20 shows the positive SSIMS spectra for samples 3A-3D. Although the

qualitative characteristics show similarities, the intensity patterns differ between the

samples. The block copolymer attached slide (Figure 3.20d) shows substantial

reduction in the yield of the Si+ ion (m/z = 28).

The negative mass SSIMS spectra for samples 3A-3D are shown in Figure 3.21.

When PMOEP and P(MOEP-b-AAEMA) (samples 3C and D, respectively) were

attached, the spectra clearly showed the appearance of peaks at m/z = 63 and m/z =

79, which correspond to the PO2- and PO3

- fragments. This clearly confirms the

presence of phosphorous-containing moieties on the surface.

As described in detail in Section 3.1.4.2, the complex ToF-SIMS data sets were

subjected to both PCA and Analysis of Means (graphical procedure).101,102 Each

sample was characterised by 8 positive and 5 negative spectra: 49 positive and 49

negative peaks were selected for sample characterisation using MVA (Table 3.13).

Table 3.13: Positive and negative fragments used in PCA.

Positive Fragments Negative Fragments

C+, CH+, CH2+, CH3

+, C2H2+, C2H3

+, C2H4

+, C2H5+, C2H6

+, C3H2+, C3H3

+, C3H4

+, C3H5+, C3H6

+, C3H7+, C3H8

+, C4H3

+, C4H4+, C4H5

+, C4H6+, C4H7

+, C4H8

+, C4H9+, C4H10

+, C5H5+, C5H6

+, C5H7

+, C5H8+, C5H9

+, C5H510+, C5H11

+, Al+, Si+, SiH+, CH2Si+, CH3Si+, SiOH+, CHO+, CH3O+, C2H2O+, C2H3O+, C2H4O+, COOH+, C2H5O+, C4H5O+, CH2NO+, CH3NO+,CH4N+, C2H4N+.

C-, CH-, CH2-, CH3

-, C2-, C2H-, C2H3

-, C3-

, C3H-, C3H2-, C3H3

-, C3H4-, C4

-, C4H-, C4H2

-, C4H3-, C5H-, O-, OH-, Si-, SiH-,

CHSi-, SiO-, SiOH-, SiO2-, SiO2H-, SiO3

-, SiO3H-, P-, CP-, PO-, POH-, PO2

-, PO3-,

NO-, NO2-, NO3

-, CN-, CHN-, C2HN-, CNO-, CHNO-, CH3O-, C3O-, C2HO-, C2H3O-, COOH-, C3H3O-, C3H4O-.

Figure 3.22 shows the scores for both the positive and negative ion mass spectra. In

both cases, there are 4 distinct clusters corresponding to the 4 investigated samples.

Each group represents multiple spectra taken from different areas of the sample. The

first two PCs captured 98.89 and 97.24% of the data variance present in the positive

and the negative mass spectra respectively, indicating most of the original spectral

information has been retained. Since samples are different, the differences were most

143


pronounced between the untreated aminated slide (sample 3A) and the block

copolymer attached slide (sample 3D).

+ SIMS (a)

- SIMS (b)

Figure 3.22: Score plots on PC1 and PC2 for aminated glass slide and its modifications. (a) scores derived from the positive fragments; (b) scores derived from the negative fragments.

Block attached PMOEP attached PAAEMA attached Aminated slide

Block attached PMOEP attached PAAEMA attached Aminated slide

The relationship between the original variables and the principal components is

illustrated by the loading plots, which are given in Figure 3.23. The score and the

relevant loading plots should be evaluated together following the principal: peaks

with positive loadings are relatively more intense in spectra with positive scores and

relatively less intense in spectra with negative scores (and vice versa).75 Only the

loadings of the first PCs are given here. Both the positive and the negative

hydrocarbon fragments loaded negatively on the relevant first PCs. This in fact

indicates the opposite physical meaning. The positive hydrocarbon ions are

characteristic of the P(MOEP-b-AAEMA) modification, whilst the negative

144


hydrocarbon lines are associated with the APS treated glass slide. These plots are not

shown here due to the large number of the CxHy+/- ions selected. Considering the

peak intensities and their specificities, the probing of the orientation was performed

using 2 lines: Si+ was selected as the best mark of APS, whilst PO3- was selected as

the best marker for the PMOEP segment.

+ SIMS

APS marker

(a)

- SIMS

PMOEP marker

(b) x 0.1

Figure 3.23: Loadings of selected positive (a) and negative (b) fragments on PC1s.

Figure 3.24 shows the statistically evaluated values of the Si+ and PO3- intensities for

samples 3A-D. Figure 3.24a indicates that the block copolymer was the most

efficient in masking the substrate signal. Figure 3.24b shows that the surface

exposure of the PO3- line was higher for the block copolymer compared to that for

the PMOEP polymer when attached alone. This observation is in good agreement

145


with the quantitative XPS data. These results clearly indicate that the PMOEP

segment at the ASP surface was more exposed when associated with the PAAEMA

block. The combination of data from these two important characterisation techniques

suggest that the attachment of the P(MOEP-b-AAEMA) block copolymer onto the

aminated glass slide resulted in a conformation of the PMOEP moiety extending

from the surface surrounded by the PAAEMA fragments. The PAAEMA blocks in

the attached copolymer are lying along the substrate surface thus exposing the

phosphate groups more to the vacuum (See Figure 3.18B).

(a)

(b)

Figure 3.24: Normalised intensities of Si+ and PO3- for APS-treated aminated glass

slide and its modifications. (a) Si+ intensity reflects the APS coverage with polymers; (b) PO3

- intensity is a sign of the surface density of the PMOEP segment.

146


Figure 3.25 shows the negative mass spectrum of sample 3D (P(MOEP-b-AAEMA)

attached aminated slide) and the corresponding distribution of the PO3- fragment

(peak at 79 m/z) across the examined area (100×100 μm). The image reveals that the

lateral distribution of phosphate groups was uniform across the surface.

10 μm

Total (-)

PO3

-

Figure 3.25: Negative mass spectrum of P(MOEP-b-AAEMA) attached aminated slide (sample 3D) and the lateral distribution of the terminating phosphate groups (PO3

-) across the sample (analysed area → 100x100 μm).

147


3.3.3 Adsorption of Fluorinated Polymers onto PTFE

The adsorption of fluorinated homo and copolymers onto PTFE was carried out by

soaking the PTFE films in 1 mg/mL polymer solution in the solvent of choice

(dichloromethane (DCM), fluorobenzene (FB), methylethyl ketone (MEK) or

dimethylformamide (DMF)) for about 16 hours then washing three times (~10

seconds each) with the same solvent used for soaking and then dried. The films were

analysed by XPS and contact angle measurements. Dynamic light scattering (DLS)

was also used to investigate whether the amphiphilic block copolymers are forming

aggregates in the deposition solvents.

3.3.3.1 Effect of monomer structure and solvents for adsorption

Adsorption of three types of fluorinated homopolymers was investigated: PFS,

PTFPMA and PTFPA onto PTFE using different solvents (DCM, FB and MEK). All

polymers easily dissolved in these solvents. The structure and properties of the

polymers used are shown in Scheme 3.2 and Table 3.14, respectively.

Scheme 3.2: Structures of fluorinated homopolymers.

Table 3.14: Polymer characteristics.

Sample Polymer Mn n PDI No. F a

4B PFS 27417 140 1.06 700

4C PTFPMA 38951 193 1.13 773

4D PTFPA 39959 213 1.06 853 a: Number of fluorines on the polymer chain was calculated from the Mn and structure of repeating units

OO

F

F

FF

H

** *

n

F

FF

F

F* *

n

OO

F

F

FF

H

* n

PFS PTFPMA PTFPA PTFE

*

FF

*F

Fn

148


Figure 3.26 shows the XPS survey scans of untreated (sample 4A), PFS-adsorbed

(sample 4B3), PTFPMA-adsorbed (sample 4C2), and PTFPA-adsorbed (sample 4D2)

PTFE films.


270280290300500520540560

(A) (B) (C) (D)

C1s O1s

F1s

F2s

FKLL

Figure 3.26: XPS Survey spectra of PTFE films (A) untreated, (B) PFS-adsorbed (sample 4B3), (C) PTFPMA-adsorbed (sample 4C1), (D) PTFPA-adsorbed (sample 4D1).

The untreated PTFE (sample 4A) and PFS-adsorbed (sample 4B3) films showed only

F and C, whereas a small amount of O was observed on the PTFPMA- and PTFPA-

adsorbed PTFE films (samples 4C1 and 4D1, respectively). This is in agreement with

the chemical structures. The atomic concentrations of all samples are summarized in

Table 3.15.

The theoretical values of the F/C ratios of PTFE, PFS, PTFPMA and PTFPA are 2,

0.63, 0.57 and 0.67, respectively. Therefore adsorption of these polymers should

decrease the F/C ratios. The untreated PTFE showed an F/C ratio of 2.2 which is

slightly higher than the theoretical.

149


Table 3.15: Atomic % of elements from XPS survey scans and atomic C% of different C elements from the high resolution C1s scans.

Sample Polymer Solvent Atomic % from Survey

XPS

Atomic % from

C1s Narrow Scan

F O C F/C C-F2 C-others

4Aa ― ― 68.8 ― 31.2 2.2 99.3 0.7

4B1 DCM 59.2 ― 49.8 1.5 78.2 21.8

4B2 FB 53.7 ― 46.4 1.2 73.3 26.7

4B3

PFS

MEK 57.3 ― 42.7 1.3 67.9 32.1

4C1 DCM 55.5 1.3 43.2 1.3 85.4 14.6

4C2 FB 59.9 0.8 39.3 1.5 97.5 2.5

4C3

PTFPMA

MEK 62.5 0.0 37.5 1.7 99.2 0.8

4D1 DCM 54.7 1.1 44.3 1.2 87.1 12.9

4D2 FB 81.3 0.7 39.1 1.6 98.3 1.8

4D3

PTFPA

MEK 63.9 0.0 36.0 1.8 99.6 0.4 a: Untreated PTFE

Adsorption of PFS reduced the F/C ratios to 1.2-1.5, depending on the solvent used.

In the case of PTFPMA and PTFPA, the presence of O can also be used to identify

polymer adsorption. Both polymers showed the appearance of O (1.1-1.3%) when

adsorbed in DCM with a concomitant reduction in F/C ratios to 1.2-1.3. When FB

was used, there was also some O (0.7-0.8%) seen as well as a reduction in the F/C

ratios (1.5-1.6). But these changes were smaller than those obtained in DCM. In the

case of MEK, there was no O, and the F/C ratios were high (1.7-1.8).

The high resolution C1s XPS spectra were found to be useful for identifying the

polymer adsorption. Figure 3.27 shows the high resolution C1s XPS spectra of

untreated and PFS adsorbed PTFE films. The peak at 292.5 eV corresponds to the C-

F2 from PTFE. Untreated PTFE showed trace amounts of aliphatic carbon (0.7%) at

285.3 eV, indicating the presence of a small hydrocarbon impurity. The C1s

spectrum of the PFS adsorbed PTFE surface can be curve-fitted using five peak

components with binding energies at about 286.1, 286.7, 286.8, 288.9, and 292.5 eV.

These are attributed to C*-H (peak 1), C*-C6H5 (peak 2), C*-CF (aromatic, peak 3),

C*-F (aromatic, peak 4) and C*-F2 (PTFE, peak 5) species, respectively.103

150



F

FF

F

F

** n

(A)


(B)

1 2 3

4

5

3

4

2 1

**

F F

F Fn 5

5

5

Figure 3.27: C1s narrow scans of PTFE films (A) untreated (sample 4A) and (B) PFS adsorbed (sample 4B3).









PFS

PTFPMA

PTFPA

DCM FB MEK

Sample 4B1 Sample 4B2 Sample 4B3

Sample 4C1 Sample 4C2

Sample 4D1 Sample 4D2

Sample 4C3

Sample 4D3


Figure 3.28: C1s narrow scans of PTFE films after fluorinated homopolymer adsorption using different solvents.

151


Figure 3.28 shows the C1s spectra of all samples studied. The atomic % of C-F2 and

carbons from polymers (C-others) are shown in Table 3.15. PFS adsorption onto

PTFE occurred in all solvents and it was the highest in MEK (32% C-others). The

best adsorptions of PTFPA and PTFPMA were observed when DCM was used as the

solvent. Only a very small adsorption of these polymers was observed in FB, and

none in MEK.

The contact angle measurements of sample 4C1 (PTFPMA adsorbed PTFE in DCM)

was carried out. The advancing and receding angles were found to be 97 ± 3° and 91

± 4°, respectively that were slightly lower than those of PTFE (~110°) (shown in

Table 3.19).

3.3.3.2. Adsorption of PFS with varying Mn’s

The effect of PFS Mn on adsorption onto PTFE was examined using three different

molecular weights (Table 3.16). MEK was the adsorption solvent of choice, since it

showed the best PFS adsorption. The atomic % of C-F2 and C-others from the high

resolution C1s peaks are shown in Table 3.16. The results are the average of four

samples.

Table 3.16: Properties of PFS and atomic % from C 1s scans of PFS adsorbed PTFE films.

Sample Adsorbed PFS Atomic % from C 1s scans

Mn n PDI no. F C-F2 STD C-others STD

4D 6514 32 1.09 161 88.4 2.2 11.6 2.2

4E 12093 61 1.05 305 83.9 2.4 16.1 2.4

4B 27417 140 1.06 700 67.9 3.1 32.1 3.1

Figure 3.29 shows the plot of units of PFS against atomic % of C-others. The

adsorption of PFS onto PTFE was proportional to the molecular weight in the range

investigated.

152


0

5

10

15

20

25

30

35

40

0 20 40 60 80 100 120 140 160

Unit of PFS C

oth

ers

XP

S

Figure 3.29: Relationship between molecular weight of PFS and adsorbed amounts of PFS onto PTFE films.

3.3.3.3. Adsorption of P(FS-b-tBA) and P(FS-b-AA) block copolymers onto

PTFE

Table 3.17 shows the properties of the P(FS-b-tBA) and P(FS-b-AA) block

copolymers used in this study. Two different PtBA block segments were used:

sample 4F has a 1.7 times longer PtBA segment than that of sample 4G. Samples 4H

and 4I were obtained from the hydrolysis of samples 4F and 4G, respectively. All

copolymers were synthesised from the same PFS macro-RAFT (i.e. the same Mn and

PDI for the first block).

Table 3.17: Properties of PFS block copolymers.

Polymer PFS

1st block

PtBA or PAA

2nd block

Mn N no. F PDI Mn n

PDI of

full block

4F P(FS-b-tBA) 19869 101 505 1.04 30377 237 1.11

4G P(FS-b-tBA) 19869 101 505 1.04 18099 141 1.07

4H P(FS-b-AA) 19869 101 505 1.04 17078 1 237

4I P(FS-b-AA) 19869 101 505 1.04 10160 1 141 1: calculated by assuming 100% cleavage of tBA side-chains

The attachment of P(FS-b-tBA) and PtBA was carried out in MEK. Since sample 4H

(P(FS101-b-AA237)) with a long PAA chain was only soluble in DMF, the attachment

153


was carried out in this solvent. Sample 4I was soluble in DMF and MEK hence both

solvents were examined. Figure 3.30 shows the high resolution C1s XPS spectra of

these samples and the atomic % of C-F2 (from PTFE) and C-others (from polymer)

are shown in Table 3.18.






(A) (B)

(C) (D)

(E)

1 2 3 4 5

6

7

1 2 3 4 5

6

7

Figure 3.30: C1s narrow scans of block copolymer adsorbed PTFE films (A) P(FS101-b-tBA237) (sample 4F), (B) P(FS101-b-tBA141) (sample 4G), (C) P(FS101-b-AA237) (sample 4H), (D) P(FS101-b-AA141) (sample 4I1) and (E) P(FS101-b-AA141) (sample 4I2).

154


Table 3.18: Atomic % of C-F2 and C-others.

Sample Polymer Solvent C-F2 C-others

4F P(FS101-b-tBA237) MEK 78.7 21.3

4G P(FS101-b-tBA141) MEK 78.4 21.6

4H P(FS101-b-AA237) DMF 44.1 55.9

4I1 P(FS101-b-AA141) DMF 49.2 50.8

4I2 P(FS101-b-AA141) MEK 63.3 36.7

The C1s high-resolution spectra of both P(FS-b-tBA) and P(FS-b-AA) adsorbed to

the PTFE surface can be curve-fitted into seven peak components. Peaks 1-5 with

binding energies at about 285.0 285.6, 286.1, 286.6, and 286.8 eV are attributed to

C*-H (PtBA or PAA, peak 1), C*-C(=O)O (PtBA or PAA, peak 2), C*-H (PFS, peak

3), C*-C6H5 (PFS, peak 4), and C*-CF (aromatic of PFS, peak 5) species. Peak 6 at

288.8 eV is attributed to both C*-F (aromatic of PFS) and C(=O)OH/C(=O)O- (from

PtBA/PAA) species. Peak 7 at 292.5 eV is from C*-F2 (PTFE).

