Laser Modification of the Wettability Characteristics of Engineering Materials

Laser Modification of the Wettability Characteristics of Engineering Materials This page intentionally left blank E...

Author: J. Lawrence | L. Li

125 downloads 625 Views 13MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Laser Modification of the Wettability Characteristics of Engineering Materials

This page intentionally left blank

ENGINEERING

RESEARCH SERIES

Laser Modification of the Wettability Characteristics of Engineering Materials J Lawrence and L Li

Series Editor Duncan Dowson

Professional Engineering Publishing Limited, London and Bury St Edmunds, UK

First published 2001 This publication is copyright under the Berne Convention and the International Copyright Convention. All rights reserved. Apart from any fair dealing for the purpose of private study, research, criticism, or review, as permitted under the Copyright Designs and Patents Act 1988, no part may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, electrical, chemical, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owners. Unlicensed multiple copying of this publication is illegal. Inquiries should be addressed to: The Publishing Editor, Professional Engineering Publishing Limited, Northgate Avenue, Bury St Edmunds, Suffolk, IP32 6BW, UK. Fax: +44(1)284705271.

© Lawrence and Li

ISBN 1 86058 293 1 ISSN 1468-3938 ERS3 A CIP catalogue record for this book is available from the British Library.

Printed and bound in Great Britain by

The publishers are not responsible for any statement made in this publication. Data, discussion, and conclusions developed by the Authors are for information only and are not intended for use without independent substantiating investigation on the part of the potential users. Opinions expressed are those of the Authors and are not necessarily those of the Institution of Mechanical Engineers or its publishers.

About the Authors Dr Jonathan Lawrence gained his PhD on laser materials processing from University of Manchester Institute of Science and Technology (UMIST) in 1999. Prior to this, he studied at the University of Bradford for a BEng in Mechanical Engineering and at Stockport College where he gained an ONC and an HNC in Mechanical and Production Engineering. He is widely published in various technological and scientific journals on the subject of employing lasers for the modification of the wettability characteristics of materials. His current research interests include: laser modification of the wettability characteristics of bio-materials, laser treatment of building materials, laser enamelling, thin and thick film deposition using lasers and laser forming ([email protected]). Professor Lin Li is the Director of the Laser Processing Research Centre at UMIST. His current research interests include laser machining, laser drilling, marking/engraving, rapid prototyping, laser surface engineering, as well as mathematical modelling, monitoring and closed-loop control of laser materials processing ([email protected]).

Related Titles IMechE Engineers' Data Book Second edition

C Matthews

ISBN 1 86058 248 6

A Guide to Presenting Technical Information - Effective Graphic Communication

C Matthews

ISBN 1 86058 249 4

Guide „ . ,to .Wear . , Problems and Testing for Industry (Tribology in Practice Series)

U T JI VNeale T i and m,M Gee ^ M

TC^XT 1i 86058 o^nco 10-7-7 ISBN 287 7

Journal of Engineering Tribology

Part J of the Proceedings of the IMechE

ISSN 1350-6501

Journal of Materials: Design and Applications

Part L of the Proceedings of the IMechE

ISSN 1464-4207

Other titles in the Engineering Research Series Industrial Application of Environmentally Conscious Design (ERS1)

T C McAloone

ISBN 1 86058 239 7 ISSN 1468-3938

Surface Inspection Techniques Using the Integration of Innovative Machine Vision and Graphical Modelling Techniques (ERS2)

ML Smith

ISBN 1 86058 292 3 ISSN 1468-3938

Fatigue and Fracture Mechanics of Offshore Structures (ERS4)

L S Etube

ISBN 1 86058 312 1 ISSN 1468-3938

Adaptive Neural Control of Walking Robots(ERSS)

M J Randall

ISBN 1 86058 294 X ISSN 1468-3938

Strategies for Collective Minimalist Mobile Robots (ERS6)

C Melhuish

ISBN 1 86058 3180 ISSN 1468-3938

For the full range of titles published by Professional Engineering Publishing contact: Sales Department Professional Engineering Publishing Limited Northgate Avenue Bury St Edmunds Suffolk IP32 6BW

UK Tel: +44(0)1284724384 Fax: +44(0)1284718692 www.pepublishing.com

Contents Series Editor's Foreword

xi

Acknowledgements

xiii

Dedication

xiv

Introduction

1

Chapter 1 Laser Fundamentals and Contemporary Industrial Lasers 1.1 Laser operating principles 1.1.1 Introduction 1.1.2 Stimulated emission and amplification 1.1.3 Population inversion and non-equilibrium pumping 1.1.4 Resonant cavity 1.2 Contemporary industrial lasers 1.2.1 Introduction 1.2.2 The CO2 laser 1.2.3 The Nd:YAG laser 1.2.4 The excimer laser 1.2.5 The diode laser 1.2.6 Comparison of the general operating features 1.3 Laser beam properties 1.3.1 Monochromaticity 1.3.2 Wavelength 1.3.3 Polarization 1.3.4 Coherence 1.3.5 Mode structure 1.3.6 Divergence 1.3.7 Numerical aperture for fibre-delivered beams 1.3.8 Brightness 1.3.9 Beam dimensions 1.3.10 Focused beam characteristics 1.3.11 Energy and power output 1.3.12 Energy transfer and efficiency

5 5 5 7 8 9 9 9 11 11 12 21 22 22 22 23 24 25 26 27 28 28 30 31 32

Chapter 2 Basic Background Theory of Wettability, Adhesion, and Bonding 2.1 Introduction 2.2 Wetting and contact angle 2.2.1 Contact angle

33 34 34

Contents

2.3 2.4

2.2.2 Types of wetting 2.2.3 Spreading and the spreading coefficient 2.2.4 Contact angle hysteresis 2.2.5 Static and dynamic contact angles 2.2.6 The effect of surface roughness on contact angle 2.2.7 The effect of chemical composition on contact angle 2.2.8 The effect of temperature on contact angle Surface energy and the dispersive/polar characteristics The bonding of liquids and solids 2.4.1 Physical bonding 2.4.2 Mechanical bonding 2.4.3 Chemical bonding

Chapter 3 Laser Surface Modification of Selected Composite Materials for Improved Wettability Characteristics 3.1 Introduction 3.2 Experimental procedures 3.2.1 Materials 3.2.2 Laser processing procedure 3.2.3 Wettability characteristics analysis procedure 3.3 The effects of high-power diode laser radiation on the wettability characteristics of ordinary Portland cement 3.3.1 The general effects of high-power diode laser radiation 3.3.2 Wettability and surface energy characteristics 3.4 The effects of laser radiation on the wettability and bonding characteristics of the Al2O3/SiO2-based composite material 3.4.1 The general effects of laser radiation 3.4.2 Wettability and surface energy characteristics 3.4.3 High-power diode laser-induced solidification microstructures and the resulting effects on wettability characteristics 3.4.4 Bonding mechanisms between the high-power diode laser-treated Al2O3/SiO2-based composite material and enamel 3.5 Summary Chapter 4 High-Power Diode Laser Modification of Selected Engineering Ceramic Materials for Improved Wettability Characteristics 4.1 Introduction 4.2 Experimental procedures 4.2.1 Preparation of materials 4.2.2 Laser processing procedure 4.3 High-power diode laser modification of the wettability characteristics of selected engineering ceramic materials 4.3.1 Contact angle and surface energy analysis procedure 4.3.2 Effects of high-power diode laser radiation on contact angle characteristics

viii

35 37 38 39 41 41 42 42 45 45 45 46

49 51 51 51 52 54 54 56 59 59 66 76 79 81

83 84 84 84 84 84 84

Contents

4.4 4.5

4.3.3 Surface energy and the dispersive/polar characteristics

87

Discussion of high-power diode laser wettability characteristics modification Summary

90 90

Chapter 5 Laser Modification of Selected Metallic Materials for Improved Wettability Characteristics 5.1 Introduction 5.2 Experimental procedures 5.2.1 Materials and laser processing procedures 5.2.2 Wettability characteristics analysis procedure 5.3 Effects of laser radiation on the wettability characteristics 5.3.1 Variations in surface roughness characteristics 5.3.2 Variations in surface O2 content 5.3.3 Surface energy and the dispersive/polar characteristics 5.4 Discussion of laser-effected modification of wettability characteristics 5.5 Bonding mechanisms between the high-power diode laser-treated mild steel and enamel 5.6 Summary

102 103

Chapter 6 Laser Modification of Selected Polymer Materials for Improved Wettability Characteristics 6.1 Introduction 6.2 Experimental procedures 6.2.1 Materials and laser processing procedure 6.2.2 Wettability characteristics analysis procedures 6.3 Effects of laser radiation on wettability characteristics 6.3.1 Variations in surface roughness characteristics 6.3.2 Variations in surface O2 content 6.3.3 Surface energy and the dispersive/polar characteristics 6.4 Discussion of laser-effected wettability characteristics modification 6.5 Summary

105 107 107 108 109 110 113 113 116 118

91 92 92 93 94 94 96 97 98

Chapter 7 Practical Applications of Lasers for the Modification of Wettability Characteristics 7.1 Introduction 119 7.2 A two-stage ceramic tile grout sealing process 120 7.2.1 Application background 120 7.2.2 Sealing process development 122 7.2.3 Seal characterization and performance testing 123 7.2.4 Discussion of results 136 7.2.5 Summary 144 7.3 The enamelling of ordinary Portland cement by means of high-power diode laser radiation 146

ix

Contents

7.4

Chapters

7.3.1 General results of high-power diode laser enamelling 7.3.2 Mechanical, physical, and chemical properties 7.3.3 Summary The enamelling of carbon steel by means of high-power diode laser radiation 7.4.1 Enamel glaze characteristics 7.4.2 Mechanical, physical, and chemical characteristics 7.4.3 Summary

146 148 153 154 154 155 157

Conclusions

159

References

163

Index

175

x

Series Editor's Foreword The nature of engineering research is such that many readers of papers in learned society journals wish to know more about the full story and background to the work reported. In some disciplines this is accommodated when the thesis or engineering report is published in monograph form - describing the research in much more complete form than is possible in journal papers. The Engineering Research Series offers this opportunity to engineers in universities and industry and will thus disseminate wider accounts of engineering research progress than are currently available. The volumes will supplement and not compete with the publication of peer-reviewed papers in journals. Factors to be considered in the selection of items for the Series include the intrinsic quality of the volume, its comprehensive nature, the novelty of the subject, potential applications, and the relevance to the wider engineering community. Selection of volumes for publication will be based mainly upon one of the following; single higher degree theses; a series of theses on a particular engineering topic; submissions for higher doctorates; reports to sponsors of research; or comprehensive industrial research reports. It is usual for university engineering research groups to undertake research on problems reflecting their expertise over several years. In such cases it may be appropriate to produce a comprehensive, but selective, account of the development of understanding and knowledge on the topic in a specially prepared single volume. Authors are invited to discuss ideas for new volumes with Sheril Leich, Commissioning Editor in the Books Department, Professional Engineering Publishing Limited, or with the Series Editor. The third volume in the Series comes from UMIST and is entitled Laser Modification of the Wettability Characteristics of Engineering Materials by Jonathan Lawrence and Lin Li Manufacturing Division, Department of Mechanical Engineering, UMIST, Manchester. This multi-disciplinary study embraces accounts of the fundamental operation of lasers and the phenomena of wettability, adhesion, and bonding. The work originated from an in-depth study of new approaches to the problem of ceramic tile grouting, but this has been widened to include improvements to the wettability of engineering ceramics, metals, and polymers. The very breadth of materials considered should be of interest to engineers from various disciplines.


