Laser Surface Treatment of Bio-Implant Materials [1 ed.] 0470016876, 9780470016879

Citation preview

Laser Surface Treatment of Bio-Implant Materials

Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

Laser Surface Treatment of Bio-Implant Materials

Liang Hao Loughborough University, UK Jonathan Lawrence Nanyang Technological University, Singapore

Copyright ß 2005

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone (þ44) 1243 779777

Email (for orders and customer service enquiries): [email protected] Visit our Home Page on www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (þ44) 1243 770620. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1 Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Library of Congress Cataloging-in-Publication Data Hao, Liang. Laser surface treatment of bio-implant materials/Liang Hao, Jonathan Lawrence. p. cm. Includes bibliographical references and index. ISBN-13 978-0-470-01687-9 (cloth : alk. paper) ISBN-10 0-470-01687-6 (cloth : alk. paper) 1. Biomedical materials–Effect of lasers on. 2. Biomedical materials–Biocompatibility I. Lawrence, J. (Jonathan), 1970-. II. Title. [DNLM: 1. Lasers–therapeutic use. 2. Laser Surgery–methods. 3. Biocompatible Materials. WB 117 H252L 2005] R857.M3H37 2005 2005019354 6100 .280 4–dc22 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13 978-0-470-01687-9 ISBN-10 0-470-01687-6 Typeset in 11/13pt Palatino by Thomson Press (India) Limited, New Delhi, India. Printed and bound in Great Britain by Antong Rowe Ltd, Chipperham, Wiltshire This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

To Yan-jun and My Parents For the world they bring to me.

To Louise and Ethan, Always there; beyond compare.

Contents Acknowledgements

xiii

Introduction

xv

Bio-Implants and Surface Modification of Biomaterials Wettability in Biomaterials Science and Modification Techniques Lasers and Their Application for Modification of the Biomaterials 1

2

Bioactivity and Biointegration of Orthopaedic and Dental Implants 1.1 Introduction 1.1.1 Biocompatibility 1.1.2 Host Response to Biomaterials 1.1.3 In vitro Models of Biological Response to Implants 1.2 Bioactivity of Bone Implants 1.2.1 The Mechanism of Apatite Formation 1.2.2 Functional Group 1.3 Biointegration of Orthopaedic and Dental Implants 1.3.1 Osseointegration 1.3.2 Bone Cell Adhesion [44] 1.3.3 Osteoblast–Material Interactions 1.4 Controlling the Bone–Implant Interface 1.4.1 Physicochemical Methods 1.4.2 Biochemical Methods [9] Surface Modification of Biomaterials 2.1 Introduction 2.1.1 Orthopaedic and Dental Implants 2.1.2 Surface Properties of Biomaterials

xv xvi xvii

1 1 2 2 2 3 3 4 5 5 5 6 6 7 7 11 11 11 12

Contents

viii

2.2

2.3

2.4

2.5

3

2.1.3 Surface Analysis of Biomaterials Ceramic Implants [65] 2.2.1 Nearly Bioinert Ceramics [64, 69] 2.2.2 Alumina 2.2.3 Zirconia Ceramics Metallic Implants 2.3.1 Mechanical Properties 2.3.2 Corrosion Surface Modification of Biomaterials 2.4.1 Introduction 2.4.2 Radiation Grafting and Photografting [76] 2.4.3 Plasma Surface Modification of Biomaterials 2.4.4 Ion Beam Processing 2.4.5 Other Methods [65] Laser Surface Modification of Biomaterials 2.5.1 Introduction 2.5.2 Laser Patterning and Microfabrication 2.5.3 Pulsed Laser Deposition (PLD) of Biocompatible Ceramics 2.5.4 Matrix-Assisted Pulsed Laser Evaporation and MAPLE Direct Write 2.5.5 Other Laser Surface Treatments

Wettability in Biomaterials Science and Modification Techniques 3.1 Introduction 3.2 Wettability, Adhesion and Bonding: Theoretical Background 3.2.1 The Wetting Process 3.2.2 Contact Angle and Work of Adhesion 3.2.3 Surface Energy and the Dispersive/Polar Characteristics 3.2.4 Physical Bonding 3.2.5 Mechanical Bonding 3.2.6 Chemical Bonding 3.3 Wettability in Biomaterial Science 3.3.1 Biomaterial Interfaces [110] 3.3.2 Tensiometry 3.3.3 Interfacial Biophysics 3.3.4 Thermodynamic Concepts in Biomaterials Science 3.4 Current Methods of Wettability Modification 3.4.1 Chemical Reactions 3.4.2 Plasma Surface Modification

12 13 13 14 14 15 16 16 17 17 17 18 18 19 19 19 20 20 21 21

23 23 23 23 24 25 28 28 28 29 29 29 30 31 33 33 33

Contents

ix

3.4.3 3.4.4 3.4.5 3.4.6 3.4.7 3.5 Laser 3.5.1

Ion Beam Processing Radiation Grafting UV and Ozone Corona Discharge Electrowetting Wettability Characteristics Modification Laser Surface Modification of Ceramic Materials for Improved Wettability 3.5.2 Laser Surface Modification of Metallic Materials for Improved Wettability

4

5

CO2 Laser Modification of the Wettability Characteristics of Magnesia Partially Stabilised Zirconia 4.1 Introduction 4.2 Experimental Procedures 4.2.1 Material Specifications 4.2.2 CO2 Laser Experimental Arrangement 4.2.3 Morphological, Chemical and Phase Analysis Procedures 4.2.4 Wettability Characteristics Analysis Procedure 4.3 The Effects of CO2 Laser Radiation on Wettability Characteristics 4.3.1 Contact Angle 4.3.2 The Effect of Surface Oxygen Content 4.3.3 The Effect of Surface Roughness 4.3.4 The Effects of Solidified Microstructures and Surface Melting on Wettability Characteristics 4.4 Surface Energy and Its Component Parts 4.5 Identification of the Predominant Mechanisms Active in Determining Wettability Characteristics 4.6 The Role Played by Microstructures in Terms of Crystal Size and Phase in Effecting Surface Energy Changes 4.6.1 The Role of Crystal Size on Surface Energy 4.6.2 The Role of Phase Change on Surface Energy 4.7 Investigation of Wettability and Work Adhesion Using Physiological Liquids 4.8 Summary In vitro Biocompatibility Evaluation of CO2 Laser Treated Magnesia Partially Stabilised Zirconia 5.1 Introduction 5.2 Sample Preparation 5.3 Bone-Like Apatite Formation

34 34 34 35 35 35 35 36 37 37 38 38 39 39 40 41 41 42 43 45 47 52 56 56 61 61 63

65 65 67 67

Contents

x

5.4

5.5

5.6 5.7

6

5.3.1 Experimental Procedures 5.3.2 Spectral Analysis and Hydroxyl Group 5.3.3 The Correlation between OH Groups and Wettability Characteristics 5.3.4 The Effects of CO2 Laser Treatment on the MgO–PSZ in Simulated Body Fluids Protein Adsorption 5.4.1 Experimental Procedures 5.4.2 Albumin and Fibronectin Adsorption on CO2 Laser Treated MgO–PSZ Osteoblast Cell Response 5.5.1 Experimental Procedures 5.5.2 Osteoblast Cell Response on the CO2 Laser Treated MgO–PSZ 5.5.3 The Effect of CO2 Laser Treatment on the Osteoblast Cell Response Predictions for Implantation in an in vivo Clinical Situation Summary

The Effects of CO2 Laser Radiation on the Wettability Characteristics of a Titanium Alloy 6.1 Introduction 6.2 Experimental Procedures 6.2.1 Material Specifications and Preparation 6.2.2 CO2 Laser Surface Treatment 6.2.3 Morphological, Chemical and Phase Analysis Procedures 6.2.4 Wettability Characteristics Analysis Procedure 6.3 The Effects of CO2 Laser Radiation on Wettability Characteristics 6.3.1 Contact Angle 6.3.2 Morphological Analysis and Its Effect on Wettability Characteristics 6.3.3 Phase and Chemical Analysis and Its Effects on Wettability Characteristics 6.4 Surface Energy and Its Component Analysis 6.5 Identification of the Predominant Mechanisms Active in Determining Wettability Characteristics 6.6 Investigation of Wettability and Work Adhesion Using Physiological Liquids 6.7 Summary

68 69 71 73 75 76 77 80 81 83 89 95 98

99 99 101 101 102 102 102 103 103 103 105 109 111 114 116

Contents

7

8

In vitro Biocompatibility Evaluation of CO2 Laser Treated Titanium Alloy 7.1 Introduction 7.2 Sample Preparation 7.3 Bone-Like Apatite Formation on Titanium Alloys 7.3.1 Experimental Procedures 7.3.2 The Effects of CO2 Laser Treatment on the Ti–6Al–4V in Simulated Body Fluid 7.4 Protein Adsorption 7.4.1 Experimental Procedures 7.4.2 Albumin and Fibronectin Adsorption on CO2 Laser Treated Titanium Alloy 7.5 Osteoblast Cell Adhesion 7.5.1 Experimental Procedure 7.5.2 Osteoblast Cell Response on CO2 Laser Treated Titanium Alloy 7.5.3 The Effect of CO2 Laser Treatment on the Osteoblast Cell Response 7.6 Predictions for Implantation in an in vivo Clinical Situation 7.7 Summary Enquiry into Possible Generic Effects of the CO2 Laser Treatment on Bone Implant Biomaterials 8.1 Introduction 8.2 Ascertaining the Generic Effects of CO2 Laser Treatment on Bioinert Ceramics 8.2.1 Experimental Procedures 8.2.2 Modification of the Surfaces Properties and Wettability Characteristics of a Y–PSZ Bioinert Ceramic 8.2.3 Identification of the Predominant Mechanism Active in the Wettability Characteristics Modification of a Y–PSZ Bioinert Ceramic 8.2.4 Generic Effects of CO2 Laser Treatment on the Wettability Characteristics of Bioinert Ceramics 8.2.5 CO2 Laser Induced Effects on the Cell Response on a Y–PSZ Bioinert Ceramic 8.2.6 Generic Effects of CO2 Laser Treatment on the Cell Response on Bioinert Ceramics 8.3 Ascertaining the Generic Effects of CO2 Laser Treatment on Metal Implants 8.3.1 Experimental Procedures

xi

117 117 119 120 121 121 123 123 124 127 127 128 131 135 138

141 141 142 143 144

149 150 153 157 157 158

Contents

xii

8.3.2 Modification of Surfaces Properties and Wettability Characteristics of a 316 LS Stainless Steel 8.3.3 Identification of the Predominant Mechanism Active in the Wettability Characteristics Modification of a 316 LS Stainless Steel 8.3.4 Generic Effects of CO2 Laser Treatment on the Wettability Characteristics of Biometals 8.3.5 CO2 Laser Induced Effects on Protein Adsorption and the Cell Response on a 316 LS Stainless Steel 8.3.6 Generic Effects of CO2 Laser Treatment on Protein Adsorption and the Cell Response on Biometals 8.4 Summary

159

166 168 170 175 176

Conclusions

179

References

185

Index

209

Acknowledgements First, we gratefully appreciate the understanding and encouragement of our families who live on opposite sides of the world in China and England. We are indebted to all the technicians in the Materials Laboratory at Nanyang Technological University for their advice and assistance on optical microscopy, XRD, SEM, EDX, contact angle analysis, XPS analysis and sample preparation. Many thanks also to the Doctoral Research students in the Materials Laboratory for sharing their knowledge on material treatment and analysis. We acknowledge the tremendous contribution made by the first-rate work of the 2003–2004 Final Year Project students: Y.F. Phua, T.L. Tan, M.W. Koo and T.H. Wang. On numerous occasions we obtained superb instruction and assistance from Mr Ma Dong Rui on the subject of osteoblast cell culture, for which we are extremely grateful.

Introduction Bio-Implants and Surface Modification of Biomaterials There is archaeological evidence that lost teeth were replaced by handcarved ivory or wood ‘implants’ as long ago as ancient Egypt. In the early 1950s Swedish orthopaedic surgeon Per Ingvar Branemark began studying the healing process of titanium anchoring screws, which proved to be a seminal point for modern dental and orthopaedic implants [1]. His work showed that fusion between bone and the titanium implant could take place, a phenomenon he called ‘osseointegration’. Orthopaedic implants to treat joint degradation due primarily to osteoarthritis, osteoporosis or injury are now commonplace. These include hip (around 325 000 US implants in 2001) and shoulder, wrist and knee (around 300 000 US implants in 2001). Biomaterial applications make use of all classes of materials, metals, ceramics, polymers and composite. These are divided roughly into three user types [2]: (a) inert or relatively inert with minimal host response; (b) bioactive, which actually stimulates bonding to the surrounding tissue; and (c) biodegradable, which resorb in the body over a period of time. Events leading to integration of an implant into bone, which in turn determine the performance of the device, take place largely at the tissue– implant interface. The main requirements for a biomaterial to function properly in an osseous site include good biocompatibility favouring bone apposition, adequate mechanical properties and the ability to ensure skeletal functions [3–5]. Bioactivity and biointegration are the two essential aspects of these interactions. Bioactivity and the maintenance of skeletal functions are usually attributed to the ability to induce an apatite layer on a material’s surface in physiological conditions [6–8]. The close apposition between bone and an implant surface, or osseointegration, presented as the ability to promote bone cells anchorage, attachment, spreading, growth and differentiation [9, 10], is another key factor for successful implantation of a biomaterial for dental and orthopaedic applications [3–5].

xvi

Introduction

Biointegration is the ideal outcome expected of an artificial implant. This implies that the phenomena that occur at the interface between the implant and host tissues do not induce any deleterious effects such as chronic inflammatory response or formation of unusual tissues. It is, therefore, of paramount importance to design biomaterials used in implants with the best surface properties. Meanwhile, these biomaterials must possess bulk properties that meet other requirements, especially mechanical properties, in order to function properly in a bioenvironment. As it is quite difficult to design biomaterials fulfilling both needs, a common approach is to fabricate biomaterials with adequate bulk properties followed by a special treatment to enhance the surface properties. Hence surface modification of biomaterials is becoming an increasingly popular method to improve device multifunctionality, tribological and mechanical properties, as well as biocompatibility of artificial devices while obviating the needs for large expenses and a long time to develop brand new materials [11]. Materials can be surface modified by using biological or physicochemical methods.

Wettability in Biomaterials Science and Modification Techniques The wetting of a surface by a liquid and the ultimate extent of spreading of that liquid are very important aspects of practical surface chemistry. Even with all the new information of the last 20 years, however, there still remains a great deal to learn about the mechanisms of movement of a liquid across a surface and the factors that govern such movement [12]. Biomaterial scientists have long sought a single, material-related parameter that effectively measures biocompatibility and might serve as a practical design guide. The theories of surface energy and wetting for such parameters present an attractive means to do this as surface properties are important determinants of a biomaterial function [12]. The ability to control the surface wettability of solid substrates is important in many situations. Various surface processes are used for modifying the surfaces of materials depending on the actual material and the application. A number of laser-based techniques for altering the wettability characteristics of engineering materials have been investigated [13]. Surface sensitivity is of critical importance in biomaterials surface science because only the uppermost layers are in direct physicochemical contact with the biological environment; consequently only the upper few molecular layers determine biocompatibility. Thus the interfacial chemistry of concern to biomaterial scientists is determined by material composition within the upper nanometre or so. Tensiometry encompasses a broad range of related ‘wetting’ techniques that measure surface energy. These include the observation of contact angles, which is perhaps the most familiar and widely

Lasers and Their Application for Modification of the Biomaterials

xvii

applied method. Tensiometric methods have singular potential in biomaterials surface science based on the criteria of surface sensitivity, kind of analytical information obtained and relevance of that information to biomedical problems. First, with respect to surface sensitivity, wetting measurements are sensitive only to the upper 0.5 nm or so of a surface [14, 15] and are therefore among the most surface-sensitive techniques available. Second, tensiometry directly measures the fundamental energy at an interface that drives important processes such as adsorption and adhesion. This kind of information must be particularly pertinent to biomaterial problems because of the overwhelming importance of protein adsorption and cell/tissue adhesion. Third, wetting measurements can be made using proteinaceous saline solutions that are particularly relevant to biomedical applications. Special high-vacuum preparation techniques that might introduce experimental artefacts are not required. Various methods are used to improve the surface wettability of materials and their adhesion to other materials. Hydrophilicity is a characteristic of materials exhibiting an affinity for water. Hydrophilic literally means ‘waterloving’ and such materials readily adsorb water. Hydrophobic describes materials possessing a characteristic that has the opposite response to water interaction compared to hydrophilic materials. Smaller water contact angles correspond to more hydrophilic surfaces and higher surface free energies. At present, the processes available to engineers for the modification of a material’s wettability characteristics are invariably complex and consequently somewhat difficult to control. Lasers, on the other hand, can offer the user not only an exceedingly high degree of process controllability but also a great deal of process flexibility. There is a growing amount of published work that testifies to the potential of lasers for altering the surface properties of materials in order to improve their wettability characteristics. Laser radiation was found to effect significant changes in the wettability characteristics of materials. Lawrence and Li [13] have amply demonstrated the practicability of employing different types of lasers to effect changes in the wettability characteristics of composites and ceramics, metals and plastics for improved adhesion and bonding.

Lasers and Their Application for Modification of the Biomaterials The laser has some unique properties for surface treatment. The electromagnetic radiation of a laser beam is absorbed within the first atomic layers for opaque materials, such as metals, and there are no associated hot gas jets, eddy currents or even radiation spillage outside the optically defined beam area. In fact, the applied energy can be placed precisely on the surface only where it is needed. Thus it is a true surface heater and a unique tool for

xviii

Introduction

surface engineering. Common advantages of laser surfacing compared to alternatives are [16]:
chemical cleanliness;
controlled thermal penetration and, therefore, distortion;
controlled thermal profile and, therefore, shape and location of the heat affected region;
less after-machining, if any, is required;
remote noncontact processing is usually possible; and
relatively easy to automate. Surface treatment is a subject of considerable interest at present as it seems to offer the chance to save strategic materials or to allow improved components with idealised surfaces and bulk properties. At present, the lasers are being used in the following surface modifications of the biomaterials:
Laser patterning and microfabrication
Pulsed laser deposition (PLD) of biocompatible ceramics
Matrix-assisted pulsed laser evaporation (MAPLE) and MAPLE direct write (MDW)
Laser surface treatment for improving corrosion
Laser grafting
Laser treatment of plasma sprayed hydroxyapatite coatings However, little work has been carried out to investigate employing lasers to modify the surface properties of biomaterials in order to improve their biocompatibility. Having said that, it is recognised within the currently published work that laser irradiation of material surfaces can affect changes in the cell adhesion on biomaterials. Lately, several publications have investigated the modification of biocompatibility of a biomaterial’s surface following laser irradiation. A CO2 pulsed laser was used to graft a polymer [17] and a rubber [18]. The results showed a marked reduction of the platelet adhesion and aggregation for the modified polymer surface and cell attachment, with a greater degree of spreading and flattening on the unmodified rubber surface. L929 fibroblast cells attached and proliferated extensively on the CO2 and KrF laser treated films [19] in comparison with unmodified PET (polyethylene teraphthalate), with surface morphology and wettability being found to affect cell adhesion and spreading. However, so far, no work has investigated the use of laser surface modification of bioinert ceramics and biograde metals for improved biocompatibility. With a view to improve the biocompatibility of the dental and orthopaedic implant, a CO2 laser was used to modify the surface properties of the widely used bioinert ceramics and biograde metals. By varying the CO2 laser power densities, the work investigated how the CO2 laser affected the surface

Lasers and Their Application for Modification of the Biomaterials

xix

properties of magnesia–partially stabilised zirconia (MgO–PSZ), a bioinert ceramic, such as morphology, surface roughness, surface oxygen content and rapidly solidified microstructure. It particularly analysed the change in the wettability characteristics and basic mechanisms governing this modification [20–22], since wettability is widely recognised as an important determinant of cell adhesion and biomaterial’s function. Then, the bioactivity evaluation of the MgO–PSZ was conducted to find whether the bone-like apatite could form on the MgO–PSZ following CO2 laser irradiation in the stimulated body fluid and what were the functional groups for apatite nucleation [23]. Furthermore, it investigated how the CO2 laser modified surface properties influence the protein adsorption [24, 25] and osteoblast cell adhesion [26] that would manipulate the biointegration between the implant and tissue. Moreover, the effects of the CO2 laser on the modification of the wettability characteristics of a titanium alloy (Ti–6Al–4V) and thereof the adhesion with the simulated physiological liquids were analysed. Further, it observed the apatite formation on the CO2 laser treated Ti–6Al– 4V alloy after soaking in simulated body fluid and the significant difference of albumin and fibronectin adsorption between the untreated sample and CO2 laser treated samples. More importantly, osteoblast cell adhesion and proliferation performed on the CO2 laser treated Ti–6Al–4V was compared with untreated titanium alloy. With the aim of establishing the laser as the innovative technique for the surface modification of the biomaterials, the generic effects of CO2 laser irradiation on the biocompatibility of a yttria–partially stabilised zirconia (Y–PSZ) and a biograde stainless steel are investigated. Similar effects of the CO2 laser treatment on the wettability characteristics, protein adsorption and osteoblast cell were observed on the Y–PSZ and stainless steel. The study therefore proved the generic and great potential application of the laser surface process of widely used bioinert ceramic and biograde metals.