Both P(FS-b-tBA) block copolymers (samples 4F and 4G) had similar by adsorbed

amounts (C-others: ~21%) regardless of the PtBA chain length. P(FS-b-AA) block

copolymers (samples 4H, 4I1, and 4I2) showed a higher adsorption onto PTFE

compared to those of the P(FS-b-tBA) block copolymers. Samples 4H and 4I1 (DMF)

showed similar adsorption amounts (C-others: ~50%). Sample 4I2 (MEK) showed

less adsorption (C-others: 37%) compared to sample 4I1 (DMF).

The effect of block copolymer adsorption on the surface wettability of the modified

PTFE substrates was also characterized using water contact angle measurements.

Figure 3.31 shows photographic images of advancing and receding water droplets

(both 5 μL) on untreated PTFE, P(FS101-b-tBA141)-adsorbed (sample 4G), and

P(FS101-b-AA141)-adsorbed (sample 4I1) films. The advancing and receding angles

are shown in Table 3.19 that also contains literature values of the advancing angles

of PFS, PtBA, and PAA.

155


Advancing Receding

(A)

(F) (E)

(C) (D)

(B)

Figure 3.31: Water droplet (5 μL) profiles on the surfaces of untreated PTFE (sample 4A) (A) advancing and (B) receding, P(FS101-b-tBA141) adsorbed PTFE (sample 4G) (C) advancing and (D) receding, and P(FS101-b-AA141) adsorbed PTFE (sample 4I1) (E) advancing and (F) receding.

156


Table 3.19: Advancing and receding water contact angles of polymer-adsorbed PTFE films.

Sample Adsorbed

Polymers

Deposition

Solvent

Advancing

Angle (°)

Receding

Angle (°)

Hysteresis

(°)

4A a ― ― 109 ± 6 111 ± 6

4C1 PTFPMA DCM 97 ± 3 91 ± 4 6 ± 5

4F P(FS101-b-tBA237) MEK 88 ± 2 86 ± 5 2 ± 5

4G P(FS101-b-tBA141) MEK 98 ± 3 90 ± 6 8 ± 7

4H P(FS101-b-AA237) DMF 91 ± 6 31 ± 3 60 ± 7

4I1 P(FS101-b-AA141) DMF 88 ± 5 28 ± 3 60 ± 6

4I2 P(FS101-b-AA141) MEK 101 ± 4 49 ± 3 52 ± 5

PFSb 99104,105

PtBAc 88 ± 3106

PAAd 1554, 48107 a: Untreated PTFE film, b: PFS homopolymer (Ref: 104) and PFS grafted surface (layer thickness = 89 nm) (Ref: 105), c: PtBA homopolymer, d: spin coated PAA thin film (Ref: 54) and PAA grafted polyethylene (Ref: 107)

The advancing contact angle of water on a PTFE surface (sample 4A) was

determined to be 109 ± 6°. No hysteresis was observed, indicating a homogeneous

and stable surface. When P(FS101-b-tBA237) with the long tBA segment was adsorbed

(sample 4F), the advancing angle decreased to 88°. This is the same as the literature

value for the PtBA contact angle. For the P(FS101-b-tBA141) adsorbed surface

(sample 4G), the advancing angle was 98° which is the same as that of PFS (99°).

However, it showed a slightly larger although still small hysteresis (8°).

Both P(PFS-b-AA) polymers adsorbed to PTFE surfaces using DMF as the

deposition solvent (samples 4H and 4I1) showed advancing angles of ~90°. Receding

angles were 31° and 28°, respectively, indicating reorganization of the block

copolymer chains during the contact angle measurment. The difference in chain

length of the PAA segments did not show any significant effect on the contact angles.

Sample 4I2 (MEK) showed advancing and receding angles of 101 ± 4° and 49 ± 3°,

respectively: these are higher than those of sample 4I1(DMF).

157


To further investigate the chain reorganisation phenomenon, modified films were

soaked in MilliQ water at room temperature for 2 days and blot dried before contact

angle measurements (Table 3.20). Samples 4H and 4I1 (DMF) showed ~17° lower

advancing angles (73 ± 3°) compared to the dry samples. Receding angles also went

down slightly to 21 ± 2°. Again, regardless of the AA segment length, the results

were the same. The advancing angle of sample 4I2 (MEK) was 92 ± 4°. This was a 9°

decrease compared to the dry sample. However, it is still highly hydrophobic. The

receding angle was 50 ± 6°, which did not change from that of the dry sample.

Table 3.20: Advancing and receding contact angles of P(FS-b-AA) adsorbed surfaces after soaking in MilliQ water for 2 days.

Sample Adsorbed

Polymer

Deposition

Solvent

Advancing

Angle (°)

Receding

Angle (°)

Hysterisis

(°)

4H P(FS101-b-AA237) DMF 73 ± 3 21 ± 2 52 ± 4

4I1 P(FS101-b-AA141) DMF 73 ± 4 21 ± 2 52 ± 4

4I2 P(FS101-b-AA141) MEK 92 ± 4 50 ± 6 42 ± 7

The characterisation of block copolymers formed in DMF and MEK were performed

using dynamic light scattering (DLS). The number-average hydrodynamic diameter

(DH) is shown in Table 3.21. In DMF, sample 4H with the long PAA segment

showed a larger DH (30 nm) compared to sample 4I1 (24 nm). Sample 4I2 (MEK)

showed a slightly smaller DH (21 nm) compared to the DMF sample.

Table 3.21: DH of P(FS-b-AA) in DMF and MEK.a

Sample Polymer Solvent DH (nm)

4H P(FS101-b-AA237) DMF 30

4I1 P(FS101-b-AA141) DMF 24

4I2 P(FS101-b-AA141) MEK 21 a: All samples were run 10 times and the STD = ± 1 nm

158


3.4 Discussion

3.4.1 Layer-by-Layer (LbL) Assembly

LbL assembly is a well-recognised and robust surface modification technique. There

is no limitation in the substrate shapes and sizes, and indeed it has found a wide

range of applications including in biomaterials. In this study, carboxylate- and

phosphate-containing surfaces were prepared using PEI as the polycation and PAA,

PMAEP or PMOEP as the polyanion.

3.4.1.1 LbL assembly of PAA

The LbL assemblies of PEI and PAA were produced under three different pH

conditions: pH 5.5, 7 and water without adjustment. As expected for a weak

polyanion, the pH of the solution strongly affected the deposition of the PAA film by

altering the degree of ionisation of the PAA.

At low pH (unadjusted water, pH 3.3) where PAA is less than half ionised, the

polymer chains are in a coiled conformation which contains a high numbers of loops.

Hence more chains are required to form layers on the surface and thus results in a

thick film. Choi and Rubner suggested that when the degree of ionization falls below

a critical value, dramatic increases in bilayer thickness occur, driven by the less

densely charged polymer being in contact with the high charge density polymer of

opposite charge.108 Moreover, at low pH, the unionized carboxylic acid groups can

form strong hydrogen bonding between the PAA chains and that makes for an even

thicker layer (shown in Figure 3.32A).

At pH 7, the degree of ionisation is 100 %. The chains can be more elongated (and

have fewer loops) and hence they interact strongly with the PEI film resulting in

thinner PAA films (Figure 3.32B). There should be no hydrogen bond formation at

this pH.

159


+ +

+ + + +

+ +

+ + +

+ +

+ +

+ +

+ +

+ +

-

-

-

-

- -

- -

- - -

- -

- - -

- -

- -

- -

- - COOH COO- COOH COO-

COOH

-

- -

- -

COOH

-

- -

- -

COOH

+ +

+ + + +

+ +

+ + +

+ +

+ +

+ +

+ +

+ - - -

- -

- -

-

-

- - - - -

- - - -

- - -

- - -

- -

-

A B

pH 7 pH 3.3

- -

-

-

- -

Figure 3.32: Schematic representation of PAA deposition onto PEI deposited surface at different pH.

Hydrogen bonding plays an important role in LbL formation in many systems. There

are cases that use pure hydrogen bonding instead of electrostatic interactions to form

polyelectrolytes layers.109-112

Kim et al.19 showed that the thickest dendrimer/PAA film was obtained by the

deposition of dendrimers at pH 8 and PAA at pH 4. This study showed similar results

in that the thickest film was from PEI deposition at pH 9.9 and PAA at pH 3.3.

3.4.1.2 LbL assembly of PMAEP and PMOEP

The XPS investigation of the LbL assemblies of both PEI-PMAEP and PEI-PMOEP

prepared in unadjusted water showed a large increase in atomic C%, indicating thick

film formation. The pHs of the PMAEP and PMOEP solutions were 2.7 and 2.8,

respectively. Since the pKa values of dihydrogen phosphates are 1.7 and 7.01 (values

for hydroxymethyanephosphonic acid113), at least one proton on the phosphate

groups of PMAEP and PMOEP are deprotonated. The ionic interaction of the

protonated amines and the phosphates was evident from the XPS high resolution N1s

spectra, as well as from the IRRAS data.

AFM images revealed a smooth topography for the PMOEP LbL film surface,

whereas PMAEP layer was patchy, showing aggregates of polymers. The PMAEP

chains contain a high degree of carboxylic acid groups, most likely randomly

distributed. The degree of ionization of PAA at pH 2.5 was measured to be 20-30

160


%.20 In comparison to PMOEP, the PMAEP chains would be less ionised. Therefore

the conformation of the polymer chains should be more coiled and hydrogen bonding

between the chains would be also possible. This explains the topography observed by

AFM.

Since the aim of this experiment was only to create a carboxylate- or phosphate-

containing surface, only one layer of each-polyanion and polycation-was deposited.

However, in the literature it has been shown that a consistent surface charge

distribution (i.e. uniform polymer layer coverage) does not fully develop until after

the adsorption of a few bilayers.19,114,115 Therefore, it is not impossible that the

surfaces created here may also show PEI properties as well as some Si-OH.

Moreover, the PEI-PMAEP LbL film would contain large amounts of free-carboxylic

acid groups (see Figure 3.33). Nevertheless, the XPS investigation revealed that

these LbL systems produced surfaces that contain high amounts of carboxylate or

phosphate groups.

Figure 3.33: Schematic representation of PEI-PAA LbL showing free-functional groups.

In summary, the LbL assembly approach successfully immobilised carboxylate- and

phosphate-containing polymers on the substrate surface. Since strong ionic

interaction of polyelectrolytes is stable in solution, the mineralisation study of these

water-soluble polymers is made possible in SBF by this technique. Although these

LbL surfaces may show small amounts of hydroxyl and amine groups from either the

substrate or PEI, respectively, it is expected to have dominant carboxylate and

phosphate groups.

OH OH NH2NH2NH2 NH2

PO4H2

COOH COOH HOOC COOH PO4H2PO4H2

PO4H2

161


3.4.2 Block Copolymer Attachment onto Aminated Slides

3.4.2.1 Qualitative analysis of attached polymers

The XPS investigation revealed successful immobilization of the polymers onto the

aminated slides. From the survey scans, the increase in C% and the concomitant

decrease in N% were evident when polymers were attached.

The attached amounts of PAAEMA and PMOEP homopolymers were similar, based

on the increase in C% (samples 3B and 3C, respectively). The high resolution spectra

of the N1s peak of the PAAEMA attached slide showed an increase in the amine

peak, indicating PAAEMA was attached through reductive amination of the keto

group. The PMOEP attached slide showed an increase in the protonated amine peak,

which indicated PMOEP attachment was through ionic interactions. Although the

attachment reaction was carried out in the aprotic solvent DMF and hence is not

expected to cause deprotonation of the phosphate groups of the PMOEP moieties, it

is possible that proton transfer between the phosphate groups of PMOEP and the

amine groups of APS can occur.

From the atomic C% from the survey scans, the degree of functionalisation was

higher for all block copolymers compared to homopolymers, possibly due to the

much higher Mn’s of the block copolymers. Since both the PAAEMA and PMOEP

homopolymers are capable of attaching to the aminated slides, in the case of the

block copolymers, it was envisaged that a competition between reductive amination

(leading to attachment of the PAAEMA block) and proton transfer (leading to

attachment of the PMOEP block), might exist.

One interesting finding is the fact that the amounts of attached PMOEP copolymers

(samples 4D and 4E) were higher than those for the PMAEP copolymers (samples 4F

and 4G). Since PMAEP contains high amounts of carboxylic acid groups, the

differences observed between the PMOEP and PMAEP copolymer behaviour may

indicate better proton transfer of phosphates to amines compared to carboxylic acid

groups.

The ionic interaction of these block copolymers was also evident from the high

resolution N1s spectra. The atomic % of amines and protonated amines of block

copolymer-functionalised slides were somewhere between those of the PAAEMA

162


and PMOEP-functionalised surfaces. This indicates that both reactions are occurring

for these samples.

3.4.2.2 Conformation of attached block copolymers

In this study where the block copolymers contain phosphate groups, ToF-SIMS is the

“gold-standard” surface characterization technique because these groups have been

found to be ideal markers for the conformational analysis of phospholipids.77 ToF-

SIMS provides a higher surface specificity (1~2 nm) and sensitivity (107 – 1011

atoms/cm2) than XPS. Together with PCA, it makes a powerful tool for the

elucidation of both the conformation and orientation of adsorbed molecules such as

proteins and lipids on the surface.74,75,79,116

XPS revealed that both PAAEMA and PMOEP/PMAEP segments are capable of

interacting with amines through reductive amination and proton transfer, respectively.

Therefore, the conformation of the adsorbed block copolymers depends ultimately on

the relative reactivity of these two functional groups. The three types of

conformational models are shown again in Figure 3.34:

Figure 3.34: Possible conformations of block copolymers reacted with aminated slide.

where block copolymers were attached through (A) covalent interaction of

PAAEMA block, (B) ionic interaction of PMOEP block and (C) combined

interactions.

The positive SSIMS spectra of these samples showed differences in intensity of the

Si+ ion (m/z = 28). The block copolymers showed substantial reduction in this peak,

O

NH

O

PAAEMA

PMAEP/PMOEP

(A) (B) (C)

O

OP O

O-O

NH3+

163


which indicates higher polymer coverage. The negative SSIMS spectra of PMOEP

and block copolymer functionalised surfaces clearly showed evolution of the PO2-

(m/z = 63) and PO3- (m/z = 79) peaks.

Both score plots for positive and negative ion mass spectra showed distinctive

differences between the samples investigated and the most pronounced differences

are found between the untreated and block copolymer-attached aminated slides.

From the loadings plots, Si+ and PO3- are selected as the best markers for APS (i.e.

aminated slide) and the PMOEP segment, respectively.

From the statistically evaluated values of the Si+ intensities, the block copolymer was

found to be the most efficient in attenuating the substrate signal. The PO3- intensity

for the block copolymer was higher than that for PMOEP homopolymer attachment

alone. This observation is in agreement with the quantitative XPS data (1.3 and 1.5

atom % of P for PMOEP and block copolymer immobilized slides, respectively).

These data strongly support the view that the PMOEP segment on the aminated slide

more exposed when associated with the PAAEMA block. When the conclusions

from the data generated by the different techniques are combined, together they

suggest that the immobilization of the P(MOEP-b-AAEMA) block copolymer onto

the aminated slide resulted in a conformation where the PMOEP moieties are coiled

away from the surface and surrounded by the PAAEMA fragments situated much

closed to the surface (Figure 3.34A).

The surfaces created from this technique provide phosphate-containing polymer

chains and hence are suitable for mineralisation study of soluble PMAEP and

PMOEP. It is in contrast to the LbL films that employ deprotonated phosphate

groups for film formation by ionic interactions.

164


3.4.3 Adsorption of Fluorinated Polymers onto PTFE

The adsorption of fluorinated homo and block copolymers onto PTFE was carried

out by soaking the PTFE in the relevant polymer solution. This technique has the

great advantage that it does not require any special instrument, is robust, and can be

applied to any shapes such as microporous expanded PTFE (ePTFE). This is highly

relavant in the event of its finding a commercial viable application.