The role of lasers in joining and material removal processes is well established, but further developments have revealed their potential in wider aspects of materials processing. These range from the modification of surface properties to the synthesis of new materials, layered manufacture, and fabrication processes. Stereo lithography, which is currently attracting much attention, has benefited enormously from the application of laser technology. The reader will value the authors' comprehensive account of contemporary lasers and their fundamental operating characteristics, together with their discussion of the principles of such basic concepts in surface engineering as wettability, adhesion, and bonding. The accounts of the application of laser processing techniques to a wide range of engineering materials will also be of interest to many readers.

Professor Duncan Dowson Series Editor Engineering Research Series

xii

Acknowledgements First, and most importantly, I wish to express my heartfelt gratitude to my dearest and most cherished friend Louise, without whose steadfast, unconditional support and encouragement this whole venture, and everything before it, would have been meaningless and impossible. Always there, beyond compare. In my work I have received considerable help from many people within BNFL. In particular, I would like to thank my industrial supervisor, Dr Julian Spencer, for his exceptional support of the work, superb technical advice, and invaluable literary criticism. I would also like to thank: Carl Hogarth and Carol Duckworth for assisting and sharing their time and knowledge of SEM, ESEM, and EDX procedures; Sarah Hawley for her assistance and advice on XRD techniques; and Paul Gilchrist for his help in obtaining and interpreting the TG-DTA results. I have been privileged to have received substantial assistance from numerous people in the various fields of expertise investigated in this work. On many occasions I obtained superb advice on vitreous enamels from Max Richens of Vitreous Enamels Limited and Edward Simpkins of Ferro Group Limited, for which I am most appreciative. I am very grateful to Dr Liu Zhu who generously gave her valuable advice on the subject of laser surface treatment and solidified structures. Many thanks go also to Dr Andrew Robinson of Birmingham University for his expert advice on the subject of wettability. Within the Department of Mechanical Engineering at UMIST I would like to thank the members of the Royce Laboratory who have assisted in the design and manufacture of the specialist equipment required during this work. In particular, Peter Baldwin, Richard Sanders, and especially Arthur Sumner, who helped in any way he could. I am very grateful to John Howe, the department photographer, who readily gave me so much of his time and so many consistently high-quality SEM images. Many thanks also to David Low for his help, advice, encouragement, and agreeable company during the 'all-nighters'. Finally, I would like to record my extreme indebtedness and gratitude to the 'supervisor supreme', Professor Lin Li, for his invaluable guidance, endless patience, excellent advice, and boundless confidence in me throughout the whole of this work. He is a man with whom I am most fortunate and pleased to have worked.

Dr Jonathan Lawrence

Entirely to Louise, Semper Fidelis

Introduction

Lasers and materials processing Since the first successful demonstration of the laser in 1960, the laser's potential for materials processing has been clearly recognized and actively pursued. The first laser materials processing application in 1965 was the use of a ruby laser to drill holes through diamonds for wire dies. Other materials processing applications such as micro-welding and jet-assisted cutting followed a short time later. In these initial tasks the laser was shown to offer unique processing capabilities as well as fairly reliable performance. The development of more reliable, higher output laser systems has significantly enhanced the laser's capability for large-scale materials processing tasks to the point where today, lasers are used in a whole host of areas within engineering as shown in Fig. 1. As Fig. 1 indicates, lasers are increasingly being recognized by the modern engineer as merely just another machine among the already available wide array. Indeed, in recent years the capabilities of the laser have been recognized to such an extent that not only are lasers being used to perform tasks uniquely suited to them, such as laser vapour chemical deposition (LCVD) and laser physical chemical deposition (LPCD), but they are also being employed in applications that were considered to be the exclusive domain of the more traditional machine tools, such as metal cutting and forming. The choice of a particular laser to suit an intended application is largely dependent upon the laser beam characteristics, the laser's operating features as well as the economic considerations of the laser. At present, the lasers most commonly used in industry for many of the materials processing applications detailed in Fig. 1 are the CO2, the Nd:YAG (Nd:YAG is defined in Chapter 1, Sub-section 1.2.3), and the excimer laser. Arguably, one of the most significant occurrences in the field of laser technology is the development of the diode laser.

1


Fig. 1 Laser applications in materials processing

The high-power diode laser Notwithstanding the constant advances in laser materials processing, since its advent in 1962 the diode laser, despite its inherent advantages, has up until recently only found its way into a limited number of applications. Such applications were, and still are, generally those where only a small amount of laser power is required. These applications are dominated by the longer wavelength devices (950-1500 nm) which are used almost entirely in telecommunications systems, by far the most predominant application area comprising 60 per cent of all diode laser sales. Shorter wavelength devices (650-800 nm) account for almost all of the remaining market (35 per cent), being used chiefly in items such as CD audio systems, CD-ROM, and laser printers. Nonetheless, today the diode laser forms by far the largest single segment of the worldwide laser market, some 57 per cent [1].

2

Introduction

Rapid and continuous enhancement of diode laser material and fabrication technology over the last decade has increased the average power output of the devices by two-fold each year [2], giving rise today to the commercial availability of diode lasers delivering output powers in excess of 50 mW - high-power diode lasers (HPDLs). Indeed, output powers of up to 120 W for single 10 mm experimental diode bars have been reported [3]. Further, commercial HPDLs can now reach 2.5 kW from a single unit of stacked diode bars. Already HPDLs are beginning to replace technologically mature lasers in many application areas, and evidence suggests that this trend will almost certainly occur in the materials processing sector; a lucrative market worth over £210 million per annum [4]. Despite such advances, the proliferation of HPDLs into more application areas, especially in materials processing, has been held in check by its prohibitively high price; currently around £50 per watt at high level, compared with £35 and £90 per watt at the same level for COi and Nd:YAG lasers respectively [5]. However, it is an active field of research that will surely expand as the 30-50 per cent per annum price reduction continues. Because semiconductors, and hence diode lasers, can be mass produced, large economies of scale, fuelled by demand, can be realized by manufacturers. As a result diode lasers have the potential to become extremely inexpensive (3000 per cent increase in life) the diode laser has over a relatively short period of time almost replaced the traditional pumping agent, the arc or flashlamp [6]. HPDLs are also found extensively within medicine, where they are used in the treatment of many ailments, from simple ear, nose, and throat operations to cancer ablation [7, 8]. Despite the fact that the current maximum output power of commercially available multimode HPDL is only 2.5 kW in the continuous wave (CW) mode, giving an intensity of around 2 x 105 W/cm2 when the beam is re-imaged, a relatively small number of direct HPDL materials processing applications do exist, such as soldering, metal-ceramic powder sintering, transformation hardening, foil cutting, marking, and the welding of plastics [5, 9-14].

Lasers and their application for the modification of wetting and bonding characteristics Much work has been carried out to study the effects of laser wavelength variation for medical and surgical applications, revealing clear differences in the performance and effectiveness of many different lasers for such applications. In contrast, comparisons of the differences in the beam interaction characteristics with various materials of the predominant materials processing lasers, the COi, the Nd:YAG, and the excimer laser, are limited. Previously the main fundamental differences resulting from wavelength

3


variations of CO, COa, Nd:YAG, and excimer lasers for a number of materials processing applications have been detailed [15-18]. Likewise, such practical comparisons between these traditional materials processing lasers and the more contemporary HPDL are even fewer in number. Previously Schmidt and Spencer [19] compared the performance of COa, excimer, and HPDL in the removal of chlorinated rubber coatings from concrete surfaces, noting wavelength dependent differences in the process performance. Additionally, Bradley and Scott [20] compared COa and HPDL for the treatment of AlaOa-based refractory materials in terms of microstructure, observing wavelength-dependent microstructural characteristics unique to each laser. In more comprehensive investigations, Lawrence and Li compared the effects of COi, Nd:YAG, excimer, and HPDL radiation on the wettability characteristics of an AliOa/SiOa-based composite [21, 22] a mild steel [23, 24] and a number of biocompatible polymer materials [25, 26], noting that changes in the wettability characteristics of the materials varied depending upon the actual material itself and the laser type. Both scientists and engineers alike have a great interest in understanding the interfacial phenomena between liquids and solid substrates, since in many practical applications where liquids are coated or fired on to solid substrates, the performance of the article is directly linked to the nature of the liquid-solid interface. At present, the processes available to engineers for the modification of a material's wettability characteristics are invariably extremely 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. Lawrence and Li have amply demonstrated the practicability of employing different types of lasers to effect changes in the wettability characteristics of composites and ceramics [21, 22, 27], metals [23, 24] and plastics [25, 26] for improved adhesion and bonding.

4

Chapter 1

Laser Fundamentals and Contemporary Industrial Lasers

This chapter details the fundamental theory of laser technology and the basic operating principles and characteristics of the most prevalent industrial lasers in common usage today. Owing to its unique characteristics and potential for many materials processing applications within industry, particular emphasis is placed on the contemporary HPDL.