1 Bioactivity and Biointegration of Orthopaedic and Dental Implants 1.1 Introduction Events leading to integration of an implant into bone, which in turn determine the performance of the device, take place largely at the tissue– implant interface. The main requirements for a biomaterial to function properly in an osseous site include good biocompatibility favouring bone apposition, adequate mechanical properties and the ability to assure skeletal functions [3–5]. Bioactivity and biointegration are the two essential aspects of these interactions. Bioactivity and the maintenance of skeletal functions are usually attributed to the ability to induce an apatite layer on a material’s surface in physiological conditions [6–8]. The close apposition between bone and an implant surface, or osseointegration, presented as the ability to promote bone cells anchorage, attachment, spreading, growth and differentiation [9, 10], is another key factor for successful implantation of a biomaterial for dental and orthopaedic applications [3–5]. Recent years have seen considerable interest in the systematic investigation of the relationship between surface chemistry [27, 28] and morphology [29] and biological interfacial reactions. Ultimately, such studies may lead to an enhanced understanding of the surface chemical cues that guide the combination of apatite induction ability with attachment, growth and differentiation of bone cells and to the design of improved synthetic materials for dental and orthopaedic applications.

Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

2

Bioactivity and Biointegration

1.1.1 Biocompatibility A biocompatible material has been defined as a material that does not induce an acute or chronic inflammatory response and does not prevent a proper differentiation of implant surrounding tissues [30]. It is recognised that some adverse tissue reaction around the implanted biomaterial is inevitable, owing to surgical trauma during insertion. This definition implies that biocompatibility depends on the purpose of the implant. Terms such as ‘bioinert’ or ‘bioactive’, rather than ‘biocompatibility’, more accurately describe the features required of an ideal biomaterial or device. These terms better describe the action or nonaction required from surrounding tissue and are thus related to the choice of materials and material characteristics (hydrophilic/hydrophobic). 1.1.2 Host Response to Biomaterials The biocompatibility of the implant material is closely related to the reactions between the surface of the biomaterial and the inflammatory host response [31]. The implantation response in bone differs in some ways from that taking place in soft tissue. There is an inflammatory and a reparative response which occur one on the other. The reparative response starts 2–3 days after the implantation. The stem cells of bone develop into osteoblasts, which form a layer near the implant together with fibroblasts. Fibroblasts, osteoblasts and capillaries penetrate into the blood clot, replacing it, and fill the space between the implant and bone [32]. After the formation of a collagen-rich extracellular matrix (ECM), mineralization follows. Normally, there are vesicles in the ECM and some of them include calcification focuses. The presence of vesicles with biomaterial in the early period is a sign of good primary acceptance. When the membranes of these vesicles rupture, the erupted apatite crystals unite and form calcifying structures [33]. Early trabecles grow and continue to mineralise, and some of them reach the implant surface. In an optimal situation, the material is covered by bone tissue and not by fibrous capsule. The healing of bone tissue continues like fracture healing. Remodeling bone tissue begins after two weeks and continues for the lifetime. When the material is biocompatible, there is an abundance of the ECM and osteoblasts. This is confirmed by the close attachment and fast proliferation of these cells [34]. 1.1.3 In vitro Models of Biological Response to Implants In vitro models have been widely used to investigate biological responses to biomaterials, and the materials adopted for implantology have been no exception. Given their relative ease of use and lack of expense, in vitro models would appear to investigate the biological response to implant

Bioactivity of Bone Implants

3

materials [35]. Cell culture methods have been used to evaluate the biological compatibility of materials for more than two decades [36]. Bone cell culture models are increasingly employed to study bone–biomaterial interactions. Most of the cultures have utilized osteoblastic cells (reviewed by Cooper et al. [37]), with only a few using osteoclastic cells [38, 39]. In vitro models have the potential to help elucidate events at the bone–implant interface (reviewed by Davies [40]), by providing morphological, biochemical and molecular information regarding osteoblastic development and synthesis of the matrix at the interface with various biomaterials.

1.2 Bioactivity of Bone Implants For an artificial material to bond to living bone, it is essential that the material has the ability to form a biologically active, bone-like, apatite layer on its surface in the human body. Under normal conditions, the body fluid is already supersaturated with respect to apatite, and once apatite on its nuclei form on the surface of a material, they can spontaneously grow by consuming the calcium and phosphate ions from the body fluid. The nucleation of apatite on the surface of a material is induced by the functional groups on its surface. Naturally, the development of bioactive materials that have improved and ultimately the bone-like mechanical properties is desirable [41]. 1.2.1 The Mechanism of Apatite Formation The biological activity of most orthopaedic and dental biomaterials is related to their ability to promote the formation of a neoformed layer of carbonate apatite crystals analogous to bone mineral. This layer also associates specific bone proteins and is the starting point of bone reconstruction [42]. Orthopaedic biomaterials may be classified as ‘passive’ or ‘active’ with regard to their propensity simply to allow the nucleation and growth of carbonate apatite crystals from body fluids (hydroapatite, titanium oxide, neutral hydrogels, collagen) or to supply ions to build and develop this layer (bioglasses, alkaline hydroxides, Ca–P compounds, calcium carbonate). Bioactive ceramics have a common characteristic at the interface with bone after integration. In fact, bioactive ceramics, including bioglass and hydroxyapatite (HA) reveal a layer of apatite at the interface, mediating integration with bone. Histological examination in vivo shows that this apatite layer is formed on the ceramic surface early in the implantation period and thereafter the bone matrix integrates into the apatite. Detailed characterisation indicated that this apatite layer consists of nanocrystals of carbonateion-containing apatite that has a defective structure and low crystallinity. These features are, in fact, very similar to those of the mineral phase in bone;

4

Bioactivity and Biointegration

hence bone-producing cells (osteoblasts) can preferentially proliferate on the apatite and differentiate to form an extracellular matrix composed of biological apatite and collagen. As a result, the surrounding bone comes into direct contact with the surface apatite layer. When this process occurs, a chemical bond is formed between the bone mineral and the surface apatite to decrease the interfacial energy between them. It can be concluded that an essential requirement for an artificial material to bond to living bone is the formation of a layer of biologically active bone-like apatite on its surface in the body. 1.2.2 Functional Group Bioactivity can be induced on surfaces of nonbioactive materials either by the formation of the functional groups that are able to induce apatite formation or by forming thin ceramic phases that have the potential to form the functional groups on exposure to a body environment [41]. The catalytic effect of the Si–OH groups and Ti–OH groups for apatite nucleation has been proven from the observation that silica and titania gels produced by the sol-gel method form apatite on their surfaces in simulated body fluids (SBF), and these functional groups are abundant on their surfaces. Zirconia, niobium oxide and tantalum oxide gels have also been shown to form apatite on their surface in SBF, as shown in Figure 1.1. This indicates that Zr–OH, Nb–OH and Ta–PH groups are effective for apatite nucleation. Other assessments using self-assembled monolayers (SAM) in SBF have indicated that COOH and PO4H2 groups are also effective for apatite nucleation [41].

Figure 1.1 Scanning electron microscopy (SEM) photographs of the surfaces of silica (A), titania (B), zirconia (C), niobium oxide (D) and tantalum oxide (E) gels after soaking in an SBF for 14 d [41]

Biointegration of Orthopaedic and Dental Implants

5

These groups have specific structures revealing negatively charge, and induce apatite formation via formations of an amorphous calcium compound, e.g. calcium silicate, calcium titanate and amorphous calcium phosphate. Moreover, the efficacy of apatite nucleation of the above functional groups is determined, not by their composition alone but in a complicated fashion that is dependent on their concentration and structural arrangement.

1.3 Biointegration of Orthopaedic and Dental Implants 1.3.1 Osseointegration Integration of the implant into the bone is a property of paramount importance in the proper functioning of the implant; consequently extensive studies have been carried out using different techniques to improve osseointegration. Osseointegration was defined by Branemark [43] as: ‘A direct structural and functional connection between living bone and the surface of a load-carrying implant.’ The integration of a biomaterial to bone involves essentially two processes: interlocking with bone tissue and chemical interactions with bone constituents. It is essential for the efficacy of orthopaedic or dental implants to establish a mechanically solid interface with complete fusion between the material’s surface and the bone tissue with no fibrous tissue interface. 1.3.2 Bone Cell Adhesion [44] The term ‘adhesion’ in the biomaterial domain covers different phenomena: the attachment phase, which occurs rapidly and involves short-term events like physicochemical linkages between cells and materials involving ionic forces, van der Waals forces, etc., and the adhesion phase, occurring in the longer term and involving various biological molecules: extracellular matrix proteins, cell membrane proteins and cytoskeleton proteins which interact together to induce signal transduction, promoting the action of transcription factors and consequently regulating gene expression. It is widely acknowledged that a major determinant of the bone–biomaterial interfacial response is the initial attachment, spreading and growth of osteoblasts on the implant surface and that improvements in these processes may lead to faster and long-term stability [3, 45]. It has been possible to demonstrate cellular attachment to implant surfaces and there is good evidence for new bone formation being initiated at a distance from but towards the surface of an implant rather than on the surface itself [3, 45]. The biocompatibility of biomaterials is very closely related to cell behaviour on contact with them and particularly to cell adhesion to their surface.

6

Bioactivity and Biointegration

1.3.3 Osteoblast–Material Interactions Osteoblast–material interaction depends on the surface aspects of materials, which may be described according to their topography, chemistry or surface energy. These surface characteristics determine how biological molecules will adsorb to the surface and more particularly determine the orientation of adsorbed molecules [46]. They also determine the cell behaviour on contact. It is known that osteoblast cells initially respond in a differential manner to the material surface. The comparison of the behaviour of different cell types on materials shows that they react differently according to surface roughness [28, 47]. Scanning electron microscopic examination of bone cells on materials with various surface roughness generally demonstrated that cell spreading and continuous cell layer formation were better on smooth surfaces compared to rough surfaces [48–50]; however, higher levels of cellular attachment have been found on rough surfaces of titanium with irregular morphologies [51–53] in vitro. Similarly, recent studies have shown that alkaline phosphatase (ALP) specific activity is enhanced on rough Ti and Ti–6Al–4V [54, 55]. Cells grown on rougher surfaces exhibited increased production of collagen [52, 54] and transforming growth factor b [54]. These differences of cell response could be attributed to either the microstructure, crystalline or chemistry, as different methods were used to obtain different roughnesses. A contact guidance phenomenon has also been described on osteoblastic cells. On smooth surfaces, bone cells were randomly oriented although they were aligned parallel to the direction of the grooves in an end-to-end fashion in 5 mm deep grooves [47]. Osteoblast attachment is also affected by the chemistry of the biomaterial’s surface, but a conclusive picture has not yet emerged. Early in vitro cytocompatibility studies focused on the morphological aspect, growth capacity and the state of differentiation of cells on materials with various chemical compositions. The diversity of cell responses to the different materials tested highlighted the capacity of cells to distinguish the effects of subtle changes in substratum surface chemistry. For primary bovine osteoblasts, the wettability of the surface has been shown to be one of the important factors [56]. On the other hand, the functional groups present on the surface have been demonstrated to be of even greater importance than the wettability for the adhesion of MC3T3-E1 osteoblasts [57].

1.4 Controlling the Bone–Implant Interface Different approaches are being used in an effort to obtain the desired bone– implant interface. As Kasemo and Lausmaa [58], among others, have described, biological tissues interact mainly with the outermost atomic

Controlling the Bone–Implant Interface

7

layers of an implant; consequently, much effort is being devoted to methods of modifying surfaces of existing biomaterials to achieve desired biological responses. The approaches can be classified as physicochemical and biochemical methods. 1.4.1 Physicochemical Methods Wettability characteristics are among the physicochemical characteristics that have been altered with the aim of improving the bone–implant interface [9]. Polymer surfaces were modified by glow discharge to study the effect of surface treatment on cell adhesion using polyethylene, polytetrafluoroethylene, poly(ethylene terephthalate), polystyrene and polypropylene films. The surface wettability of all the films decreased with respect to the length of plasma treatment. For each of the polymers, a different dependence of cell adhesion on the length of plasma treatment was observed, but, in each case, the optimal water contact angle for cell adhesion was approximately 70
[59]. Alterations in surface morphology and roughness have been used to influence cell and tissue responses to implants [9]. In addition to providing mechanical interlocking, surfaces with grooves can induce ‘contact guidance’, whereby the direction of cell movement is affected by the morphology of the substrate. For osteoblast-like cloned mouse cells (MC3T3-E1) cultured on titanium plates roughened by wire-type electric discharge machining or plasma coating, proliferation and ALP activity were enhanced on the roughened surfaces [60]. Considering the role of electrostatic interactions in many biological events, charged surfaces have been proposed as being conducive to tissue integration. Calcium phosphate coatings have been extensively investigated because of their chemical similarity to bone mineral [9]. The osteoblast-like cells cultured on RKKP- and AP40-bioglass layer coated zirconia showed a higher proliferation rate, leading to confluent cultures with higher cell density and a generally better expression of osteoblast ALP activity in comparison with zirconia substrate [61]. Each approach, however, has drawbacks. Contradictory results with charged materials in bone have been reported: indeed both positively and negatively charged surfaces were observed to promote bone formation. Although short-term clinical results have been encouraging, dissolution of coatings as well as cracking and their separation from metallic substrates remain concerns [9]. 1.4.2 Biochemical Methods [9] Biochemical methods of surface modification offer an alternative or adjunct to physicochemical and morphological methods. Biochemical surface modification endeavours to utilise current understanding of the biology and biochemistry of cellular function and differentiation. Much has been learned

8

Bioactivity and Biointegration

about the mechanisms by which cells adhere to substrates, and major advances have been made in understanding the role of biomolecules in regulating differentiation and remodelling of cells and tissues, respectively. The goal of biochemical surface modification is to immobilise proteins, enzymes or peptides on biomaterials for the purpose of inducing specific cell and tissue responses or, in other words, to control the tissue–implant interface with molecules delivered directly to the interface. One approach to controlling cell–biomaterial interactions utilises cell adhesion molecules. Since identification of the arginine–glycine–aspartic acid (RGD) sequence as mediating attachment of cells to several plasma and extracellular matrix proteins, including fibronectin, vitronectin, type I collagen, osteopontin and bone sialoprotein, researchers have been depositing RGD-containing peptides on biomaterials to promote cell attachment. A second approach to biochemical surface modification uses biomolecules having demonstrated osteotropic effects. A large amount of information has been obtained about biomolecules involved in bone development and fracture healing. By delivering one or more of these molecules, which normally play essential roles in osteogenesis, directly to the tissue–implant interface, bone formation may be promoted. To control exposure and concentration, retention and/or release of biomolecules from implant surfaces can be altered using different methods, including adsorption, covalent immobilisation and release from coatings (Figure 1.2). The simplest way to deliver biomolecules to the tissue–implant interface is by dipping the device in a solution of protein before inserting it. Studies using simple adsorption indicate that delivery of TGF-b to the tissue–implant interface can improve bone formation in the periprosthetic

Figure 1.2 Schematic illustration of methods for controlling retention and/or release of biomolecules at the tissue–implant interface [9]

Controlling the Bone–Implant Interface

9

gap and can enhance bone ingrowth into porous coatings [62]. Using a similar approach, ALP adsorbed on titanium implants enhanced periprosthetic bone formation [63]. One drawback with the adsorption method, however, is that it provides little control over the delivery, including release/retention and orientation, of molecules. Proteins are initially retained on the surface by weak physisorption forces; then, depending on the implant microenvironment, which varies between anatomical sites and between patients, they desorb from the surface in an uncontrolled manner to initiate desired responses. Considering the necessity of specific receptor– ligand interactions for activity of many relevant biomolecules, appropriate presentation of protein may also be needed. Although positive responses have been observed using this simple approach, there is no indication that they are optimal for clinical applications. Bonding biomolecules to implants is an alternate way of delivering them to the tissue–implant interface, albeit protein will not be released. This approach is more complicated than adsorption, because of the chemistry involved, but the activity of molecules immobilised on plastics has been shown to equal or exceed that of soluble protein [64]. For orthopaedic and dental applications, metal surfaces possess a relative paucity of functional groups needed for immobilising molecules. However, the passivating oxide film on these materials does have surface hydroxyl groups that provide locations for bonding using silane chemistry. This approach has been used to immobilise peptides, enzymes and adhesive proteins on different biomaterials, including Co–Cr–Mo, Ti–6Al–4V, Ti and NiTi [9].

2 Surface Modification of Biomaterials 2.1 Introduction Biomaterials have been studied for many years and have been defined by Ratner [65] as being nonviable materials used in a medical device and intended to interact with a biological system. There is a big demand for biomaterials to assist or replace organ functions and to improve patients’ quality of life. Biomaterial applications make use of all classes of materials, metals, ceramics, polymers and composite. These are divided roughly into three user types [2]: (a) inert or relatively inert with minimal host response; (b) bioactive, which actually stimulates bonding to the surrounding tissue; and (c) biodegradable, which resorb in the body over a period of time. The biological responses to biomaterials and devices are largely controlled by their surface chemistry and structure. That is to say, the surface characteristics play a role in the functioning of biomaterial. The rationale for the surface modification of biomaterials is straightforward. Either biological or physicochemical methods are often employed to modify the material surface. Various physicochemical methods will be introduced in this chapter; among them, laser surface treatment has proved to have good potential for it is a unique feature. 2.1.1 Orthopaedic and Dental Implants There is archaeological evidence that lost teeth were replaced by handcarved ivory or wood ‘implants’ as long ago as ancient Egypt. In the early 1950s Swedish orthopaedic surgeon Per Ingvar Branemark began studying

Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

12

Surface Modification of Biomaterials

the healing process of titanium anchoring screws, which proved to be a seminal point for modern dental and orthopaedic implants [1]. His work showed that fusion between bone and the titanium implant could take place, a phenomenon he called ‘osseointegration’. Orthopaedic implants to treat joint degradation due primarily to osteoarthritis, osteoporosis or injury are now commonplace. These include hip (around 325 000 US implants in 2001) and shoulder, wrist and knee (around 300 000 US implants in 2001). These types of implant have undergone continual advancement of their metal and plastic components in order to improve wear resistance and fixation in the insertion site; nevertheless, research continues for more wear-resistant joint interface materials, for materials with improved mechanical properties, and to improve long-term reliability, especially as it relates to fixation. The objective, of course, is to develop implants that can serve patients for an indefinite period. 2.1.2 Surface Properties of Biomaterials In the case of medical implants the importance of surface science is quite obvious [66, 67]. It has been hypothesised that tissue–biomaterial interactions are governed by surface properties and that the important interactions occur within around 1 nm of the biomaterial surface [58]. Natural biological structures appear to be able to interact selectively with relevant biomolecules while resisting nonspecific interactions. Furthermore, biological interfaces are highly dynamic. As pointed out by Blawas and Reichert [68], simply trapping cells at a particular point on a surface is not enough. Cells must first be encouraged to ‘differentiate’ (i.e. change their behaviour to perform as required) and once their function is complete their activity must be ‘turned off’ again. 2.1.3 Surface Analysis of Biomaterials Scanning electron microscopy (SEM) has been the traditional method for studying the microscopic ‘surface structure’ of biomaterials. While allowing visualisation and chemical analysis of the specimen, SEM is not a surfacespecific technique in the strictest sense. Elemental microanalysis (EDAX) may also be carried out using SEM. This relies on the detection of X-rays of characteristic energy, which are emitted on interaction with the electron beam. With the recent, rapid growth of methods for preparing spatially welldefined materials, the focus of biomedical surface science is now on high spatial resolution surface chemical state analysis. The driving forces for developing biomedical surface chemical state imaging techniques are addressed below. X-ray photoelectron spectroscopy (XPS), time-of-flight secondary ion mass spectrometry (ToF SIMS), scanning probe microscopy

Ceramic Implants [65]

13

(SPM) and near-edge X-ray absorption fine structure (NEXAFS) each has its own strengths and weaknesses with respect to generating surface chemical state information at high spatial resolution, but together they provide a powerful set of complementary techniques. For example, XPS and ToF SIMS can be used to improve the level of chemical state information obtainable with SPM, while SPM can be used to improve the spatial resolution obtainable with XPS and ToF SIMS.