3.4.3.1 Homopolymer adsorption

PTFE is well known for its low surface tension (of 18.6 dynes/cm117), which

provides its non-stick property. In this study, three types of fluorinated

homopolymers were tested for adsorption onto PTFE in different solvents (DCM, FB

and MEK). PFS showed the greatest adsorption onto PTFE in all solvents. From XPS,

the highest PFS adsorption was found when MEK was used as the solvent. For

PTFPMA and PTFPA, some adsorption was observed only when DCM was used.

Since these polymers had higher amounts of fluorine atoms in each polymer chain

(773 and 853, respectively) compared to PFS (700), this would appear to indicate

that the structure of the repeating unit affects the adsorption.

There are two factors that need to be considered regarding the solvent: swelling of

PTFE and solubility of the polymer (which affects both adsorption and desorption).

Since the same solvent was used for adsorption and washing, a good solvent for

polymer enhances the desorption during washing. A good swelling solvent for PTFE

allows the fluropolymers to entangle and anchor well. However, it is also possible

that this solvent will remove adsorbed polymers in the washing step.

Nasef investigated the swelling of PTFE film in different solvents.118 The sorbed

liquid was found to be 0.35 wt% for DCM, 0.15% for methanol, and 0.23% for

benzene. Thus DCM was the best solvent. From this trend, it can be predicted that

PTFE swelling is higher in DCM than in MEK. It is also possible that FB is a better

swelling agent than benzene.

In this study, the adsorption of PTFPMA and PTFPA was only observed in DCM.

Although both DCM and fluorobenzene swell PTFE, the results may indicate that

interaction between the fluorinated solvent (i.e. FB) and the polymers (PTFPMA and

165


PTFPA) are so strong, they prevent adsorption onto the PTFE. There was no

adsorption of these polymers when MEK was used as the solvent, this is in

agreement with the prediction that MEK is a low-swelling solvent for PTFE.

The contact angle of PTFPMA-adsorbed PTFE was higher than that of PtBA. If only

the fluorine atoms of the PTFPMA were adsorbed into the PTFE chains (Figure

3.35A), the contact angle of this surface would be expected to be close to that of

PtBA (88°), assuming uniform polymer coverage since XPS showed that the total

atomic C% from the adsorbed polymer is over 10%. The fact that the contact angle

obtained was higher (θadv = 97°), could indicate that the whole PTFPMA chain is in

fact entangled (Figure 3.35B).

Figure 3.35: Schematic representation of fluoropolymer adsorption onto PTFE by (A) fluorine adsorption and (B) chain entanglement.

In the case of PFS, the highest adsorption was observed when MEK was used.

Jankova et al. investigated the solubility of linear PFS synthesised by ATRP and

found that FB, MEK, and DCM dissolve 0.417, 0.143, and 0.009g of PFS per gram

at room temperature, respectively.119 This supports the suggestion that in FB, PFS is

well solvated and in a stretched conformation, whereas in DCM, the chains are more

coiled. In MEK, although the chains are possibly only slightly coiled, the swelling of

the PTFE film is less, therefore more chains are packed on the surface, compared to

those adsorbed from DCM (Figure 3.36).

F F F F F F F F F F F F F F

F F F

F

F

F F

F F

F

F F

F F F F

(A) (B)

PTFE

166


Figure 3.36: Possible adsorption behaviour of PFS onto PTFE from different solvents.

The effect of PFS chain length on adsorption onto PTFE was investigated using three

different molecular weights of polymers. The chain entanglement Ne of PTFE is

~110.67 The polymers below this length which were studied showed much less

adsorption compared to the highest one. However, more measurements using

polymers with molecular weights closer to the Ne are required to confirm if the

adsorption is in fact by entanglement. Since the MEK used for this study is a low-

swelling solvent for PTFE, the entanglements may be minimal. The solubility of

polymers in MEK may also play a role: the higher the molecular weight, the less

soluble, hence the higher the adsorption.

The most important interactions for fluoropolymer adsorptions are hydrophobic and

fluorine-fluorine interactions. PFS was the most hydrophobic polymer studied and it

showed the best adsorption. It is also possible that in PFS, fluorine atoms are easier

to polarise due to the presence of the aromatic ring when in close contact with the

PTFE. Therefore, the fluorine-fluorine interaction between PFS and PTFE becomes

stronger than for the other polymers. Since PTFPMA and PTFPA contain carbonyl

groups (i.e. they are more hydrophilic), some repulsion between these polymers and

highly hydrophobic PTFE may occur. This results in competition with the fluorine-

fluorine interactions.

3.4.3.2. Block copolymer adsorption onto PTFE

XPS analysis showed that the amounts of adsorbed P(FS-b-tBA) from MEK with

different PtBA chain lengths were similar. However, the contact angle measurements

of these samples were different. The advancing angle of the copolymer-adsorbed

DCM FB MEK

(A) (B) (C)

PTFE

Swollen Region

167


PTFE with the longer PtBA segment (n = 237, sample 4F) was close to that of PtBA,

whereas the shorter PtBA (n = 141, sample 4G) was closer to that of PFS. The

homopolymer of PtBA did not adsorb to PTFE (data not shown). The experimental

results indicate that PFS segments adsorb onto the PTFE and PtBA segments are

extending from the surface. The block copolymer with the long PtBA segments

covers the surface well, whereas the short tBA segments do not fully cover the

surface, leaving some PFS regions. Since both PFS and PtBA are soluble in MEK,

there should be no polymer-aggregates in this solvent.

A higher adsorption of P(FS-b-AA) copolymers was observed compared to that of

P(FS-b-tBA), which suggests possible aggregation of P(FS-b-AA). As mentioned in

the Introduction (Section 3.1), micelles are known to adsorb readily onto surfaces.

Whittaker et al.120 investigated the micellisation of P(S-b-AA) in DMF, which is a

good solvent for both chains, and toluene, which is only good for PAA. In their study,

the number-average hydrodynamic diameters (DH) of P(S153-b-AA175) and P(S153-b-

AA234) in DMF were 6.2 and 6.3 nm, respectively, whereas in toluene, they were

42.5 and 44.5 nm. In this study, DLS analyses showed that the DH of the block

copolymers in DMF and MEK were between 21-30 nm, indicating formation of

some kind of aggregates. DMF is a good solvent only for the PAA segments,

whereas MEK is good only for the PFS segments. Therefore, it could be predicted

that inverse-structures of aggregates in DMF and MEK occur with the outer layer

PAA segments in DMF, and PFS segments in MEK.

Both P(FS-b-AA)-attached films prepared in DMF showed advancing angles of~ 90°.

This suggests that the aggregates with PAA segments as the outer layer possibly

reorganized when adsorbed and dried onto PTFE exposing the PFS segments on the

surface in order to reduce surface tension (shown in Figure 3.37). Another possibility

is that when these aggregates were adsorbed, they may have lost their aggregate

structures.

168


DMF

MEK

Drying

Drying

PAA

PFS

Figure 3.37: Aggregate adsorption onto PTFE surfaces in different solvents.

Receding angles of these surfaces were ~30°, indicating some reorganization of the

polymer chains during the contact angle measurement. After soaking these samples

in water, the advancing angles reduced to 73° (~17° decrease). This also suggests the

existence of flexible polymer chains on the PTFE surfaces.

P(FS101-b-AA141) adsorbed onto PTFE in MEK showed the highest advancing angle

(101°). In MEK, aggregates with the PFS segments in the outer layer possibly

adsorbed onto PTFE without structural change. The receding angle was 50°, and this

did not change after soaking in water, which indicates that full rearrangement occurs

during the contact angle measurement.

Hydrophilicity/hydrophobicity of the PFS containing amphiphilic block copolymers

adsorbed PTFE is found to be reversibly tuneable depending on the environment (e.g.

water vs air). Since the carboxylic acid groups on these chains are free, it is suited for

mineralisation study of these groups.

169


3.5 Conclusions

Three techniques for the fabrication of modified surfaces using well-defined

functional polymers were presented in this study:

• Layer-by-Layer (LbL) assembly of soluble phosphate- and carboxylate-

containing homopolymers

• Coupling reactions of keto-containing block copolymers onto aminated slides

• Adsorption of fluorinated homo and block copolymers containing carboxylic

acid groups onto PTFE

LbL assembly was monitored by XPS. The thickness of PAA film was found to be

strongly dependent on the pH during deposition. This is in agreement with the

literature reports for weak polyanions. Phosphate-containing LbL films were

successfully prepared using soluble PMAEP and PMOEP with PEI. AFM

investigations showed the formation of a patchy PMAEP film, whereas the PMOEP

film was smooth. Since PMAEP contains large amounts of carboxylic acid groups,

these groups are not deprotonated at a low deposition pH and hence this affected the

film morphology.

Block copolymers consisting of PAAEMA and either PMAEP or PMOEP were

attached onto aminated glass slides, by the coupling reaction of the keto and amine

groups to form imines which were then reduced to form stable secondary amines.

Thus free-phosphate polymers immobilised on a surface was successfully

accomplished. XPS investigations revealed successful attachments of block

copolymers, as well as both PAAEMA and PMOEP homopolymers. Since the

PMOEP homopolymer was found to be capable of attaching onto aminated slides

through electrostatic interactions, it is concluded that the conformation of the

attached block copolymer depends on the relative reactivity of the two functional

groups. ToF-SIMS investigations of the P(MOEP-b-AAEMA)-attached aminated

slide showed that the PMOEP segment was more extended to the vacuum and

PAAEMA segment more attached to the aminated slide surface.

Adsorption of three types of fluorinated homopolymers (PFS, PTFPMA, and PTFPA)

was investigated and it was found that PFS showed the best adsorption behaviour

170


onto PTFE. The attractive interactions governing the adsorption are thought to be

hydrophobic and fluorine-fluorine interactions. Because PFS is the most hydrophobic,

and the fluorine atoms in the aromatic rings are possibly easier to polarise, this

creates stronger fluorine-fluorine interactions. Both PTFPMA and PTFPA contain

ester groups which may create some repulsive force with the PTFE. These polymers

were only found to adsorb in DCM, possibly by entanglement. P(FS-b-AA) block

copolymers were also successfully adsorbed onto PTFE. Contact angle

measurements showed large hysteresis indicating a fast reorganisation of the

adsorbed block copolymer chains.

These techniques successfully created surfaces consisting of carboxylate and

phosphate groups. The resulting surfaces are suitable for the in vitro assessments

such as for mineralisation. The surface modification approach using well-defined

polymers by these techniques have shown to be robust and hence will find wider

applications.

171


3.6 References (1) Decher, G., and Hong, J. D. Makromol Chem Macromol Symp 1991, 46, 321-

327. (2) Decher, G., Hong, J. D. and Schmidt, J. Thin Solid Films 1992, 210, 831-835. (3) Decher, G., Lvov, Y. and Schmitt, J. Thin Solid Films 1994, 244, 772-777. (4) Decher, G. Science 1997, 277, 1232-1237. (5) Caruso, F., Lichtenfeld, H., Giersig, M., and Möhwald, H. J Am Chem Soc

1998, 120, 8523-8524. (6) Gong, Y., Zhu, Y., Liu, Y., Ma, Z., Gao, C., and Shen, J. Acta Biomaterialia

2007, Available on line July 2007. (7) Kotov, N. A., Dekany, I., and Fendler, J.H. J Phys Chem 1995, 99, 13065-

13069. (8) Caruso, F., Lichtenfeld, H., Giersig, M., and Mohwald, H. J Am Chem Soc

1998, 120, 8523-8524. (9) Lvov, Y., Decher, G., and Sukhorukov, G.B. Macromolecules 1993, 26,

5396-5399. (10) Sukhorukov, G. B., Montrel, M.M., Petrov, A.I., Shabarchina, L.I., and

Sukhorukov, B.I. Biosensors and Bioelectronics 1996, 9, 913-922. (11) Pei, R., Cui, X., Yang, X., and Wang, E. Biomacromolecules 2001, 2, 463-

468. (12) Michel, M., Arntz, Y., Fleith, G., Toquant, J., Haikel, Y., Voegel, J-C.,

Schaaf, P., and Ball, V. Langmuir 2006, 22, 2358-2364. (13) Lvov, Y., Ariga, K., Ichinose, I. and Kunitake, T. J Am Chem Soc 1995, 117,

6117-6123. (14) Derbal, L., Lesot, H., Voegel, J.C. and Ball, V. Biomacromolecules 2003, 4,

1255-1263. (15) He, P., and Hu, N. J Phys Chem B 2004, 108, 13144-13152. (16) Salloum, D. S., and Schlenoff, J.B. Biomacromolecules 2004, 5, 1089-1096. (17) Rojas, O. J., Ernstsson, M., Neuman, R.D., and Claesson, P.M. Langmuir

2002, 18, 1604-1612. (18) Claesson, P. M., Poptoshev, E., Blomberg, E., and Definaite, A. Adv Colloid

Interf Sci 2005, 114-115, 173-187. (19) Kim, B. Y., and Bruening, M.L. Langmuir 2003, 19, 94-99. (20) Shiratori, S. S., and Rubner, M.F. Macromolecules 2000, 33, 4213-4219. (21) Mendelsohn, J. D., Barrett, C.J., Chan, V.V., Pal, A.J., Mayes, A.M., and

Rubner, M.F. Langmuir 2000, 16, 5017-5023. (22) Morgan, S. E., Jones, P., Lamont, A.S., Heidenreich, A., and McCormick,

C.L. Langmuir 2007, 23, 230-240. (23) DeLongchamp, D. M., and Hammond, P.A. Langmuir 2004, 20, 5403-5411. (24) Kharlampieva, E., Kozlovskaya, V., Tyutina, J., and; Sukhishvili, S. A.

Macromolecules 2005, 38, 10523-10531. (25) Sukhishvili, S. A., and Granick, S. Macromolecules 2002, 35, 301-310. (26) Caruso, F., Caruso, R.A., and Möhwald, H. Chem Mater 1999, 11, 3309-3314. (27) Such, G. K., Tjipto, E., Postma, A., Johnston, A.P.R., and Caruso, F. Nano

Letters 2007, 7, 1706-1710. (28) Johnston, A. P. R., Read, E.S., and Caruso, F. Nano Letters 2005, 5, 953-956. (29) Li, Q., Quinn, J.F., Wang, Y., and Caruso, F. Chem Mater 2006, 18, 5480-

5485.

172


(30) Michel, M., Vautier, D., Voegel, J.C., Schaaf, P., and Decher, G. Langmuir 2004, 20, 4835-4839.

(31) Zhang, X., Wu, T., Sun, J. and Shen, J. Colloids and Surface A: Physicochemical and engineering aspects 2002, 198-200, 439-442.

(32) Harris, J. J., DeRose, P.M. and Bruening, M.L. J Am Chem Soc 1999, 121, 1978-1979.

(33) Lawrie, G., Grøndahl, L., Battersby, B., Keen, I., Lorentzen, M., Surawski, P., and Trau, M. Langmuir 2006, 22, 497-505.

(34) Takemoto, K., Ottenbrite, R.M., and Kamachi, M., Ed. Functional monomers and polymers; Marcel Dekker Inc: NY, 1997; Vol. Second Edition.

(35) Benetti, S., and Romagnoli, R. Chem Rev 1995, 95, 1065-1114. (36) Mastrorilli, P., Nobile, F.C., Suranna, G.P., Rosaria, M., and Latronico, M.

Inorganica Chimica Acta 2002, 335, 107-112. (37) Powell, K. R., McCleskey, T.M., Tumas, W., and DeSimone, J.M. Ind. Eng.

Chem. Res. 2001, 40, 1301-1305. (38) Krasia, T., Soula, R., Borner, H.G., and Schlaad, H. Chem Commun 2003,

538-539. (39) Schlaad, H., Krasia, T., and Patrickios, C.S. Macromolecules 2001, 34, 7585-

7588. (40) Moszner, N., Salz, U., and Rheinberger, V. Polymer Bulletin 1994, 32, 419-

426. (41) Park, J. P., Monteiro, M.J., van Es, S., and German, A.L. Euro Polym J 2001,

37, 965-973. (42) Yu, Z., Alesso, S., Pears, D., Worthington, P.A., Luke, R.W.A., and Bradley,

M. Tetrahedron Letters 2000, 41, 8963-8967. (43) Boyko, V., Lu, Y., Richter, A., and Pich, A. Macromol Chem Phys 2003, 204,

2031-2039. (44) Kim, Y.-B., Kim, H.K., Nishida, H., and Endo, T. Macromol Mater Eng 2004,

289, 923-926. (45) Schlaad, H., Krasia, T., and Antonietti, M. J Am Chem Soc 2004, 126, 11307-

11310. (46) Sharma, R. In Surfactant adsorption and surface solublization; Sharma, R.,

Ed.; American Chemical Society: Washington, 1995. (47) Pabon, M., and Corpart, J.M. J Fluorine Chem 2002, 114, 149-156. (48) Rojas, O. J., Macakova, L., Blomberg, E., Emmer, A., and Claesson, P.M.