1.1 Laser operating principles [28-32] 1.1.1 Introduction In order for a laser to function a variety of necessary prerequisites must be in place. These working requirements include the presence of an active medium to support lasing, the accomplishment of stimulated emission and amplification, the attainment of a population inversion, a non-equilibrium pumping mechanism, and in most lasers, a resonant cavity. 1.1.2 Stimulated emission and amplification For simplicity, if an atom is considered as having only two possible energy states, an upper state, E2, and a lower state, E,, then if an atom in the upper state makes a transition to the lower state, the energy of the electromagnetic radiation or photon emitted is given by [28]

5


where h is Plank's constant (6.626 x 10~34 J s), c is the velocity of light (2.99 x 108 m/s), A is the wavelength of the electromagnetic radiation and v is the frequency of the electromagnetic radiation. The frequency of the emitted radiation is given by [28]

Also, if an atom is initially in the lower energy state and makes a transition to the higher energy state then energy in the form of photons or electromagnetic radiation of frequency given in equation (1.2), must be absorbed by the atom or the molecule. Such energy level transitions of atoms from the lower energy states to the higher energy states can be achieved through the absorption of electromagnetic radiation with an energy level of AE, as shown in Fig. 1.1 (a). According to Einstein [33], once in the excited state there are only two independent processes through which the excess energy may be radiatively emitted. Firstly, the excited atom will eventually decay spontaneously of its own accord to a state of lower energy, Ej, emitting an electromagnetic wave (photon) with an energy of AE = vh in a random manner as shown in Fig. l.l(b). This process is known as spontaneous emission and is in direct contrast to absorption.

Fig. 1.1 The phenomenon of electromagnetic radiation in terms of (a) absorption, (b) spontaneous emission, and (c) stimulated emission

Secondly, the atom can be impelled to decay to the lower energy level before its natural time by a photon having an energy equal to AE interacting with the atom in the upper state and causing it to change to the lower state with the creation of a second photon [Fig. l.l(c)]. These two photons not only have the same energy (AE) but also travel in the same direction with the same frequency and exactly in phase with each

6


other. As such, the light wave representing the stimulated photon adds to the incident light wave on a constructive basis, thereby increasing its amplitude, and thus giving rise to the possibility of light amplification by stimulated emission radiation (LASER). 1.1.3 Population inversion and non-equilibrium pumping In reality, the likelihood of a photon with an energy of A£ incident upon an atom in an upper energy level inducing stimulated emission or absorption are of equal probability. Thus, in a system containing a very large number of atoms the dominant process will depend upon the relative number of atoms in the upper and lower states. A large population of atoms in the upper state will result in stimulated emission dominating, while if there are more atoms in the lower state there will be a greater probability of absorption rather than stimulated emission. Under normal conditions of thermal equilibrium, the population of atomic energy levels obey the Boltzmann distribution. Therefore, for any two levels of energy E\ and £2 (assuming £2 > EI) and population NI and N2 [28]

where kb is Boltzmann's constant (1.38 x 10-23 J/°K) and T is absolute temperature. For most metallic atoms the first level above the lowest possible (the ground state) is separated by a gap of 1.25 eV [28]. This amount of energy corresponds to that of a photon of visible light. Therefore, under normal equilibrium conditions and at room temperature, the population of the excited state, A^, will be very small compared with that of the lower energy state, NI, according to equation (1.3). For this reason any photons of visible light are very much more likely to be absorbed rather than give rise to stimulated emission. For stimulated emission to dominate it is necessary to increase the population of the upper energy level so that it is greater than that of the lower energy level. This situation is known as a population inversion. To create a population inversion the atoms within the laser medium must be excited or pumped into a non-equilibrium distribution through the application of a large amount of energy to the medium from an external source. There are a number of ways of pumping a collection of atoms into a higher energy level, these include: pumping by optical radiation, collisions induced by an electrical discharge, resonant energy transfer, plasma generation, passage of an electrical current, electron bombardment, and the release of chemical energy. Yet in the steady state, it is almost impossible to produce a population inversion with the two-level system, involving only NI and A^, due to the fact that when the condition is reached such that the populations are equal (N2 = Nj), then the absorption and stimulation processes will compensate one another. This situation is often referred to as 'two-level saturation'. Consequently, to avoid such a problem and to create a population inversion successfully, three- or four-level systems are employed as Fig. 1.2 shows.

7


In a three-level laser system [Fig. 1.2(a)], the atoms are pumped from the ground level, EI, to the highest level, £2. If the material is such that, after an atom has been raised to level £2, it decays rapidly in approximately 1CT7 s [28] to level EI, then a population inversion can be obtained between levels EI and EQ. Eventually, when the population inversion is achieved, amplification of radiation can take place by means of stimulated emission. As the level £2 is not directly involved in the amplification process it can be broad, and thus a wide band of wavelengths in the pumping radiation are effective in pumping, increasing the efficiency of the pumping operation. However, the three-level system requires very high pumping power because the terminal state Eo of the laser transition is the ground state and thus rather more than half of the total number of atoms must be pumped from the ground state into the excited state before a population inversion can be achieved. As such, the energy required to pump half of the total number of atoms in the system into EI via £2 is wasted, thus resulting in the inherent inefficiency of the three-level system. The pumping requirements can be greatly reduced by a four-level system as shown in Fig. 1.2(b). If EI - EO is large compared with the thermal energy, kT, at the temperature of operation, then the populations of the levels EI, £2, and £3 are all effectively zero before pumping commences. Thus a population inversion between levels £2 and EI can be readily achieved, with again, level £3 possibly being broad for effective pumping. Typically the atoms are pumped from the ground level, EO, to the highest level, £3, from where they decay rapidly to the metastable level £2 and a population inversion is obtained between levels £2 and Ej. If the lifetime of the transition from level EI to EO is short, then the population inversion can be maintained easily with modest pumping.

Fig. 1.2 Schematic representation of (a) three-level and (b) four-level laser schemes

1.1.4 Resonant cavity To achieve a continuous laser beam without constantly being dependent upon a spontaneous emission to initiate lasing, an optical cavity with feedback end-mirrors is required. The basics of this resonant cavity are a tube for the laser medium and two parallel mirrors, one being totally reflective (-99.99 per cent) and curved, so as to

8


reduce diffraction losses and facilitate easy alignment, the other being either flat or curved, and partially transitive to allow some of the oscillating power to emerge as the laser beam. Such a resonant cavity is known as a Fabry-Perot resonator. The mirrors are separated at either end of the cavity by a distance, L, of [29]

where, q is the cavity or longitudinal laser mode integer and A is the laser beam wavelength. The mirror curvatures at either end of the cavity can only lie within certain values or the resonant cavity will become 'unstable' [34] by losing power around the edges of the output mirror. Resonant cavities can be distinguished as being either 'stable' or 'unstable' depending on whether or not they cause the oscillating beam to converge into the cavity or spread out from the cavity.

1.2 Contemporary industrial lasers 1.2.11ntroduction Lasers are classified, according to the lasing medium used, into: gas lasers such as HeNe lasers, liquid lasers such as the dye laser, solid-state lasers such as the ruby laser, and semiconductor lasers such as the diode laser. However, not all lasers are suitable for industrial purposes due to the beam properties and output power, etc. The most commonly used industrial lasers today are the CC>2 laser, the Nd:YAG laser, the excimer laser and, more increasingly, the high-power diode laser (HPDL). 1.2.2 The CO2 laser [35-41] The lasing medium in a CC>2 laser is generally composed of a mixture of carbon dioxide, nitrogen, and helium gases. The nitrogen is active in the excitation process and is added to increase the excitation efficiency, and the helium, which comprises approximately three-quarters of the total gas mixture, is included because it acts as an internal heat sink. The lasing transitions are due to energy levels resulting from the electronic motions, vibrational motions, and rotational motions of the COi molecule. Just as in single atoms, electrons in molecules can be excited to higher energy levels. Independent of the electronic state, the atomic nuclei, which are held together by molecular binding forces, will vibrate about their equilibrium position, giving rise to quantized vibrational energy levels. The energy separation between vibrational levels of the same electronic energy state generally corresponds to frequencies in the near and middle infrared range, thus each of the widely spaced electronic levels will be split into numerous vibrational sub-levels. The three possible quantized modes of vibration in the CO2 molecule in the lowest electronic state are symmetric, bending, and asymmetric vibrational motions. A molecule in any electronic-vibrational level can be capable of rotation about a number of axes in space. This rotational action will in turn lead to closely spaced, discrete energy levels, further subdividing each vibrational energy level into a series of levels with energy separations corresponding to far infrared frequencies. Lasing transition occurs between a rotational sub-level of one vibrational level and

9


another rotational sub-level of a lower vibrational level, where the quantized angular momentum numbers of the two rotational levels differ by exactly unity. When a gas discharge is established, COi molecules are excited into higher electronic levels by colliding with excited A^ molecules, from where they proceed to decay to a lower level. Favourable relaxation rates permit the establishment of a population inversion leading to lasing between numerous rotational levels around 10.6 urn. There are basically three major CC>2 laser designs: slow flow CC>2 lasers (SF), fast axial flow CC>2 lasers (FAF), and transverse flow (TF) CO2 lasers, which are all based on variations of the methods of cooling the CC>2 gas mixture. 7.2.2.1 Slow flow CO2 lasers SF COi lasers rely on thermal conduction through the walls of the cavity for the cooling of the gases, which flow through at around 20 1/min and are usually below atmospheric pressure. Consequently, the cross-sectional area of the discharge region is restricted and relatively narrow bore tubes must be used. Because of the narrowness of tubes the gain per metre length is comparatively small, around 50 W/m, and thus SF COi lasers are either very long (>20 m) or not very powerful (2 lasers are among the best lasers for materials processing operations such as cutting. However, the employment of SF CC>2 lasers in a wider range of materials processing applications is limited by its maximum output power capabilities, typically less than 1 kW. 1.2.2.2 Fast axial flow CO2 lasers The gas flow rate in FAF CC>2 lasers is typically around 400 1/min and is usually below atmospheric pressure. The cooling of the gas mixture is achieved by convection through the discharge zone and as a result the gain per metre length is around 600 W/m, allowing the laser to be fairly compact. Due to the cavity length fairly low Fresnel numbers are achieved giving a low-order beam mode ideal for many materials processing operations. Pumping is either by DC or RF discharge and currently maximum output powers of 5 kW are not uncommon. Because of these advantages, the FAF CC>2 laser is the most commonly used CC2 lasers are the most powerful design, having a gain per metre length of around 2.5 kW/m resulting in maximum output powers in excess of 25 kW. Indeed, currently output powers of 100 kW have been reported for some TF CO2 lasers. The high output powers are due to the electrical discharge being applied transverse to the optical axis and a blower rapidly circulating the gas transverse to the laser cavity, allowing gas operating pressures of several atmospheres. The suitability of the CO2 laser for materials processing operations is due to the laser possessing a variety of beneficial characteristics. The overall operating efficiency of the laser is relatively high, up to 12 per cent, while the efficiency of the conversion of