2.2 Ceramic Implants [65] It is essential to recognise that no one material is suitable for all biomaterial applications. As a class of biomaterials, ceramics, glasses and glass–ceramics are generally used to repair or replace skeletal hard connective tissues. Their success depends upon achieving a stable attachment to connective tissue. The mechanism of tissue attachment is directly related to the type of tissue response at the implant–tissue interface (see Table 2.1). 2.2.1 Nearly Bioinert Ceramics [65, 69] Bioceramics are compatible because they are composed of ions commonly found in the physiological environment (calcium, potassium, magnesium, sodium, etc.) and of ions showing limited toxicity to body tissue (zirconium and titanium). Two nearly inert ceramics most used in surgical implants are

Table 2.1 Types of bioceramic–tissue attachment and their classification [65] Type of attachment

Example

1. Dense, nonporous, nearly inert ceramics attach by bone growth into surface irregularities by cementing the device into the tissues or by press-fitting into a defect (termed ‘morphological fixation’) 2. For porous inert implants, bone ingrowth occurs that mechanically attaches the bone to the material (termed ‘biological fixation’) 3. Dense, nonporous surface-reactive ceramics, glasses and glass–ceramics attach directly by chemical bonding with the bone (termed ‘bioactive fixation’) 4. Dense, nonporous (or porous) resorbable ceramics are designed to be slowly replaced by bone

Al2O3 (single crystal and polycrystalline) ZrO2 (partially stabilised zirconia)

Al2O3 (polycrystalline) Hydroxyapatite-coated porous metals Bioactive glass Bioactive glass–ceramics Hydroxyapatite Calcium sulfate (plaster of Paris) Tricalcium phosphate Calcium–phosphate salts

Surface Modification of Biomaterials

14 Table 2.2

Properties of bioinert ceramics [70]

Property

Units

Chemical composition Density Bending strength Compression strength Young modulus Fracture toughness KIC Hardness

Alumina 99.9 %MgO

g/cm3 MPa MPa GPa MPa/m HV 0.1

3.97 >500 4100 380 4 2200

MgO–PSZ Al2O3 þ ZrO2 þ MgO (8–10 mol %) 5.74–6 450–700 2000 200 7–15 1200

Y–PSZ (TZP) ZrO2 þ Y2O3 (3 mol %) >6 900–1200 2000 210 7–10 1200

alumina and zirconia. The characteristics of bioinert ceramics for biomedical application are shown in Table 2.2. High-strength ceramics used for implants are very inert in the body and exhibit minimal ion release. Inert bioceramics undergo little or no chemical change during long-term exposure to the physiological environment. 2.2.2 Alumina High-density, high-purity (>99.5%) alumina is used in load-bearing hip prostheses and dental implants because of its excellent corrosion resistance, good biocompatibility, high wear resistance and high strength [69, 71]. Although some dental implants are single-crystal sapphires most Al2O3 devices are very fine-grained polycrystalline a-Al2O3 produced by pressing and sintering at T ¼ 1600–1700  C. Alumina has been used in orthopaedic surgery for nearly 20 years. Its use has been motivated largely by two factors: its excellent biocompatibility and very thin capsule formation, which permits cementless fixation of prostheses and its exceptionally low coefficients of friction and wear rates. 2.2.3 Zirconia Ceramics Zirconia is also exceptionally inert in the physiological environment and zirconia ceramics have an advantage over alumina ceramics of higher fracture toughness and higher flexural strength and lower Young’s modulus [69]. Partially stabilised zirconia (PSZ) is a ceramic that has found wide usage in medical and dental surgery. During a heating process, zirconia will undergo a phase transformation process. The change in volume associated with this transformation makes the usage of pure zirconia in many applications impossible. Addition of some oxides, such as calcia (CaO), magnesia (MgO) and yttria (Y2O3), into the zirconia structure in a certain degree

Metallic Implants

15

Figure 2.1 Medical-grade zirconia used as (a) femoral balls, (b) thumb and (c) dental implant [69]

results in a solid solution, which is a cubic form and has no phase transformation during heating and cooling. This solid solution material is termed ‘stabilised zirconia’. Within medicine it is commonly used to fabricate hip ball joints, knee, thumb, etc., while in dentistry it is used to manufacture dental implants, dental posts, brackets and inlays. Some zirconia implants for medical and dental applications are shown in Figure 2.1. The published results of in vitro wear tests demonstrated that zirconia has a superior wear resistance. Saikko [72] showed no wear of zirconia femoral heads on his hip simulator wear test against the 10.9 mm ultra high molecular weight polyethylene (UHMWPE) cup, and Oka et al. [73] demonstrated the high wear resistance of zirconia against UHMWPE and the superiority of zirconia ceramics even over alumina ceramics in terms of low wear and low friction.

2.3 Metallic Implants Metals have been the primary materials in the past for damaged human bones due to their superior mechanical properties [74], albeit dangerous ions that are released in vivo from these alloys. Although pure metals are sometimes used, alloys (metals containing two or more elements) frequently provide improvement in material properties, such as strength and corrosion resistance. Three material groups dominate biomedical metals: 316 L stainless steel, cobalt–chromium–molybdenum alloy, and pure titanium and titanium alloys. The main considerations in selected metals and alloys for biomedical applications are biocompatibility, appropriate mechanical properties, corrosion resistance, and reasonable cost.

Surface Modification of Biomaterials

16 Table 2.3

Selected properties of metallic biomaterials

Material Stainless steel Cobalt–chromium (Co–Cr) alloys Titanium (Ti) Ti–6Al–4V Cortical bone

Young’s modulus, E (GPa)

Yield strength, sy (MPa)

Tensile strength, sUTS (MPa)

Fatigue limit, send (MPa)

190 210–253

221–1213 448–1606

586–1351 655–1896

241–820 207–950

110 116 15–30

485 896–1034 30–70

760 965–1103 70–150

300 620

2.3.1 Mechanical Properties The mechanical properties of materials are of great importance when designing load-bearing orthopaedic and dental implants. Some mechanical properties of metallic biomaterials are listed in Table 2.3. With a few exceptions, the high tensile and fatigue strength of metals, compared with ceramics and polymers, make them the material of choice for implants that carry mechanical loads. The elastic moduli of the metals list in Table 2.3 are at least seven times greater than that of natural bone. This mismatch of mechanical properties can cause ‘stress shielding’, a condition characterised by bone resorption (loss of bone) in the vicinity of implants. The key problems associated with the use of these metallic femoral stems are thus the release of dangerous particles from wear debris, the detrimental effect on the bone remodelling process due to stress shielding and also loosening of the implant tissue interface. It has been shown that the degree of stress shielding is directly related to the difference in stiffness of bone and implant material [75]. Titanium alloys are favourable materials for orthopaedic implants due to their good mechanical properties. Titanium, however, does not bond directly to bone, resulting in loosening of the implant. Undesirable movements at the implant–tissue interface results in failure cracks of the implant. 2.3.2 Corrosion The physiological environment is typically modelled as a 37  C aqueous solution, at pH 7.3, with dissolved gases (such as oxygen), electrolytes, cells and proteins. Immersion of metals in this environment can lead to corrosion, which is deterioration and removal of the metal by chemical reactions. During the electrochemical process of corrosion, the metallic biomaterial can release ions, which may reduce the biocompatibility of materials and jeopardise the fate of implants. For example, the type and concentration of released corrosion products can alter the functions of cells in the vicinity of implants as well as of cells as remote locations after transport of the

Surface Modification of Biomaterials

17

corrosion by-products to distant sites inside the body. These circumstances become stronger possibilities in the bodies of sick and elderly patients, who are the largest group of recipients of prostheses. Titanium and its alloys, as well as cobalt–chromium alloys, have more favourable corrosion resistance for long-term implant applications such as joint and dental prostheses. 2.4 Surface Modification of Biomaterials 2.4.1 Introduction Biointegration is the ideal outcome expected of an artificial implant. This implies that the phenomena that occur at the interface between the implant and host tissues do not induce any deleterious effects such as chronic inflammatory response or formation of unusual tissues. It is, therefore, of paramount importance to design biomaterials used in implants with the best surface properties. Meanwhile, these biomaterials must process bulk properties that meet other requirements, especially mechanical properties, in order to function properly in a bioenvironment. As it is quite difficult to design biomaterials fulfilling both needs, a common approach is to fabricate biomaterials with adequate bulk properties followed by a special treatment to enhance the surface properties. Hence surface modification of biomaterials is becoming an increasingly popular method to improve device multifunctionality, tribological and mechanical properties, as well as biocompatibility of artificial devices while obviating the need for large expenses and a long time to develop brand new materials [11]. Materials can be surfacemodified by using biological or physicochemical methods. A few of the more widely used physicochemical methods are briefly described here. 2.4.2 Radiation Grafting and Photografting [76] Radiation is widely used in biomaterials science for surface modification, sterilization and to improve bulk properties. The use of gamma, ultraviolet (UV) and electron beam radiation has enabled the biomaterial scientist to perform bulk and surface modifications that improve the biological response of materials and, subsequently, the performance of many medical devices. Radiation grafting has proven to be a simple technique that enables control placement of bioactive molecules on a polymer surface. Radiation-induced cross-linking has allowed the tailoring of the composition and properties of hydrogels to meet numerous biomedical applications. In addition, photocross-linking can serve to enhance the tribological properties of load-bearing components of the total artificial knee and hip. UV radiation appears to have the potential to facilitate in situ curing of adhesives, in situ production and modification of devices and the generation of smart biomedical devices such as biochips.

18

Surface Modification of Biomaterials

2.4.3 Plasma Surface Modification of Biomaterials In the plasma surface modification process, glow discharge plasma is created by evacuating a vessel, usually quartz because of its inertness, and then refilling it with a low-pressure gas. The gas is then energised using techniques such as radiofrequency energy, microwaves and alternating current or direct current. The energetic species in gas plasma include ions, electrons, radicals, metastables and photons in the short-wave ultraviolet (UV) range. These energy transfers are dissipated within the solid by a variety of chemical and physical processes, to result in surface modification. Plasma-based techniques combining the advantages of conventional plasma and ion beam technologies are effective methods for medical implants with complex shapes [77]. In particular, modification of the surface energetics of the materials can improve the adhesion strength, surface and coating properties, and biocompatibility, to name just a few [78]. However, the apparatus used to produce plasma depositions can be expensive [65] and the chemistry produced on a surface can be ill-defined.

2.4.4 Ion Beam Processing Biomaterial modification by ion beam processing is becoming popular for improving medical device function, biocompatibility and as a new mutation breeding method. Ion beam base processes, such as ion implantation and ion beam assisted deposition (IBAD), can provide beneficial surface layers with desirable properties without detrimentally affecting the bulk properties. The ion beam method injects accelerated ions with energies ranging from 101 to 106 eV (1 eV ¼ 1.6  1019 J) into the surface zone of a material in order to alter its surface properties. Important potential applications for biomaterials include modification of hardness (wear), lubricity, toughness, corrosion, conductivity and bioreaction [79]. The primary advantage of ion implantation is selective surface modification without detrimentally affecting bulk properties. The drawbacks of this process are the high cost and the relatively shallow depth of modification. Ion beam assisted deposition (IBAD) is a vacuum deposition process that combines physical vapour deposition (PVD) with ion beam bombardment. The major feature of IBAD is bombardment with a certain energy (ranging from several hundred to several thousand eV) ion beam during the deposition of coating. IBAD is used in hydroxyapatite coating preparation, diamond-like carbon (DLC) film, C–N film and other coatings. Another ion beam process is ion beam texturing (IBT). IBT has the ability to create desirable microfeatures and macrofeatures on the biomaterials to meet the requirement of biocompatibility in vivo.

Laser Surface Modification of Biomaterials

19

2.4.5 Other Methods [65] Silanization Silane reactions can be used to modify hydroxylated or amine-rich surfaces. Since glass, silicon, germanium, alumina and quartz surfaces, as well as many metal oxide surfaces, are all rich in hydroxyl groups, silanes are particularly useful for modifying these materials. Langmuir–Blodgett Deposition The Langmuir–Blodgett (LB) deposition method covers a surface with a highly ordered layer. Each of the molecules that assemble into this layer contains a polar head group and a nonpolar region. Self-Assembled Monolayers Self-assembled monolayers (SAMs) are surface coating films that spontaneously form highly ordered structures (two-dimensional crystals) on specific substrates. Surface-Modifying Additives Certain components can be added in low concentrations to a material during fabrication and will spontaneously rise to and dominate the surface. Conversion Coating Conversion coatings modify the surface of a metal into a dense oxide-rich layer that imparts corrosion protection, enhanced adhesivity and sometimes lubricity to the metal. Parylene Coating Parylene (para-xylylene) coatings occupy a unique niche in the surface modification literature because of their frequent application and the good quality of the thin-film coatings formed.

2.5 Laser Surface Modification of Biomaterials 2.5.1 Introduction Lasers can rapidly and specifically induce surface changes in organic and inorganic materials. The advantages of using lasers for such modification are the precise control of the frequency of the light, the wide range of frequencies available, the high energy density, the ability to focus and raster the

20

Surface Modification of Biomaterials

light, the possibilities for using both heat and specific excitation to effect change and the ability to pulse the source and control reaction time. Treatments are pulsed (100 nanoseconds to picoseconds pulse times) and continuous wave (CW), with interaction times often less than 1 microsecond. Laser-induced surface alterations include annealing, etching, deposition and polymerisation. 2.5.2 Laser Patterning and Microfabrication Laser patterning is based on the possibility of focusing an intense laser beam at certain spots on a surface, where the high beam intensity causes evaporation of the material. By this approach, pits can be produced down to 1 mm, in the size range of interest to match cell sizes. By controlled motion of the beam (either by using clever optics or by sample motion), predesigned patterns can be made. With a kinoform, it is possible to ‘laser-machine’ multiple pits in a surface at once [80]. A new method combines microphotolithographical techniques with laser excimer beam technology to create surfaces with well-defined three-dimensional microdomains in order to delineate critical microscopic surface features governing material–cell interaction [81]. Most laser-based patterning techniques use UV photoablation to micromachine biological substrates in order to generate mesoscopic patterns, and arrays of viable cells are required to fabricate next-generation tissue-based sensing devices to build three-dimensional cellular structures for advanced tissue engineering and to separate selectively and culture differentially microorganisms for a variety of basic and applied research applications [82]. Recently, laser etching (or laser ablation) and microlithography have been adopted to achieve micrometer dimensions with high precision in order to develop a large number of miniaturised systems for the analysis of biological tissues. Excimer laser etching was used to microtexture a biocompatible substratum for high-contrast microscope cell analysis [83]. 2.5.3 Pulsed Laser Deposition (PLD) of Biocompatible Ceramics Pulsed laser deposition (PLD) is a new technique for the deposition of thin films of biocompatible ceramics [84]. Pulsed laser deposition is especially well suited to the deposition of bone-like ceramics (e.g. hydroxyapatite (HA) and calcium phosphates) on to metal, ceramic, semiconductor or polymer substrates for potential application in medical implants, prosthetic devices and biocompatible probes or sensors. The degree of control over film characteristics offered by PLD exceeds that of other known deposition techniques presently applied to production of thin films of biocompatible ceramics. It is anticipated that PLD will develop into the technique of choice

Laser Surface Modification of Biomaterials

21

for the manufacture of implant or prosthetic devices comprising biocompatible films on structurally robust substrates. Pulsed laser ablation is a new method for deposition of thin layers of HA on to biomaterial surfaces. Differences in cell spreading were apparent which were correlated with the fluence used to deposit the HA [85]. Pulsed laser ablation and deposition of bioactive glass [86, 87] have been performed and the plume and film compositions have been characterised. All the elements present in the target have been found in the gaseous phase. 2.5.4 Matrix-Assisted Pulsed Laser Evaporation and MAPLE Direct Write Two techniques, matrix-assisted pulsed-laser evaporation (MAPLE) and MAPLE direct write (MDW) were developed to deposit biomaterial thin films [88]. MAPLE involves dissolving or suspending the biomaterial in a volatile solvent, freezing the mixture to create a solid target and using a low fluence pulsed laser to evaporate the target for deposition of the solute inside a vacuum system. Using simple shadow masks, i.e. lines, dots and arrays, pattern features with length scales as small as 20 mm can be deposited using multiple materials on different types of substrates. MAPLE utilises a low fluence pulsed UV laser and a frozen target consisting of a dilute mixture of the material to be deposited and a high vapour-pressure solvent. MDW uses pulsed laser radiation to directly transfer material from a ribbon to a substrate. Patterns with a spatial resolution of approximately 10 mm can be written directly. Biomaterials ranging from polyethylene glycol to eukaryotic cells (Chinese hamster ovaries) were deposited with no measurable damage to their structures or genotype. Deposits of immobilized horseradish peroxides (an enzyme) in the form of a polymer composite with a protective coating, i.e. (polyurethane) retained their enzymatic functions. A dopamine electrochemical sensor was fabricated by MDW using a natural tissues/graphite composite. The novelty of the MDW process is that the interaction of the incident laser pulse with the coating on the ribbon can transfer the micrometre-size powder, nanopowders and especially the chemical precursors to form a densely packed composite on the receiving substrate [89]. 2.5.5 Other Laser Surface Treatments Laser Surface Treatment for Improving Corrosion It was found that the corrosion behaviour of NiTi samples was improved by excimer laser surface melting [90]. Laser treatment improvement resistance is explained by a combination of the homogenisation of the surface by melting, the hardening due to N incorporation and the thickening of the

22

Surface Modification of Biomaterials

oxide layer. Moreover, excimer laser surface treatment in air showed a remarkable improvement in pitting corrosion resistance for 316LS biograde stainless steel. The results show that excimer laser surface melting can effectively eliminate carbides and second phases alike, while also serving the function of homogenising the microstructure. N2 induced into the lasertreated surface could promote new precipitates and as a result lowered the corrosion resistance of 316LS stainless steel [91] and Ti–6Al–4V alloy [92]. Laser Grafting For the purpose of improved surface hydrophilicity and biocompatibility of ethylene–propylene rubber, 2-hydroxyethyl methacrylate (HEMA) and N-vinyl pyrrolidone (NVP) have been grafted on to the surface of this polymer using a CO2 pulsed laser at different fluence (output power J/cm2) as the excitation source [17]. Moreover, ethylene–propylene rubber (EPR) based vulcanizates have been surface-grafted with acrylamide (AAm) and HEMA using a CO2 pulsed laser as the excitation source [18]. Surface hydrophilicity (measured by the water drop contact angle) increased for the grafted samples and comparative results indicate that the adhesion of macrophages to EPR samples modified with AAm and HEMA, with no respiratory burst and cellular damage, is significantly lower than their adhesion on unmodified surfaces, which show an activated state of the attached macrophages. Laser Treatment of Plasma Sprayed Hydroxyapatite Coatings [93] The three requirements generally expected of biomaterials coating are: crystallinity, porosity and adhesion. Crystallinity is essential because amorphous coatings are more resorbable. The study found that laser treatment of plasma-sprayed coatings led to a wide range of microstructures. The porosity of the coatings was reduced significantly. Nd:YAG (neodymiumdoped yttrium aluminium gainet) laser (pulsed) treatment significantly changes the characteristics of the plasma-sprayed coating microstructure in several ways. It ranges from a flat and smooth surface profile containing fine grains to an irregular surface comprising re-melted particles, spherical pores and tracks. Laser treatment of plasma-sprayed HA coatings basically generates a molten layer that rapidly solidifies.