Langmuir 2002, 18, 8085-8095. (49) Hamley, I. Block copolymers in solution: fundamentals and applications;

John Wiley & Sons Ltd: Chichester, 2005. (50) Leibler, L. Makromol Chem Macromol Symp 1988, 16, 1-17. (51) Leibler, L. Physica A 1991, 172, 258-268. (52) Cho, D., Kim, Y.J., Erkey, C., and Koberstein, J.T. Macromolecules 2005, 38,

1829-1836. (53) Koberstein, J. T. J Polym Sci B Polym Phys 2004, 42, 2942-2956. (54) Pan, F., Wang, P., Lee, K., Wu, A., Turro, N.J., and Koberstein, J.T.

Langmuir 2005, 21, 3605-3612. (55) Wang, S., and Marchant, R.E. Macromolecules 2004, 37, 3353-3359. (56) Larsen, C. C., Kligman, F., Kottke-Marchan, K., and Marchant, R.E.

Biomaterials 2006, 27, 4846–4855. (57) Larsen, C. C., Kligman, F., Tanga, C., Kottke-Marchant, K., and Marchant,

R.E. Biomaterials 2007, 28, 3537–3548.

173


(58) Birdi, K. S. In Handbook of Surface and Colloid Chemistry; Birdi, K. S., Ed.; CRC Press: Boca Raton, 2003.

(59) Silberberg, M. Chemistry: The molecular nature of matter and change; Mosby-Year Book Inc.: USA, 1996.

(60) http://www.sp.uconn.edu/~bi107vc/fa02/terry/Lect3.html. 16/8/2007 (61) Alkorta, I., and Elguero, J. Structural Chem 2004, 15, 117-120. (62) Matta, C., Castillo, N., and Boyd, R.J. J PHys chem A 2005, 109, 3669-3681. (63) Minko, S., Mu¨ ller, M., Motornov, M., Nitschke, M., ; Grundke, K., and

Stamm, M. J Am Chem Soc 2003, 125, 3896-3900. (64) Graessley, W. W. In Physical properties of polymers, Third ed.; Mark, J.,

Ngai, K., Graessley, W., Mandelken, L., Samulski, E., Koenig, J., and Wignall, G., Ed.; Cambridge University Press: Cambridge, 2003.

(65) Wu, S. J Polym Sci B Polym Phys 1989, 27, 723-741. (66) Tuminello, W. H., Treat, T.A., and English, A.D. 1988, 21, 2606-2610. (67) Wu, S. Polym Eng Sci 1988, 28, 538-543. (68) Matsuo, K., Stockmayer, W.H., and Needham, G.F. J Polym Sci Polym Symp

1984, 71, 95-100. (69) Tobolsky, A. V., amd Takahashi, M. J Appl Poly Sci 1963, 7, 1341-1347. (70) Ratner, B. D. In Biomaterials science : an introduction to materials in

medicine; Ratner B.D., H., A.S., Schoen, F.S., and Lemons, J.E., Ed.; Academic Press: San Diego, 1996.

(71) Cumpson, P. J. Surf Interf Anal 2001, 31, 23-34. (72) Reed, N., and Vickerman, J. In Surface characterization of advanced

polymers; Sabbatini, L., and Zambonin, P., Ed.; VCH: Weinheim, 1993. (73) Eynde, X. V., Weng, L.T., and Bertrand, P. Surf Interf Anal 1997, 25, 41-45. (74) Bertrand, P. Appl Surf Sci 2006, 252, 6986-6991. (75) Wagner, M. S., and Castner, D.G. Appl Surf Sci 2004, 231-232, 366-376. (76) Wang, H., Castner, D.G., Ratner, B.D., and Jiang, S. Langmuir 2004, 20,

1877-1887. (77) Pacholski, M. L., Cannon, D.M., Ewing, A.G., Winograd, N. J Am Chem Soc

1999, 121, 4716-4717. (78) Wagner, M. A., Graham, D.J., and Castner, D.G. Appl Surf Sci 2006, 252,

6575-6581. (79) Graham, D. J., Wagner, M.S., and Castner, D.G. Appl Surf Sci 2006, 252,

6860-6868. (80) Kattner, J., and Hoffmann, H. In Handbook of vibrational spectroscopy:

Sampling techniques; Chalmers, J. M., and Griffiths, P.R., Ed.; John Wiley & Sons Ltd.: Chichester, 2002; Vol. 2, p 1009.

(81) Lowman, G. M., Tokuhisa, H., Lutkenhaus, J.L, and; Hammond, P. T. Langmuir 2004, 20, 9791-9795.

(82) Kulkarni, S. A., Mirji, S.A., Mandale, A.B., Vijayamohanan, K.P. Thin Solid Films 2006, 496, 420 – 425.

(83) de Boer, B., Meng, H., Perepichka, D.F., Zheng, J., Frank, M.M., Chabal, Y.J., and Bao, Z. Langmuir 2003, 19, 4272-4284.

(84) Ganesh, V., and Lakshminarayanan, V. J Phys Chem B 2005, 109, 16372-16381.

(85) Roach, P., Farrar, D., and Perry, C.C. J Am Chem Soc 2005, 127, 8168-8173. (86) Werner, C., and Jacobasch, H.-J. Int J Artificial Organs 1999, 22, 160-176. (87) Chittur, K. K. In Biopolymers at interfaces; Malmsten, M., Ed.; Marcel

Dekker: New York, 1998.

174

http://www.sp.uconn.edu/%7Ebi107vc/fa02/terry/Lect3.html


(88) Andrade, J. D., Smith, L.M., and Gregonis, D.E. In Surface and interfacial aspects of biomedical polymers; Andrade, J. D., Ed.; Plenum Press: N.Y., 1985; Vol. Volume 1: Surface chemistry and physics.

(89) Keen, I., Broota, P., Rintoul, L., Fredericks, P., Trau, M., and Grøndahl, L. Biomacromolecules 2006, 7, 427-434.

(90) Beamson, G., and Briggs, D. High resolution XPS of organic polymers; The Scienta ESCA300 Database; John Wiley & Sons Ltd: West Sussex, 1992.

(91) Choosakoonkriang, S., Lobo, B.A., Koe, G.S., Koe, J.G., and Middaugh, C.R. J Pharmaceutical Sci 2003, 92, 1710-1722.

(92) Nitschke, M., Schmack, G., Janke, A., Simon, F., Pleul, D., and Werner, C. J Biomed Mater Res 2002, 59, 632–638.

(93) Beck, A. J., Whittle, J.D., Bullett, N.A., Eves, P., Neil, S.M., McArthur, S.L., Shard, A.G. Plasma Process Polym 2005, 2, 641-649.

(94) Vidyadhar, A., Hanumantha Rao, K., Chernyshova, I.V., Pradip, and Forssberg, K.S.E. J Colloid and Interface Sci 2002, 256, 59-72.

(95) Xu, Z., Liu, Q., and Finch, J.A. Appl Surf Sci 1997, 120, 269-278. (96) Kovtyukhova, N. I., and Mallouk, T.E. J Phys Chem B 2005, 109, 2540-2545. (97) Zhang, L., Liu, J., Yang, J., Yang, Q., and Li, C. Microporous Mesoporous

Mater 2007, Accessed online on 10/08/2007. (98) Boonaert, C. J. P., and Rouxhet, P.G. Appl Envronm Microbiol 2000, 66,

2548-2554. (99) Gengenbach, T. R., Chatelier, R.C., and Griesser, H.J. Surf Interf Anal 1996,

24, 611-619. (100) Barth, G., Ed. Defining a "clean" surface; Surface Science Laboratories:

Mowntain View, California, USA, 1985; Vol. Winter. (101) Miller, J. C., and Miller, J.N. Statistics for Analytical Chemistry,; Ellis

Horwood Ltd: NY, 1993. (102) Zuwaylif, F. H. General Applied Statistics; Addison-Wesley Publishing

Company, Inc., 1980. (103) Zhang, F., Xu, J., Kang, E.T., and Neoh, K.G. Ind. Eng. Chem. Res. 2006, 45,

3067-3073. (104) Lemieux, M., Minko, S., Usov, D., Stamm, M., and Tsukruk, V.V. Langmuir

2003, 19, 6126-6134. (105) Lemieux, M., Usov, D., Minko, S., Stamm, M., Shulha, H., and Tsukruk, V.V.

Macromolecules 2003, 36, 7244-7255. (106) Walters, K. B., and Hirt, D.E. Polymer 2006, 47, 6567-6574. (107) Ikada, Y. Biomaterials 1994, 15, 725-736. (108) Choi, J., and Rubner, M.F. Macromolecules 2000, 38, 116-124. (109) Quinn, J. F., and Caruso, F. Langmuir 2004, 20, 20-22. (110) Sukhishvili, S. A., and Granick, S. J Am Chem Soc 2000, 122, 9550-9551. (111) Wang, L., Fu, Y., Wang, Z., Fan, Y., and Zhang, X. Langmuir 1999, 15,

1360-1363. (112) Stockton, W. B., and Rubner, M.F. Macromolecules 1997, 30, 2717-2725. (113) Martell, A. E., and Smith, R.M. Critical stability constants; Plenum Press:

NY, 1977; Vol. 3. (114) Losche, M., Schmitt, J., Decher, G., Bouwman, W.G., and Kjaer, K.

Macromolecules 1998, 31, 8893-8906. (115) Hoogeveen, N. G., Cohen Stuart, M.A., Fleer, G.J., and Bohmer, M.R.

Langmuir 1996, 12, 3675-3682. (116) Michel, R., and Castner, D.G. Surf Interf Anal 2006, 38, 1386–1392.

175


(117) Sperati, C. A. In Polymer handbook; Brandrup, J., and Immergut, E.H., Ed.; John Wiley & Sons: USA, 1989.

(118) Nasef, M. M. Polym Int 2001, 50, 338-346. (119) Jankova, K., and Hvilsted, S. Macromolecules 2003, 36, 1753-1758. (120) Whittaker, M. R., Urbani, C.N., and Monteiro, M.J. Langmuir 2007, 23,

7887-7890.

176



4.1 Introduction

4.1.1 Mineralisation

An ideal implant surface for hard tissue applications should provide optimal

conditions for osseointegration. One feature, which has been demonstrated to be

important, is rapid calcium phosphate layer deposition. Small changes in surface

properties have been shown to improve calcium phosphate (CaP) nucleation and

growth, and the presence of anionic functional groups on the materials surface

especially phosphates (one of the components of hydroxyapatite; HAP) has been

shown to enhance mineral growth.1-4

The first step in mineralisation, before growth proceeds, is nucleation.

Heterogeneous nucleation occurs on a substrate surface under normal body fluid

conditions. The interaction between the surface and the ions lowers the interfacial

energy; therefore, nucleation can proceed at a lower ionic concentration than that of

supersaturation.5 This process has important consequences for biomaterials as well as

in biomineralisation. In the case of bone formation through the behaviour of the

biomineralisation process, matrix proteins such as osteonectin and phosphoproteins

attach to the collagen matrix and act as nucleators.5,6 Synthetic materials, such as

polymeric biomaterials, generally do not possess nucleation sites. However, it has

been suggested that the presence of polar functional groups on the surface is a key

factor in nucleation. In such cases, nucleation is thought to be triggered by

electrostatic interactions.7 Once the apatitic nuclei are formed, the apatite crystals

grow by uptake of calcium and phosphate ions from the solution.

There are many different inorganic phases formed from calcium and phosphate

(usually referred to as the calcium phosphates, CaP). Some are listed in Table 4.1,

together with the atomic calcium to phosphate ratios (Ca/P ratio). Substitution of

other ions within the apatite lattice commonly occurs, for example, biological apatite

is usually substituted with other ions.8 Bone-like apatite is, in fact, carbonated HAP.

Carbonate ions can be substituted into the apatite lattice in two different positions:

substitution of either the OH, or phosphate groups.9 In simulated body fluid (SBF)

studies, partial substitution of Ca2+ with Mg+ ions is often observed.10 These

177


substitutions have been found to reduce both the crystallinity of the apatitic crystal11,

and inhibit the crystal growth.12

Table 4.1: Different calcium phosphate structures.8

4.1.2 Simulated Body Fluid (SBF)

The need to predict how a material will behave in vivo has led to the use of the so-

called simulated body fluid techniques (SBF) as the most common method for testing

CaP mineralisation in vitro.13 In such studies, the test material is immersed in a

solution containing the inorganic ions found in blood plasma, controlled at

physiological pH, temperature and concentration. Calcium phosphate growth on the

surface of the material is subsequently investigated. Currently, several different SBF

techniques have been developed (Table 4.2). Corrected SBF (c-SBF) has been

proposed to the International Organization for Standardization as the most suitable

solution for in vitro measurement of the apatite-forming ability of materials.14

Revised SBF (r-SBF), which has equal ion concentrations to that of human blood

plasma, was chosen for this study to mimic in vivo conditions as well as to maintain

consistency with the previous study.15

178

halla



Table 4.2: Ion concentrations of human blood plasma, different SBF solutions,14 and SPF.16

Ion Concentration (mM)

Na+ K+ Mg2+ Ca2+ Cl- HCO3- HPO4

2- SO42-

Human Blood Plasma 142.0 5.0 1.5 2.5 103.0 27.0 1.0 0.5

Original SBF 142.0 5.0 1.5 2.5 148.8 4.2 1.0 0

Corrected SBF (c-SBF) 142.0 5.0 1.5 2.5 147.8 4.2 1.0 0.5

Revised SBF (r-SBF) 142.0 5.0 1.5 2.5 103.0 27.0 1.0 0.5

Newly improved SBF (n-SBF) 142.0 5.0 1.5 2.5 103.0 4.2 1.0 0.5

Simulated physiological fluid

(SPF)

152.0 5.0 1.5 2.5 136.0 27.0 2.5 0.5

Although this test only studies inorganic apatite growth, without any interaction of

biomolecules such as proteins, the formation of calcium-phosphate (CaP) minerals or

a bone-like apatitic layer in SBF is accepted as an indication of the bone bonding

ability of the material in vivo.17 Kokubo and Takadama14 has recently reviewed the

quantitative correlation of apatite formation in SBF with in vivo bone bioactivity.

Studies in this article included the evaluation of: bioglasses and other ceramic

materials such as sintered HAP, HAP/β-tricalcium phosphate and calcium sulfate.

For all of these materials, apatite formation on their surfaces in SBF is correlated

well with their in vivo bone bonding ability.

This type of correlation is also observed for polymeric materials. MOEP-grafted

polyethylene (PE) and poly(ethylene terephthalate) (PET) both form HAP in SBF

and have been shown to have bone bonding ability in vivo.1,18,19 Copolymers of

MOEP and 1-vinyl-2-pyrrolidinone,20,21 MOEP or MAEP and 2-hydroxyethyl

methacrylate (HEMA)22 failed to induce CaP nucleation in SBF. These copolymers

were subsequently reported as having performed poorly in in vivo tests. These

observations also appear to confirm the correlation between apatite formation in SBF

with in vivo behaviour.

In some cases, dynamic SBF is used. This is thought to create a more similar

environment to the living body than that of static SBF tests.23-26 The topology of the

material becomes more relevant than that of the chemistry in this system. For

example, an alkaline-treated micropatterned surface of a titanium alloy showed HAP

growth over all of the surface in static SBF, but was more pronounced in the bottom

179


of the microholes than the flat surfaces in dynamic SBF.27 This micropatterned

surface showed the best bone bonding ability in vivo compared to those without

patterned surfaces.

CaP phases other than HAP have also been found to form in SBF, especially in the

early stage of mineralisation. Brushite and octacalcium phosphate (OCP) are thought

to be precursors of HAP.8,28,29 OCP is known to promote osteoblast differentiation

and proliferation, and it has been reported previously that this phase has a higher

probability of precipitation from r-SBF at a pH of 7.4.30 Amorphous CaP phases have

also often been found to form initially in SBF.31 Using 31P solid state NMR

spectroscopy, Lin et al.32 identified an amorphous phase with substantial hydration as

the initial phase formed on a bioglass surface in SBF. It has been shown that

amorphous CaP is then transformed into HAP which is the most thermodynamically

stable phase.