10


electrical energy to optical energy efficiency is typically around 15-20 per cent [35]. The laser operates in the infrared region of the spectrum at around 10.6 yon in both pulsed and continuous wave (CW) mode giving the beam reasonable material coupling properties. However, the processing of highly reflective materials such as copper, silver, and gold is very difficult. The output power is one of the highest available with CW outputs of several tens of kilowatts easily achievable. At present the CO2 laser dominates the laser market for materials processing use, with applications from cutting, welding, soldering, and drilling to a variety of surface treatment processes. 7.2.3 The Nd:YAG laser [35] The Nd:YAG laser consists of a standard cavity design with neodymium (Nd) doped in a yttrium-aluminium-garnet (YAG) crystal rod. The rod is mounted at one of the foci of an elliptical cavity, while at the other focus is a krypton or xenon flash lamp used to excite the crystal rod. Krypton or xenon flash lamps are used because the necessary wavelength to cause laser excitation of the crystal rod lies within the broad spectrum of their output. However, because the range of excitation wavelengths is very narrow, the part of the spectrum remaining is wasted as heat. Thus the overall operating efficiency of Nd:YAG lasers tends to be relatively low, only 0.1-2 per cent. But, Nd:YAG lasers that incorporate a diode laser to excite (pump) the crystal rod can have an efficiency of up to 30 per cent due to the fact that the diode laser output can produce the narrow range of wavelengths most efficient for exciting solid-state laser materials, therefore producing much less waste heat [6, 42]. The output wavelength of Nd:YAG lasers is 1.06 (im, which is close to the visible spectrum, allowing the use of conventional optics as well as optical fibre delivery of the output beam. The laser can be operated in both CW and pulsed mode with output powers currently ranging from a few watts to several thousand watts, with peak pulse powers of around 50 kW at kiloHertz rates and up to 5 kW in CW mode. Typical materials processing applications of Nd:YAG lasers tend to be marking, scribing (e.g. thick- or thin-film resistor trimming), drilling, cutting, and hermetic seam welding of thin materials. 1.2.4 The excimer laser [43,44] The lasing media used in excimer lasers are the rare gas halides such as ArF, KrF and XeCl. These materials are used because the diatomic molecules, while being stable in the excited state, are not so in the ground state. Thus 'excimer' is contrived from the contradiction of excited dimer [43, 44] Because of the molecules' instability under equilibrium conditions the population of the ground state is low, thus enabling a population inversion to be achieved quite easily. As such, the conversion of electrical energy to optical energy efficiency of excimer lasers is relatively high, typically 10-15 per cent. However, the overall operating efficiency of excimer lasers is only around 2 per cent. Gaseous discharges, electron beams, and photon beams can all be used for pumping. However, because the time duration of the excited state is very short the excimer laser is operated in pulsed mode. Pulse durations are in the order of tens of nanoseconds but the average output power of the laser is generally low, typically 10-200 W, with peak energies per pulse of up to 1 J. However, 1 kW excimer lasers are currently available. Because excimer lasers emit in the short wavelength ultraviolet

11


region of the spectrum which ranges from 193 to 351 nm, the output beams can be focused to extremely small spots, obtaining high intensities at the workpiece. Excimer laser materials processing applications are mainly within the semiconductor industry for the fabrication of premium solar cells, doping, etching, photolithography and laser chemical vapour deposition. Excimer lasers are also finding applications in surface treatment and thin materials micromachining. 1.2.5 The diode laser Von Neumann [45] first proposed carrier injection across the junction of two semiconductor materials that had been doped with either a pentavalent or a trivalent element to create a positive charge (p-type) or a negative charge (n-type) as a means for achieving stimulated emission in semiconductors. The first semiconductor injection lasers were demonstrated in several laboratories in 1962 [46^-9]. However, because of the complexity of the physical mechanisms and the technology involved in semiconductor lasers, more than three decades of world-wide research has been necessary in order to create the present generation of practical semiconductor, or diode lasers. In particular, the rapid and continuous enhancement of diode laser material and fabrication technology over the last decade has increased the average power output of the devices by two-fold each year [2], giving rise today to the commercial availability of diode lasers delivering output powers in excess of 50 mW - HPDL. Indeed, output powers of up to 120 W for single 10 mm experimental diode bars have been reported [1]. Already HPDLs are beginning to replace technologically mature lasers in many application areas, and evidence suggests that this trend will almost certainly occur in the materials processing sector; a lucrative market worth over £120 million per annum [50]. Such research efforts have been primarily motivated by the unique features and possibilities that semiconductor lasers offer, such as high-power conversion efficiency (50-80 per cent), high reliability (lifetimes in excess of 20 000 h), exceedingly low running costs and extensive flexibility potential, and true portability [9]. Also, because semiconductors, and hence diode lasers, can be mass produced, large economies of scale, fuelled by demand, can be realized by manufacturers. As a result diode lasers have the potential to become extremely inexpensive (100 W/cm) that is possible with certain semiconductor laser materials.

Fig. 1.5 Schematic construction of GaAs homojunction diode laser. The emission is confined within the narrow junction region of width, d

1.2.5.2 Stimulated emission and lasing in semiconductors The electronic transitions which give rise to optical amplification in diode lasers occur between a continuum of levels in the conduction band, and a continuum of levels in the valence band. This is profoundly different from the situation in other types of lasers, such as CC>2 and Nd:YAG lasers, in which the radiative transitions take place between two 'sharp' atomic levels. This is shown clearly in Fig. 1.6, where the allowed electron energies are plotted against the crystal momentum, Km, along the [1 0 0] and [1 1 1] crystal directions. The direct-bandgap, F, and the indirect gaps, X and L, are also indicated. GaAs, and other semiconductor compounds used in diode lasers, is a direct-bandgap semiconductor in which the conduction band minimum and the valence band maximum occur at the same point in k-space (the /"point in the case of GaAs). Hence electronic transitions between these two band extrema do not involve the participation of phonons for crystal momentum conservation, which results in highly efficient radiative transitions. As such, this characteristic makes direct-bandgap semiconductors, under appropriate excitation, efficient emitters of photons.

14


Fig. 1.6 The band structure of GaAs at ambient temperature [51]

1.2.5.3 Material systems and fabrication techniques Diode lasers, in particular the active region, can only be fabricated from semiconductor compounds which are direct-bandgap in nature in order to have a high radiative recombination efficiency. The semiconductor compounds surrounding the active region should be lattice matched to the active region to allow suitable confinement. The most extensively studied group of materials for diode lasers has been the ffl-V semiconductor compound system. This system includes binary, ternary, and quaternary solid solutions of group IIIA atoms, Al, Ga and In, and group VA atoms, P As and Sb. There are a number of methods for fabricating diode lasers, based principally on the epitaxial growth of the thin crystalline layers which constitute the laser structure. These include, liquid phase epitaxy (LPE), molecular beam epitaxy (MBE) and vapour phase epitaxy: metalorganic vapour phase epitaxy and metalorganic chemical vapour deposition [52]. 1.2.5.4 Structure and configuration technology Homojunction diode lasers Homojunction diode lasers, whose band diagram is shown in Fig. 1.3 and schematic structure is shown in Fig. 1.5, exhibit high threshold current densities, typically around 400 A/mm2 for a GaAs homostructure at ambient temperature [31]. There are basically two factors influencing this high threshold current density. Firstly, the inherent lack of carrier confinement in the structure. Diffusion of the injected carriers normal to the pn junction plane results in relatively thick inverted regions (typically a few microns), which in turn give rise to high threshold current densities. The second factor is the poor optical confinement of the structure. The photons that are generated by stimulated emission are coupled into an optical field distribution that extends beyond the edges of the active region. As a consequence, only a small fraction of the optical field can experience gain. Furthermore, the fraction of the optical field which extends beyond the edges of the inverted region is strongly absorbed by the un-pumped material.

15


Heterojunction diode lasers Fabrication techniques such as MOCVD and CVD-diamond allow the close control of material composition, layer interfaces, and device geometry [53, 54]. Such developments in material and fabrication technology have lead from the homoj unction diode laser to the modem and more efficient heterojunction and double-heterojunction diode lasers. Significant reduction in the threshold current density and improvement in the efficiency of diode lasers have been realized by embedding the recombination region in semiconductor material of higher bandgap and lower refractive index to create a heterojunction, usually GaAlAs, sandwiching the active GaAs layer as shown in Fig. 1.7.

Fig. 1.7 End-on view of the front facet of the common double-heterojunction diode laser [55]

The potential barriers thus formed at the edges of the recombination region serve to confine the injected carriers to an active region whose dimensions can be made much smaller than the diffusion length of the minority carriers. In this way the carriers are physically prevented from diffusing away from the junction by the potential barriers created by the differing energy gaps of the various layers. Clearly, the more densely packed together the electrons and holes are within the gain medium, the higher the population inversion. Additionally, the refractive indices differences between the active layer and the surrounding layers (the active region having a higher refractive index) results in confinement of the optical field due to the phenomenon of total internal reflection taking place at the boundaries of the active region. Figure 1.8 shows a number of current heterojunction configurations. For each configuration the energy band diagram under forward-bias (top) and optical field distributions (bottom) are illustrated schematically (note that the hatched regions in the bandgap diagrams indicate filled electron states). Because the optical and transverse carrier confinement increases the density of electrons within the gain medium, a lower current density is required to create a similar population inversion, 10 A/mm2 compared with 400 A/mm2 [31]. Indeed, the development of these heterostructure diode lasers has made possible the CW operation of diode lasers at room temperature.