3 Wettability in Biomaterials Science and Modification Techniques 3.1 Introduction The wetting of a surface by a liquid and the ultimate extent of spreading that liquid are very important aspects of practical surface chemistry. Even with all the new information of the last 20 years, however, there still remains a great deal to learn about the mechanisms of movement of a liquid across a surface and the factors that govern such movement [12]. Biomaterial scientists have long sought a single, material-related parameter that effectively measures biocompatibility and might serve as a practical design guide. The theories of surface energy and wetting for such parameters present an attractive means to do this as surface properties are important determinants of a biomaterial function [12]. The ability to control the surface wettability of solid substrates is important in many situations. Various surface processes are used for modifying the surfaces of materials depending on the actual material and the application. A number of laser-based techniques for altering the wettability characteristics of engineering materials have been investigated [13]. 3.2 Wettability, Adhesion and Bonding: Theoretical Background 3.2.1 The Wetting Process The term wetting in its most general sense is used to denote the displacement of air from a liquid or solid surface by water or any aqueous or molten Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

Wettability in Biomaterials Science

24

solution [94]. Wetting is fundamentally a thermodynamic process and the changes in free energy that may occur determine whether or not wetting will happen, at what rate it will proceed and how far it will progress against the external forces. 3.2.2 Contact Angle and Work of Adhesion When a drop of liquid is in free space it is drawn into a spherical shape by the tensile forces of its surface tension, which results from the attractive and repulsive forces that exist between the molecules of the liquid. When such a drop of liquid is brought into contact with a flat solid surface, the final shape taken by the drop, and thus whether it will wet the surface or not, depends upon the relative magnitudes of the molecular forces that exist within the liquid (cohesive) and between the liquid and the solid (adhesive) [95]. The index of this effect is the contact angle, y, which the liquid subtends with the solid. In practice, for wetting to occur the contact angle should be less than 90
. If the contact angle is greater than 90
then the liquid does not wet the solid surface and no adhesion takes place [95]. Figure 3.1 shows a schematic view of a liquid droplet on a solid surface. The contact angle is related to the solid and liquid surface energies, gsv and glv , and the solid–liquid interfacial energy, gsl , through the principal of virtual work expressed by Young’s equation: gsv ¼ glv cos y þ gsl

ð3:1Þ

If an equilibrium for the droplet of liquid melt shown in Figure 3.1 is established, then the relation of y to gsv , glv and gsl is described by the rearranged Young’s equation: cos y ¼

gsv  gsl glv

ð3:2Þ

Clearly, to achieve wetting gsv should be large, while gsl and glv should be small. Hence liquids of a lower surface tension will always spread over a solid surface of higher surface tension in order to reduce the total free energy of the system [96, 97]. Whether the drop of liquid spreads across the solid γlv

γsv

θ

γsl

Figure 3.1 Schematic of the wetting of a solid medium by a liquid melt [95]

Wettability, Adhesion and Bonding

25

surface to wet the surface and provide a coating or remains as a finite drop with an equilibrium angle is dependent upon the spreading coefficient S. For spreading to occur spontaneously S ¼ glv ðcos y  1Þ > 0

ð3:3Þ

The adhesion intensity of a liquid to a solid surface is known as the work of adhesion Wad and is given by the Young–Dupre equation Wad ¼ glv ð1 þ cos yÞ

ð3:4Þ

Based on the nature of the attractive forces existing across the liquid–solid interface, wetting can be classified into the two broad categories of physical wetting and chemical wetting. In physical wetting the attractive energy required to wet a surface is provided by the reversible physical forces, such as the van der Waals and dispersion forces. In chemical wetting adhesion is achieved as a result of reactions occurring between the mating surfaces, giving rise to chemical bonds [98]. 3.2.3 Surface Energy and the Dispersive/Polar Characteristics The intermolecular attraction that is responsible for surface energy, g, results from a variety of intermolecular forces whose contribution to the total surface energy is additive [99]. The majority of these forces are functions of the particular chemical nature of a certain material, and as such the total surface energy comprises gp (polar or nondispersive interaction) and gd (dispersive component, since van der Waals forces are present in all systems regardless of their chemical nature); therefore, the surface energy of any system can be described by [100] g ¼ gd þ gp

ð3:5Þ

Similarly, Wad can be expressed as the sum of the different intermolecular forces that act at the interface [101]:
1 p
1 p d þ Wad ¼ 2 gdsv gdlv 2 þ 2 gpsv glv 2 Wad ¼ Wad

ð3:6Þ

If a liquid that has both dispersive and polar forces is in contact with a solid surface where the surface energy is due to dispersion forces only, then the relationship between the contact angle and the surface energies of the liquid and solid are given by [101, 102] 2 gdsv gdlv cos y ¼ glv


12

1

ð3:7Þ

Wettability in Biomaterials Science

26 1

Cos θ

0.5

0

−0.5

−1 0

Figure 3.2 substrate

0.04

0.08

0.12

0.16

1 2

Plot of cos y against ðgdlv Þ =glv for a theoretical liquid system on any solid

However, by equating Equation (3.7) with Equation (3.4), the contact angle for solid–liquid systems where both dispersion forces and polar forces are present can be related to the surface energies of the respective liquid and solid by
1 p p
1 2 gdsv gdlv 2 þ 2 gsv glv 2 1 cos y ¼ glv

ð3:8Þ

Therefore, from Equation (3.8), one can estimate the dispersive component of a solid substrate surface energy, gdsv , by plotting the graph of cos y against 1 (gdlv )2 =glv . This is shown in Figure 3.2 for a theoretical liquid system on any solid substrate. Thus, according to Fowkes [100], the value of gdsv is estimated by the 1 gradient (¼ 2 [(gdsv ) ]2 ) of the line (----) that connects the origin (cos y ¼ 1) 1 with the intercept point (cos y against (gdlv )2 =glv ) of the straight line (—) correlating the data point with the abscissa at ‘cos y ¼ 1 In contrast’, it is not possible to determine the value of the polar component of a solid substrate 1 p surface energy, gsv , directly from cos y against (gdlv )2 =glv . This is because the p p 1 intercept of the straight line (cos y against (gdlv )1=2 =glv ) is at 2(gsv glv )2 =glv , and thus only refers to individual control liquids and not the control liquid system as a whole. However, it has been established that the entire amount of the surface energies due to dispersion forces either of the solids or the liquids are active in the wettability performance [100, 103]. As such, it is d , possible to calculate the dispersive component of the work of adhesion, Wad by using only the relevant part of Equation (3.6). Thus d ¼ 2 gdsv gdlv Wad


12

ð3:9Þ

Wettability, Adhesion and Bonding

27

d If one plots a graph of Wad against Wad for the solid substrate, then for each particular liquid in a given system in contact with the solid surface, Wad , which was determined from Equation (3.4), can often be correlated with d , which was determined from Equation (3.9), by the straight line Wad relationship d þb Wad ¼ aWad

ð3:10Þ

Therefore, for a solid substrate the constants a and b can be deduced respectively by calculating the gradient of the best-fit straight line and by extrapolating the best-fit straight line to find the intercept point on the p axis. Also, if one plots a graph of (glv ) against (gdlv ), then for the liquids in a p given liquids system, (glv ) can often be correlated with (gdlv ) by the straight line relationship p

glv


12


1 ¼ c gdlv 2 þ d

ð3:11Þ

Again, for a solid substrate the constants c and d can be deduced respectively by calculating the gradient of the best-fit straight line and extrapolating the best-fit straight line to find the intercept point on the axis. By introducing Equation (3.10) into Equation (3.6) and rearranging, then p

d þb Wad ¼ ða  1ÞWad

ð3:12Þ

or, alternatively, gpsv


12

p

glv


12

¼ ða  1Þ gdsv


12


1 b gpsv 2 þ 2

ð3:13Þ

By introducing Equation (3.11) into Equation (3.13) and differentiating 1 1 p 1 with respect to (gdlv )2 , considering that (gdsv )2 and (glv )2 are constant, then the following can be derived: gpsv


12


1 gdsv 2 ða  1Þ ¼ c

ð1:14Þ

Since gdsv for the solid substrate can be determined previously directly from 1 the plot of cos y against (gdlv )2 =glv , then it is possible to calculate gdsv for the solid substrate Equation (3.14) directly. By employing this approach it is possible to determine, from y measurements and the control liquid surface energy properties, the changes in the wettability characteristics of the materials.

28

Wettability in Biomaterials Science

3.2.4 Physical Bonding Physical bonding is essentially the effect that occurs when two perfectly flat surfaces are brought together to atomic interaction distances, resulting in local atomic rearrangement and consequently adhesion. A typical example of physical bonding is that of van der Waals bonding. The energy difference between the specific surface energy of one material and that of the other is the work of adhesion. The work of adhesion can yield a theoretical breaking stress similar to the strength of either of the materials used. Physical bonding provides a useful guideline to the selection of materials that will bond well together. 3.2.5 Mechanical Bonding Mechanical bonding basically refers to the interlocking microstructure of rough surfaces to provide tensile strength and, in the case of shear, frictional strengthening. During the bonding process, the liquid or melt can flow with varying degrees of ease into cavities and asperities; a ductile metal or glass melt can conform to a rough solid substrate surface, or a vapour can deposit in surface asperities. The solid substrate surface may be roughed by means of acid or base chemical attack, grinding, grit or sand blasting, or laser treatment to enhance mechanical bonding. The effectiveness of these different surface-roughening techniques is entirely dependent upon their optimum application as well as on the specific methods and materials being used. In addition, chemical interaction between materials that are being bonded can lead to mechanical bonding. Further, the increase in the surface area of a mechanically roughened surface can affect an increase in the level of physical bonding. 3.2.6 Chemical Bonding Considerable research into the various chemical mechanisms that can be present during the bonding process is in process. Although most of the research is qualitative or semi-quantitative in nature, it is providing a useful background of chemical data that is contributing to a basic understanding of the principles of chemical bonding. A chemical bond is formed at an interface when a balance of bond energies and a continuous electronic structure are present across the interface for any two dissimilar phases. This structure occurs when a thermodynamically stable chemical equilibrium exists at the interface and is essentially achieved by chemical reactions at the interface. Generally, equilibrium compositions (which can be determined if an equilibrium phase diagram of the two phases being bonded is available) at the interface are attained at the reaction temperature very rapidly.

Wettability in Biomaterial Science

29

3.3 Wettability in Biomaterial Science From a historical perspective, Baier’s proposal that critical surface energy can be directly linked to biocompatibility is perhaps the most penetrating concept among the few generalities offered to explain rules of biocompatibility. This theory, in its most general form, recognises that surface energy must control the way biologic fluids interact with materials and that this interaction, in turn, must primarily influence tissue and cell reactions. As examples, Baier pioneered the use of Zisman’s critical surface tension as an indicator of blood compatibility [104, 105] and bioadhesion [106, 107]. Neumann et al. employed their ‘equation-of-state’ approach to calculate interfacial tensions from contact-angle measurements that, in turn, were used to predict cell adhesion [108] and thromboresistance [109]. Whereas concepts such as these have served as useful general guidelines or ‘rules of thumb’ for biomaterials design, each has fallen far short of being the desired quantitative predictor of biocompatibility, particularly when applied to proteinaceous environments. 3.3.1 Biomaterial Interfaces [110] Surface sensitivity is of critical importance in biomaterials surface science because only the uppermost layers are in direct physicochemical contact with the biological environment; consequently, only the upper few molecular layers determine biocompatibility. Chemical events such as acid–base reactions, hydrogen bonding and ion exchange occur over atomic bondlength distances. Longer-range hydrophobic forces can extend up to about 10 nm and are responsible for nonspecific adsorption, adhesion and surfaceinduced water structure within this zone. Thus the interaction of a material with the biological environment occurs at or through the narrow region termed the interface. Therefore, the interfacial chemistry of concern to biomaterial scientists is determined by the material’s composition within the uppermost few nanometres. 3.3.2 Tensiometry Tensiometry encompasses a broad range of related ‘wetting’ techniques that measure surface energy. These include the observation of contact angles, which is perhaps the most familiar and widely applied method. Tensiometric methods have singular potential in biomaterials surface science based on the criteria of surface sensitivity, kind of analytical information obtained and relevance of that information to biomedical problems. First, with respect to surface sensitivity, wetting measurements are sensitive only to the upper 0.5 nm or so of a surface [14, 15] and are therefore among the most surface-sensitive techniques available. Second, tensiometry directly

Wettability in Biomaterials Science

30

measures the fundamental energy at an interface that drives important processes such as adsorption and adhesion. This kind of information must be particularly pertinent to biomaterial problems because of the overwhelming importance of protein adsorption and cell/tissue adhesion. Third, wetting measurements can be made using proteinaceous saline solutions that are particularly relevant to biomedical applications. Special highvacuum preparation techniques that might introduce experimental artefacts are not required. 3.3.3 Interfacial Biophysics The physicochemical nature of biomaterial interfaces was considered, leading to the conclusion that interfacial energy is a primary determinant of biocompatibility. A colloid science theory that quantifies interactions at small distances and has biomaterial applications is the so-called Derjaguim Landau Verwey Overbek (DLVO) theory [111, 112], shown in the Figure 3.3. DLVO illustrates the relationship between particle (cell) distance from the surface and repulsive (electrostatic) and attractive (namely van der Waals) interaction energies. The basis of this theory is that attractive van der Waals potentials and repulsive electrostatic forces are additive. Formulation of these interaction potentials can be quite detailed for each case, but the qualitative predictive aspects for a macroscopic substrate are quite straightforward. From a physicochemical point of view, the kinetics of adhesion can be described as long-range interactions [100] and short-range interactions (acid–base, hydrogen bonds) [113]. Others describe these forces as dispersive and polar [114]. G GE 10 G 50

100

D(Å)

−10 GVDW

Figure 3.3 Interaction energies between a particle (cell) approaching a solid surface [65]

Wettability in Biomaterial Science

31

The total interaction energy (G, solid line) is composed of attractive van forces (GE ). A secondary der Waals forces ðGVDW Þ and repulsive electrostatic
minimum can be
observed at approximately 100 A and a primary minimum
at a distance 0.5 kW/cm2) intercept the ordinate considerably high above the origin. The highest intercept point is found for the sample CO2 laser treated with 1.6 kW/cm2 power density. An interception of the ordinate above the origin is indicative of the action of polar forces across the interface, in addition to dispersion forces, and hence improved wettability and adhesion is promoted [100, 101]. Furthermore, because none of the best-fit straight lines intercept below the origin, it can be said that the development of an equilibrium film pressure of adsorbed vapour on the MgO–PSZ surface (untreated and CO2 laser treated) did not occur [101]. p It is not possible to determine the gsv value of the MgO–PSZ directly from Figure 4.8. This is because the intercept of the straight line (cos y against 1 1 p (gdlv )2 =glv ) is at 2(gsv )(gdlv )2 =glv , and so only refers to individual control liquids and not the control liquid system as a whole. Still, it has been established that the entire amount of the surface energies due to dispersion forces either of the solids or the liquids are active in the wettability performance [100, 103]. As such, it is possible to calculate the dispersive component of the work d , by Equation (3.9). of adhesion, Wad Table 4.3 shows the values of Wad calculated using Equation (3.4) and the d values of Wad calculated using Equation (3.9) for both the untreated and CO2 laser treated MgO–PSZ with various power densities. Figures 4.9 and 4.10 d show the best-fit straight line plots of Wad against Wad for the MgO–PSZ when it is both untreated and CO2 laser treated. From the plots of Wad d against Wad one can see that the experimental results reveal that for each

Surface Energy and Its Component Parts

49

d Table 4.3 Values of Wad and Wad for the control test liquids and the determined d constant, a, from the plots of Wad against Wad for untreated (UT) and CO2 laser treated MgO–PSZ with various power densities
d Power Work of adhesion ðWad Þ Dispersive work of adhesion Wad density a Glyc Form Ethel P1 P2 Glyc Form Ethel P1 P2 (kW/cm2)

UT 0.5 0.9 1.6 1.9 2.5

2.41 2.37 3.03 4.25 3.42 3.17

76.2 79.4 92.8 113.3 105.0 101.8

75.2 77.5 89.8 105.5 98.5 95.6

71.5 72.5 80.7 90.3 86.5 84.5

69.6 70.9 77.0 82.7 80.0 78.7

66.6 67.3 68.8 71.4 70.6 70.3

76.1 77.2 77.7 80.9 80.5 77.7

74.2 75.2 75.7 78.8 78.4 75.2

70.7 71.6 72.1 75.1 74.7 71.6

69.3 70.3 70.7 73.7 73.3 70.3

66.6 67.5 67.9 70.8 70.4 67.5

Note: Glyc, glycerol; Form, formamide; Ethel, Etheneglycol; P1, polyglycol e-200; P2, polyglycol 15-200.

d Figure 4.9 Plot of Wad against Wad for the untreated MgO–PSZ

particular control liquid in contact with both the untreated and CO2 laser treated MgO–PSZ surfaces, Wad , determined from Equation (3.4), can be d , determined from Equation (3.9), by the straight line correlated with Wad relationship presented by Equation (3.10). Consequently, the values of a for various CO2 laser parameters shown in Table 4.3 were determined by the d best-fit straight line of Wad against Wad . p As Figure 4.11 shows, a linear relationship between gdlv and glv of the control test liquids’ surface energies was observed which satisfied Equation (4.2) and thereby allowed the constant, c, to be calculated. Since gdsv has already been determined for the untreated and CO2 laser treated

CO2 Laser Modification of the Wettability Characteristics

50

d Figure 4.10 Plot of Wad against Wad for the CO2 laser treated MgO–PSZ (1.6 kW/cm2 and 2000 mm/min)

8 7 6

d lv

(γ )

1/2

5 4 3 2 1 0 0

1

2

3

4

5

6

7

8

1/2

(γ ) d lv

Figure 4.11 Plot of (gplv )2 against (gdlv )2 for the control test liquids 1

1

p

MgO–PSZ from Figure 4.8, then it is possible to calculate gsv for untreated and CO2 laser treated MgO–PSZ using Equation (3.14). The values for gsv , gdsv p and gsv of the untreated and CO2 laser treated MgO–PSZ are given in Figure 4.12. As one can see from Figure 4.12, CO2 laser treatment of the

Surface Energy and Its Component Parts

51

Figure 4.12 Relationship between surface energy (gdsv , gpsv and gsv ) for the CO2 laser treated MgO–PSZ and power density

surface of the MgO–PSZ leads to an overall increase in the gsv , while, more p importantly, also significantly increasing gsv . The increase in gsv of the MgO– p PSZ was primarily attributed to the increased gsv value, since gdsv was almost p similar for all the samples. The increase, in particular the increase in gsv , had a positive effect upon the action of wetting and adhesion [103] as primarily both dispersion and polar forces were active to a greater extent [100, 155]. The changes in the surface energy are thought to be due to the fact that the CO2 laser treatment of the MgO–PSZ results in the melting of the surface, a p transition that is known to cause an increase in gsv [141] and hence an improvement in the wettability characteristics. p Moreover, as can be seen from Figure 4.12, gsv and gsv changed depending on the microstructures obtained at different CO2 laser parameters. When a hexagonal microstructure was obtained on the surface of the MgO–PSZ, a p marked increase in gsv and gsv occurred, as is evident from Figure 4.12. When the MgO–PSZ was treated with a relatively medium CO2 laser power density, cell formation on the MgO–PSZ surface was induced and the p maximum value of gsv and gsv was achieved. Figure 4.12 suggests that the onset of melting initiated the cell formation. With relatively high CO2 laser power densities, coral and dendritic microstructures were apparent on the

52

CO2 Laser Modification of the Wettability Characteristics

surface of the MgO–PSZ. The formation of these microstructures was p accompanied by a reduction in gsv and gsv from the maximum value. 4.5 Identification of the Predominant Mechanisms Active in Determining Wettability Characteristics Regardless of the CO2 laser power density employed, noticeable differences in surface roughness, microstructure, surface oxygen content and surface energy of the MgO–PSZ were occasioned simultaneously. All these surface properties will have been a factor in determining the wettability characteristics of the MgO–PSZ. First, CO2 laser treatment of the surface of the MgO– PSZ generated a rougher surface and thereby reduced y. Second, the increase in the surface oxygen content of the MgO–PSZ resulting from CO2 laser treatment will be influential in the promotion of wetting, because an increase in surface oxygen content inherently effects a decrease in y and vice versa. p Lastly, an increase in gsv resulting from the melting and re-solidification of the surface of the MgO–PSZ occurred. This naturally created a different microstructure that quite possibly improved the action of wetting and adhesion. Still, the foregoing sections have revealed that each of these factors did not play an equal role in governing the wettability characteristics of the MgO–PSZ. It is, therefore, essential to identify the effect of each factor and find the predominant mechanism active in governing the wettability characteristics of the MgO–PSZ. Figure 4.13 shows the relationship between the wettability characteristics (denoted by cos y for glycerol) p and the influential factors of surface roughness, gsv and surface oxygen content. As is evident from Figure 4.13, the rougher surface of the modified sample has a higher value of cos y than the smooth, untreated sample. Nevertheless, the change in cos y is not proportional to the alteration in the surface roughness, with cos y increasing sharply up to a surface roughness of 0.717 mm, then decreasing despite a considerable increase in surface roughness. This signifies that other mechanisms, namely the surface oxygen p content and gsv , may play a more predominant role in influencing the wettability characteristics of the CO2 laser treated MgO–PSZ than surface roughness. It is apparent from Figure 4.13 that cos y increased with increasing surface oxygen content below 63.4 at %. This is in accord with the established theory for the relationship between the surface oxygen content of a material and their propensity for wetting. However, when the value of the surface oxygen content exceeded 63.4 at %, cos y decreased, indicating that the surface oxygen content is not a major factor active in changing the wettability of the MgO–PSZ.

Identification of the Predominant Mechanisms Active

53

Figure 4.13 Relationship between cos y for glycerol and the surface roughness, the surface oxygen content, gpsv and microstructures of the untreated and CO2 laser treated MgO–PSZ p

A clear relationship between the value of cos y and gsv is observed in p Figures 4.13. Figure 4.14 reveals that an increase in gsv will in turn cause a p rise in cos y. From this it is evident that gsv influences the wettability characteristics of the MgO–PSZ. Indeed, it was found by Lawrence [154] that surface energy was the most predominant factor governing the wetting characteristics of a SiO2/Al2O3-based ceramic following irradiation with an HPDL. From the above discussion it is not clear whether the surface roughness, the surface energy (by way of microstructural changes) or the surface oxygen content alone, or a combination thereof, are the principal factors influencing the observed changes in the wettability characteristics of the MgO–PSZ after CO2 laser surface treatment. Therefore, several stages of fine grinding were used to isolate the various influential factors detailed above and thus analyse and establish qualitatively the effect each one had on the wettability characteristics of the MgO–PSZ. In the first stage, the surfaces of the untreated MgO–PSZ and CO2 laser treated MgO–PSZ (1.6 kW/cm2) with the largest change in y were ground with grinding paper (180 grit SiC)

CO2 Laser Modification of the Wettability Characteristics

54

Figure 4.14 Relationship between cos y for glycerol and gpsv on the untreated and CO2 laser treated MgO–PSZ

for 3 minutes, while still retaining the CO2 laser treated microstructure. In this way it was possible to investigate the effects of the surface oxygen content, which exists within the first atomic layers of the material. In order to evaluate the influence of the CO2 laser induced microstructure, an intermediate grinding stage using grinding paper (400 grit SiC) for 3 minutes was used to remove the microstructure. In the final stage, both the untreated and the CO2 laser treated samples were ground down further with grinding paper (800 grit SiC) for 3 minutes to study the effect of surface roughness. The observed changes to surface roughness, O2 content and y (glycerol) effected by these steps are given in Table 4.4. After the first grinding stage, a large difference in y for glycerol was observed between the untreated and the CO2 laser treated samples, with y

Table 4.4 The contact angle (for glycerol), surface roughness and surface oxygen content of the untreated and CO2 laser treated MgO–PSZ following the fine grinding stages

Polishing steps Unpolished 180 grit SiC 400 grit SiC 800 grit SiC

Untreated ———— ———————————————— Ra (mm) O2 (at %) y (deg) 0.30 0.22 0.08 0.06

41.6 41.5 41.7 41.8

79.1 81.9 82.3 82.3

CO2 Laser Treated ——————————————— Ra (mm) O2 (at %) y (deg) 0.72 1.90 1.46 1.23

64.3 42.0 41.8 41.7

39.4 43.7 77.3 77.9

Identification of the Predominant Mechanisms Active

55

increasing slightly from 39.4 to 43.7 for the CO2 laser treated sample, while y for the untreated sample increased from 79 to 82 (Table 4.4). From this observation it is reasonable to suggest that the retained CO2 laser induced microstructure in the CO2 laser treated MgO–PSZ could be the mechanism responsible for the large different y from the untreated sample. Furthermore, the surface oxygen content of the CO2 laser treated sample was found to have reduced from 64.3 to 42.0 at %, a level similar to that of the untreated sample, 42.2 at %. It is, therefore, quite possible that the decreased surface oxygen content could be the factor influencing the general increase in y. It is interesting to note that although the surface roughness of the CO2 laser treated sample increased from 0.72 to 1.90 mm, the value of y increased, thus implying that surface roughness does not have as great an influence on the wetting characteristics of the MgO–PSZ as that of the surface oxygen content. This proposition was borne out somewhat when the samples were ground further. In this subsequent stage, the CO2 laser induced microstructure and heat affected zone (HAZ) were removed from the CO2 laser treated sample. The surface oxygen content on the untreated and CO2 laser treated samples were practically the same as the original untreated value of 41.6 at %. Significantly, y for the CO2 treated samples was 77.3 , a value close to the original untreated value of 79.1 . Basically, the removal of the CO2 laser induced microstructure alone appears to have brought about an increase in y to around the original level, since the surface oxygen content was almost the same value in both ground stages. Such findings reveal unequivocally that the microstructure is by far the predominant mechanism governing the wettability characteristics of the MgO–PSZ. The effect of the surface roughness was studied through a further step. This step generated a smoother surface by reducing the Ra from 1.72 to 1.40 mm. For the CO2 laser treated sample, this step caused y to change slightly from 77.3 to 77.9 . For the untreated sample, the final ground stage caused Ra to change considerably, reducing from 0.30 to 0.08 mm. However, as seen from Table 4.4, y only increased very marginally from 79.1 to 82.3 . Despite the magnitude change in surface roughness, y had varied only slightly in this range of surface roughness. Such a finding indicates that changes in surface roughness are insignificant, as reflected by the corresponding change in y. It is therefore reasonable to assume that the surface energy difference brought about by microstructural changes is the primary influential factor governing changes in y and, in turn, the wettability characteristics of the MgO–PSZ. What is more, the surface oxygen content was also found to influence changes in the wettability characteristics of the MgO– PSZ but to a much lesser extent, while surface roughness was shown to play a very minor role in inducing changes in the wettability characteristics of the MgO–PSZ.