4.1.3 Negatively Charged Groups

Many researchers have shown that negatively charged groups on a material’s

surfaces are better heterogeneous nucleators of apatite-like minerals in SBF than

positive or neutral surfaces. Hence, it can be interpreted that the accumulation of

Ca2+ ions near the negatively charged surfaces, due to chelation, increases the

likelihood of supersaturation, and as a result, initial nucleation is preferentially

triggered at such sites.33 Among these negatively charged groups, phosphates are

found to have the largest effects.

Self assembled monolayers (SAMs) have been used to study the effect of different

functional groups on the nucleation of calcium phosphate in SBF. Tanahashi and

Matsuda 3 showed significant apatite formation on negatively charged SAMs, but not

on positively charged, or neutral surfaces. The growth rate determined using quartz

crystal microbalance (QCM) measurements decreased in the order: -PO4H2 > -

COOH » -CONH2 ≃ -OH > -NH2 » -CH3 ≃ 0.3 Interestingly, -COOH terminated

surfaces exhibited an induction period of almost 10 days before apatite growth, with

a linear growth rate, whereas, -PO4H2 surfaces did not exhibit this induction period.

180


Another study of calcium phosphate growth induced by SAMs with different

functional groups on titanium showed similar results.4 Nucleation occurred on the

surfaces with -PO4H2 and -COOH groups, but not on the surfaces with -OH or -

CH=CH2. Again, -PO4H2 was found to exhibit stronger propensity for nucleation

than the -COOH.

Incorporation of negatively charged functional groups such as carboxylate and

phosphate ions on the surface of biomaterials or within a polymeric framework is a

popular strategy for attempting to enhance the deposition of calcium-phosphate

minerals. The effect of these groups on the calcification of polymeric materials has

recently been the subject of an excellent review by Chirila.34 Although the

incorporation of phosphorous-containing moieties has been shown to lead, in vitro, to

the enhanced calcification of naturally occurring materials, such as: cotton,35,36

bamboo37,38 and chitin,39 the calcification results for synthetic polymers are less well-

studied and somewhat controversial.40 As Chirila pointed out in his review “the

effect of the phosphate group may be more complicated than initially thought”.

Polymers containing carboxylic acid groups have shown induction periods for apatite

nucleation, no mineralisation or even inhibitory effects.41-44 However, apatite

formation can generally be achieved, even without an induction period, if the

polymer is pre-treated with Ca(OH)2 or CaCl2 before SBF immersion.45 It has been

shown that bound Ca2+ ions get released when the polymer is soaked in SBF. The

high concentration of Ca2+ ions enhances chelation with the carboxylic acid groups

and hence accelerates apatite formation.41 In some cases, such as NaOH42 or O2

plasma46 treated poly(ε-caprolactone) (PCL), alternative dipping of samples into Ca2+

and PO42- solutions was used before the SBF study.

In the case of carboxylated polyphosphazenes (structure shown in Scheme 4.1),

which contains two carboxylic acid groups per repeating unit, nucleation of CaP has

been shown within 24 hours in SBF, without Ca2+ treatment.47

Scheme 4.1: Structure of carboxylated polyphosphazene

181


* N P *

O

O

COOH

COOH

n

After 5 days, the crystalline composition of this mineral was similar to that of

octacalcium phosphate, then changed to either calcium deficient apatite, or tricalcium

phosphate after 7 days. This indicates that the composition of ionic groups and the

structure of the polymer are also important factors in nucleation.

In a series of elegant experiments, Ikada et al.1,18 demonstrated that MOEP-grafted

PE and PET exhibit significantly improved mineralisation in SBF. Expanded PTFE

(ePTFE) membranes have been surface modified with MAEP,48 or MOEP49 by

radiation induced grafting. MAEP grafted ePTFE with an external surface coverage

of 44 % or more has shown secondary growth of apatite-like mineral formation in

SBF, although unmodified ePTFE did not mineralise.2 MOEP-grafted samples also

showed the formation of carbonated HAP and other types of CaP.15,49 These MAEP-

or MOEP-grafted materials all showed mineralisation in SBF without Ca2+

pretreatment.

When incorporated into hydrogels, the effect of phosphate groups on mineralisation

is not straight forward. Chirila et al.22 showed that after 9 weeks in SBF, CaP

deposition on copolymers of PHEMA and 10 mol % MAEP, or MOEP was four

times lower than that found on the PHEMA homopolymer. Subsequent animal

studies revealed no CaP deposition on PHEMA copolymers whereas PHEMA

homopolymer was extensively mineralised. This clearly demonstrates an inhibitory

effect of phosphate groups in hydrogels on mineralisation.

Stancu et al.20 copolymerized MOEP with (diethylamino)ethyl methacrylate

(DEAEMA) and 1-vinyl-2-pyrrolidinone (VP) (Structures shown in Scheme 4.2).

Globular apatite formation was only observed on P(MOEP-co-DEAEMA) with less

than 60 mol % MOEP, and on PDEAEMA after 15 days in SBF. In P(MOEP-co-

DEAEMA), MOEP alternates with DEAEMA showing uniform distribution,

whereas PMOEP forms “islands” in P(MOEP-co-VP) due to the different reactivity

ratios of these monomers. It was suggested that in PMOEP rich areas, Ca2+ ions are

coordinated by two phosphate groups from adjacent MOEP units. These strongly

182


bound Ca2+ ions are no longer capable of attracting phosphate ions from SBF, thus

inhibiting further nucleation. However, this explanation cannot be applied to the

mineralisation of MAEP- and MOEP-grafted materials, and phosphate terminated

SAMs.

Scheme 4.2: Structures of DEAEMA and VP.

ON

O DEAEMA

N

O

VP

In this study, the effects of polymer structure on mineralisation were investigated in

SBF using a series of phosphate containing polymers (soluble and cross-linked

PMAEP and PMOEP). The mineralisation of fabricated surfaces (LbL films

containing phosphate or carboxylate groups, block copolymers attached to the

aminated slide, and fluorinated block copolymers attached to a PTFE substrate) were

also studied. Since PMAEPs have been found to contain some carboxylic acid groups,

the effect of this group on mineralisation was also investigated using poly(acrylic

acid) (PAA) gels.

183


4.2 Experimental

4.2.1 Materials

Acrylic acid (99%), stabilised with 200 ppm hydroquinone monomethyl ether

(MEHQ), was purchased from Aldrich and used as received. The purities of the

chemicals used to prepare the SBF solution were as follows: NaCl (99.9%), NaHCO3

(99.0%), Na2CO3 (99.9%), KCl (99.0%), K2HPO4 (99.0%), MgCl2·6H2O (99.0%),

CaCl2·2H2O (99.5%), Na2SO4 (99.0%), and HEPES (2-[4-(2-Hydroxyethyl)-1-

piperazinyl]ethanesulfonic acid) (99.9%).

Samples subjected to mineralisation studies are as follows: The synthesis of gels and

soluble PMAEP and PMOEP are described in Chapter 2. The material properties of

samples investigated are listed in Table 4.3. Fabrication of LbL films and block

copolymers coupled onto aminated slides are discussed in Chapter 3, and their

properties are listed in Tables 4.4 and 4.5, respectively.

4.2.2 Methods

4.2.2.1 Synthesis of cross-linked PAA gels

Acrylates are known to cross-link when polymerised under radiation, due to

hydrogen abstraction. To obtain cross-linked PAA gels for mineralisation study,

gamma radiation-induced polymerisation of AA was carried out. A solution of 10

w/v % acrylic acid in MilliQ water was prepared in a glass tube and sealed with a

Suba cap. Dissolved oxygen in the monomer solution was removed by bubbling

nitrogen gas through for 20 minutes in an ice-bath. The sample was then subjected to

a 60Co gamma radiation source to obtain a total dose of 16 kGy (dose rate of 2.8

kGy/h) using a 220 Nordian Gamma-cell (Canada) at room temperature. The

obtained gel was soaked in MilliQ water for 3 days with regular water exchanges to

remove any residual monomer. The sample was then dried in a vacuum oven to

constant weight.

184


4.2.2.2 Simulated body fluid (SBF) experiments

The simulated body fluid was prepared according to the method described by Kim et

al. 13. Chemicals were dissolved in MilliQ water (that had been boiled for one hour

prior to preparation) and buffered with HEPES (2-[4-(2-Hydroxyethyl)-1-

piperazinyl]ethanesulfonic acid) and 1M NaOH at pH = 7.4 at 36.5 ºC. The insoluble

gels were used as prepared. The soluble PMOEP was cast from methanol solution

onto the glass substrate where it was left for the mineralisation study. The soluble

PMAEP was cast onto PTFE and was used without this substrate. Approximately 10

mL of SBF solution were added to a 15 mL polystyrene tube containing a polymer

sample. The tubes were immersed in a water bath at 36.5 + 0.2 ºC for a period of

seven days. The SBF solution was changed every 2 days. After 7 days, the materials

were washed by soaking in MilliQ water for 10 minutes, this was repeated three

times. The materials were subsequently dried in a vacuum oven at 40 ºC to constant

weight.

4.2.2.3 Scanning electron microscopy with energy dispersive x-ray analysis

SEM/EDX (FEI Quanta 200 Environmental SEM equipped with an Evarhart

Thomley secondary electron detector) was performed at 10 kV to examine the

morphology of the calcium phosphate deposit and to obtain the elemental

composition of the CaP minerals on the surface. Before SEM analysis, the surface of

the sample was coated with a very thin layer of carbon using Cressington Carbon

Coater to reduce sample charging. In some cases, where it is stated, the sample was

coated with gold using a Biorad Gold Sputter Coater.

4.2.2.4 Fourier transform infrared spectroscopy – attenuated total reflectance

A Nicolet Fourier Transform Infrared Spectrometer equipped with a diamond ATR

(refractive index of 2.41 at 1000 nm and an average angle of incidence of 50 º) was

used to analyse the mineral deposits. Spectra, 64 scans at 4 cm-1 resolution, were

collected over a range of 4000 – 525 cm-1. The depth of penetration was calculated to

be between 2.85 (525 cm-1) and 0.37 μm (4000 cm-1), for an estimated refractive

index of the polymer/CaP phase of 1.5.

185


4.2.2.5 XPS

XPS spectra were recorded using a Kratos Axis Ultra X-ray photoelectron

spectrometer with a monochromated Al Kα X-ray source (1486.6 eV) running at 150

W (15 kV, 10 mA emission current). The survey scans were collected at 1200-0 eV

with 1.0 eV steps at a pass energy of 160 eV; narrow scans were taken at 0.1 eV

steps and a pass energy of 20 eV. Vision 2 software was used for data acquisition and

processing. The binding energies were charge-corrected using the C1s hydrocarbon

peak (285.0 eV).

186


4.3 Results

4.3.1 SBF studies of PMAEP and PMOEP

Mineralisation of a series of phosphate-containing polymers was investigated using

SBF in order to evaluate the effects of cross-linking and phosphate content. Sample

characteristics are shown in Table 4.3.

Table 4.3: Properties of PMAEP and PMOEP polymers subjected to SBF studies.

Samp

le

Expt a

Polymer [RAFT] (mol/L) Conv b

(%)

Characteristic Total

% P cSurface

% P d

1A 1 PMAEP ― 90 Gel 9.5 4.39

1B 2 PMAEP 1 × 10-2 (PEPDA) 83 Soluble 7.0 2.99

1C 7 PMOEP 2 × 10-2 (PEPDA) 95 Gel e e

1D 6 PMOEP 1 × 10-2 (PEPDA) 81 Gel 11.5 e

1E 5 PMOEP ― 96 Gel 11.2 4.33

1F 8 PMOEP 1 × 10-2 (CDB) 74 Soluble 9.9 5.33

1G PAA ― ― Gel ― ―

a: Expt no. from Table 2.1, Chapter 2, b: conv = conversion obtained from Raman spectroscopy, c: determined by ICP, w/w %, d: determined by XPS, atomic %, e: did not determined

Cross-linked gel polymers and soluble PMOEP cast on a glass surface were

immersed in SBF for seven days with regular solution changes. The films obtained

from the soluble PMAEP dissolved in SBF, hence a modified SBF solution 1.5 times

the concentration of normal SBF, was used. After seven days, the polymers were

washed and dried to constant weight. CaP depositions were characterised by

SEM/EDX and ATR-FTIR.

For identification of the mineral phases, XRD is generally used in conjunction with

SEM/EDX. However, Grøndahl et al.2 has demonstrated that minerals formed in

SBF over a period of 7-14 days are often amorphous which do not show XRD

patterns. Instead, this study used FTIR since the previous studies had already

demonstrated it to be an useful technique for mineral identification.2,15

187


Ca

O

C

NaMg P

Ca

A

B

C

D

E Ca/P = 0.88 (Ca+Mg/P) = 1.37

Ca/P = 1.14 (Ca+Mg)/P = 1.34

Au-Coated

Untreated

Ca/P = 3.95 (Ca+Mg)/P = 4.54

F

Figure 4.1: SEM images of sample 1A (PMAEP gel) (A) before treatment, and after SBF immersion for 7 days with (B) carbon coating and (C) gold coating, (D) sample 1B (the film obtained from soluble PMAEP after 7 days in 1.5 SBF), (E) sample 1G (PAA gel) after 7 days immersion in SBF, and (F) EDX spectrum of the mineral on sample 1G.

188


D A Untreated Untreated

B E Ca/P = 0.73 (Ca+Mg)/P = 0.98

Ca/P = 0.94 (Ca+Mg)/P = 1.16

Ca/P = 1.41 Ca+Mg/P = 1.51

Ca/P = 1.51 Ca+Mg/P = 1.55

C F Carbon Tape

Ca/P = 1.49 Ca+Mg/P = 1.54

Ca/P = 1.48 Ca+Mg/P = 1.52

Au-Coated

Figure 4.2: SEM images of sample 1C (PMOEP gel) (A) before treatment, and after SBF immersion for 7 days with (B) carbon coating and (C) gold coating, sample 1D (PMOEP gel) (D) before treatment and (E) after SBF immersion for 7 days and (F) minerals dislodged from sample 1D.

189


O

Na Mg

P

Ca Ca

D

C

Ca/P = 0.73 (Ca+Mg)/P = 0.99

Ca/P = 1.44 (Ca+Mg)/P = 1.55

Untreated Ca/P = 1.72 (Ca+Mg)/P = 1.77

C

B

A

Figure 4.3: SEM images of sample 1E (PMOEP gel) (A) before, and (B) after SBF immersion for 7 days, (C) sample 1F (soluble PMOEP cast on glass) after SBF immersion for 7 days, and (D) EDX spectrum of the mineral on sample 1E.

Figures 4.1-4.3 show SEM images of the samples before and after SBF immersion.

EDX elemental analysis of the minerals revealed the presence of C, O, P, Ca and

small amounts of Mg and Na in all homopolymer samples after SBF immersion. The

EDX spectra of minerals on sample 1E and 1G are shown in Figure 4.3D and 4.1F,

respectively. The presence of carbon is due mainly to the sample coating used for

SEM/EDX analysis but may also be from the polymer and/or carbonate ions. In

mineralisation studies, Mg2+ substitution for Ca2+ within a CaP lattice is often

observed10 and it is therefore necessary to evaluate both the Ca/P and (Ca+Mg)/P

ratios. These ratios are shown as inserts in the following SEM images.

SEM images of sample 1A (PMAEP gel) before and after SBF immersion for 7 days

are shown in Figure 4.1 A and B, respectively. The untreated polymer surface was

190


smooth, whereas after soaking in SBF, the surface was covered with thin layers of

CaP mineral, which had multiple cracks due to sample preparation. EDX analysis of

this mineral yielded Ca/P and (Ca+Mg)/P ratios of 0.88 and 1.37, respectively

(Inserts in Figure 4.1B). Sample 1B (the film obtained from soluble PMAEP) also

showed the same tile-like morphology with Ca/P and (Ca+Mg)/P ratios of 1.14 and

1.34, respectively (Figure 4.1D). These values indicate that this mineral phase is

possibly either brushite (CaHPO4·H2O), monetite (CaHPO4), or octacalcium

phosphate (Ca8H2(PO4)6·5H2O). However, it is important to note that since these

mineral layers were thin, the ratios obtained from EDX are associated with large

errors due to the large sampling depth of EDX (~ 5 μm). It is also important to note

that when the SBF-treated PMAEP gel was coated with gold on top of a carbon

coating so as to obtain a clear image, this tile-like structure was completely covered

and the surface morphology was very smooth (Figure 4.1C). The thickness of gold

coatings is typically 25 nm or more, whereas carbon coatings are generally 10-15 nm,

and less dense. This suggests that the tile-like mineral layer is very thin.