16


Fig. 1.8 Heterojunction diode laser configurations: (a) homojunction; (b) single-heterojunction; (c) double-heterojunction; (d) large optical cavity (LOG); (e) separate confinement heterojunction; and (f) multiple quantum well (MQW) [52]

1.2.5.5 Combining of individual diode lasers for increased power Currently, the maximum power of a single diode laser emitter is only around 60 mW and the power density of the emitter facet is around 2 MW/cm2 [2]. The power density of a single emitter is limited to 1-10 MW/cm2 by the catastrophic optical damage (COD) facet threshold requirement [13]. Currently, the only feasible way of increasing diode laser output power is to increase the number of emitters in the system. Commercially this is achieved in a number of ways. One-dimensional and two-dimensional diode laser arrays A typical example of a 10 mm diode bare emitting 50 W Q-cw from an array of 55, 5x1 jam, emitting stripes with a centre-to-centre spacing of 180 (im is shown in Fig. 1.9. The brightness of these one-dimensional (1-D) arrays is typically less than 20 MW/cm2. sr. An increase in power (up to 5 kW) is achieved by stacking these bars to form a two-dimensional (2-D) structure with a layer-to-layer spacing of 0.4-2 mm, depending upon the heat dissipation requirements. The beam divergence ranges from 30 x 10 degrees to 40 x 10 degrees. However, initial work by Snyder et al. [56] which led to the development of fast axis diffraction limited microlenses suitable for onedimensional arrays, and subsequent research by a number of workers, who have made commercially available stacked microlenses for 2-D arrays [13], has meant that a

17


collimated beam with a divergence of 1.5 x 10 degrees full wave half maximum (FWHM) and a brightness of 20 MW/cm2sr is possible. A symmetrized and re-imaged beam of numerical aperture (NA) 0.4 from such a brightness can give an intensity of about 10 MW/cm2, suitable for some materials processing applications.

Fig. 1.9 Typical structure of a 1-D commercial diode laser array [57]

Monolithic diode laser arrays have been fabricated that operate at greater than 120 W, with total efficiencies in excess of 60 per cent [58]. Table 1.1 summarizes the operating characteristics of the current commercially available 1-D and 2-D diode laser arrays. Table 1.1 Typical operating characteristics of commercially available linear and stacked diode laser arrays Operating mode Q-cw CW

Power Wavelength Threshold Operating Operating Conversion (W) (nm) current (A) current (A) volts (V) efficiency (%) 785-810 20-35 2-96 1-5000 75-125 25 .W C , then S is positive (6=0 degrees) and the liquid will spread spontaneously over the solid to form a thin film. If, on the other hand, Wa < Wc, then S is negative (9 > 0 degrees) and the liquid will not spread over the surface but will instead form droplets with a finite contact angle. So, by equating equation (2.1) with equation (2.9), the condition for spreading to occur spontaneously in terms of S is given by

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 [71]. In either case, the driving force for wetting is the reduction of the surface free energy of the solid by the liquid (YW - Ys/)- Spreading requires the additional contribution to the driving force of the free energy of the interfacial reaction [72]. 2.2.4 Contact angle hysteresis For a given liquid-solid system it has been found that a number of stable angles can be measured. Two reproducible angles are the largest, known as the advancing angle, OA, and the smallest, known as the receding angle, 0R [73]. The angles are so named because the advancing angle is often obtained by advancing the periphery of a sessile drop through the addition of more liquid; the height of the drop is subsequently increased while its base remains the same. Thus the contact angle will be greater than

38

Basic Background Theory of Wettability, Adhesion, and Bonding

before the addition of the liquid. The receding angle is obtained by pulling the periphery, thereby removing liquid from the drop. Again the base will not change but the drop will reduce in height and the contact angle will therefore reduce. The difference between 0A and OR is called contact angle hysteresis. Both advancing and receding angles can be seen on a drop when the solid substrate is placed at an angle as depicted in Fig. 2.3.

Fig. 2.3 Schematic of contact angle hysteresis

2.2.5 Static and dynamic contact angles Static angles are defined as those angles which do not change with time. They are in equilibrium and they represent either stable or metastable situations [65, 66]. Angles that change with time do so because the system has not reached equilibrium and are consequently known as dynamic angles [65, 66], which are both advancing and receding. If a mass of liquid comes into contact with a solid surface, the liquid-vapour interface at the point of the three-phase line [liquid-solid-vapour (LSV) line] will tend to advance over the solid by means of the self-spreading mechanism until an equilibrium contact angle, 9e, is reached [65]. At the same time the interface will tend to reach a state of constant shape and minimum area. These changes are illustrated schematically in Fig. 2.4.

Fig. 2.4 The stages involved during the self-spreading of a liquid drop to form dynamic angles and then a static angle in equilibrium

39


In Fig. 2.4(a) the drop has just come into contact with the solid surface and the contact angle, 0, is almost 180 degrees. After some time 6 has become smaller [Fig. 2.4(b)] until finally the system reaches a state of complete equilibrium [Fig. 2.4(c)]. In this state the LSV line does not move and 6 has now become a static contact angle [66]. The angles 6a and 6b formed while the LSV line is in motion along the solid surface are both dynamic angles and are represented in general terms by 6^. In Fig. 2.4 the liquid-vapour interface is shown in all three stages in its equilibrium constant-curvature minimum-energy shape, that is, that of a zone of a sphere. Yet this happens only when the motion of the LSV line is very slow. In reality, however, the liquid-vapour interface will not be in a state of constant curvature while the liquid is in motion. The driving forces by which the liquid spreads across the solid surface, dynamically changing its contact angle until it is eventually static and in equilibrium, are referred to as self-spreading forces and can be analysed in terms of Young's equation [equation (2.1)]. Thus, for a system where yjv and ys/ are in equilibrium but Od > Qe the force on the LSV line, Fisv, is

A more detailed illustration of the self-spreading drop is given schematically in Fig. 2.5. Figure 2.5(a) depicts the drop shortly after it has come into contact with the solid surface. The pull of the solid-vapour interface has caused a thin layer of liquid, F, to move out from the bulk of the drop. Since a strong pull of the solid-vapour interface cannot extend above the surface more than a very few molecular diameters, the layer Y is extremely thin and as such, possibly may not possess liquid properties [66]. After some time a secondary region of liquid develops, B, which is thick enough to have liquid properties [Fig. 2.5(b)]. When complete equilibrium is reached, the regions Y and B disappear due to their absorption by the bulk of the drop.

Fig. 2.5 The primary (Y) and the secondary (B) layers in the self-spreading of a liquid drop to equilibrium

40


2.2.6 The effect of surface roughness on contact angle For an ideal solid surface, that is, plane, homogeneous, and uniform, the liquid has only one angle with the solid. In practice, however, this is rarely the case since for a given solid-liquid system a number of stable angles can be measured. As discussed above, the two most clearly discernible angles are 0A and 6R, the difference between the two angles (&& - OR) being known as contact angle hysteresis. Hysteresis of the contact angle has been shown to be primarily the result of surface roughness [73-78]. As such, it is important to consider also the influence of the substrate surface roughness on the wetting contact angle. Rough grooves on a surface, which may contribute to the influence on contact angles, can be categorized as either radial or circular grooves. Any actual rough surface can be represented by a combination of these two cases [74]. In fact two roughness parameters can be defined: the Wenzel type, DR [79] and the Cassie/Baxter type, FR [80]. In the instance where wetting spreads radially, as is the case where the solid surface is laser treated, then the resulting radial contact angle, Qmd, is related to the theoretical contact angle, 6th, by [80]

According to Neumann [81, 82], only if FR is equal to zero, then a model similar to that for heterogeneous solid surfaces can be developed in order to account for surface irregularities, being given by Wenzel's equation

where ra is the roughness factor defined as the ratio of the real and apparent surface areas and 6W is the contact angle for the wetting of a rough surface. A rearrangement of Wenzel's equation gives

It is important to note that Wenzel's treatment is only effective at the position of the wetting triple line [74]. Nevertheless, equation (2.13) shows that %/ is inversely proportional to the smoothness of the solid surface, ra. Thus if ra is small, that is the solid surface is rough, then fsi will become small. Therefore, according to equation (2.13) an increase in the contact angle will be inherently realized by the liquid. 2.2.7 The effect of chemical composition on contact angle Generally the wettability of solids is influenced to some extent by the nature and packing of the atoms or groups of atoms which compose the solid's surface, being independent of the nature and arrangement of the sub-surface atoms. This is because the surface atoms for both solids and liquids attract each other by means of highly

41


localized attractive force fields such as van der Waals forces which decrease in intensity with increased distance. As such, these types of forces become negligible at distances of only a few atom diameters, therefore there is little contribution to the force of adhesion by atoms below the surface layer of atoms in both the solid and the liquid [83]. 2.2.8 The effect of temperature on contact angle Much work has been conducted as to the effects of atmospheric and liquid temperature variations on the liquid contact angle. Moderate temperatures, that is those between 20 and 80 °C have been seen to have little or no effect whatsoever upon the contact angle of most liquids with most solids [84, 85]. Nevertheless, research has shown that for some liquids in contact with certain solids, temperatures close to 0 °C and around and above 100 °C do have some sort of effect upon the liquid contact angle. For almost all vitreous enamels, however, even temperature variations well beyond 100 °C have not been observed significantly to affect the wettability performance and thus the contact angle [86].

2.3 Surface energy and the dispersive/polar characteristics The intermolecular attraction which is responsible for surface energy, y, results from a variety of intermolecular forces whose contribution to the total surface energy is additive [85]. The majority of these forces are functions of the particular chemical nature of a certain material, and as such the total surface energy comprises of yp (polar or non-dispersive interaction) and yd (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 [87]

Similarly, Wad can be expressed as the sum of the different intermolecular forces that act at the interface [87]:

If a liquid which 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 [88, 89]

However, by equating equation (2.16) with equation (2.3), 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

42


Consequently, from equation (2.18), one can estimate the dispersive component of a solid substrate surface energy,/ d sv , by plotting the graph of cos 9 against ( y f v )1/2/7;v. This is shown in Fig. 2.6 for a theoretical liquid system on any solid substrate. Thus, according to Fowkes [87], the value of 7 dsv is estimated by the gradient (= 2(y dv )1/2) of the line (—) which connects the origin (cos 9 = -1) with the intercept point of the straight line [cos 9 against (ydv )1/2/7;v] (—) correlating the data point with the abscissa at cos 0=1.