56

CO2 Laser Modification of the Wettability Characteristics

4.6 The Role Played by Microstructures in Terms of Crystal Size and Phase in Effecting Surface Energy Changes The surface energy has been identified as the main mechanism governing the modification of wettability characteristics of the MgO–PSZ and is varied with the surface microstructure. It is believed that the changes in surface energy of the MgO–PSZ were attributed to the change in microstructures in the form of crystal sizes and phase changes. 4.6.1 The Role of Crystal Size on Surface Energy After CO2 laser radiation, the modified surface of the MgO–PSZ exhibits a typical microstructure of rapid solidification. As the XRD patterns of the untreated and CO2 laser treated MgO–PSZ given in Figure 4.15 show, after

Figure 4.15 XRD analysis of the MgO–PSZ surface (a) before and (b) after CO2 laser treatment (1.6 kW/cm2) (c ¼ cubic, t ¼ tetragonal, m ¼ monoclinic)

The Role Played by Microstructures in Terms of Crystal Size and Phase

57

CO2 laser treatment a distinct phase change took place. The diffraction patterns of 2-theta between 28 and 32 are of particular relevance (shown in Figure 4.15) and should be emphasised [156]. The peak at 29 belongs to the overlap diffraction of the (111) cubic phase and (101) tetragonal phase (denoted as c(111) and t(101)). The peaks of the cubic and tetragonal phases are all overlapped with others for the untreated and CO2 laser treated samples (see Figure 4.15). The peaks at 28.2 and 31.5 belong to the (111) monoclinic phase (m (111)). An increase in the peak at 29 signifies the increase in the tetragonal phase. After CO2 laser treatment, the peak at 29 , which shifts to 30.5 , was seen to rise while the peaks at 28.2 and 31.5 disappeared (see Figure 4.15), indicating that the tetragonal phase was increasing while the monoclinic phase was decreasing in the MgO–PSZ surface. As shown in Figure 4.16, the relative intensity of the tetragonal phase on the MgO–PSZ treated with a CO2 laser power density below 1.6 kW/cm2 increased with the power density and then decreased slightly as the power density increased beyond 1.6 kW/cm2. In contrast, the monoclinic phase decreased as the tetragonal phase increased.

Figure 4.16 The XRD pattern of the MgO–PSZ with various power densities between 27 and 32 2-theta angle

58

CO2 Laser Modification of the Wettability Characteristics

According to the equilibrium phase diagram for MgO–ZrO2 [157], the monoclinic phase is stable below 1240  C and the tetragonal phase is stable between 1240 and 1400  C. Above 1400  C, tetragonal and cubic phases coexist, with increasing temperature, tetragonal transforms to the cubic phase. From 2370 C to the melting temperature (2600  C) the stable phase is a cubic structure. With the increase in CO2 laser power density, the surface temperature of the MgO–PSZ will increase and create the phase transformation on the surface. As is evident from Figure 4.16, the decrease in the monoclinic phase is obvious while the tetragonal phase increases greatly when the power density is above 0.9 kW/cm2. Indeed, pyrometer readings in the range 250–2000  C showed that the surface temperature of the MgO–PSZ was above the 2000  C when the CO2 laser power density was 1.6 kW/cm2. Since the monoclinic–tetragonal transformation temperature is 1240  C, then under these conditions the monoclinic phase transformed to the tetragonal phase and the tetragonal phase reached the highest density. It has been reported that the CO2 laser cladding of ZrO2 composite coating showed a higher tetragonal phase in the laser-clad ZrO2 ceramic layer than that in the original ZrO2 powder [158]. Moreover, the plasma sprayed 8 wt % Y2O3–PSZ coating generated the metastable tetragonal phase (peak at 29 ) in the as-sprayed condition after Nd:YAG laser treatment [159]. Moreover, the intensity of the peak at 30.5 assigned as the (111) plane of the tetragonal lattice overwhelms the others, which indicated the tendency of crystal orientation. The crystal size in the direction perpendicular to the hkl plane, Dhkl , is expressed by the Scherrer equation [160] Dhkl ¼

Kl b cos a

ð4:2Þ 

where l is the wavelength of the X-ray (1.54056 A for the Ka line of Cu in this experiment), a is the Bragg angle, b is the expansion of the XRD peak caused by the crystal size and K is the Scherrer constant. We take the full-width halfmaximum (FWHM) of the peak at (111) in the XRD analysis as b, the crystallite size as Dhkl and K as 0.91. The crystallite sizes in the untreated and CO2 laser treated MgO–PSZ at various power densities are listed in Table 4.5. As shown in Table 4.5, the crystal sizes in the MgO–PSZ following CO2 laser radiation are consistently larger than the untreated sample, implying that the crystal grew after CO2 laser radiation. A high heat input from a laser beam facilitates surface localized melting at a very high efficiency. Its ability to maintain a cold substrate while melting a thin surface layer of material results in rapid quenching of the molten layer once the beam is removed. Thermal gradients at the liquid–solid interface layer are very steep and cause crystal growth taking place along the thermal gradient. The power

The Role Played by Microstructures in Terms of Crystal Size and Phase

59

Table 4.5 Crystallite size calculated from the XRD of the untreated and CO2 laser treated MgO–PSZ on the basis of the Scherrer equation Power Density (kW/cm2 ) Untreated 0.5 0.9 1.6 1.9 2.5

FWHM

a (deg)

D (nm)

4.187 3.489 2.318 1.396 2.415 2.268

15.26 15.25 15.22 15.20 15.24 15.18

34.7 41.5 65.4 103.1 60.1 64.0

density of the CO2 laser treatment has a significant effect on the crystal size. Generally, the crystal sizes in the MgO–PSZ increase with the increasing power density, with the largest crystal size of around 103.1 nm occurring at a power density of 1.6 kW/cm2. Furthermore, it was found that the crystal size varies similarly as surface energy with power density, as shown in Figure 4.17, signifying that the crystal size is correlated with the surface energy of the MgO–PSZ. According to the classical theory of nucleation and growth in solids [161], a crystallite nucleate in the form of critical embryos grows by accreting atoms from the surroundings. Assuming the embryo has a spherical shape

Figure 4.17 Relationship between surface energy and crystal size of the CO2 laser treated MgO–PSZ and power densities

60

CO2 Laser Modification of the Wettability Characteristics

of radius, r, the free surface energy of an unstrained spherical particle, G, is expressed as GðrÞ ¼ 43pr3 GV þ 4pr2 gt

ð4:3Þ

where G is the free energy of an spherical particle, GV is the free energy per unit volume of a crystal and gt is the surface energy of the tetragonal crystal as Figure 4.16 showed, the crystals of the CO2 laser treated MgO–PSZ are tetragonal. Therefore, by assuming the number of crystals in the CO2 laser treated surface to be n in unit area, the total surface energy of MgO–PSZ, gsv (considering that half of the surface of the crystal is covered by neighbour crystals), may be estimated by the expression gsv ¼ 12npr2 gt ¼ 18npD2 gt

ð4:4Þ

where D is crystal size. Since the main crystals remain as tetragonal structures in the MgO–PSZ, it is reasonable to assume that gt does not change with the crystal size. During CO2 laser processing, the n could remain the same as the crystal grows. If this is so, then gsv is proportional 1 to D2 , which is in agreement with the linear relationship between g2sv and D as shown in Figure 4.18. Therefore, the larger crystal size can be attributed to the increase in the surface energy of the MgO–PSZ. Indeed, Man et al. [153, 162] found that excimer laser treatment induced a conical structure, a microscale of the peak and valley structure, providing an extra adherent surface area for a strong adhesion joint on Si3N4 and LT35 surfaces. The joint

1

Figure 4.18 Relationship between g2SV and crystal size ðDÞ of the MgO–PSZ

Investigation of Wettability and Work Adhesion

61

strength increased with the height of the cones and the laser energy density. Kappel [128] showed that the improved adhesion strength occasioned by excimer laser texturing of ceramics is due to the formation of raised microscopic protrusions over the surface. 4.6.2 The Role of Phase Change on Surface Energy The previous XRD analysis revealed that the (101) tetragonal phase increased and the (111) monoclinic phase decreased in the CO2 laser interaction layer on the MgO–PSZ. Furthermore, the relative intensity of the tetragonal phase increased when power density was below 1.6 kW/cm2 and then decreased slightly as the power density increased further. The MgO–PSZ CO2 laser treated at 1.6 kW/cm2 power density has the highest surface energy (108.9 mJ/m2), as discussed previously, and corresponds to the highest intensity of the tetragonal phase. When the power density is at 1.9 and 2.5 kW/cm2, the surface temperature of the MgO–PSZ increased to well above the melting temperature, thereby generating more cubic phase and reducing the tetragonal phase. Associated with this phenomenon, the surface energy begins to decrease from 108.9 to 80.7 mJ/m2 and then finally to 74.9 mJ/m2. These phase changes could be represented by the different microstructures on the CO2 laser treated MgO–PSZ with various power densities. It is noticeable that tetragonal intensity varies, in a similar manner to the surface energy, with the CO2 laser power density. This indicates that the tetragonal intensity is closely correlated with the surface energy of the MgO–PSZ. In fact, it has been found that at T ¼ 0 K, the surface energy of the (101) tetragonal phase is 45 % higher than that of the (111) monoclinic phase and 95 % higher than that of the (111) cubic phase [163]. Based on this it is believed that the phase change (an increase in the tetragonal phase and a decrease in the monoclinic phase) resulted in the higher surface energy of the MgO–PSZ following the CO2 laser irradiation. 4.7 Investigation of Wettability and Work Adhesion Using Physiological Liquids In order to simulate the biological environment, the physiological fluids and simulated physiological liquids used for the wetting experiments were human blood, human blood plasma, simulated body fluid (SBF) and SBF þ BSA (bovine serum albumin). The SBF was prepared by dissolving reagentgrade chemicals, NaCl, NaHCO3, KCl, K2HPO4 3H2O, MgCl2 6H2O, CaCl2 and Na2SO4 in ion-exchanged and distilled water, buffered at pH 7.25 at 37  C with tris(hydroxymethyl) aminomethane ([CH2OH]3CNH2) and 1 m hydrochloric acid (HCl). The SBF has an inorganic ion concentration close to that found in human blood plasma, as shown later in Table 5.1.

62

CO2 Laser Modification of the Wettability Characteristics

Table 4.6 Mean values of contact angles formed between the selected and simulated physiological test liquids and the untreated and CO2 laser treated MgO–PSZ

MgO–PSZ Untreated CO2 laser (1.6 Kw/cm2 )

Contact angle, y (deg) ————————————————————————————— Human blood Human blood plasma SBF SBFþBSA 54.7 35.8

57.9 39.2

78.8 56.2

67.9 48.5

SBF þ BSA was prepared using BSA (Sigma A-9306, lot 106H0300) dissolved in SBF at pH 7.25 with a concentration of 4 mg/ml, to be used for protein studies to simulate human albumin because it is very similar to the sequence of amino acid units. The values of y formed between the selected and simulated physiological test liquids and untreated and CO2 laser treated MgO–PSZ at power density of 1.6 kW/cm2 alloy are shown in Table 4.6. It clearly reveals that the y values of all body fluids on the CO2 laser treated MgO–PSZ are lower than the untreated specimens, indicating the wettability characteristics of the material with the body fluids improved obviously after CO2 laser treatment. Further, according to Equation (3.4), the decrease in y resulted in an increase in the Wad of the MgO–PSZ towards the physiological and simulated physiological liquids. Using the referenced glv value of human blood (47.5 mJ/m2), human blood plasma (50.5 mJ/m2) [141], SBF (72.5 mJ/m2) and SBFþBSA (54.0 mJ/m2) [164], the work adhesion, Wad , of the Ti–6Al–4V alloy towards these body fluids were determined through Equation (3.4), as shown in Figure 4.19. A discernable increase in Wad of body fluids can be

Figure 4.19 Work adhesion of body fluids for untreated, mechanically roughened and CO2 laser treated MgO-PSZ

Summary

63

seen on the MgO–PSZ following CO2 laser treatment. Moreover, Wad increased as the CO2 laser power density increased. Since biomaterials first contact a proteinaceous liquid phase, almost aqueous in nature, leading to surface reorganization of proteins followed by cell attachment on biomaterials, wettability characteristics, by controlling the interaction with physiological fluids, would primarily influence cell behaviour on biomaterials. Wetting of the solid surface is a predictive index of cytocompatibility [165]. Moreover, the improvements of Wad towards these fluids would imply better suitability of titanium in a biological environment.

4.8 Summary The results presented in this chapter are a clear indication that CO2 laser surface treatment of the MgO–PSZ brought about a reduction in the y formed between the MgO–PSZ and the control test liquids, indicating that the wettability characteristics of the material were modified. The extent of this wettability characteristics modification was varied by manipulation of the CO2 laser parameters. Changes in the wettability characteristics of the MgO–PSZ were attributed to the following factors: (a) an increase in surface roughness; (b) incorporation of oxygen at the MgO–PSZ surface resulting from CO2 laser treatment; p and (c) the increase in the polar component, gsv , of the surface energy resulting from the melting and re-solidification of the MgO–PSZ surface. The p changes in gsv resulted from the melting and solidification of the MgO–PSZ surface, with the value varying with the solidified microstructure. The cellular microstructure obtained by the CO2 laser induced rapid solidificap tion corresponded to the maximum value of gsv . Indeed, it was found that the surface energy of the MgO–PSZ increased as the crystal size and tetragonal phase present increased after the CO2 laser treatment. Further analysis revealed that surface energy, by way of microstructure, was the primary influential factor governing changes in y and hence the wettability characteristics of the MgO–PSZ. Incorporation of oxygen at the surface was also shown to influence, to a lesser extent, changes in the wettability characteristics, while surface roughness was found to play a varying minor role in inducing changes in the wettability characteristics of the MgO–PSZ.

5 In vitro Biocompatibility Evaluation of CO2 Laser Treated Magnesia Partially Stabilised Zirconia This chapter is concerned with comparatively evaluating the biocompatibility of the CO2 laser treated MgO–PSZ. An investigation of the bioactivity of the CO2 laser modified MgO–PSZ in simulated body fluids (SBF) was conducted. Thereafter the effect of the CO2 laser treatment on the OH groups, the correlation between OH groups and polar surface energy and the effect of the OH groups on the apatite formation were studied. In addition, ellipsometry was used to investigate the albumin and fibronectin adsorption on the untreated and CO2 laser treated MgO–PSZ bioceramic. The relationship between the protein adsorption and surface properties of the MgO–PSZ was discussed. Finally, the in vitro behaviour of human fetal osteoblastic (hFoB) cells on the untreated and CO2 laser treated MgO–PSZ was studied. An evaluation of osteoblast cell adhesion and proliferation was also conducted to determine the effect of surface properties on the osteoblast cell adhesion and growth, elucidating the mechanisms active in the osteoblast cell response and thus deducing that the main factors are active. 5.1 Introduction The biological activity of most orthopaedic and dental biomaterials is related to their ability to promote the formation of a neoformed layer of carbonate apatite crystals analogous to bone mineral. This layer also associates specific bone proteins and is the starting point in bone reconstruction [42]. The

Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

66

In vitro Biocompatibility Evaluation

nucleation of apatite on the surface of a material is induced by the functional groups on its surface [166]. The adsorption of proteins on to a biomaterial surface from the surrounding fluid phase is rapid, with the surface properties of the biomaterial determining the type, amount and conformation of the adsorbed proteins [167]. The structure and composition of the adsorbed protein layer determine the type and extent of the subsequent biological reactions, such as osseointegration [168]. The adsorption of the protein layer may also be critical in terms of providing attachment sites for bone cells such as osteoblasts and their progenitors [169]. The cellular behaviour on a biomaterial is an important factor determining the biocompatibility. Osteoblasts are anchorage-dependent cells that must adhere to substrate surfaces prior to undergoing subsequent cell functions such as proliferation, synthesis of collagen and other extracellular matrix proteins, etc. Cell adhesion is one of the initial events essential to subsequent proliferation and differentiation of cells before tissue formation. The whole process of adhesion and spreading of the cell after contact with biomaterials consists of cell attachment, growth of filopodia, cytoplasmic webbing and flattening of the cell mass, and the ruffling of peripheral cytoplasm, which progress in a sequential fashion [170]. With the aim being to improve the biocompatibility (bioactivity and biointegration) of a magnesia–partially stabilised zirconia (MgO–PSZ), CO2 laser radiation was used to generate surface properties that would promote a better biological response. For an artificial material to bond to living bone, it is essential that the material has the ability to form a biologically active, bone-like, apatite layer on its surface in the human body. On account of this the bioactivities of the untreated and CO2 laser treated MgO–PSZ at different laser power densities were evaluated by observing the bone-like apatite formation on their surface after soaking in simulated body fluids (SBF). Protein adsorption and osteoblast cell response were performed on the untreated and CO2 laser treated MgO–PSZ in order to assess their propensity for biointegration, because protein adsorption is an almost immediate event occurring upon implantation of metals and mediates, prior even to cell response and tissue–implant interactions [9]. In addition, it is widely acknowledged that a major determinant of the bone–biomaterial interfacial response is the initial attachment, spreading and growth of osteoblasts on the implant surface and that improvements in these processes may lead to faster and more extensive implant integration and higher long-term stability [171]. Indeed, Vitale Brovarone et al. [172] investigated the in vitro behaviour of samples coated with a glass-matrix/zirconia particle composite by means of soaking in SBF followed by scanning electron microscopy (SEM) observation and X-ray diffraction (XRD) analysis. Rosengren et al. [173] studied in vitro the adsorption of proteins from diluted human plasma on hydroxyapatite,

Bone-Like Apatite Formation

67

alumina and zirconia with regard to total protein binding capacity, relative binding capacity for specific proteins and flow-through and desorption patterns. Josset et al. [174] evaluated the biocompatibility of two implantable materials, zirconia and alumina ceramics, in vitro using human osteoblast cell cultures. Furthermore, Bosetti et al. [61] used methods of soaking in SBF, protein adsorption and cell culture for an in vitro characterisation of a biomedical device.