The effect of carboxylic acid (-COOH) groups on mineralisation was tested using a

poly(acrylic acid) (PAA) gel (sample 1G). Figure 4.1E shows the SEM image of

PAA after 7 days soaking in SBF. Again a tile-like morphology was observed. The

Ca/P and (Ca+Mg)/P ratios were 3.95 and 4.54, respectively.

All the PMOEP gels (sample 1C, 1D and 1E) had smooth surface morphologies

before SBF (Figure 4.2 and 4.3). After 7 days in SBF, the surfaces were covered with

a tile-like mineral layer similar to those observed on the PMAEP samples. The Ca/P

and (Ca+Mg)/P ratios of this layer were 0.73-0.94 and 0.98-1.16, respectively. In

addition, a secondary growth of round mineral nodules of various sizes (ø ~ 2-5 μm)

including some large clusters, was observed. These round mineral nodules had Ca/P

and (Ca+Mg)/P ratios of 1.41-1.51 and 1.51-1.55, respectively. Figure 4.2F shows

minerals dislodged onto carbon tape during SEM sample preparation. Here, two sizes

of spherical mineral clusters were observed, one of which had particles of much

smaller diameter. Nevertheless, both of them had similar Ca/P and (Ca+Mg)/P ratios

(1.48-1.49 and 1.52-1.54, respectively). The tile-like morphology was masked when

coated with gold (Figure 4.2C), also shown on the PMAEP gel.

191


After 7 days, the SBF-treated soluble PMOEP cast on a glass surface (sample 1F)

was also covered with a large amount of spherical mineral clusters (Figure 4.3C). Its

clusters were much smaller in diameter than those formed on the PMOEP gels: they

were, however, similar to one of the mineral types dislodged from the PMOEP gel

(sample 1D) shown in Figure 4.2F. EDX analysis revealed Ca/P and (Ca+Mg)/P

ratios of 1.72 and 1.77, respectively; slightly higher than the theoretical value for

HAP (i.e., 1.67).

To summarise SEM results, PMAEP gel and the film obtained from soluble PMAEP

and PAA gel showed tile-like mineral morphology. PMOEP gels also showed similar

tile-like morphology covering the polymer surfaces as well as secondary growth of

globular HAP. The best mineralisation was observed on soluble PMOEP cast on

glass that was covered with high amounts of HAP like globular minerals.

192


1069

1723

1721

995

1075

99

3

B A

1565

1567

Wavenumber (cm-1) Wavenumber (cm-1)

17

10

1069

1645

991

959 C D

1715

1070

991

959

1652


1482

14

18

1021

872

1710

E F

1647

1067

990

959


1541

G

1406

14

51

1043

Wavenumber (cm-1)

Figure 4.4: ATR-FTIR spectra of polymer samples after 7 days immersion in SBF (solid line) and initial, untreated polymers (dotted line). (A) sample 1A (PMAEP gel); (B) sample 1B (PMAEP film) (Experiment done in 1.5×SBF.); (C) sample 1C (PMOEP gel); (D) sample 1D (PMOEP gel); (E) sample 1E (PMOEP gel); (F) sample 1F (PMOEP film on a glass surface), and (G) sample 1G (PAA gel). y-axis is absorbance.

193


Figure 4.4A and B shows the ATR-FTIR spectra of cross-linked gel and the film

obtained from soluble PMAEP (sample 1A and 1B), respectively, before (dotted line)

and after SBF immersion (solid line). After SBF treatment, both spectra showed the

presence of a large peak at 1069-1075 cm-1 corresponding to a phosphate vibration

band. The peak at 993-995 cm-1 which is also in the phosphate region, together with a

shoulder on the higher wavenumber side of the band ~1070 cm-1, indicate possible

brushite formation.50 Another important feature of FTIR spectra of the SBF-treated

PMAEP samples is the new band at 1565-1567 cm-1. This corresponds to the

carbonyl vibration of a carboxylate group, indicating that large amounts of the

PMAEP side chains were hydrolysed at the C-O-C ester bond. In the case of the

soluble PMAEP sample (Figure 4.4B), the band at 1069 cm-1 has a shoulder at 1036

cm-1. However, it is not possible to identify a mineral phase based on one band. The

bands around 1650 cm-1 observed in these samples were assigned to water bound to

either the CaP mineral or the polymers in their deprotonated state.51

ATR-FTIR spectra of all PMOEP gels showed a new band around 1070 cm-1 after

SBF-treatment (Figure 4.4C-E). This is similar to that observed for the PMAEP

samples. This band also showed a shoulder at higher wavenumbers, and the band at

990-991 cm-1 may indicate the presence of brushite; again similar to the PMAEP

samples.

According to the literature, to the best of my knowledge, the phosphate band around

1070 cm-1 does not correspond to any previously identified CaP phase.50 To

investigate whether this band can be ascribed to the phosphate groups of the

polymers with calcium ions, PMOEP gel was soaked in Ca(OH)2 solution in order to

chelate the Ca2+ ions. The resulting ATR-FTIR spectrum of this sample shows that

such an interaction results in the large broad acidic phosphate vibration around 975

cm-1 disappearing, and giving rise to two strong bands at 1060 and 962 cm-1,

corresponding to the out-of-phase stretching and in-phase stretching vibrations,52

respectively (Figure 4.5). Therefore it can be concluded that the band around 1070

cm-1 which appears after SBF-treatment does not correspond to this particular

interaction.

194


1163

974

1060

Wavenumber (cm-1)

961

Figure 4.5: ATR-FTIR spectra of Sample E (PMOEP gel) reacted with Ca(OH)2 (solid line) and untreated (dotted line). y-axis is absorbance.

Figure 4.4F shows the ATR-FTIR spectra of soluble PMOEP sample (sample 1F)

before and after SBF treatment. After SBF, the spectra showed a dramatic decrease

of the polymer bands, indicating a large amount of mineral formation on this sample.

There was a broad band at 1021 cm-1, which can be assigned to the P-O stretching of

the phosphates in HAP. This sample also showed bands at 1482, 1418 and 872 cm-1

corresponding to the carbonate vibrations of carbonated HAP.

The ATR-FTIR spectrum of sample 1G (PAA gel) shows a large peak at 1541 cm-1

corresponding to carboxylate-bound calcium (Figure 4.4G). This spectrum shows a

broad phosphate vibration band around 1018-1080 cm-1, possibly due to a mixture of

CaP phases.

4.3.2 SBF studies of Layer-by-Layer (LbL) films

Mineralisation of the layer-by-layer ultrathin films fabricated from soluble PMAEP

or PMOEP and PEI on silicon wafers was investigated in r-SBF for 7 and 14 days.

An LbL film with PEI, only, and the silicon wafer substrate were also used as

controls. After the periods of immersion, sample surfaces were characterised using

SEM/EDX and XPS. Mineralisation of PAA and PEI LbL films on glass slides was

also investigated by 7 days immersion in SBF, and characterised by XPS. Table 4.4

summarizes results for the LbL samples in this study.

195


Table 4.4: LbL samples for SBF and their atomic % of Ca and P after 7 days in SBF (obtained from XPS survey scans).

XPS XPS After 7 days in SBF Sample Substrate Polymer

P % C(=O)O % Ca % P % Ca/P

2A Si PEI-PMAEP 2.28 6.89 1.12 1.52 0.74

2B Si PEI-PMOEP 2.62 6.69 1.39 2.62 0.53

2C Si PEI ― ― 0.60 ― ―

2D Si ― ― ― 0.87 ― ―

2E Glass PEI-PAA ― 3.55 0.89 ― ―

2F Glass PEI ― ― 0.64 ― ―

2G Glass ― ― ― 0.38 ― ―

Figure 4.6A-D shows SEM images of the LbL films, and the silicon wafer, after 7

days in SBF. Sparse and patchy CaP mineral formation was observed on both

samples 2A (PMAEP) and 2B (PMOEP) LbL films (Figure 4.6A and B,

respectively). EDX analyses of these minerals showed the Ca/P ratios of 1.77 and

1.65 for samples 2A and 2B, respectively, indicating possible HAP formation

(Inserts in Figure 4.6). No Mg was detected in these minerals. The control sample 2C

(PEI) and sample 2D (silicon substrate) surfaces showed sparse and patchy calcium

mineral formation, possibly CaCO3 after 7 days (Figure 4.6C and D, respectively).

196


A E

B

Ca/P = 1.77

F

Ca/P = 1.65

Ca/P = 2.06

Ca/P = 1.96

GC

D H

Figure 4.6: SEM images of minerals formed on the LbL surfaces after SBF immersion for 7 days: (A) sample 2A (PEI-PMAEP), (B) sample 2B (PEI-PMOEP), (C) sample 2C (PEI only), and (D) sample 2D (silicon wafer), and 14 days: (E) sample 2A, (F) sample 2B, (G) sample 2C, and (H) sample 2D.

197


Table 4.4 shows atomic % of Ca and P from XPS survey scans of samples after 7

days immersion in SBF. None of the samples contained calcium before the SBF

studies according to XPS analysis. The stability of the LbL films in SBF was also

confirmed from the carbon and silicon % by comparison to those before the SBF

study. The presence of Ca was observed in all samples. P was only observed in

sample 2A (PMAEP) or 2B (PMOEP) LbL films. Due to the sampling area of XPS

(0.7 × 0.3 mm2) both the mineral-containing and the non-mineral-containing areas

are analysed by this technique. The rationale for this statement is taken from the

spatial distribution of mineral deposits observed in the SEM images.

XPS analysis of samples 2A and 2B showed that the atomic concentration of P is

higher than that of Ca. This P comes both from the mineral and the polymers. It is

proposed that not all the phosphate groups of the polymers are involved in the

chelation of Ca2+ due to strong interactions with the protonated amines in the LbL

films.

In the case of sample 2C, the PEI surface, initial attraction is through the chelation of

phosphate ions by the protonated amines. XPS did not detect any P on this surface,

although there was a small amount of Ca (0.60 %) possibly from the minerals

identified from the SEM/EDX.

The SBF-treated sample 2D, the silicon wafer surface, had slightly higher amounts of

Ca (0.87 %) compared to that of the PEI surface. There was no detectable P on this

surface; this is in agreement with the EDX analysis of the minerals. Here, the initial

interaction is the chelation of Ca2+ ions by –OH groups.

Figure 4.6E-H presents SEM images of the LbL films and the silicon wafer surface

after 14 days in SBF. No significant mineral growth was observed on the surfaces of

sample 2A (PMAEP) and 2B (PMOEP) films (Figure 4.6E and F, respectively). In

the case of samples 2C (PEI) and 2D (silicon surface), sparse and patchy CaP

mineral formation was observed after 14 days (Figure 4.6G and H, respectively), but

not after 7 days (Figure 4.6C and D) indicating there was more than a 7 day

induction period for CaP formation on these samples. Although amine and -OH

groups are not as efficient at inducing CaP growth as phosphate and carboxylate

groups, CaP nucleation has previously been observed on amine and -OH terminated

self assembled monolayers.3,16

198


In the case of PEI-PAA LbL film fabricated on a glass surface (sample 2E), XPS

revealed slightly higher amounts of Ca (0.9 %) than the controls (0.64 and 0.38 % for

a samples 2F (PEI) and 2G (a bare glass slide), respectively) after 7 days in SBF.

This indicates that chelation of Ca2+ ions with carboxylic acid groups is stronger than

with the Si-OH (a glass surface), and protonated amines (PEI). XPS analysis showed

no P on all of these slides.

4.3.3 SBF studies of block copolymers coupled to aminated slides

Mineralisation of block copolymers consisting of PMAEP or PMOEP and PAAEMA

fixed on aminated slides (Table 4.5) were investigated in SBF for up to two weeks. A

slide with PAAEMA attached and a slide without any polymer were also used as

controls. The samples were characterised using SEM/EDX.

Table 4.5: Block-copolymer-attached aminated slides for SBF.

Sample Coupled Block Copolymer

m a n b PDI

XPS Atomic

P % c

3A P(MOEPn-b-AAEMAm) 109 97 1.38 1.53

3B P(AAEMAm-b-MOEPn) 22 132 1.41 0.66

3C P(AAEMAm-b-MAEPn) 22 99 1.38 0.22

3D P(AAEMAm-b-MAEPn) 22 160 1.38 0.27

3E PAAEMAm 22 ― 1.13 ―

3F d ― ― ― ― ― a: unit of PAAEMA segment, b: unit of PMOEP/PMAEP segment, c: obtained from XPS survey scans of samples before SBF treatment, d: untreated aminated slide

Figures 4.7 and 4.8 show, respectively, SEM images of the samples, and the controls,

after SBF treatment.

199


A E Mg = 1.57 Ca = 2.85 P = 0.68

Mg = 1.62 Ca = 5.29 P = 2.04

B F Mg = 1.51 Ca = 2.85 P = 0.95

Mg = 1.86 Ca = 1.83 P = 0

C G

Mg = 1.66 Ca = 2.14 P = 0.42

Mg = 1.57 Ca = 1.75 P = 0

H Mg = 1.63 Ca = 1.79 P = 0

D

Mg = 1.25 Ca = 5.69 P = 2.69

Figure 4.7: SEM images of minerals formed on the block copolymer functionalized aminated slides after SBF immersion for 7 days: (A) sample 3A, (B) sample 3B, (C) sample 3C and (D) sample 3D, and 14 days: (E) sample 3A, (F) sample 3B, (G) sample 3C and (H) sample 3D.

200


A C Mg = 1.89 Ca = 2.07 P = 0

NaCl

B D Mg = 0.97 Ca = 6.96 P = 3.40

Figure 4.8: SEM images of minerals formed on the PAAEMA functionalized and untreated aminated slides after SBF immersion for 7 days: (A) sample 3E (PAAEMA) and (B) sample 3F (untreated aminated slide), and 14 days: (C) sample 3E (PAAEMA) and (D) sample 3F.

EDX spectra of the mineral on sample B (P(AAEMA-b-MOEP)) and the non-

mineral area are shown in Figure 4.9A and B, respectively.

Figure 4.9: EDX spectra of the mineral (A) and the non-mineral area (B) on sample 3B (see Figure 4.7F).

The EDX analysis of the aminated slide indicated that it contains high amounts of Ca

(1.83 %), Mg (1.86 %) and Na (8.07 %), as well as C (16.47 %), O (54.08 %), Si

C

B

O Na Mg

Si

P Cl Ca C

O Na Mg

Si

Ca

A

201


(17.29 %), and K (0.47 %) (Figure 4.9A). There was no detectable P in the non-

mineral area of the slide. In contrast, the EDX spectrum of the mineral showed P

(0.95 %), Ca (2.85 %), Mg (1.51 %) and Na (5.81 %), as well as Cl (0.98 %) (Figure

4.9B). The presence of Cl may be from NaCl. The other elements detected were C

(29.23 %), O (44.42 %) and Si (14.28 %). Due to the large sampling depth of EDX, it

was not possible to quantitatively evaluate the Ca/P ratios of the minerals. However,

all the minerals showed an increase in Ca compared to that of the slides, and the

atomic concentrations of Ca, Mg, and P are shown as inserts in SEM images (Figure

4.7 and 4.8)).

The SEM images of block copolymer-functionalized aminated slides showed sparse,

patchy CaP mineral formation after 7 days in SBF (Figure 4.7A-D). After two weeks,

the mineral formation was slightly increased but still patchy (Figure 4.7E-H). Only

some areas of sample D (P(AAEMA-b-PMAEP) attached slide) showed some large

clusters of CaP minerals (Figure 4.7H).

CaP mineral formation was not observed on the control PAAEMA even after two

weeks of SBF treatment (Figure 4.8A and C). However, the aminated slide induced

CaP formation after two weeks (Figure 4.8D). Interestingly, this mineral phase

showed a distinctive morphology not seen in any of the phosphate-containing

samples. Such non-spherical mineral formations have also been observed on amine-

terminated SAMs have in simulated physiological fluid (SPF).16

4.3.4 SBF studies of fluorinated amphiphilic block copolymers attached onto

PTFE Films

The mineralisation of fluorinated amphiphilic block copolymers (P(FSm-b-AAn))

attached to PTFE was investigated in SBF for up to two weeks. Untreated PTFE film

was used as a control. The samples were characterised using SEM/EDX and XPS.

Table 4.6 shows properties of P(FSm-b-AAn) attached PTFE films.

202


Table 4.6: Properties of fluorinated block copolymer absorbed PTFE films.