1/2

/ d \ Fig. 2.6 Plot of cos 9 against \y lv ) /y/v for a theoretical liquid system on any

solid substrate

In contrast, it is not possible to determine the value of the polar component of a solid substrate surface energy, ypsv» directly from a plot of cos 0 against (ydlv )1/2/Y/v This is because the intercept of the straight line (cos 0 against (ydlv )'/2/7;v) is at 2(7 y7 M /y/v, and therefore 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 [87, 90]. As such, it is possible to calculate the dispersive component of the work of adhesion, Wdd, by using only the relevant part of equation (2.16) thus

43


If one plots a graph of Wad against Wd for the solid substrate, then for each particular liquid in a given system in contact with the solid substrate surface, Wad, which was determined from equation (2.3), can often be correlated with Wd, which was determined from equation (2.19), by the straight line relationship

Consequently, 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 axis. Also, if one plots a graph of yp against yd, then for the liquids in a given liquid system, yp can often be correlated with Y?V by the straight line relationship

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 (2.20) into equation (2.16) and rearranging, then

or, alternatively

By introducing equation (2.21) into equation (2.23) and differentiating with respect to j \ 1/2

( Yivj

/

, \l/2

, considering that (j;v j

/

\l/2

and (y^ J

are constant, then the following can be

derived:

Since 7 dsv for the solid substrate can be determined previously directly from the plot of cos 9 against (ydlv )1/2/7/v, then it is possible to calculate 7 fv for the solid substrate equation (2.24) directly. By employing this approach it is possible to determine, from contact angle measurements and the control liquid surface energy properties, the changes in the wettability characteristics effected by laser treatment of a number of engineering materials in terms of surface energy.

44


2.4 The bonding of liquids and solids An essential condition for bonding is good wettability of the substrate surface by a liquid or a melt. As such, the surface tension of the liquid or melt should not be excessively high, while the substrate surface itself should be free from coarse irregularities and dirt. However, many coatings are held more firmly on a rough surface than on a completely smooth surface. In the development of the bond, the composition of the liquid or melt is of importance since it determines the surface tension and the coefficient of thermal expansion. The achievement of strong and reliable bonds between any two dissimilar materials depends on the mechanisms that come into play during the bonding process. The mechanisms for considering and promoting bonding between liquids or melts and solid substrates can be categorized generally as: physical bonding; mechanical bonding; and chemical bonding. Within each of these somewhat broad categories of bonding mechanisms a number of theories exist with regard to mechanisms involved, with the category of chemical bonding providing by far the most by virtue of the vast number of possible chemical reactions that can occur between the range of dissimilar materials that can be bonded together. In practice, however, complex combinations of the various bonding mechanisms actually occur, varying according to the types of materials used [91]. Even so, enhancement of all the mechanisms involved when preparing a material for bonding and when joining is a very important consideration in order to engineer a strong and lasting bond. 2.4.1 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. In the case of enamelling steel, either the breaking stress of the steel or the fired enamel glaze. However, the actual failure stress is often several orders of magnitude smaller because of flaws in the bond itself. Physical bonding provides a useful guideline to the selection of materials that will bond well together. 2.4.2 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 roughened 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

45


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 effect an increase in the level of physical bonding. The degree of mechanical bonding is often modelled on the interface ratio or undercut density. The former is the ratio of actual bonded surface to that of a flat, smooth interface and is often taken on a cross-section sample. The undercut density measures the number of re-entrant features on a cross-section per linear distance. An increase in either number usually enhances the bond.

2.4.3 Chemical bonding Considerable research into the various chemical mechanisms that can be present during the bonding process is in progress. 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. A chemical bond is represented by an electronic structure and a balance of bond energies across the interface whether the bonding is ionic, covalent, or metallic. These factors influence the bond microstructure.

2.4.3.1 Theory of formation of the intermediate oxide layer With regard to steel substrates, it is known that bonding of the enamel and the steel is obtained when the firing is performed in oxidizing conditions. Before the enamel melts, the oxygen in the air simply penetrates through the porous enamel, oxidizing the steel surface. A film of scale forms, insecurely attached to the metal. In order to develop bonding, it is necessary for it to be completely dissolved in the enamel melt in the form of FeO. Only when a contact between the melt and a clean steel surface is produced, is the growth of a new intermediate oxide layer on the boundary observed. This layer is a bonding layer, and holds firmly to the steel surface until a certain optimum thickness is exceeded. If firing is continued the layer of FesC^ at the interface between the metal-enamel boundary becomes excessively thick, resulting in spallation. Many researchers consider that the bonding occurs as a result of the oxide film being partially dissolved in both the steel and the enamel [92-94].

2.4.3.2 Theory of bonding due to oxygen bridging This theory, in contrast to the theory of formation of the intermediate oxide layer, is based on the observation that bonding of the enamel with a substrate (usually steel) is achieved after the scale, formed before the fusion of the melt, has completely dissolved. The addition to the enamel of bonding oxides, such as cobalt oxide, accelerates considerably the solution of the scale, as well as reducing the firing time and enhancing the bond strength.

46


After solution of the originally formed scale, the enamel melt comes into close contact with the surface of the substrate. The added oxide, being an oxygen carrier, gives oxygen ions to the surface of the substrate. These ions are bonded not only with the substrate, but also with the melt. An actual oxide film is not formed, since the ions of oxygen are common to both the substrate and the enamel, in effect creating a bridge of oxygen ions between the two. 2.4.3.3 The electrochemical theory Arguably, in terms of molten liquids (usually enamels) on metallic substrates, the electrochemical theory is the most promising explanation for the development of bonds between these two materials. When the oxide layer on a metal is completely dissolved by a molten glass (enamel) layer, then a redox reaction has to occur at the interface in order to form more metal oxide. The principal aspect of this theory is that, during firing of the enamel, ferric oxides within the substrate react with other metallic oxides in the enamel. Thus, for instance, with cobalt, a reduction in the cobalt by the metallic iron gives

Equation (2.25) is the redox reaction forming FeO and occurs at the interface. The metallic iron is dissolved and passes into the melt in the form of a silicate, while the cobalt precipitates from the melt and is deposited on the surface of the substrate. Between the parts of the surface not coated with deposited cobalt and the particles of cobalt, local short-circuited cells are formed. The iron gives the cobalt its valence electrons in the form of Fe2+. The cobalt, in turn, gives the electrons to the depolarizers, ions of Fe3+, or to molecular oxygen. The cell functions until the excess electrons are transferred to the depolarizers. Continuous solution of metallic iron occurs in separate parts of the surface of the substrate. The surface under these conditions is badly corroded, and the depressions formed are immediately filled with molten enamel which is held extremely tightly. An electrochemical system for a vitreous enamel and a mild steel is shown in Fig. 2.7.

Fig. 2.7 Schematic representation of the electrochemical bonding process [95]

47


Modification of the solid substrate surface as a result of electrochemical reactions has been shown to be more effective in terms of bond generation and bond strength than preliminary mechanical roughening [95]. This is primarily because the surface area on which bonding of the melt with the solid substrate surface is greatly enlarged. Also, the air film does not prevent penetration of the melt into the pores and crevices, thereby allowing more complete contact. 2.4.3.4 Other chemical bonding theories A number of other widely recognized and accepted theories exist with regard to the bonding of the enamel coating with a substrate. These can be summarized as [96]: 1. The hydrogen reduction theory. This theory is concerned principally with the enamelling of steel but is, nevertheless, applicable to other systems. The theory states that the oxide of cobalt is a sacrificial material, in that it is reduced to cobalt by the hydrogen in order to permit an intimate contact and adherence of the enamel to the substrate. 2. The dendrite theory. This theory is mainly concerned with metallic substrates, but as above, could reasonably be applied to other systems providing certain elements are present within the liquid and the solid substrate. The theory suggests that the dendrites formed by the reduction of metallic iron, or the growth of iron crystals at the interface, serve as projections to hold the enamel and the iron. 3. The atomic attraction theory. This theory claims that any glass which contains a large amount of the lowest valence oxide of the substrate to which it is applied, at the interface, adheres to the substrate.

48

Chapter 3

Laser Surface Modification of Selected Composite Materials for Improved Wettability Characteristics

This chapter describes the interaction ofCO2, Nd:YAG, excimer, andHPDL radiation with the surface of a number of composite materials and the subsequent effects on the wettability and bonding characteristics of the materials. Laser radiation was found to effect significant changes in the wettability characteristics of the materials. In order to give the work practical validity, the changes in the wettability characteristics of the composite materials are quantified in terms of firing a vitreous enamel coating on to the laser-treated surfaces of the composite materials. The work presented in this chapter demonstrates that the wettability characteristics of the ordinary Portland cement (OPC) and the Al2O3/SiO2-based composite material could be controlled and/or modified with laser surface treatment.

3.1 Introduction Wetting is often the primary factor governing whether a coating will adhere and bond to a substrate in practical applications such as enamelling, painting, etc. Indeed, in many technological applications where vitreous enamels are fired on to ceramic substrates, the performance of the article is directly linked to the nature of the enamelceramic interface. Many studies to investigate these phenomena have been carried out, however, they have been principally concerned with the wettability of zirconia and other oxide ceramics on metals [97-101] as well as the adhesion of silicone sealants to aluminium [102], and the coating of aluminium alloys with ceramic materials [103, 104]. The interfacial mechanisms investigated have centred principally around the thermodynamic criterion [98, 99, 101], the electronic theory [100] and the occurrence of oxidation [97, 105].