5.2 Sample Preparation The MgO–PSZ, with properties as described in Section 4.2.1, was subjected to various in vitro evaluations to determine the effects of CO2 laser surface treatment on bioactivity. In order to perform all of the in vitro tests, the MgO–PSZ sheet was cut into 30 blocks each of 50
12
12:15 mm3 with a cutting machine (Miniton; Struers, GmbH) using a diamond-rimmed cutting blade, used as received prior to CO2 laser treatment. The 30 blocks were then divided into two groups of 15 samples, with the groups being untreated and CO2 laser treated. The CO2 laser processing of the materials was conducted in the same manner as described in Section 4.2.2. For the in vitro apatite formation test, the samples used were CO2 laser treated power densities ranging from 0.6 to 2.5 kW/cm2. For the protein adsorption test, only samples that were CO2 laser treated with laser power densities of 0.9 and 1.6 kW/cm2 were used. Samples CO2 laser treated with power densities ranging from 0.6 to 2.5 kW/cm2 were used in the in vitro osteoblast cell adhesion and proliferation test, while samples CO2 laser treated with power densities of 0.9 and 1.6 kW/cm2 were used in the evaluation of cell functions. Untreated samples were used as control in all of the in vitro tests. 5.3 Bone-Like Apatite Formation Histological examinations in vivo show that an apatite layer is formed on the ceramic [41] surface early in the implantation period and thereafter the bone matrix integrates into the apatite. This apatite layer consists of nanocrystals of carbonate-ion-containing apatite that has a defective structure and low crystallinity. These features are, in fact, very similar to those of the mineral phase in bone and hence bone-producing cells (osteoblasts) can preferentially proliferate on the apatite and differentiate to form an extracellular matrix composed of biological apatite and collagen. As a result, the surrounding bone comes into direct contact with the surface apatite layer. When this process occurs a chemical bond is formed between the bone mineral and the surface apatite to decrease the interfacial energy between them. It can be

In vitro Biocompatibility Evaluation

68

concluded from these findings that an essential requirement for an artificial material to bond to living bone is the formation of a layer of biologically active bone-like apatite on its surface in the body [41]. There have been considerable efforts for improving the bioactivity of zirconia inert bioceramics. It has been revealed that a zirconia gel forms an apatite on its surface in SBF, indicating that the Zr–OH group is able to induce apatite nucleation [166]. The investigation of the apatite-forming ability of zirconia gels with different structures shows that specific structures of the Zr–OH group in tetragonal or monoclinic zirconia are effective for inducing apatite nucleation [175–177]. For the purpose of the apatite formation, the chemical treatment has been used to produce the Zr–OH group on a zirconia/alumina composite subjecting the composite to H3PO4, H2SO4, HCl or NaOH aqueous solution treatments [178] and on zirconium metal treated with aqueous NaOH [179]. 5.3.1 Experimental Procedures FTIR Analysis The optical adsorption spectra were measured at room temperature by means of a Fourier transfer infrared (FTIR) (FTS135; Bio-Rad, Inc.) spectrometer over the 500–5000 cm1 range at a resolution of 1 cm1. Soaking in Simulated Body Fluid The samples were soaked in an acellular SBF [41], having an ion concentration nearly equal to that of human blood plasma. This solution, whose composition is reported in Table 5.1, was prepared by dissolution of highpurity reagents in distilled water, and was buffered at 7.25 with 50 mM trishydroxymethyl amino ethane and 45 mM hydrochloric acid. The untreated and CO2 laser treated samples (treated with various power densities) were immersed in 30 mL SBF in a polyethylene bottle at 37  C, without stirring. After 14 days they were removed from the solution, gently washed in distilled water and dried at room temperature. The soaked samples were then characterised by SEM and EDX, the details of which

Table 5.1 Ionic concentration and pH of SBF in comparison with those in human blood plasma [41] Concentration Ion SBF Blood plasma

Naþ

Kþ

Mg2þ

Ca2þ

Cl

HCO 3

HPO2 4

SO2 4

pH

142.0 142.0

5.0 5.0

1.5 1.5

2.5 2.5

148.8 103.8

4.2 27.0

1.0 1.0

0.5 0.5

7.40 7.40

Bone-Like Apatite Formation

69

Absorbance (arb. units)

0.6 arb. units 2.5 kW/cm2 1.9 kW/cm2 1.6 kW/cm2 0.9 kW/cm2 0.6 kW/cm2 Untreated 1000

1500

2000

2500

3000

3500

4000

Wavenumber (cm−1)

Figure 5.1 Infrared spectra of the untreated and CO2 laser treated MgO–PSZ (treated with different power densities)

are given in Section 4.2.3. The samples for SEM observations were simply dried and covered by a thin gold layer to guarantee the conductivity. 5.3.2 Spectral Analysis and Hydroxyl Group The main regions in the 750–950 cm1 range are ascribed to ZrO2 stretching modes, as shown in Figure 5.1, owing to the fact that it is similar to infrared (IR) peak of the 20 mol % Al2O3-doped ZrO2 nanoparticles reported previously [180] and the IR peak of the Al2O3–ZrO2 nanopowders after laser ablation [181]. The vibration around 3300–3500 cm1 in the FTIR spectra (see Figure 5.2) could be attributed to the OH groups. Indeed, OH stretching vibrations around this region have been observed by other workers on Al2O3–ZrO2 nanopowders after Nd:YAG laser ablation [181] and on the Fe-doped crystals after laser irradiations [182]. As one can see from Figure 5.2, the absorption coefficient of the OH group on the MgO–PSZ in this region increased after CO2 laser irradiation and varied with the power density employed. For the untreated sample and the CO2 laser treated sample (power density of 0.6 kW/cm2), the absorption peaks from 3200 to 3600 cm1 are not obvious, indicating that no OH groups bonded on these samples. In contrast, the absorption peaks at this region can be clearly observed on the samples following the CO2 laser irradiation with power densities of 0.9, 1.6 and 1.9 kW/cm2, denoting that OH groups existed on these samples. The highest absorption coefficient of OH groups was obtained on the sample that had been treated at 1.9 kW/cm2. This finding shows that the OH groups increased with the CO2 laser power density used. This relationship was also seen by Zeng, Yung and Xie [183] on the OH

In vitro Biocompatibility Evaluation

70

2.5 kW/cm2 Absorbance (arb. units)

0.05 arb.units

1.9 kW/cm2 1.6 kW/cm2 0.9 kW/cm2 0.6 kW/cm2 Untreated 3000 3500 Wavenumber (cm−1)

4000

Figure 5.2 Infrared spectra of the hydroxyl groups presents on the surface of the untreated and CO2 laser treated MgO–PSZ (treated with different power densities)

groups bonded on to copper after CO2 laser treatment; nevertheless, the absorption peaks in this region on the sample treated with a CO2 laser power density of 2.5 kW/cm2 was not obvious, suggesting that the OH bond does not increase linearly with the CO2 laser power density. The explosive evaporation due to the super-high temperature on the MgO–PSZ surface treated at this power density caused water vaporisation and disappearance of the OH band. The phenomena of losing OH groups was also found on the human dentine after Er:YAG (erbium-doped YAG) laser irradiation [184]. CO2 is a weak acid and is known to absorb on ZrO2 in the form of both carbonate and bicarbonate species [185]. Carbonate structures are formed via the interaction of CO2 with zirconium cations in the lattice, as well as with a surface oxygen atom, whereas bicarbonate structures are formed via the interaction of CO2 with a hydroxyl group. The peak from 2800 to 3000 cm1 testified to the existence of carbonate structures on the MgO–PSZ surface. The change of carbonate structures has the same trend as the OH groups discussed above. The formation of the hydroxyl groups on the MgO–PSZ is due to the reactions of the zirconia with water vapour in air during CO2 laser processing. The hydroxyl ion is a common impurity in insulating crystals and by interacting with other impurities it gives rise to new complexes. The OHstretching frequency is a very sensitive probe of the hydroxyl environment. Proper thermal treatments and isotopic substitutions allowed the stretching mode absorption lines to be assigned to the defects in which OH is embedded and to supply possible models for them [185]. Crystal growth

Bone-Like Apatite Formation

71

from the melt is commonly carried out in air atmosphere as air always contains a certain degree of humidity from which OH ions are incorporated into the lattice [186]. CO2 laser irradiation is a thermal process. When the laser fluence exceeds the ablation threshold, the irradiated surface experiences melting, followed by evaporation, whereupon the particles emit from the surface. At a higher fluence, the amount of particles increases and they break out quickly from the superheated surface to produce a high-density vapour plume wherein a portion of the particles are ionised due to the thermal ionisation. The main reactions to occur in the melt ceramic and vapoured flume are ZrO2 , Zr4þ þ 2O 2

ð5:1Þ

and the following oxidoreduction reactions occur at the melt–atmosphere interface: O 2 ! 12 O2 þ 2e

ð5:2Þ

H2 O þ e ! 12 H2 þ OH

ð5:3Þ

O 2 þ 2H2 O ! 2OH þ H2 þ 12 O2

ð5:4Þ

the whole reaction being

Finally, the OH ions produced according to Reaction (5.4) would be incorporated with one, two and three and four surface Zr4þ respectively. According to the classification proposed by Tsyganenko and Filimonov [187], the OH groups bonded in the spectral ranges at 3770 and 3680 cm1 are typical one and three surface Zr4þ cations, respectively, in tetragonal zirconia while the OH groups at 3775 and 3675 cm1 bonds are one and more than one (possibly three) surface Zr4þ ions, respectively. It has been speculated that surface melting was occasioned on the MgO– PSZ treated by the CO2 laser treatment with 1.6 kW/cm2 power density. In turn the Zr4þ ion and OH were produced and reaction between these ions brought about the Zr–OH group on the MgO–PSZ. The relatively high amounts of the hydroxyl groups bonded on to the modified samples with 1.6 and 1.9 kW/cm2 were associated with the melting and chemical reaction on the MgO–PSZ surface. 5.3.3 The Correlation between OH Groups and Wettability Characteristics The values of the surface energy have been calculated for the MgO–PSZ treated by the various power densities in detail (see Chapter 4) and are

In vitro Biocompatibility Evaluation

72

Table 5.2 Determined surface energy values for the MgO–PSZ before and after CO2 laser treatment (treated with various power densities and traverse speed of 2000 mm/min) Surface Energy (mJ/m2) gdsv p gsv gsv

Untreated 42.7 10.1 52.8

CO2 laser treated (kW/cm2) ———————————— ——————————————————————— 0.5 0.9 1.6 1.9 2.5 43.8 10.4 54.2

44.4 21.9 66.3

48.2 60.7 108.9

47.5 33.2 80.7

48.2 26.7 74.9

shown in Table 5.2. It is found that the absorption coefficient of the p OH group (see Figure 5.2) and gsv on the MgO–PSZ increased after CO2 laser irradiation (see Table 5.2) and varied with the power density employed. The untreated sample and the samples treated with the lower power densities had relatively low absorption peaks of the OH groups and lower p gsv . On the other hand, the relatively high absorption peaks of the OH groups and higher surface energy existed on the samples following the CO2 laser irradiation with power densities of 1.6 and 1.9 kW/cm2. Moreover, when the OH groups decreased on the sample treated at the power density p of 2.5 kW/cm2, there was a corresponding decrease in gsv . It is also interesting to notice that the melting of the MgO–PSZ was the fundamental reason for the induction of the OH groups and improvement of the surface energy. This finding implied that there was a correlation between the OH p groups and gsv . Indeed, Takeda et al. found that the surface OH groups governed the wettability of commercial glasses [188] and adsorption properties of metal oxide films [189]. A previous study [190] indicated that in the case of cassiterite its wettability strongly depends on the acid–base interactions (polar component) resulting from the presence of OH groups and physically adsorbed water on it. For the surface of the ‘dry’ cassiterite its surface free energy practically results only from Lifshitzvan der Walls (dispersive component) intermolecular interactions. Most metal oxides are hydroxylated under normal conditions, i.e. at room temperature and when water or its vapour has had access to the surface. It was stated that the acid– base component of surface energy of the zirconia probably depends on the density of OH groups on the surface of the solids studied [191]. Indeed, the acid–base component of surface energy presented the majority of the forces as the functions of the particular chemical nature of a certain material p corresponding to gsv [192]. As such, it is believed that the CO2 laser induced p hydroxyl groups could be a major factor influencing gsv and, in turn, the wettability characteristics of the MgO–PSZ.

Bone-Like Apatite Formation

73

5.3.4 The Effects of CO2 Laser Treatment on the MgO–PSZ in Simulated Body Fluids As one can see from Figure 5.3, very small amounts of sediment are apparent on the surface of the untreated MgO–PSZ (Figure 5.3(a)) and the 0.6 kW/ cm2 CO2 laser treated MgO–PSZ (Figure 5.3(b)). In contrast, considerable amounts of sediment were clearly discernible on the samples that were CO2 laser treated with power densities of 0.9 (Figure 5.3(c)), 1.6

Figure 5.3 SEM images of the MgO–PSZ soaked in the SBF: (a) untreated, (b) 0.6 kW/ cm2, (c) 0.9 kW/cm2, (d) 1.6 kW/cm2, (e) 1.9 kW/cm2 and (f) 2.5 kW/cm2

74

In vitro Biocompatibility Evaluation

Figure 5.4 SEM image and EDX analysis of the apatite formed on the surface of the MgO–PSZ when treated with power densities of (a) 1.6 kW/cm2 and (b) 1.9 kW/cm2

(Figure 5.3(d)), 1.9 (Figure 5.3(e)) and 2.5 kW/cm2 (Figure 5.3(f)), with the highest amount of sediment being observed on the sample that was CO2 laser treated at 1.9 kW/cm2. The EDX analysis shows that most particles on the soaked surface are NaCl sediments as the elements of Na and Cl, as shown in the Figure 5.4(a). Apatites were only found on the samples treated at power densities of the 1.6 and 1.9 kW/cm2. As shown in Figure 5.4(a), only a few apatites formed on the sample treated at a power density of 1.6 kW/cm2, with only a small amount of Ca element shown in the EDX analysis given in Figure 5.4(a). However, on the sample treated with a CO2 laser power density of 1.9 kW/ cm2, some apatites were observed. One of them is shown in Figure 5.4(b) with the Ca:P ratio about 1.65. This Ca:P ratio exhibits the calcium phosphate transforms into apatite [41]. The Effect of OH Groups There was no occurrence of apatite formation on the untreated and certain CO2 laser treated samples (0.6, 0.9 and 2.5 kW/cm2) that displayed few hydroxyl groups. On the other hand, some apatites formed on other CO2

Protein Adsorption

75

laser treated samples (1.6 and 1.9 kW/cm2) that did display any hydroxyl groups. This finding suggests that the hydroxyl group on the MgO–PSZ could be the predominant factor governing the formation of the apatites. The hydroxyl groups on the MgO–PSZ surface certainly generate Zr–OH groups, which have been shown to be functional groups for the formation of the apatite [166]. It is suggested that Zr–OH functional groups formed on the samples in the CO2 laser processing at certain parameters and that such functional groups naturally brought about the nucleation of the apatite on these samples in the simulated body fluid environment. The nucleation of the apatite could yield the apatite formation and bone-bonding ability to the MgO–PSZ modified to have Zr–OH groups on the surface. The Effect of Wettability It was found that there were more sediments and apatites on the surface with the higher surface energy than on the surface with the lower surface energy. In the process of Ca–P precipitation, the variations of the Gibbs function ð
GÞ of the MgO–PSZ with the higher surface energy should be greater, compared to that of the MgO–PSZ surface with lower surface energy. This finding, agreeing with the study by Feng et al. [193], suggested that the adsorption and reaction would more easily have occurred on the surface with the higher surface energy, especially the polar component of surface energy, which would be beneficial to the chemical force and bonding. 5.4 Protein Adsorption The molecules involved in cell adhesion and spreading include extracellular matrix molecules, transmembrane receptors and intracellular cytoskeletal components. Among the extracellular matrix proteins shown to mediate cell attachment to substrates, fibronectin is protein found in many extracellular matrices and in blood plasma and serves as an attachment molecule between the substrate and cell membrane of anchorage-dependent cells. It is known that the ligand fibronectin connects to the cell membrane via integrin receptors. The activation of integrins triggers cytoplasmic reactions, and thereby stimulates the intracellular signalling pathway and subsequently cellular functions such as proliferation and differentiation [169]. On the other hand, human albumin is a nonadhesive protein for osteoblasts [194]. Albumin is the major protein component of serum and dominates the adsorption of phenomena on medical implants in the first stage of contact with body fluids. Human serum albumin or bovine serum albumin (BSA) coatings are often used as a passivating agent to prevent the adhesion of cells and thrombus formation [195].

76

In vitro Biocompatibility Evaluation

An important fact to highlight is that different surfaces provide very different opportunities for protein binding. The investigation of competitive protein adsorption showed that BSA in a single-component solution adsorbed on to a hydrophobic surface two times more than that on to a hydrophilic surface [196]. The surface wettability of biomaterials affects the ability of cells to reorganise pre-adsorbed fibronectin and to form their own matrix by secreted fibronectin. Moderate wettable and hydrophilic surfaces are ideal for better interactions with cells while hydrophobic substrata inhibit early and late matrix formations. It is possible that there is a critical value in the strength of fibronectin adsorption that regulates the ability of cells to construct a fibronectin matrix [197]. 5.4.1 Experimental Procedures Protein Adsorption The proteins used for this study were human serum albumin and human plasma fibronectin (Calbiochem, Inc.). Prior to the adsorption of 1 mg/ml albumin in phosphate buffered salines (PBS), MgO–PSZ samples were rinsed with deionised water. The individual samples were transferred into a 24-well tissue culture plate. Thereafter, 2.5 ml of prepared albumin solution was added into each well. Adsorption proceeded for 1 hour in an incubator at 37  C. After adsorption was complete, the samples were dried with N2 and immediately transferred to an ellipsometer for measurement of the adsorbed protein layer. The above procedure was repeated with a 0.2 mg/ml concentration of fibronectin in PBS. Ellipsometric Measurement Human plasma fibronectin was measured using an automatic ellipsometer equipped with a 633 nm helium–neon laser (L117F; Gaertner, Inc). The thickness and refractive indices of protein films were determined using an  ellipsometer computer program with an accuracy of 3 A. Four ellipsometer measurements at different locations on each sample were taken and the average value was calculated. Statistics Statistical analysis was performed with an SPSS v.12 software package (SPSS/PC, Inc.). Data are reported as mean SD (standard deviation) at a significance level of p < 0:05. After having verified normal distribution and homogeneity of variances, one-way ANOVA and Scheffe´’s post hoc multiple comparison tests were done.

Protein Adsorption

77

5.4.2 Albumin and Fibronectin Adsorption on CO2 Laser Treated MgO–PSZ The thickness of the absorbed fibronectin layer was less on the untreated sample than that on the CO2 laser treated sample, as shown in Figure 5.5. The statistical analysis reveals that the thickness of absorbed fibronectin on the untreated sample was similar to that on the sample CO2 laser treated at the power density of 0.9 kW/cm2, but significantly less than that on the sample CO2 laser treated at the power density of 1.6 kW/cm2 ðp < 0:01Þ. On the other hand, the thickness of the absorbed albumin layer on the untreated MgO–PSZ was higher than that on the CO2 laser modified sample, as shown in Figure 5.5. The statistical analysis reveals that the thickness of the absorbed albumin layer on the untreated sample was significantly higher than on the CO2 laser treated samples ðp < 0:01Þ. From Figure 5.5 it is apparent that the CO2 laser power density applied in the experiments was negatively correlated to the amounts of albumin, but positively correlated with the fibronectin. The results showed that the CO2 laser treatment promoted the adsorption of the fibronectin on the MgO–PSZ and the amount of the adsorbed fibronectin was positively correlated with the CO2 laser power density applied in the experiments, as shown in 700

Thickness of Adsorbed Protein Layer (Å)

Fibronectin

Albumin

600 * *

*

500

*

400

300

Untreated

CO2 laser CO2 laser 0.9 kW/cm2 1.6 kW/cm2

Untreated

CO2 laser CO2 laser 0.9 kW/cm2 1.6 kW/cm2

Figure 5.5 Thickness of the adsorbed fibronectin and albumin layer on the untreated and CO2 laser treated MgO–PSZ (treated with different power densities). For the fibronectin adsorption, there was a significant statistical difference in thickness between the untreated MgO–PSZ and the sample CO2 laser treated at 1.6 kW/cm2, and no statistical difference between the untreated MgO–PSZ and the sample CO2 laser treated at 0.9 kW/cm2. For the albumin adsorption, there was a significant statistical difference in thickness between the untreated MgO–PSZ and the samples that were CO2 laser treated at 0.9 and 1.6 kW/cm2, and no statistical difference between the CO2 laser treated samples ð p < 0:05Þ

In vitro Biocompatibility Evaluation

78

Figure 5.5. However, the thickness of the adsorbed human serum albumin layer on the untreated MgO–PSZ is higher than that on the CO2 laser modified MgO–PSZ and was negatively correlated with the CO2 laser power densities. This finding is perhaps not so surprising as the various CO2 laser power densities used brought about changes in wettability characteristics and surface roughness of the MgO–PSZ: it is known that protein adsorption is influenced by the surface topography (roughness) [198] and the surface chemistry (wettability characteristics) [199]. The Effects of Surface Roughness By altering the CO2 laser power density, it was possible to obtain the narrow range of surface roughness values detailed in Figure 5.6 (see also Section 4.3.3). The experimental results given in Figure 5.6 reveal that the amount of fibronectin adsorption increased, while the amount of albumin adsorption decreased with the surface roughness of the MgO–PSZ. These trends in absorption for the fibronectin and the albumin were verified to a large extent by the results of the statistical analysis.