Sample P(FSm-b-AAn) Absorption solvent Contact Angle (°)

m n PDI b Advancing Receding

4A 101 237 1.11 MEK 91 ± 6 31 ± 3

4B 101 141 1.07 MEK 88 ± 5 28 ± 3

4C 101 141 1.07 DMF 101 ± 4 49 ± 3

4D a ― ― ― ― 109 ± 6 111 ± 6

a: Untreated PTFE film, b: PDI of corresponding P(FSm-b-tBAn) block copolymers before hydrolysis of tBA groups

Figure 4.10 represents SEM images of samples after 7 and 14 days in SBF. The

PTFE film used in this study was not smooth, and SEM revealed scratch marks.

However, these topological features did not affect mineralisation since untreated

PTFE film did not mineralise in SBF for up to 2 weeks, in agreement with the

previous study.2 SEM images of all the P(FSm-b-AAn) attached surfaces showed no

mineral formation after 7 days in SBF. EDX spectra of these samples only showed

the carbon and fluorine. After 14 days, the modified samples showed very small

patchy mineral depositions.

203


E A

D

C

B F

G

H

Figure 4.10: SEM images of minerals formed on P(FSm-b-AAn) attached and untreated PTFE films after SBF immersion for 7 days: (A) sample 4A, (B) sample 4B, (C) sample 4C and (D) sample 4D (untreated PTFE), and 14 days: (E) sample 4A, (F) sample 4B, (G) sample 4C and (H) sample 4D.

204


Table 4.7: Atomic % of Ca and P from XPS survey scans and the Ca/P ratios.

After 7 days in SBF After 14 days in SBF a

Ca P Ca/P Ca P Ca/P

4A 1.3 0.6 2.2 0.9

0.9

0.8

0.5

0.7

0.6

1.8

1.3

1.3

4B 1.2 0.5 2.4 2.5

0.8

0.6

1.1

0.7

0.4

2.3

1.1

1.5

4C 1.2 0.5 2.4 0.9

0.9

1.0

0.5

0.3

0.4

1.8

3.0

2.5

4D 0 0 ― 0 0 ― a: three randomly chosen areas were analysed on each samples.

XPS was also used to identify any absorbed ions. Table 4.7 summarises atomic % of

Ca and P of the samples from XPS survey scans. Block copolymer attached PTFE

films (samples 4A-4C) showed ~1% calcium after 7 days in SBF. Small phosphorous

peaks were also identified from XPS survey scans of these samples. The Ca/P ratios

are also shown in Table 4.7. Calcium ions are most likely bound to the carboxylic

acid groups of the blocks as no mineral was observed on these samples from SEM.

For the samples after 14 days in SBF, three randomly chosen spots were analysed by

XPS, since SEM showed patchy mineralisation. The results showed small variations

in these three spots as expected. Compared to samples after 7 days, all samples

showed no significant increase in Ca % and P %, except one area of sample 4B

which had 2.5% Ca. Untreated PTFE (sample 4D) up to 14 days in SBF did not show

any calcium or phosphorous.

205


4.4 Discussion

4.4.1 SBF studies of PMAEP and PMOEP

In the present study, the mineralisation ability of a series of phosphate-containing

soluble polymers and cross-linked gels was investigated after 7 days immersion in

SBF. We have already shown that the different mineralisation outcomes of these

polymers are not simply due to the phosphate content, but a combination of factors,

including: polymer structure and phosphate distribution, as well as the degree of

cross-linking. This study also showed good correlation with previous mineralisation

studies of MAEP and MOEP-grafted ePTFE.

On the surfaces of the PMAEP and PMOEP gels, and the film obtained from soluble

PMAEP, thin layers of CaP mineral were observed. This tile-like morphology has

been observed previously to form on PMAEP-g-ePTFE.2 The Ca/P ratio from EDX

and FTIR analysis indicate that this mineral phase is possibly brushite

(CaHPO4·H2O). Brushite, which is a thermodynamically less stable phase, has been

proposed to be a precursor of the more stable HAP in an in vitro mineralisation

mechanism.53 On the PMOEP gel, a secondary growth, of presumably calcium-

deficient HAP (based on the Ca/P ratio and the crystal morphology), was observed.

A similar secondary growth has also been observed on PMAEP-g-ePTFE.2 Soluble

PMOEP cast on a glass surface formed large amounts of carbonated HAP which was

confirmed by FTIR. This is analogous to our earlier results, a sample where MOEP

was grafted onto ePTFE in methanol.15 Interestingly, PMOEP-g-ePTFE formed in

other solvents did not induce HAP nucleation, but rather different CaP phases, thus

suggesting that only the graft copolymer formed in methanol is structurally similar to

the soluble PMOEP of this study.

FTIR spectra of SBF-treated PMAEP samples also revealed the presence of

carboxylate groups. This indicates that large amounts of the PMAEP side chains

were hydrolysed at the C-O-C ester bond (shown as 2 in Scheme 4.3). Deprotonation

of these carboxyl groups occurs as a consequence of Ca2+ binding similar to that

observed previously for PAA-g-ePTFE 2 and for PAA gel in this study.

206


O

OP O

OHOH

O*

R

n

O

OH

O

R

m

OHO

*

R

l

O

OP O

OHOH

O**

R

n

1

2

Scheme 4.3: Possible hydrolysis sites on the side-chain of the polymer and structure

of resulting polymer (R = H: MAEP, R = CH3: MOEP).

The effect of carboxylic acid groups on mineralisation was tested using a

poly(acrylic acid) (PAA) gel. Formation of similar tile-like CaP mineral layer on this

sample after 7 days immersion in SBF was evident from the SEM image. The FTIR

spectrum of this sample showed a broad phosphate band indicating a mixture of CaP

phases. The broad phosphate peak was also observed on the SBF-treated PAA-g-

ePTFE.2

The carboxylate band was not present in the ATR-FTIR spectra of anye PMOEP

samples. This indicates that the loss of phosphate in these polymers was due to the

cleavage at C-O-P linkage and not at the C-O-C linkage, which was the case in

PMAEP polymers. Therefore, it indicates that PMOEP contains some hydroxyl

groups. Another important observation is that since all PMOEP gels (samples 1C, 1D

and 1E) showed very similar mineralisation patterns, it can be assumed that these

gels are physically and chemically the same. This is especially supported by the fact

that the PEPDTA did not work as a RAFT agent for this monomer.

PMOEP samples generally showed better mineralisation compared to PMAEP.

Chirila et al.22 observed slightly better calcification of P(HEMA-co-MOEP)

compared to that of P(HEMA-co-MAEP) when these polymers were immersed in

SBF for up to 9 weeks, although PHEMA itself calcified much better. The difference

between MAEP and MOEP is that MOEP has an extra methyl group, which makes

this monomer more hydrophobic. The water solubility of MAEP is at least 30 % w/w

(in this study), whereas MOEP is only 4 % w/w. The effect of the hydrophobicity of

these polymers on mineralisation is not completely clear. A previous study of

207


PMOEP-g-ePTFE has shown that the most hydrophobic surface had the best HAP

formation.15

However, the structures of PMAEP and PMOEP are further complicated by the fact

that this study showed that the hydrolysis of the side-chains in these polymers occurs

at different linkages. PMAEP polymers contain large amounts of carboxylic acids,

which can affect the mineral formation observed on PAA. On the other hand,

PMOEP does not contain this group, but has hydroxyl groups. The distribution of

phosphates in these polymers would also be different due to the cleavage of side-

chains.

Large amounts of carbonated HAP were only observed for the soluble PMOEP

sample despite the larger phosphate loss compared to that of the PMOEP gels. This

strongly supports the hypothesis that the flexibility of polymer chains, and hence the

degree of cross-linking, also influences mineralisation and not just the phosphate

content.

4.4.2 SBF Studies of Layer-by-Layer (LbL) Films

Layer-by-Layer films were fabricated using soluble PMAEP and PMOEP with PEI

on silicon wafers. Mineralisation of these films was investigated in SBF for up to 14

days. The PAA LbL film was also studied in SBF.

Both PMAEP and PMOEP films showed sparse and patchy CaP mineral formation

after 7 days in SBF. No further significant mineral growth was observed on these

materials after 14 days. The control PEI film and silicon wafer surface both showed

sparse and patchy calcium mineral formation after 7 days and CaP formation after 14

days. The PAA surface showed some Ca2+ uptake after 7 days in SBF but no

phosphorous was detected.

The PMAEP and PMOEP LbL films results are in contrast to the observed

mineralisation of the films obtained from soluble PMAEP and PMOEP films, which

yielded brushite-like minerals and a thick layer of HAP, respectively, in SBF after 7

days. The PAA gel also formed CaP minerals after 7 days in SBF, whereas the PAA

LbL film did not. AFM images of the PMAEP and PMOEP LbL films showed a

smooth topography indicating a uniform polymer film coverage (described in

208


Chapter 3, Section 3.3.1), although the PMAEP film was slightly rough at the scales

analysed (1×1 μm, and 10×10 μm). Therefore, the patchy CaP formation cannot be

related to patchiness of the films.

It was evident from Infrared Reflection-Absorption Spectroscopy (IRRAS) analysis

of the PMOEP LbL film that a large amount of phosphate groups were interacting

with amine groups and, therefore, there were not many free phosphates available for

Ca2+ chelatation (described in Chapter 3, Section 3.3.1). This was also observed from

the XPS analysis of the PMAEP or PMOEP LbL films after 7 days in SBF.

Therefore, it can be concluded that LbL films do not contain enough free phosphate

groups to promote the nucleation of CaP mineral. These results also seem to support

the conclusion that the phosphates, or carboxylates, interacting with protonated

amines in the LbL films are incapable of chelating Ca2+.

For this study, only one layer of each polycation and polyanion was deposited, and

the layers are not expected to be uniformly distributed. There may be effects of

amines and hydroxyl groups from PEI and the substrate, respectively. In future work,

mineralisation of multi-layer films should be tested.

4.4.3 SBF Studies of Block Copolymers Coupled to Aminated Slides

Block copolymers consisting of PMAEP or PMOEP and PAAEMA fixed on

aminated slides (samples 3A-D) showed sparse, patchy CaP mineral formation after

up to two weeks in SBF. Although the control PAAEMA attached slide (sample 3E)

did not induce any CaP mineral formation, CaP was observed on the aminated slide

(sample 3F) after two weeks. This mineral phase had a distinctive morphology not

seen in any of the phosphate-containing samples. Although amine groups are not as

efficient at inducing CaP growth as phosphate and carboxylic acid groups, CaP

nucleation has previously been observed on amine terminated self assembled

monolayers.3

These results show that sparse mineralisation observed on the block-copolymer-

functionalized aminated slides is not due to the inhomogeneous dispersion of

phosphate groups, since the ToF-SIMS has shown a uniform distribution of these

groups (Chapter 3, Section 3.3.2). It is however possible that the total density of the

209


phosphate groups on the surfaces may not be sufficient enough for mineral

formations. Another likely explanation is that ionic interactions between the

phosphate segment and the protonated amine groups leads to ion pairs that prevent

chelatation of calcium ions from the SBF solution. Although static ToF-SIMS

revealed that the most likely confirmation of sample 3A (P(MOEP-b-AAEMA)) on

the aminated slide was the PAAEMA segment coupled with the amine groups and

the PMOEP segments extending from the surface (Figure 4.10A), that analysis was

done under vacuum. Under wet conditions (e.g. in SBF) deprotonated phosphates

may very well interact with protonated amines on the slide and, hence, once again

prevent Ca2+ chelation (Figure 4.10B).

(A)

PAAEMA PMOEP

(B)

SBF

Figure 4.11: Proposed structures of a P(AAEMA-b-MOEP) block copolymer on an aminated slide (A) in vacuum and (B) in SBF solution.

The SBF results of both LbLs and coupled aminated slides support Matsuda’s

conclusion that the chelation of calcium ions by negatively charged groups plays an

important role in biomaterial mineralisation.3

4.4.4 SBF Studies of Fluorinated Amphiphilic Block Copolymers Adsorbed onto

PTFE Films

SEM images of P(FSm-b-AAn) adsorbed PTFE surfaces (samples 4A-C) did not

show any mineralisation after 7 days. However XPS revealed chelation of Ca2+ ions

by PAA segments as well as the presence of small amounts of phosphorous. Since

LbL films of PEI-PAA showed no phosphorous after 7 days in SBF, the free-

carboxylic acid groups on these modified films are better nucleators.

210


After 14 days, very small and patchy mineral formation was observed on all the

polymer adsorbed samples (samples 4A-C). These minerals were much smaller in

size compared to those formed on the phosphate-containing surfaces. The results

emphasises that carboxylic acid groups are not as efficient nucleators as phosphates.

It is also possible that the density of carboxylic acid groups may not have been high

enough to promote mineralisation.

In literature, carboxyl-terminated SAM surfaces and polymers containing -COOH

have shown an induction period (10 day for COOH-SAM) before apatite growth, and

this study correlates with this. Longer soaking time in SBF or pre-treatment with

Ca2+ ions may be required for these surfaces.

Nevertheless, it was evident that these adsorbed polymers were highly stable in SBF,

even up to two weeks with regular solution changes. Hence fluoropolymer surface

modification using block copolymers containing PFS is promising and further

investigation would be worthwhile.

211


4.5 Conclusion

This study investigated the effects of polymeric features on mineralisation behaviour

in SBF using a series of phosphate-containing polymers, i.e. soluble and cross-linked

PMAEP and PMOEP. The effect of the carboxylic acid groups on mineralisation was

also investigated using poly(acrylic acid) (PAA) gels. The mineralisation of

fabricated surfaces (LbL films with soluble PMAEP and PMOEP polymers,

phosphate containing block copolymers attached to aminated slides, and carboxylic

acid-containing fluorinated block copolymers adsorbed to a PTFE substrate) were

studied.

Both soluble and cross-linked PMAEP samples showed the formation of brushite-

like minerals with tile morphology, whereas PMOEP gels showed secondary growth

of HAP on top of this mineral. The film obtained from soluble PMOEP showed large

amounts of HAP formation after 7 days in SBF. The FITR investigation also

revealed the presence of large amounts of carboxylic acid groups in PMAEP,

indicating cleavage of side-chains at C-O-C ester linkage. The PAA gel containing

carboxylic acid groups also formed a characteristic tile-like mineral layer in SBF

after 7 days. This leads to the conclusion that mineralisation is mainly influenced by

not only the nature of the monomer, but also by the structure of the polymer, as well

as the extent of cross-linking.

Both LbL films and block copolymers attached to aminated slides that contain

phosphate groups showed sparse and patchy mineralisation in SBF for up to two

weeks. Limited mineral growth on these samples highlights the importance of

accessible, ionic phosphate groups for calcium ion chelation and subsequent CaP

nucleation on materials designed for use in implantable materials. Fluorinated block

copolymers adsorbed PTFE films showed even smaller and patchier mineral

formation after two weeks in SBF compared to other fabricated surfaces containing

phosphate groups. This result emphasises that phosphate groups have stronger

tendency for CaP nucleation and mineralisation than the carboxylic acid groups, in

agreement with literature.

212


4.6 References

(1) Tretinnikov, O. N., Kato, K., and Ikada, Y. J Biomed Mater Res 1994, 28, 1365-1373.

(2) Grøndahl, L., Cardona, F., Chiem, K., Wentrup-Byrne, E., and Bostrom, T. J.Mater Sci Mater Med 2003, 14, 503-510.

(3) Tanahashi, M., and Matsuda, T. J Biomed Mater Res 1997, 34, 305-315. (4) Liu, Q., Ding, J., Mante, F.K., Wunder S.L., and Baran, J.R. Biomaterials

2002, 23, 3103-3111. (5) Sikavitsas, V. I., Temenoff, J.S., and Mikos, A.G. Biomaterials 2001, 22,

2581-2593. (6) Mann, S. Biomineralization : principles and concepts in bioinorganic

materials chemistry; Oxford: N.Y., 2001. (7) Wu, W., Zhuang, H., and Nancollas, G.H. J Biomed Mater Res 1997, 35, 93-

99. (8) Elliot, J. C. Structure and chemistry of the apatites and other calcium

orthophosphates; Elsevier: Amsterdam, 1994; Vol. 18. (9) Zapanta-LeGeros, R. Nature 1965, 403-404. (10) Mayer, J., Schlam, R., and Featherstone, J.D.B. J Inorg Biochem 1997, 66, 1-

6. (11) Fanovich, M. A., Castro, M.S., and Lopez, J.M.P. Ceramics Int 1999, 23,

517-522. (12) Wen, H. B., van den Brink, J., de Wijin, J.R., Cui, F.Z., and de Groot, K. J

Crystal Growth 1998, 186, 616-623. (13) Kim, H. M., Miyazaki, T., Kokubo, T., and Nakata, T. Bioceramics 2000, 13,

47-50. (14) Kokubo, T., and Takadama, H. Biomaterials 2006, 27, 2907-2915. (15) Suzuki, S., Grøndahl, L., Leavesley, D., and Wentrup-Byrne, E. Biomaterials

2005, 26, 5303–5312. (16) Toworfe, G. K., Composto, R.J., Shapiro, I.M. and Ducheyne, P.