49


To date, very little published work exists pertaining to the use of lasers for altering the surface properties of materials in order to improve their wettability characteristics. Notwithstanding this, it is recognized within the currently published work that laser irradiation of a metal surface can bring about changes in the metal's wettability characteristics. Previously Zhou and de Hosson [103, 104] carried out work on the laser coating of aluminium alloys with ceramic materials (SiC>2, AÔs, etc.), reporting on the well-documented fact that generated oxide layers often promote metal/oxide wetting. Further, Heitz et al. [106], Henari and Blau [107] and Olfert et al. [108] have found that excimer laser treatment of metals results in improved coating adhesion. The improvements in adhesion were attributed to the fact that the excimer laser treatment resulted in a smoother surface and as such enhanced the action of wetting. Yet the reasons for these changes with regard to changes in the material's surface morphology, surface composition and surface energy are not reported. However, in a number of more comprehensive investigations by Lawrence et al., which compared the effects of CC>2, Nd:YAG, excimer, and HPDL radiation on the wettability characteristics of a mild steel [23, 24] and a AÔs/SiOi-based ceramic material [21, 22] it was found that changes in the wettability characteristics of the steel varied depending upon the laser type. Furthermore, Lawrence et al. have conducted numerous studies to investigate the feasibility and characteristics of laser enamelling ceramic materials [109-112] and steels [113,114]. The unique characteristics of lasers provides them with the capability for the non-contact processing of materials which are otherwise difficult to process. Concrete is one such material since it is a composite, consisting of an array of fine and coarse aggregate pieces embedded within an ordinary Portland cement (OPC) matrix. Consequently, the processing and surface treatment of concrete can be a difficult undertaking. The laser processing of concrete is a field of ongoing research, with many studies having been carried out to investigate the technique itself and the associated phenomena. Most of the research, however, has concentrated on the laser cutting of concrete and reinforced concrete using high-power CC>2 lasers, most prominently with regard to nuclear reactor decommissioning [115-117]. Also, as part of nuclear plant decommissioning, Li et al. [118-121] conducted research to determine the workability of several laser techniques for sealing/fixing radioactive contamination on to concrete surfaces. Such techniques experimented with were: direct glazing of the concrete, single and multiple layer fusion cladding, and combined chemical/fusion cladding. Work by Sugimoto et al. [122] focused upon modifying the surface appearance and surface properties of cement-based materials using a high-power CO2 laser. The laser treatment produced novel surfaces, with surface textures, properties, and appearance unique to laser treatment. The resultant physical characteristics and mechanical behaviour of the post-process cement-based materials were later fully characterized by Wignarajah et al. [123]. Borodina et al. [124] have carried out investigations into the structural changes within the composition of zirconia concrete caused by surface exposure to CC>2 laser radiation, detailing microstructural changes, phase changes, and the absorptivity characteristics. In all of these studies, spallation and excessive cracking and porosity formation were found to be major problems undermining the performance of the laser-treated surface layer. However, Lawrence and Li [125-128]

50

Laser Surface Modification of Composite Materials for Improved Wettability Characteristics

have treated the OPC surface of concrete with both CO2 and HPDLs. The HPDLgenerated OPC glaze was shown to be more than an effective surface modification insofar as it provided superior mechanical, physical, and chemical characteristics over an untreated or CO2 laser-treated OPC surface.

3.2 Experimental procedures 3.2.1 Materials 3.2.1.1 Ordinary Portland cement The concrete studied in the experiments was the ubiquitous OPC-based concrete. For the purpose of experimental convenience the as-received concrete blocks were sectioned into squares (120 x 120 x 20 mm3) prior to HPDL treatment. The composition by volume of the concrete was as follows: 20 mm limestone aggregate (40 %), 10 mm limestone aggregate (14 %), zone M sand (28.5 %), OPC (10.5 %) and particulate fine aggregate (7 %). In order to obtain results of a practical and useful nature, the area of the concrete irradiated during the experiments was the naturally occurring 'as-cast' OPC surface of concrete. In this case the OPC surface of the concrete had a thickness of 2.5 mm. The composition by volume of the OPC was as follows: CaO (63.9%), SiO2 (21.9%), A12O3 (5.7%), Fe2O3 (2.8%), SO3 (2.7%), MgO (2.2%), K2O (0.7%), and Na2O (0.1%). 3.2.1.2 AI2O2/SiO2-based composite The Al2O3/SiO2-based composite material consisted of mixed vitrifiable oxide powders such as chamotte (mainly SiO2 (53wt%) and A12O3 (42wt%)), Fe2O3, MgO, ZrO2 and ZnO was produced. The oxide powders were sieved to ensure a particle size of less than 75 u,m, then thoroughly mixed together to ensure homogeneity, along with approximately 50wt% diluted sodium silicate solution so as to form a manageable paste. The Al2O3/SiO2-based composite material was then pasted on to an OPC substrate to a thickness of 2 mm and allowed to cure at room temperature for 12 h. 3.2.1.3 Vitreous enamel The composition of the enamel consisted mainly of the following: SiO2, B2O3, Na2O, Mn, F, and small quantities of Ba, A12O3, and Ni, while the powder size was less than 75 ujn. The enamel frit paste was allowed to cure at room temperature for 2 h and then irradiated immediately with the laser beams. 3.2.2 Laser processing procedure Figure 3.1 schematically illustrates the general laser processing experimental arrangement in which the defocused beams of the lasers were fired back and forth across the surfaces of the composite materials by traversing the samples beneath the laser beam using the x- and y-axis of the computerized numerical control (CNC) gantry table. The general operating characteristics of the lasers used in the study of the Al2O3/SiO2-based composite material are detailed in Table 3.1. Both pulsed and CW lasers were used in the study, therefore, both the average power and the peak power of each laser will differ. So, in order reasonably to compare the effects of each laser on the wettability characteristics of the selected composite materials, the laser energy

51


density (fluence) of each laser beam incident on the surface of the composite materials was set by manipulating the laser power densities and traverse speeds such that the energy density of each of the four lasers incident upon the surface was around 165 J/cm2. In the study conducted on the OPC only HPDL radiation was employed. This was principally because previous studies have shown conclusively that the HPDL is the most appropriate laser for the treatment of OPC [126, 127].

Fig. 3.1 Schematic diagram of the set-up for the COz, Nd:YAG, HPDL, and excimer laser interaction experiments with the composite materials

Table 3.1 Details of the selected industrial lasers used

Lasant Wavelength Maximum average output Maximum pulse energy Pulse width Repetition rate Fibre core diameter Mode of operation

CO2 CO2 gas 10.6 urn IkW ~ ~ ~ ~

cw

Laser HPDL Nd:YAG GaAlAs Nd:YAG crystal 810±20nm 1.06 Jim 60 W 400 W ~ 70 J ~ 0.3-10 ms ~ 1-1000 Hz 600 um 600 um CW Pulsed (rapid)

Excimer KrF gas 248 nm 5W 35 J 20ns 1-55 Hz ~ Pulsed (multiple)

In order to analyse the laser-treated specimens, they were, in some instances, sectioned with a Struers cutting machine using a diamond-rimmed cutting blade, and then polished using cloths and diamond suspension pastes down to 3 fj,m. Samples, both sectioned and aun-sectioned, were then examined using optical microscopy, scanning electron microscopy (SEM), energy disperse X-ray analysis (EDX), X-ray diffraction (XRD), and X-ray photoemission spectroscopy (XPS) techniques. 3.2.3 Wettability characteristics analysis procedure To investigate the effects of laser radiation on the wetting and surface energy characteristics of the composite materials, two sets of wetting experiments were conducted. The first set of experiments was simply to determine the contact angle between the enamel and the composite materials before and after interaction with the selected industrial lasers. The second set of experiments consisted of control

52


experiments carried out using the sessile drop technique with a variety of test liquids with known surface energy properties in order to quantify any surface energy changes in the composite materials resulting from laser interaction. The enamel-composite materials wetting experiments were carried out in atmospheric conditions with molten droplets of the enamel (600 °C). The temperature of the enamel throughout the experiments was measured using a Cyclops infrared pyrometer. The droplets were released in a controlled manner on to the surface of the composite materials (laser treated and untreated) from the tip of a micropipette, with the resultant volume of the drops being approximately 15 x 10~3 cm3. Profile photographs of the sessile enamel drop were obtained for every 60 °C fall in temperature of the molten enamel drop, with the contact angle subsequently being measured, and a mean value being obtained. The sessile drop control experiments were carried out, using human blood, human blood plasma, glycerol, and 4-octanol. The test liquids, along with their total surface energy (y 2 ), dispersive ( y f v ) , and polar (y fv) component values are detailed in Table 3.2. The experiments were conducted in atmospheric conditions at a temperature of 20 °C. The droplets were released in a controlled manner on to the surface of the test substrate materials (laser treated and untreated) from the tip of a micropipette, with the resultant volume of the drops being approximately 6 x 10~3 cm3. Each experiment lasted for 3 min with profile photographs of the sessile drops being obtained every minute. The contact angles were subsequently measured. A mean value was then determined. The standard deviation due to experimental error was calculated as being ±0.2 degrees. Table 3.2 Total surface energy (}}v) and the dispersive ( y f v ) a°d polar ( y f v ) components for the selected test liquids [102]

Liquid Human blood Human blood plasma Glycerol 4-Octanol

Y (mj/m2) 47.5 50.5 63.4 27.5

ti (mj/m2) 11.2 11.0 37.0 7.4

yP

(mj/m2) 36.3 39.5 26.4 20.1

It was observed during the wetting experiments conducted with both the enamel and the control liquids that, throughout the period of the experiments, no discernible change in the magnitude of the contact angle was observed, indicating that thermodynamic equilibrium was established at the solid-liquid interface at the outset of the experiments. This is perhaps surprising when one considers the temperature effect on surface tension as described by Mayers [129]. However, results similar to those observed in this study have been described by Agathopoulos and Nikolopoulos [130].

53


3.3 The effects of high-power diode laser radiation on the wettability characteristics of ordinary Portland cement 3.3.1 The general effects of high-power diode laser radiation The typical surface morphology of the glaze generated on the OPC surface of concrete when using the HPDL is shown in Fig. 3.2. As is evident from Fig. 3.2, crack and porosity formation were common features of the HPDL glaze.