700

Thickness of Adsorbed Protein Layer (Å)

Fibronectin

Albumin

600 *

* *

500

*

400

300

0.2

0.4

0.6

0.8

0.2

0.4

0.6

0.8

Surface Roughness, Ra (µm)

Figure 5.6 The relationship between the thickness of adsorbed fibronectin and albumin layer and Ra of the MgO–PSZ. For fibronectin adsorption, there was a significant statistical difference in thickness between the sample with Ra of 0.295 mm and the sample with Ra of 0.717 mm, and no statistical difference between the sample with Ra of 0.295 mm and the sample with Ra of 0.313 mm. For albumin adsorption, there was a significant statistical difference in thickness between the sample with Ra of 0.295 mm and the samples with Ra of 0.313 and 0.717 mm, and no statistical difference between the sample with Ra of 0.313 mm and the sample with Ra of 0.717 mm ð p < 0:05Þ

Protein Adsorption

79

The relationship between the albumin adsorption and surface roughness given in Figure 5.6 is consistent with the findings of other researchers, insofar as bovine serum was adsorbed preferentially on to the smooth substratum [200, 201], thus implying that the surface roughness of the MgO–PSZ is one of the factors active in albumin adsorption. It was explained, however, by MacDonald et al. [202] that by roughing the surface of Ti one would obtain a more hydrophilic surface. This increase in surface hydrophilicity of the Ti, according to Serro et al. [201], will consequently result in lower albumin adsorption. The relationship between the fibronectin adsorption and surface roughness given in Figure 5.6 is in agreement with previous reports. Deligianni et al. [200] found that a roughened Ti alloy sample adsorbed much more fibronectin than a smooth sample. The much higher affinity of rough substrata to fibronectin could be the driving force for preferential adsorption of fibronectin; however, other researchers have reported that the amounts of immobilised fibronectin on the rough titanium were 50 % lower than those adsorbed on the smooth one [203]. This decrease was noticed when the roughness was produced by polishing or sandblasting, followed by acid attack, which is an indication that the chemical or mechanical manufacturing process, used to achieve the surface texture, might influence the protein adsorption behaviour of a surface [203]. Hence, a simple conclusion would be difficult to execute for the relationship between the amplitude of surface roughness and protein adsorption. It must be noted that in this work the CO2 laser treatment effects change in other surface properties besides roughness. It is most likely that the surface roughness plays a role in the protein adsorption, but its effects correlate to and are less than the wettability characteristics of the MgO–PSZ. The Effects of Wettability Characteristics The previous results are a clear indication that interaction of the CO2 laser beam with the MgO–PSZ brought about a decrease in y, which in some instances was considerable. This in turn naturally gave rise to improved wettability characteristics. As one can see from Figure 5.7, as the wettability characteristics of the MgO–PSZ increased, the adsorbed amounts of fibronectin increased, while the adsorbed amounts of albumin decreased. Indeed, this observation was supported somewhat by the results of the statistical analysis. The results of the albumin adsorption are consistent with the previous finding that the increase in surface hydrophilicity of Ti results in lower albumin adsorption [201], showing that the wettability characteristics of the MgO–PSZ could be the main factor active in the albumin adsorption. The results of the adsorption of fibronectin show that it increased on the hydrophilic surface. The previous investigation [204] on the extent of fibronectin adsorption as

In vitro Biocompatibility Evaluation

80 700

Thickness of Adsorbed Protein Layer (Å)

Fibronectin

Albumin

600

*

* *

500

*

400

300 0.0 0.2 0.4 0.6 0.8

0.0 0.2 0.4 0.6 0.8

Wettability, cos θ (glycerol)

Figure 5.7 The relationship between the thickness of adsorbed fibronectin and albumin layer and wettability characteristics ðcos yÞ of the MgO–PSZ. For the fibronectin adsorption, there was a significant statistical difference in thickness between the sample with cos y ¼ 0:19 and the sample with cos y ¼ 0:77, and no statistical difference between the sample with cos y ¼ 0:19 and the sample with cos y ¼ 0:47. For the albumin adsorption, there was a significant statistical difference in thickness between the sample with cos y ¼ 0:19 and the sample with cos y ¼ 0:77, and no statistical difference between the sample with cos y ¼ 0:47 and the sample with cos y ¼ 0:77 ð p < 0:05Þ

compared to its biological activity on hydrophobic and hydrophilic surfaces suggested the possibility that fibronectin was adsorbed in two different conformations when incubated with the surfaces at low concentrations, with the more active conformation on the hydrophilic surfaces. The results showed that the antiplasma fibronectin antibody appeared to bind to the conformation of fibronectin adsorbed on hydrophilic surfaces much better than the conformation of fibronectin adsorbed on hydrophobic surfaces [204]; therefore, the wettability characteristics of the MgO–PSZ could be the predominant mechanism governing the fibronectin adsorption. It is p noticeable that considerable change in the gsv , instead of the minor difference in gdsv , was the main mechanism governing the wettability characteristics of the MgO–PSZ after CO2 laser irradiation, indicating that the albumin and fibronectin adsorption on the MgO–PSZ surfaces was probably due to the polar and chemical interactions [205].

5.5 Osteoblast Cell Response The development of bone–implant interfaces depends on the direct interactions of bone matrix and osteoblasts with the biomaterial. There is a

Osteoblast Cell Response

81

substantial body of literature based on the premise that improved initial attachment of osteoblasts or osteoblast precursor cells to orthopaedic implant surfaces may lead to improved bone integration of the implant and longer-term stability [206]. Osteoblast adhesion is a prerequisite for bone–biomaterial interaction and depends on the surface aspect of materials. A cell in contact with the surface of a material will firstly attach, adhere and then finally spread. The quality of this adhesion will influence their morphology and their future capacity for proliferation and differentiation. The attachment of anchorage-dependent cells such as osteoblasts to biomaterial surfaces is a complex process involving cell attachment and spreading [207], focal adhesion formation, and extracellular matrix formation and reorganisation [208]. 5.5.1 Experimental Procedures Osteoblast Cell and Cell Culture The human osteoblastic cell line hFOB 1.19 was obtained from American Type Culture Collection (ATCC, Inc.). The hFOB cell line was established by transfection of limb tissue obtained from a spontaneous miscarriage. The cells have the ability to differentiate into mature osteoblasts expressing the normal osteoblast phenotype and provide a homogeneous, rapidly proliferating model system for study of normal human osteoblast cells. Moreover, it overcomes the disadvantages of earlier in vitro model systems, namely the unknown species-specific phenotype characteristics of animal osteoblast cultures and the very slow rates of proliferation, as well as the short lifetime of primary cultures derived from normal human bone [209]. The cells were cultured in a medium containing a 1:1 mixture of Dulbecco’s modified Eagle’s medium without phenol red and Ham’s F-12 medium with 2.5 mM L-glutamine (D-MEM/F-12 medium), supplemented with 10 % fetal bovine serum (ATCC, Inc.) and 0.3 mg/ml G418 (Calbiochem, Inc.) at 37  C in a humidified 5 % CO2 incubator. Osteoblasts at passage numbers 2–4 were used in this experiment. Cell Cytotoxicity Cytotoxicity tests consisted of the quantification of the activity of lactate dehydrogenase (LDH) in culture medium of cells in contact with the samples. The activity of the LDH enzyme rises when cells are damaged. Thus the LDH activity induced by the untreated and selected CO2 laser treated specimens (0.9 and 1.6 kW/cm2) in triplicate were compared to that induced by a toxic agent (Triton X100 0.05 % in PBS) and to that induced by a culture polystyrene plate (NUNC, Inc.). The cell culture plate was used as a negative control and a Triton toxic agent as a positive control.

82

In vitro Biocompatibility Evaluation

Cell Adhesion and Morphology The MgO–PSZ samples were placed in a 24-well tissue culture polystyrene plate (NUNC, Inc.), sterilised in 70 % alcohol and rinsed in PBS solution. To analyse the osteoblast cell attachment and morphology, one untreated sample and one CO2 laser modified sample (power density of 1.6 kW/cm2) were used in the assessment of cell morphology. The specimens were seeded with the 0.5 ml cell suspension of 1
105 cell/mL and analysed by SEM after 24 hours of cell culture. For a 7 day osteoblast cell adhesion analysis, the samples were rinsed in PBS, whereupon they were seeded with 0.5 mL cell suspension (4
105 cell/ml). After culturing, the cells were fixed in 2.5 % glutaraldehyde solution for 1 hour, washed with PBS and then dehydrated in increasing concentrations of alcohol (70, 85 and 100 %). Thereafter, the osteoblast cells were dried in a critical point dryer (CPD030; BAL-TEC, GmbH). Then the samples were examined with by SEM after sputter gold coating. For the cell adhesion analysis, three images were taken for the each sample at different areas and a typical one was chosen for the analysis. Cell Proliferation Each group of specimens used for cell proliferation tests in triplicate was measured after cell culturing for 14 days. Osteoblast cells were cultured on specimen surfaces with the 0.5 ml cell suspension of 1
105 cell/ml in 6-well culture plates. The cell culture medium was changed every 3 days. At every harvest time point, cells were detached from specimen surfaces by incubation with trypsin/EDTA (ethylene diamine tetraacetic acid) (0.5 g/l trypsin and 0.2 g/l EDTA) (GIBCO, Ltd) for 5 minutes at 37  C and each specimen was washed with PBS. Released cells were counted with a hemocytometer, and on every specimen counting was repeated three times. Alkaline Phosphatase Assay For the staining of human osteoblast cells an alkaline phosphatase (ALP) assay kit (Sigma Diagnostics, Inc.) was used. After rinses with PBS, the samples with cells were fixed by immersing in a citrate buffered acetone for 30 seconds and rinsed gently with deionised water for 45 seconds. Alkaline dye mixture was added and the samples were incubated at 24  C for 30 minutes protected from direct light. They were then rinsed thoroughly in deionised water for 2 minutes and placed in Mayer’s hematoxylin solution for 10 minutes. Positive staining for alkaline phosphatase (red–violet) was identified by light microscopy and evaluated by scoring cell rating and count according to the characterisation method provided.

Osteoblast Cell Response

83

Statistics The statistical analysis of the results was performed with an SPSS v.12 software package in the same manner as discussed in Section 5.4.1. 5.5.2 Osteoblast Cell Response on CO2 Laser Treated MgO–PSZ Cell Cytotoxicity LDH is a kind of enzyme in the cell. When cells are damaged or broken, the LDH will leak into the culture medium. As the concentration of LDH in the medium is proportional to the numbers of dead/damaged cells, the concentration of LDH in the medium can reflect the cytotoxicity. The LDH activity in the culture media obtained from cells cultured on all the tested materials was found to be not significantly different from the negative control, as shown in Figure 5.8, indicating that untreated and CO2 laser treated MgO–PSZ were not cytotoxic. Cell Attachment Figure 5.9(a) shows that no osteoblast cells were observed on the untreated MgO–PSZ after 24 hours of cell incubation, whereas a few cells attached on

60

50

LDH Activity

40 * *

*

*

30

20

10

0

Positive Control

Negative Control

Untreated MgO-PSZ

CO2 Laser CO2 Laser MgO-PSZ MgO-PSZ 0.9 kW/cm2 1.6 kW/cm2

Figure 5.8 Results provided by assessment of cell membrane damage are expressed as LDH activity (U/L) SD at 340 nm. There was a significant statistical difference between the positive control and MgO–PSZ samples, and no statistical difference between the negative control and the untreated and CO2 laser treated MgO–PSZ ð p < 0:05Þ

84

In vitro Biocompatibility Evaluation

Figure 5.9 SEM image of hFOB human osteoblast cells after 24 hours on (a) the untreated MgO–PSZ and (b) the CO2 laser treated MgO–PSZ at power densities of 1.6 kW/cm2

to the CO2 laser treated MgO–PSZ (Figure 5.9(b)). It is quite clear that the osteoblast cell attachment on the MgO–PSZ was influenced by the CO2 laser treatment, indicating that the surface properties generated by the CO2 laser treatment were more favourable for the osteoblast cell attachment. Furthermore, the cells on the CO2 laser treated samples showed filopodia and spread well (Figure 5.9(b)), denoting good cell attachment. Moreover, it is evident from Figure 5.10 that the osteoblast cells had different morphologies in different regions of the CO2 laser treated track. Figure 5.10(a) shows that the osteoblast cells at the edge underwent initial spreading and the individual cell was found to cover an area of about 30– 40 mm, as shown in Figure 5.10(b). Short filopodia protruded and elongated about 5–10 mm from the osteoblast cell (see Figure 5.10(b)). The elongation direction of the short filopodia implies the direction of the migration process. Conversely, the osteoblast cells at the centre reached a stage where they grew and spread to cover a region of 60–150 mm (Figure 5.10(a)). One typical osteoblast cell (see Figure 5.10(c)) had spread completely and flattened, with the cytoplasmic spread to cover an area of about 30–50 mm as well as forming four filopodias, two of them elongated to a length of 50–60 mm. Likewise, another two osteoblast cells (Figure 5.10(d)) had a flat cytoplasm with two filopodias elongated to 50–60 mm. The morphologies of these osteoblast cells display the final stage of cell attachment. In general, osteoblast cells in the centre spread better and reached a higher stage of cell attachment than those at the edge of the CO2 laser treated track. Since the CO2 laser treatment exerted a higher photochemical effect at the centre than on the edge due to intensity distribution of the CO2 laser TEM01 beam mode (see Section 4.3.3), it could be concluded that the osteoblast cell spreading and attachment is influenced by the level of the CO2 laser treatment.

Osteoblast Cell Response

85

Figure 5.10 SEM images of hFOB human osteoblast cells after 24 hours (a) at the interface region, (b) at the circled area at the edge and (c) and (d) at the circled areas at the centre of the CO2 laser treated MgO–PSZ with 1.6 kW/cm2 power density

Cell Growth After a 7 day incubation period, the hFOB cells grew well and formed a layer on all of the samples (see Figure 5.11). The degree of osteoblast cell adhesion and growth in terms of cell coverage area varied with the CO2 laser power density. The cover density is defined by the ratio of the osteoblast cell adhesion area to the whole surface area and is used as an indication of the osteoblast cell adhesion and growth. It has been found that the CO2 laser power density used in the treatment had a significant influence on the cover density of the osteoblast cells (see Figure 5.12). For instance, compared with the untreated sample, a power density of 0.9 kW/cm2 generated double cover density, while the higher power densities of 1.6, 1.9 and 2.5 kW/cm2 brought about triple cover density on the CO2 laser treated MgO–PSZ. Generally, the osteoblast cell coverage area was found to increase as the power density increased when the power density is less than 1.9 kW/cm2.

86

In vitro Biocompatibility Evaluation

Figure 5.11 SEM images of hFOB cells on (a) the untreated MgO–PSZ and CO2 laser treated MgO–PSZ at power densities of (b) 0.5 kW/cm2, (c) 0.9 kW/cm2, (d) 1.6 kW/ cm2, (e) 1.9 kW/cm2 and (f) 2.5 kW/cm2

Figure 5.13 indicates that the number of osteoblast cells after 14 days on the untreated MgO–PSZ is less than the CO2 laser treated MgO–PSZ. Especially, compared with the untreated sample, the osteoblast cell grows significantly faster on the samples that were CO2 laser treated at relatively high power densities of 1.6, 1.9 and 2.6 kW/cm2. Indeed, both cell adhesion and growth in 7 days investigated by SEM and cell proliferation after

Osteoblast Cell Response

87

100

*

*

*

Cell Cover Density (%)

80

60

40

20

0

0.0

0.5

1.0

1.5

2.0

2.5

CO2 Laser Power Density (kW/cm2)

Figure 5.12 The relationship between the cover density of the hFOB cells and CO2 laser power density. There was a significant statistical difference between the untreated and CO2 laser treated MgO–PSZ samples, and no statistical difference among the samples CO2 laser treated at 1.6, 1.9 and 2.6 kW/cm2 ð p < 0:05Þ

14

Total No of Cells (×105)

12 ∗ 10 8

∗

∗

6 4 2 0

0.0

0.5 1.0 1.5 2.0 CO2 Laser Power Density (kW/cm2)

2.5

Figure 5.13 Total number of osteoblast cells on the untreated and CO2 laser treated MgO–PSZ after 14 days. There was a significant statistical difference between the untreated sample and the samples that were CO2 laser treated at 1.6, 1.9 and 2.6 kW/cm2, and no statistical difference among the untreated sample and the samples CO2 laser treated at 0.6 and 0.9 kW/cm2 ð p < 0:05Þ

88

In vitro Biocompatibility Evaluation

Figure 5.14 Optical image of the positive staining for alkaline phosphatase on (a) the untreated and (b) the CO2 laser treated samples (c) at 0.9 kW/cm2 and (d) at 1.6 kW/cm2

14 days counted by the hematocytometer show a similar trend in the cell growth on the untreated and CO2 laser treated MgO–PSZ. Alkaline Phosphatase (ALP) Activity Optical images of the positive staining for ALP on the untreated and selected samples that were CO2 laser treated at 0.9 and 1.6 kW/cm2 are shown in Figure 5.14. The cell rating was determined on the basis of quantity and intensity of precipitated dye within the cytoplasm of these cells. As can be seen from Figure 5.14, the granule on the untreated sample is small in size and shows faint to moderate intensity of staining. On the other hand, the granule is medium to large in size and shows strong intensity of staining on the sample CO2 laser treated at 0.9 kW/cm2 and brilliant intensity of staining on the sample CO2 laser treated at 1.6 kW/cm2. One granule evidenced spreading of the cell on the sample CO2 laser treated at 0.9 kW/cm2, as shown in Figure 5.14(c). The leukocyte alkaline phosphatase activity (LAPA) scores, determined by the number of cells counted and multiplying by the

Osteoblast Cell Response

89

120 100

LAPA Score

80 60 40 20 0

Untreated

CO2 laser 0.9 kW/cm2

CO2 laser 1.6 kW/cm2

Figure 5.15 LAPA scores of the untreated and CO2 laser treated MgO–PSZ. There was a significant statistical difference between the untreated sample and samples CO2 laser treated at 0.9 and 1.6 kW/cm2 at p < 0:05

value of the cell rating, are shown in Figure 5.15. The LAPA score of the CO2 laser treated sample is clearly much higher than that of the untreated sample. Furthermore, the CO2 laser treated MgO–PSZ samples had LAPA scores that were statistically significantly higher than that of untreated samples, as shown in Figure 5.15. 5.5.3 The Effect of CO2 Laser Treatment on the Osteoblast Cell Response The results show that the CO2 laser treatment brought about the considerable change in the response of osteoblast cells. Moreover, the osteoblast cell response varied on the MgO–PSZ with the power density of CO2 laser treatment applied. As discussed in Chapter 4, the variation of the power density of CO2 laser treatment resulted in the different changes in surface properties. The surface topography and wettability characteristics chemistry have been shown to be the factors influencing the osteoblast cell response in previous studies [210, 211] and are believed to effect the osteoblast cell response on the CO2 laser treated MgO–PSZ. The Effect of Topography on the Osteoblast Cell Response The topography has been shown to be one of the factors in influencing the osteoblast cell response [212]. As demonstrated in Chapter 4, CO2 laser treatment generated a consistently rougher surface on the MgO–PSZ when compared with the untreated sample and Ra increased with the power

In vitro Biocompatibility Evaluation

90

Cell Cover Density (%)

100 80 60 40 20 0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Surface Roughness, Ra (µm)

Figure 5.16 The relationship between the cover density of the hFOB cells and Ra

density of CO2 laser treatment. In this way it was possible to obtain a narrow range of surface roughness values (see Section 4.3.3). As shown in Figure 5.16, the CO2 laser treated MgO–PSZ with rougher surfaces has a higher osteoblast cell cover density compared with the smooth untreated sample. This is in agreement with some reports that the rougher surface of titanium promoted more osteoblast-like cell attachment [213]. Even so, there is no linear relationship between the osteoblast cell cover density and Ra , as shown in Figure 5.16. Furthermore, the different microstructures and increase in the crystal sizes were postulated to be the factors influencing the osteoblast cell cover density, as shown in Figure 5.17. The degrees of the cell adhesion improved markedly when an obvious microstructure change happened on the MgO– PSZ. Surface microtopography has been cited as an important factor influencing protein–surface and cell–surface interactions [80]. A number of reasons have been suggested for an increased differentiation of osteoblasts on microstructured surfaces, such as the influence of surface structure on cell shape or the fact that the surface topography creates a specific biochemical microenvironment around each cell [214]. Figure 5.17 shows that the crystal sizes in all CO2 laser treated MgO–PSZ are larger than the untreated sample. The osteoblast cell cover density generally increased with the increased crystal size when the power density was lower than 1.9 kW/cm2, indicating that crystal size could possibly influence the osteoblast cell adhesion. It is most likely that the greater nanosurface area created by the larger crystal size may promote interactions (such as adsorption, configuration, bioactivity, etc.), of select serum proteins(s), which, subsequently, enhance osteoblast adhesion. The study of osteoblast adhesion on nanophase ceramics has elucidated the fact that a critical grain size (between 49 and 67 nm for

Osteoblast Cell Response

91

Figure 5.17 The relationship between the cover density of hFOB cells, microstructure and crystal size of the MgO–PSZ (treated with various CO2 laser power densities)

alumina and between 32 and 56 nm for titania) played a crucial role in mediating osteoblast adhesion to nanophase ceramics by creating a greater surface area and promoting the interaction of protein [140]. However, a linear relationship does not exist between the cell cover density and crystal size (see Figure 5.17). Owing to this lack of a linear relationship, a simple conclusion would be difficult to elucidate the relation between the topography and cell behaviours, because the surface roughness is only one of the factors affecting cell behaviours. Indeed, Hallab et al. [215] demonstrated that surface free energy was a more important surface characteristic than surface roughness for cellular adhesion strength and proliferation. Thus it is reasonable to postulate that the surface roughness does influence the human osteoblast cell response; however, its effect is less than that of the surface energy. The Effect of Wettability Characteristics on the Osteoblast Cell Response As demonstrated in Chapter 4, the wettability characteristics of the MgO– PSZ improved after CO2 laser treatment. It is believed that the changes in wettability characteristics resulted in modification of the osteoblast cell response. The result of the one-day cell culture on the MgO–PSZ showed