Biomaterials 2006, 27, 631-642. (17) Hench, L. L. J Am Ceram Soc 1998, 81, 1705-1728. (18) Kato, K., Eika, Y., and Ikada, Y. J Biomed Mater Res 1996, 32, 687-691. (19) Kamei, S., Tomita, N., Tamai, S., Kato, K., and Ikada. J Biomed Mater Res

1997, 37, 384-393. (20) Stancu, I. C., Filmon, R., Cincu, C., Marculescu, B., Zaharia, C., Tourmen,

Y., Basle, M.F., and Chappard, D. Biomaterials 2004, 25, 205-213. (21) Stancu, I. C., Filmon, R., Grizon, F., Zaharia, C., Cincu, C., Basle, M.F. and

Chappard, D. J Biomed Mater Res 2004, 69A, 584-589. (22) Chirila, T. V., Zainuddin, Hill, D.J.T., Whittaker, A.K. and Kemp, A. Acta

Biomaterialia 2007, 3, 95-102. (23) Ramila, A., and Vallet-Regi, M. Biomaterials 2001, 22, 2301-2306. (24) Vallet-Regi, M., Pe´ rez-Pariente., J., Izquierdo-Barba, I. and Salinas, A.J.

Chem Mater 2000, 12, 3770-3775. (25) Izquierdo-Barba, I., Salinas, A.J. and Vallet-Regi, M. J Biomed Mater Res

2000, 51, 191-199. (26) Siriphannon, P., Kameshima, Y., Yasumori, A., Okada, K. and Hayashi, S. J

Biomed Mater Res 2002, 60, 175-185. (27) Lu, X., Leng, Y., Zhang, X., Xu, J., Qin, L. and Chan, C.-W. Biomaterials

2005, 26, 1793-1801.

213


(28) Johnson, M. S., and Nancollas, G.H. Critical reviews in oral biology and medicine 1992, 3, 61-82.

(29) Cheng, P. T. Calcif Tissue Int 1985, 37, 91-94. (30) Lu, X., and Leng, Y. Biomaterials 2005, 26, 1097-1108. (31) Skipper, L. J., Sowrey, F.E., Pickup, D.M., Fitzgerald, V., Rashid, R., Drake,

K.-O. Lin, Z., Saravanapavan, P., Hench, L.L., Smith, M.E. and Newport, R.J. J Biomed Mater Res 2004, 70A, 354-360.

(32) Lin, K. S. K., Tseng, Y.-H., Hsu, Y.-C., Yang, C.-M. and Chan, J.C.C. Chem Mater 2005, 17, 4493-4501.

(33) Zhu, P., Masuda, Y., and Koumoto, K. Biomaterials 2004, 25, 3915-3921. (34) Chirila, T. V., and Zainuddin. React Funct Polym 2007, 67, 165-172. (35) Mucalo, M. R., Yokogawa, Y., Toriyama, M., Suzuki, T., Kawamoto, Y.,

Nagata, F. and Nishizawa, K. J Mater Sci, Mater Med 1995, 6, 597-605. (36) Mucalo, M. R., Yokogawa, Y., Suzuki, T., Kawamoto, Y., Nagata, F. and

Nishizawa, K. J Mater Sci, Mater Med 1995, 6, 658-669. (37) Li, S., Liu, Q., de Wijn, J., Wolke, J., Zhou, B., de Groot, K. J Mater Sci,

Mater Med 1997, 543-549. (38) Li, S. H., Liu, Q., de Wijn, J.R., Zhou, B.L., and de Groot, K. Biomaterials

1996, 18, 389-395. (39) Yokogawa, Y. J Mater Sci, Mater Med 1997, 8, 407-412. (40) Swart, J. G. N., Driessen, A.A., and DeVisser, A.C. In Hydrogels for medical

and related applications. ACS Symposium Series; Andrade, J. D., Ed.; ACS: Washington D.C., 1976; Vol. 31, pp 151-161.

(41) Kawashita, M., Nakao, M., Kim, H.-M., Beppu, T., Miyamoto, T., Kokubo, T. and Nakamura, T. Biomaterials 2003, 24, 2477-2484.

(42) Oyane, A., Uchida, M., Choong, C., Triffitt, J., Jones, J. and Ito, A. Biomaterials 2005, 26, 2407–2413.

(43) Miyazaki, T., Ohtsuki, C., Akioka, Y. and Tanihara, M. J Mater Sci, Mater Med 2003, 14, 569-574.

(44) Smetana, K., Šol, M., Novák, M. and Daneš, J. Biomaterials 1995, 17, 1563-1566.

(45) Ohtsuki, C., Kamitakahara, M. and Miyazaki, T. J Tissue Eng Regen Med 2007, 1, 33-38.

(46) Oyane, A., Uchida, M., Yokoyama, Y., Choong, C., Triffitt, J. and Ito, A. J Biomed Mater Res 2005, 75A, 138–145.

(47) Brown, J. L., Nair, L.S., Bender, J., Allcock, H. and Laurencin, C.T. Materials Letters 2007, 61, 3692-3695.

(48) Grøndahl, L., Cardona, F., Chiem, K., Wentrup-Byrne, E. J Appl Poly Sci 2002, 86, 2550-2556.

(49) Wentrup-Byrne, E., Grøndahl, L., and Suzuki, S. Polym Int 2005, 54, 1581–1588.

(50) Fowler, B. O., Moreno, E.C., and Brown, W.E. Arch Oral Biol 1966, 11, 477-492.

(51) Zhou, F., and Huck, W.T. Chem Commun 2005, 5999-6001. (52) Lin-Vien, D., Colthup, N.B., Fateley, W.G., and Grasselli, J.G. The handbook

of Infrared and Raman characteristic frequencies of organic molecules; Academic Press: London, 1991.

(53) Zhang, L. J., Liu, H.G., Feng, X.S., Zhang, R.J., Zhang, L., Mu, Y.D., Hao, J.C., Qian, D.J., and Lou, Y.F. Langmuir 2004, 20, 2243-2249.

214


215



Based on the observation that the incorporation of negatively-charged functional

groups, such as phosphate groups, have shown great promise for improving the bone-

bonding ability of polymeric materials by providing mineralisation nucleation sites,

the current study was aimed at investigating new approaches for the surface

modification of materials for biomedical applications. This was achieved using well-

defined functional polymers synthesised by controlled living radical polymerisation.

The in vitro mineralisation behaviour of a suite of phosphate-containing polymers

and functionalised surfaces containing phosphate and carboxylic acid groups was

subsequently studied.

5.1 Chapter 2

A suite of well-defined phosphate- and fluorine-containing polymers was

successfully synthesised using RAFT-mediated polymerisation. Despite the fact that

during the course of the investigation it was discovered that the MAEP and MOEP

monomers contain large amounts of diene impurities, it was still possible to prevent

cross-linking by limiting the molecular weight of the desired polymers. Since the

monomers also contain orthophosphoric acids, hydrolysis of the phosphate groups

were observed for both the gel and soluble polymers; with a more pronounced effect

shown for the soluble polymers. Moreover, both soluble and cross-linked PMAEP

homopolymers were found to contain large amounts of carboxylic acid groups

indicating hydrolysis at the C-O-C ester linkages. To immobilise these polymers onto

aminated slides, keto functionalities were successfully incorporated by block

copolymerisation with AAEMA. Well-defined fluorinated block copolymers were

synthesised using the RAFT technique for subsequent use as surfactants for the

surface modification of fluorinated bulk polymers. The RAFT-mediated

homopolymerisation of TFPA and TFPMA proceeded in a well-controlled manner,

whereas the molecular weights of PFS deviated significantly from the theoretical

value. This could be attributed to the self-initiation of FS and the long polymerisation

215


times required. Chain extension of these fluorinated polymers with a range of

monomers, FS, tBA, AAEA or AAEMA was successfully carried out. Finally,

amphiphilic block copolymers containing carboxyl groups were synthesised by the

hydrolysis of tBA groups on the PtBA segments. The P(TFPMA-b-AAEMA) block

copolymer was reacted with small biomolecules (glycine or L-phenylalanyl glycine)

as a model system for the creation of biologically active surfactants.

5.2 Chapter 3

Three surface fabrication techniques were investigated: LbL assembly, coupling

reactions of keto-containing block copolymers onto aminated slides, and the

adsorption of fluorinated polymers onto PTFE. All of these techniques were found to

be robust for the polymers studied. An added advantage of these approaches is their

applicability to a wide range of materials. Surfaces containing carboxyl and

phosphate groups were produced by the LbL assembling of PEI and PAA, PMAEP

or PMOEP. The thickness of the PAA film was tuned by altering the pH of the

deposition solutions. AFM images showed that the topography of the PEI-PMAEP

LbL was patchy due to the presence of protonated carboxylate groups on the PMAEP

at low pH. On the other hand, the PEI-PMOEP LbL was found to be smooth.

Coupling reactions of the block copolymers consisting of PAAEMA and PMAEP or

PMOEP onto the aminated slides through reductive amination of the keto groups

with amines were successfully carried out. Regarding the conformation of the

attached block copolymer, ToF-SIMS revealed that the phosphate-containing

segments were more extended away from the surface whereas the PAAEMA

segments were attached to the slides. PFS was found to adsorb strongly onto PTFE in

the solvents studied, whereas PTFPA and PTFPMA were only adsorbed when DCM

was used as a solvent. The results were interpreted by considering the hydrophobicity

of the polymers, the solvent effects on the swelling of the PTFE and the solubility of

the polymers. P(FS-b-AA) with two PAA chain lengths were also adsorbed onto

PTFE. The contact angle measurements of these surfaces showed large hysteresis

indicating reorganisation of the flexible chains under the hydrated environment.

216


5.3 Chapter 4

The mineralisation behaviour of several phosphate-containing polymers and

fabricated surfaces in SBF is discussed in this chapter. Both the soluble and gel

PMAEP polymers formed brushite-like minerals that were also found on the PAA

gel. The PMOEP gel showed a secondary growth of HAP on top of this initially

formed mineral phase. Large amounts of HAP formation was observed on the film

obtained from soluble PMOEP compared to the cross-linked gel, even though this

polymer had undergone greater loss of phosphate groups than the gel. This indicates

that the flexibility of the polymer chains affects mineralisation. Both LbL films and

block copolymers attached to aminated slides showed sparse and patchy

mineralisation in SBF even after two weeks. The P(FS-b-AA) adsorbed-PTFE

surface also showed only small amounts of mineral formation even after two weeks.

For all these systems the limited mineralisation could be due to the density of the

functional groups. In the case of LbL’s and the attached block copolymer, it can also

be a result of limited accessibility of ionic phosphate (or carboxylate) groups for

calcium chelation and subsequent mineralisation. For the attached block copolymer it

is possible that under wet conditions, deprotonated phosphates may very well interact

with protonated amines on the slide and, hence, once again prevent Ca2+ chelation. In

addition, for polymers containing carboxylic acid groups (i.e. PAA and hydrolysed

PMAEP) the limited mineralisation could also be due to the immersion time since

such groups have been shown to have some induction period before nucleation.

5.4 General Discussion

Most of the polymers synthesised in the current study were shown to be excellent

candidates for the surface modification of a range of polymeric materials. It was

demonstrated that with judicious choice of surface modification techniques, using

well-defined polymers, it is possible to produce surfaces that, at least under in vitro

conditions, respond differently to a biomineralisation environment. These results are

expected to find use in a wide range of biomaterials applications.

A number of key findings in the search for improved in vitro mineralisation of

surfaces were made. Firstly, the mobility of the polymer chains appeared to have a

217


significant impact on the mineralisation outcome with the highly mobile chains

performing the best. Secondly, the availability of free phosphate groups to initially

chelate calcium ions was also found to be important and can explain the reason for

the pure outcome for the LbL assemblies and the block copolymer-attached slides. It

is possible that the density of the phosphate groups also influences the outcome

although this cannot be concluded definitely from the present data. Finally,

phosphate groups are known to exhibit stronger propensity for nucleation of CaP

than the carboxylate groups and this mineralisation study of homopolymers, LbLs

and flurosurfactants on PTFE supports this statement.

The surface modification techniques used in this study were successful in

immobilising well-defined polymers. Although LbL assembly has been previously

used in biomaterials science this study is unique in the use of well-defined

phosphate-containing polyelectrolytes. It is also the first to report the investigation of

the conformation of the attached block copolymers using ToF-SIMS. The use of

fluorinated surfactants synthesised by LRP in this area of science is novel and offers

a new direction to modifying fluoropolymer surfaces.

It is important to strongly stress that the both MAEP and MOEP monomers contain

large amounts of diene impurities and free orthophosphoric acid, and therefore the

polymerisation conditions will affect the hydrolysis of the phosphate groups as well

as preferential incorporation of dienes. Therefore, the obtained polymers must be

characterised solely for phosphate content as well as the state of the phosphates. The

contradictory mineralisation results of PMOEP-containing polymers in the literature

could be explained by this fact. This study supports the use of these monomers for

surface modifications of biomaterials to enhance bone bonding ability.

5.5 Future work

In order to optimise the syntheses developed in this thesis it would be advisable to

explore the possibility of developing a purification protocol for both the MAEP and

MOEP monomers. One approach would be to use solvent extraction to remove the

dienes and orthophosphoric acids. It would be expected that removal of the dienes

will further improve the mineralisation outcomes. The LbL and block copolymer

218


attachment techniques are applicable to any surface that contains a variety of

functional groups and amine groups, respectively. These groups can be introduced

onto the biomaterial surface by techniques such as plasma treatment and aminolysis.

A search for alternate immobilisation techniques of phosphate-containing polymers

to ensure available free phosphate groups would also be useful. One possibility for

an immobilisation technique is by using the polymer end functional groups created

by the living radical polymerisation. These surfaces will allow us to investigate the

effect of chain length of the phosphate-containing polymers on mineralisation using

different molecular weight polymers.

Regarding with RAFT polymerisation of FS, further analyses of PFS would provide

more comprehensive arguments on the effect of the self-initiation on the deviation of

the molecular weights. These include using GPC with multiple angle laser light

scattering (MALLS) detectors to obtain the absolute molecular weights. It has also

been suggested to use electrospray ionization mass spectrometry (ESI-MS) to

identify the end groups before and after storage in THF and to confirm the UV-vis

spectrometry results.

Where the fluoropolymer surfactants are concerned, future work could include

investigation of the stability of the adsorbed fluorinated surfactants under applied

shear stress. Optimisation of the adsorption conditions is needed for chain

entanglements to occur which provide stable interlocking of the surfactants.

Parameters to be investigated include the chain length of the fluorinated segments,

the adsorption solvent, the washing solvent as well as the adsorption temperature.

The dependence of the PAA segment chain length on the mineralisation is another

factor warranting investigation.

Fluorinated block copolymers with carboxylic acid groups offer the possibility of

incorporating numerous biomolecules including cell adhesion peptides and growth

factors as well as phosphate groups. It is important in biomaterial applications to

investigate cell behaviour (e.g. attachment, growth and proliferation) on these

surfaces. The stability of the fluorinated surfactants under biological conditions is

also warranted.

219


Since these adsorbed block copolymers are capable of reorganising depending on the

environmental conditions, LbL formation on top of these surfaces should provide

even more stable films. This will also allow the incorporation of biomolecules

through ionic interactions in addition to embedding them into LbLs produced from

biodegradable polymers.

The surface modification techniques presented in this study using well-defined

polymers with functional groups are suitable for use with a variety of materials and

hence in a wide range of applications. The study has also opened up the greater

opportunities to design precise surface properties in a controlled manner which

allows us to study the complicated process of mineralization as well as other

biological events such as protein adhesion and cell behaviour. Another advantage of

these modified surfaces is that further functionalisations such as the incorporation of

biomolecules to control specific cell and tissue responses can be carried out.

Finally, in vivo mineralisation tests using animal models would confirm results from

in vitro work which showed formation of different mineral phases on PMAEP and

PMOEP.

220

Documents

Shuko Suzuki Thesis (PDF 6MB)