In vitro Biocompatibility Evaluation

92

that there were no osteoblast cells attached on the untreated MgO–PSZ with low wettability characteristics, whereas some cells had already attached and spread on the CO2 laser treated MgO–PSZ (power density of 1.6 kW/cm2) with high wettability characteristics. The difference in wettability characteristics and surface energy must be the mechanism determining the difference in the osteoblast cell attachment. The CO2 laser used in the experiment is a TEM01 multimode. The level of CO2 laser beam interaction is higher in the centre than at the edge of the CO2 laser beam; consequently the modification level of surface energy would be higher at the centre than at the edge of the CO2 laser treated track. The different wettability characteristics generated by CO2 laser treatment across the track brought about the different levels of osteoblast spreading between cells at the edge and cells at the centre (see Figure 5.10). As can be seen from Figure 5.15, the enhanced cell functions represented by the LAPA increase with the wettability characteristics. It is believed that the wettability characteristics of the MgO–PSZ was the primary factor governing the cell response. The effects of wettability characteristics on cell functions could result from their influence on the protein adsorption and cell adhesion. The adsorption of the proteins is the net result of various types of interactions that depend on the nature of the protein aqueous solution. The difference in cellular response of different materials suggests that there are differences in the organization of the adsorbed protein layer. Protein adsorption mediated cell behaviours are regarded as fundamental reactions at the biomaterial–tissue interface [216]. The value of cos y (glycerol) was used to express the wettability characteristics of the MgO–PSZ. The higher the value of cos y, the higher is the wettability characteristics. It is evident from Figure 5.18 that the cell growth

Cell Cover Density (%)

100 80 60 40 20 0 0.0

0.2

0.4

0.6

0.8

1.0

Wettability, cos θ (glycerol)

Figure 5.18 The relationship between the cover density of hFOB cells and the wettability characteristics of the MgO–PSZ

Osteoblast Cell Response

93

increased when the wettability characteristics changed from a lower value to a moderate value (cos y from 0.2 to 0.65), indicating that generation of the wettability characteristics is the main mechanism accounting for the different amounts of osteoblast cell adhesion and growth. The finding agrees with previous studies showing the influence of wettability on the attachment and spreading of various cells [213, 217–219]. These studies showed good cell attachment and spreading on high-energy substrata and poor cell attachment and spreading on low-energy substrata, which accounts for the minimal energetic state of a system in equilibrium. It was noticed that the value of cos y ranged between 0.6 and 0.8 and did not present a great disparity in cell cover density. This implied that after a certain value, further increases in the wettability characteristics would not improve the cell response. This is similar to the finding that the highest levels of cell attachment were found on a moderately hydrophilic surface using a model surface [220]. As demonstrated in Chapter 4, the surface roughness, surface oxygen content and surface energy are the mechanisms governing the wettability characteristics of the MgO–PSZ. The correlation between the wettability characteristics and the osteoblast cell response implies that the surface p roughness, surface oxygen content and gsv have some bearing on the cell response owing to the fact that these surface properties influence the wettability characteristics of the MgO–PSZ. In the consideration of surface roughness in Section 5.5.3, it was found that surface roughness does influence the response of osteoblast cells, but its effect is slight. The surface oxide, among the surface characteristics, is shown to be a main factor influencing the cell response besides surface roughness [52, 221]. In vivo studies show that a high degree of bone contact and bone formation are achieved with titanium implants which are modified with respect to oxide thickness and surface topography [222]. As shown in Chapter 4, the CO2 laser processing of the MgO–PSZ with the oxygen shield gas resulted in the incorporation of oxygen atoms on the material’s surface layer. It is postulated that the surface oxygen content is one of the factors influencing cell growth. It can be seen from Figure 5.19 that the cell cover density on the material is higher when there is a higher surface oxygen content, indicating that the increase in surface oxygen content is attributed to better cell growth. However, there is no linear relationship between the surface oxygen content and cell cover density, implying that some other mechanism is more dominant in governing the cell response. In Chapter 4 it was found that changes in the wettability characteristics of p the MgO–PSZ were primarily influenced by gsv of the MgO–PSZ. As one can p see from Figure 5.20, the cell cover density statistically increases as gsv increases from 10 to 25 mJ/m2. Hence, the higher surface energy induced by the CO2 laser treatment resulted in better proliferation of human osteoblast

In vitro Biocompatibility Evaluation

94

Cell Cover Density (%)

100 80 60 40 20 0

40

45 50 55 60 65 Surface Oxygen Content (at%)

70

75

Figure 5.19 The relationship between the cover density of the hFOB cells and the surface oxygen content of the MgO–PSZ

Cell Cover Density (%)

100 80 60 40 20 0

10

20

30

40

50

60

γ svp (mJ/m2)

Figure 5.20 The relationship between the cover density of the hFOB cells and gpsv for the MgO–PSZ

cells on the MgO–PSZ surface than on the untreated MgO–PSZ. A further p increase in gsv from 25 to 60 mJ/m2 did not bring about an increase in cell p proliferation that was statistically significant. As the gsv did not change p markedly after CO2 laser treatment, the results indicated that gsv influenced the behaviour of the osteoblasts on MgO–PSZ surfaces more strongly compared to gdsv following CO2 laser treatment, which was probably attributed to the fact that the composition and the culture medium all are polar, and thus cells and the MgO–PSZ should interact mainly by polar force.

Predictions for Implantation in an in vitro Clinical Situation

95

The behaviour of osteoblastic cells at the surface of hydroxyapatite [218] p and at the surface of titanium [213] demonstrated that gsv plays a critical role.

5.6 Predictions for Implantation in an in vivo Clinical Situation A relationship between the bioactivity/biocompatibility of a material in vivo and the material’s ability to form an apatite-like layer in vitro when soaked in aqueous solutions that imitate the inorganic components of human plasma was propounded recently by Vallet-Regi [223]. This proposition extends to ceramics, for once implanted into the body, bioactive ceramics are able to bond to bone through the formation of a hydroxyapatite surface layer. These hydroxyapatite layers that are formed in vivo can be closely mimicked through in vitro testing, usually by using SBF, which contains ionic concentrations similar to that of human blood plasma. This allows for an evaluation of the potential for biocompatibility/bioactivity of a new ceramic, or a novel processing technique for a ceramic can be conducted before performing any animal tests. As the preceding sections of this chapter show, CO2 laser treatment could generate functional groups and subsequently facilitate the formation of bone-like apatites on the surface of the MgO–PSZ. Building on this finding by considering the view of Vallet-Regi [223], it is reasonable to assume that, if implanted, the in vivo performance of the CO2 laser treated MgO–PSZ would be acceptable. There is certainly a considerable body of evidence that supports this supposition. In comprehensive studies by Aldini et al. [224, 225] and Torricelli et al. [226], yttria–stabilised tetragonal zirconia (Y-STZ), either coated with a bioactive glass termed RKKP or uncoated, was evaluated in vitro using normal and osteopenic bone-derived osteoblasts and in vivo using healthy bone in female Sprague Dawley rats. The in vitro results suggested that the RKKP-coated Y–STZ was biocompatible and enhanced proliferation, activation and differentiation of normal bone-derived osteoblasts and stimulation of osseopenic bone-derived osteoblasts when compared with the uncoated Y–STZ. To assess the in vivo performance of the RKKP-coated and uncoated Y–STZ, samples of each were implanted into the distal femurs of the rats and then removed after 30 and 60 days from surgery, whereupon they were subjected to a histomorphometrical analysis to assess osseointegration and bone quality around the implants. The histomorphometrical analysis showed that the RKKP coating ensured a better osseointegration rate with higher affinity index values than the uncoated Y–STZ, even when the osseopenic rate were used. No differences were observed at the bone– biomaterial interfaces for either material. In this instance there was a direct

96

In vitro Biocompatibility Evaluation

correlation between the findings of the in vitro and the in vivo investigations in terms of biocompatibility and osseointegration, but not for the case of bone-biomaterial interfacial characteristics. In a study conducted to evaluate the biocompatibility of MgO-doped HA (hydroxyapatite)/b-TCP (tricalcium phosphate) biphasic ceramics, Ryu et al. [227] performed in vitro and in vivo tests. The in vitro tests were carried out with a murine fibroblast L929 cell culture of extract from a 1 wt % MgOdoped HA/b-TCP ceramic with the aim of establishing whether the ceramics were biocompatible or cytotoxic. The presence or absence of any cytotoxic effects was determined qualitatively using SEM analysis. The SEM analysis following the cell culture from the 1 wt % MgO-doped HA/b-TCP ceramic revealed no morphological change, no vacuolisation or cell lysis and hence the absence of any cytotoxicity. In the first of two in vivo tests small samples of the 1 wt % MgO-doped HA/b-TCP ceramic were inserted into the back muscles of mature rabbits and then removed after 8 weeks for XRD analysis. The XRD analysis showed that the 1 wt % MgO-doped HA/ b-TCP ceramic was indeed biocompatible as the b-TCP phase had completely dissolved and an apatite layer had formed on the surface by means of cellular activity and a resultant dissolution/precipitation process. The second in vivo test involved implanting the same small 1 wt % MgOdoped HA/b-TCP ceramic samples that were inserted into the back muscles of the rabbits for 8 weeks across the proximal tibial metaphysis of mature rabbits for a further 8 weeks. From a histological examination of the samples no inflammation or foreign body reaction such as the formation of an intervening fibrous tissue was observed; rather the new bone formed and bonded directly to the implanted 1 wt % MgO-doped HA/b-TCP ceramic. Clearly, there was a direct link between the in vitro results and the observations made from both of the in vivo tests, with the in vitro tests indicating that the 1 wt % MgO-doped HA/b-TCP ceramic was biocompatible, the first in vivo tests demonstrating that the 1 wt % MgO-doped HA/ b-TCP ceramic could form an apatite layer and the final in vivo tests showing that the 1 wt % MgO-doped HA/b-TCP ceramic could support the bonding of new bone to its surface. Literature exists that shows that the in vitro bioactivity of a material has been observed to be dependent upon the type of aqueous solution used and does not always correlate to the observed in vivo behaviour. For example, apatite–wollastonite (A–W) glass–ceramics, which form a hydroxyapatite surface layer when immersed in SBF, are unable to form such layers in tris buffer [228]. In addition, SBF has been reported to affect the rate at which crystalline hydroxyapatite layers develop [229]. Indeed, the work of Gil-Albarova et al. [230] studied the in vivo behaviour of an SiO2–P2O5– CaO sol-gel glass and an SiO2–P2O5–CaO–MgO glass–ceramic, both of which are bioactive when soaked in SBF but display different rates of apatite layer

Predictions for Implantation in an in vitro Clinical Situation

97

formation. The ceramics were implanted into mature and immature New Zealand rabbits and histological results after 6 and 12 weeks revealed that both ceramics allowed bone growth over their surface in similar quantities and at similar rates by means of mesenchymal cell recruitment from the surrounding bone. Similarly, the in vitro and in vivo behaviour of certain ceramics has been seen to differ on account of the ceramic material itself. For instance, sintered hydroxyapatite exhibits in vivo bioactivity although the formation kinetics of an apatite-like layer on its surface is very slow under in vitro conditions [228]. Additionally, Li et al. [8] reported that that Al2O3 gel did not induce apatite formation when immersed in SBF for 21 days, whereas both pure SiO2 gel and gel-derived TiO2 were hydroxyapatite inducers. Kobayashi et al. [231] developed a composite (designated ABC) consisting of Al2O3 bead powder as an inorganic filler and bisphenol-a-glycidyl methacrylate (bis-GMA) based resin as an organic matrix, which allows direct bone formation on its surface in vivo. Although bioactive materials such as bioglass or apatite and wollastonite-containing glass–ceramic have previously been reported to form bone-like apatite on their surfaces in vitro under acellular conditions via simple chemical reactions, ABC did not present such characteristics. Indeed, no apatite formation was detected on the surfaces of the ABC composite after soaking in SBF for 28 days in vitro. Histological examination of rat tibiae after 8 weeks revealed that the ABC composite bonded to bone directly via a layer of calcium, phosphorus and alumina with no interposed soft-tissue layer. Moreover, the amount of bone directly apposed to the ABC composite surface was seen to increase with time. These results imply that the ABC composite has the ability to bond directly with bone through a calcium–phosphorus-rich layer. It was concluded that this layer was induced by some property of the ABC composite that encouraged calcification or apatite formation due to the actions of proteins and cells in vivo. Although the aqueous solutions employed to replicate inorganic body fluids are able to reproduce the process of bone-like apatite formation on the surfaces of bioactive ceramics in vitro, they are only capable of evaluating the bone-bonding capacities of bioactive ceramics by means of simple chemical reactions. However, once a ceramic is implanted into the body, they elicit several responses from living tissue that cannot be simulated in vitro. These responses include protein adsorption, cell attachment and adhesion, as well as ionic exchange. Therefore, depending upon the actual ceramic itself and the aqueous solution used, a ceramic could well be biocompatible in vivo despite appearing to be cytotoxic in vitro. However, importantly for this work, it is clear that if a ceramic presents itself to be biocompatible after in vitro testing, then it is highly likely that it will perform satifactorily in vivo.

98

In vitro Biocompatibility Evaluation

5.7 Summary A CO2 laser was used to modify the surface of the MgO–PSZ for the purpose of acquiring surface properties favouring interaction at the implant–bone tissue interface. The bioactivity of the CO2 laser modified MgO–PSZ was investigated in SBF. Protein adsorption and hFOB cells were used to examine the in vitro biological response on the MgO–PSZ following CO2 laser treatment. It was demonstrated that CO2 laser treatment could improve the bioactivity of the MgO–PSZ surface by generating functional groups to facilitate the formation of bone-like apatites. The apatite formed readily on the MgO–PSZ with relatively high amounts of hydroxyl groups, which were generated by CO2 laser treatment with power densities of 1.6 and 1.9 kW/cm2. No apatite was observed on the untreated and CO2 laser modified samples (0.6, 0.9 and 2.5 kW/cm2), which exhibited few hydroxyl groups. These observations indicate that Zr–OH groups on the MgO–PSZ surface are the functional groups required to facilitate apatite formation. The melting and re-solidification on the surface of the MgO–PSZ induced by CO2 laser processing provides the Zr4þ ion and OH ion and therefore creates the Zr–OH group on the surface. In comparison with the untreated MgO–PSZ, CO2 laser treatment brought about a thinner adsorbed albumin layer and a thicker adsorbed fibronectin layer on the MgO–PSZ. As the wettability characteristics of the MgO–PSZ increased, the albumin adsorption decreased while the fibronectin adsorption increased, indicating that wettability is a major factor in governing p protein adsorption. Further, the correlative effect of gsv observed on the protein adsorption suggested that protein adsorption on the MgO–PSZ was probably due to the polar and chemical interactions. Better osteoblast cell responses were found on the CO2 laser treated MgO– PSZ when compared with the untreated sample. The change in topography induced by the CO2 laser treatment was identified as being one of the factors influencing the hFOB cell response, but in a minor capacity only. The improved wettability characteristics of the MgO–PSZ due to enhanced surface energy, especially the polar component, brought by the CO2 laser treatment, played a significant role in the number of initial cells that attached and spread, thereby enhancing the long-term cell adhesion and growth potential. There is a reasonable body of literature to support the concept that if a ceramic presents itself to be biocompatible after in vitro testing then it is highly likely that it will perform satisfactorily in vivo. This being the case, then it is reasonable to assume that the in vivo performance of the CO2 laser treated MgO–PSZ would be acceptable due to its excellent in vitro performance demonstrated herein.

6 The Effects of CO2 Laser Radiation on the Wettability Characteristics of a Titanium Alloy This chapter describes the modification of the wettability characteristics of a titanium alloy (Ti–6Al–4V ELI) following CO2 laser irradiation. The Ti–6Al–4V alloy is often used for the fabrication of dental and orthopaedic implants. To study the change in the wettability characteristics of the Ti–6Al–4V alloy, contact angles between selected control test liquids and the surfaces of the untreated and CO2 laser treated Ti–6Al–4V alloy were measured. The surface properties of the untreated, mechanically roughened and CO2 laser treated Ti–6Al–4V alloy were characterised and the effect of surface roughness, surface oxygen content and surface energy on the wettability characteristics of the Ti–6Al–4V alloy were analysed. It was apparent that CO2 laser treatment brought about significant changes in the wettability characteristics of the Ti–6Al–4V alloy. Furthermore, the predominant mechanisms active in determining the wettability characteristics were analysed and the primary mechanism was identified. 6.1 Introduction During the last decade, numerous materials have been used for the fabrication of dental and orthopaedic implants. The materials of choice have predominantly been metals. Stainless steel, cobalt chromium molybdenum alloy, titanium and a multitude of titanium alloys were the materials of

Laser Surface Treatment of Bio-Implant Materials L. Hao and J. Lawrence © 2005 John Wiley & Sons, Ltd ISBN: 0-470-01687-6

100

The Effects of CO2 Laser Radiation

choice. The following criteria define an ideal bone contacting material for orthopaedic surgery: a biocompatible chemical composition to avoid adverse tissue reactions, acceptable strength, a high wear resistance to minimise wear debris, excellent corrosion resistance in the physiological milieu and a modulus of elasticity similar to that of bone to minimise bone resorption around the device. Titanium and its alloys, including the Ti–6Al–4V alloy, are now being used as a common material for bone implants under biomechanical loading conditions. None of these bioinert titanium based materials, however, bonds to bone and subsequently their stable fixations to the surrounding bone have long been considered as a fundamental problem in clinical uses [232]. It is generally accepted that early surface events that occur rapidly upon implantation of a biomaterial into biological fluids determine a subsequent response. These involve wetting by physiological liquids, followed by adsorption of proteins and cells to the biomaterials surface [233]. Numerous research groups have studied the interactions of different types of cultured cells with biomaterials with different wettability characteristics to correlate the relationship between surface wettability and blood, cell or tissue compatibility for polymeric materials [59, 115, 234]. Furthermore, the surface wettability has a significant influence on the friction behaviour of a tribological system and gives an indication of its biotolerance: in a first approximation, the more wettable the material, the better the human body tolerates it [235]. Clearly, techniques to control the wettability characteristics of a biomaterial’s surface and thereby improve the material’s biocompatibility are of great interest. Due to the rapid and specific modification of organic and inorganic materials, laser surface processing has aroused growing interest and been proven to be a controllable and flexible technique for modifying the surface properties of materials. It is recognised within the currently published work that laser irradiation of material surfaces can affect their wettability characteristics. Previously Heitz et al. [133], Henari and Blau [134] and Olfert et al. [135] had found that excimer laser treatment of metals results in improved coating adhesion, attributed to the fact that the excimer laser treatment resulted in a smoother surface and as such enhanced the action of wetting. It was demonstrated that five pulses per area of CO2 laser treatment were sufficient to produce a fully wettable mild steel surface. The wettability was influenced by the surface exposing time (SET) after laser treatment [136]. Self-fluxing Fe–Cr–Ni–B–Si alloy powders with various Ni contents were laser clad on medium carbon steel substrates [137]. Lawrence and Li [138] revealed that the interaction of CO2, Nd:YAG, HPDL and excimer laser radiation with the surface of a selected mild steel gave rise to changes in the wettability characteristics of the material. It was observed that interaction of the mild steel with Nd:YAG and HPDL

Experimental Procedures

101

radiation brought about an improvement in the wettability characteristics of the steel. In contrast, interaction of the mild steel with CO2 and excimer laser radiation resulted in a depreciation of the wettability characteristics of the steel [236]. However, despite a growing amount of work conducted with metal, no work has been conducted so far on the feasibility of the laser surface treatment process for the modification of the wettability characteristics of biograde metals.

6.2 Experimental Procedures 6.2.1 Material Specifications and Preparation Medical grade titanium alloys have a significantly higher strength-to-weight ratio than competing stainless steels. The range of available titanium alloys enables medical specialist designers to select materials and forms closely tailored to the needs of the application. The natural selection of titanium for implantation is determined by a combination of most favourable characteristics, including immunity to corrosion, biocompatibility, strength, low modulus and density and the capacity for joining with bone and other tissue (osseointegration). The mechanical and physical properties of titanium alloys combine to provide implants that are highly damage tolerant. Forms and material specifications of titanium and its alloy for medical application are detailed in a number of international specifications. In this study a Ti–6Al–4V ELI alloy (F136) was used. The as-received Ti– 6Al–4V alloy (ground annealed) was in the form of a round bar with a diameter of 28.5 mm (Carpenter, Inc.). For experimental purposes, the round bar was divided into 15 sections, each of 3 mm thickness, by a cutting machine (Miniton; Struers, GmbH) using a diamond-rimmed blade. The 15 divided sections were then separated into three groups of five samples, with the groups being: untreated, mechanically roughened and CO2 laser treated. For the mechanically roughened group, the samples were roughened by evenly abrading the entire surface of the sample with grinding paper (180 grit SiC). This was achieved by applying the grinding paper to the surface of the sample with moderate pressure and drawing it across the surface in different directions eight times. The composition of the Ti–6Al–4V alloy was: 88.3–90.8 wt % Ti, 5.5–6.5 wt % Al, 3.5–4.5 wt % V,