Fig. 3.2 Typical optical surface morphology of the HPDL-generated OPC surface glaze (2.25 kW/cm2 power density, 240 mm/min traverse speed) The complex chemistry of the OPC surface of concrete and the hydration of its various constituents are a complex issue. Nonetheless, it is known that the constituents of OPC are minerals which exist as multi-component solid solution chemical compounds. Of particular importance with regard to this study, OPC contains, in relatively large proportions, a number of basic glass network formers and modifiers: SiO2, AliOa, and Fe2O3. Consequently, the intense local heating brought about by the incident HPDL beam results in melting of these compounds at around 1283 °C, thereby causing the materials to lose the retained water and form an amorphous glassy material consisting of various calcium-silicate-alumina compounds [117]. Furthermore, the fracture section of the HPDL glaze generated on the OPC surface of concrete is shown in Fig. 3.3. As can be seen from Fig. 3.3, the microstructure of the HPDL-generated glaze has no discernible structure and appears to be fully amorphous. Indeed, the amorphous nature of this glaze was verified by XRD analysis results shown in Fig. 3.4. In addition, distinct changes in the colour of the OPC surface were observed after HPDL treatment. Typically the OPC surface changed colour from grey to green. These changes are due to the resultant phase transitions and, in addition, the presence in small concentrations of metal transition ions in various oxidation states within the OPC composition - in particular, ferric ions in the Fe3+ and Fe2+ oxidation state. Fe3+ and Fe2+ ions are known to give rise to green and blue colours respectively when subjected to intense heating [131, 132]. However, if both phases are present within the composition,

54


then the colour is determined by the Fe3+/Fe2+ ion ratio, resulting in dark blue or black colours [131, 132]. Since the surface produced after HPDL treatment was green, then it is reasonable to assume that both phases were not present within the OPC.

Fig. 3.3 Typical SEM micrograph of the fracture section of the HPDL-generated OPC surface glaze (2.25 kW/cm2 power density, 240 mm/min traverse speed)

Fig. 3.4 X-ray diffraction analysis of the HPDL-treated OPC surface Petzold and Rohrs [133] have determined from differential thermal analysis (DTA) results that up to approximately 420 °C OPC remains relatively stable. Notwithstanding this, some dehydration does occur and water is also lost from the pores of the cement. This is, however, far outweighed by the dehydration of Ca(OH)2 which follows shortly after 420 °C is exceeded in accordance with

55


Furthermore, the dehydration of the Ca(OH)2 promotes the development of microcracks which begin initially around the Ca(OH)2 [134]. Moreover, this dehydration results in unslaked lime (CaO), which is effectively the generated heat-affected zone (HAZ), since the temperature of the surface of the OPC during interaction with the HPDL during glazing was measured to be well in excess of 420 °C. This generated CaO HAZ was observed located either below the glazed surface layer or around the edges of the glazes. Indeed, by using a phenolphthalein indicator followed by water misting, it was possible to discern clearly the HAZ around the HPDL-treated zone on the OPC surface of the concrete, since phenolphthalein is an indicator which is colourless in CaO, turning violet-red in the presence of Ca(OH)2 due to the change in pH. 3.3.2 Wettability and surface energy characteristics 3.3.2.1 Contact angle and wettability As was mentioned earlier, it proved impossible to fire the enamel glaze directly on to the OPC surface without prior HPDL treatment. An optical micrograph of a sessile drop of enamel (20 °C) placed on the surface of the OPC before (a) and after (b) HPDL irradiation, with the contact angle superimposed, is shown in Fig. 3.5. The experimental results showed that throughout the period of cooling of the enamel, from the molten state at 600 °C to the solid state at room temperature, no discernible change in the magnitude of the contact angle took place during the time of the experiments. This indicates that thermodynamic equilibrium was established at the solid-liquid interface at the outset of the experiment [43]. Figure 3.5 shows clearly that prior to HPDL treatment it was not possible to fire the enamel on to the surface of the OPC since the contact angle was measured as 109 degrees, and as such would prevent the enamel from wetting the OPC surface.

Fig. 3.5 Contact angles for the enamel on (a) the untreated surface of the OPC and (b) the HPDL-treated surface of the OPC One explanation for the fact that HPDL treatment of the OPC is necessary so that the enamel completely wets and adheres to the surface is that the surface resulting from the HPDL treatment is somewhat smoother, with an Ra value of 2.88 fim compared

56


with 21.91 ujn, and, according to equation (2.14), will intrinsically effect a reduction in the contact angle. Also, wetting will have certainly been influenced by the increase in the 62 content of the OPC surface as a result of the HPDL treatment, since this is known to increase the likelihood of wetting [97, 130]. Wetting is governed by the first atomic layers of the surface of a material. Thus, in order to determine element content of O2 at the surface of the OPC, it was necessary to examine the surface using XPS analysis. A difference in the surface O2 content of the OPC before and after HPDL treatment was observed, increasing from an initial value of 44.7at% to 49.2at%. 3.3.2.2 Surface energy and the dispersive/polar characteristics As was discussed in Chapter 2, it is possible to estimate reasonably accurately the dispersive component of the OPC surface energy, y dsv, by plotting the graph of cos 9 against (yf v ) 1/2 /7/v in accordance with equation (2.18), with the value of ydsv being estimated by the gradient [= 2(y dsv )1/2] of the line which connects the origin (cos 9 - 1) with the intercept point of the straight line (cos 9 against (y dlv )1/2/Y/v) correlating the data point with the abscissa at cos 0 = 1 . Figure 3.6 shows the best-fit plot of cos 9 against (y? v ) I/2 /7; v for the untreated and HPDL-treated OPC-experimental control liquids system.

/

j \l/2

Fig. 3.6 Plot of cos 0 against (yff J /y/v for the untreated and HPDL-treated OPC in contact with the wetting test control liquids

Comparing the ordinate intercept points of the untreated and HPDL-treated OPCliquid systems, it can be seen clearly from Fig. 3.6 that for the untreated OPC-liquid systems the best-fit straight line intercepts the ordinate closer to the origin. This indicates that, in principle, dispersion forces act mainly at the OPC-liquid interfaces

57


resulting in poor adhesion [87, 90]. In contrast, Fig. 3.6 shows that the best-fit straight line for the HPDL-treated OPC-liquid systems intercepts the ordinate considerably higher above the origin. This is indicative of the action of polar forces across the interface, in addition to dispersion forces, hence improved wettability and adhesion is promoted [87, 135]. Again, as was discussed in Chapter 2, it is not possible to determine the value of the polar component of the OPC's surface energy yfv directly from Fig. 3.6. If the technique discussed in Chapter 2 is used, then the values of c and d in equation (2.21) for the control liquid system used can be found. Thus equation (2.21) becomes

By introducing these values into equation (2.24), the following is derived:

From a plot of equation (2.20), a can be determined for the untreated and HPDLtreated OPC (1.3 and 1.6 respectively). Since ydsv has already been determined for the untreated and HPDL-treated OPC from Fig. 3.6, it is then possible to calculate y fv for untreated and HPDL-treated OPC using equation (3.3). Table 3.3 details the values determined for ydsv and YPSV f°r both the untreated and HPDL-treated OPC. Clearly the HPDL treatment of the surface of the OPC leads to a reduction in the total surface energy ysv, while increasing the polar component of the surface energy y psv, thus improving the action of wetting and adhesion. Such changes in the surface energy of the OPC after HPDL treatment are due to the fact that HPDL treatment of the surface of the OPC results in partial vitrification of the surface; a transition that is known to effect a reduction in jsv and an increase in y Psv [130]. Table 3.3 Measured surface energy values for the OPC before and

after HPDL irradiation

Surface energy component Dispersive component, y dsv (mJ/m2) Polar component, y psv (mJ/m2)

58

Untreated OPC 65.0

HPDL-treated OPC 73.1

3.5

15.6


3.4 The effects of laser radiation on the wettability and bonding characteristics of the AI2O3/SiO2-based composite material 3.4.1 The general effects of laser radiation In order to bond together the oxide compounds that comprise the AÔs/SiOi-based composite material, the amalgamation was mixed with 50wt% diluted sodium silicate solution. Sodium silicate solution (waterglass) is a viscous colourless solution of colloidal sodium silicate. It is a silica-containing aqueous solution that, when combined with other solutions such as the amalgamated oxide compound, forms a gel-like mass of silicate hydrate. Such a mass remains soft and malleable until it is exposed to CC>2 gas, either by means of a gas jet or through contact with the atmosphere [136]. But exposure of the hardened mass to water results in a reversing of the process and the mass returns to a gel-like state. The fact that the AliCVSiOa-based composite material in an un-heated state is hydraulically bonded, as opposed to chemically bonded, combined with the retention of chemical and mechanical water (that is water that is bonded into the materials matrix and additional free water respectively) means that the hardened mass will rehydrate when exposed to water [136, 137]. Heating of the hardened AbOs/SiOibased composite material mass fires the waterglass (similar to that of a ceramic material) [136], increasing its strength and enabling it to withstand water exposure. Thus, heating of the AlaCVSiOi-based composite material is similar in effect to the firing of ceramics, in that the heating causes gradual ceramic 'sintering' of the materials; generally bonding together and stabilizing the substances [136, 138]. As such, exposure of the AliOs/SiOz-based composite material to laser radiation results in rapid heating of the surface, for most materials typically 103-105 °C/s [139], which will lead to such sintering of the A^CVSiC^-based composite material surface with the removal of the pores between the starting particles of the compound, combined with growth together and strong bonding between adjacent particles [140], thereby creating a much more consolidated surface. Indeed, surface roughness measurements revealed that the surface roughness had decreased from an Ra value of 25.85 urn before laser treatment, to 6.27 u,m after laser treatment. Also, it was found that heating of the AÔB/SiOa-based composite material above 100 °C resulted in sufficient pyrochemical changes to prevent any rehydration. Although material ejection was not a typical feature of laser beam interaction with the AÔs/SiOi-based composite material, a loss in mass by the AliOs/SiOi-based composite material was a possibility due to the resulting heat effects of the process. In order to determine any weight loss experienced by the Al2O3/SiO2-based composite material as a result of laser irradiation, a number of samples were stored in a controlled environment for 2 days prior to laser irradiation. The samples were weighed regularly to ensure a constant mass. The samples were treated at various power densities and traverse speeds and then immediately weighed. Figure 3.7 shows the percentage loss of original mass experienced by the A^CVSiCVbased composite material in terms of HPDL power density and traverse speed respectively.

59


Fig. 3.7 Relationship betweenaAUOVSiOz-basedcomposite material percentage of original mass loss and laser power density and traverse speed

As can be seen from Fig. 3.7, the loss in mass experienced by the Al2O3/SiO2-based composite material increases almost proportionately in a linear manner with increasing power density up to approximately 2.1 kW/cm2. After this point the loss in mass can be seen to decrease in terms of the power density. It is reasonable to assume that this indicates that a level of power density saturation has been attained, beyond which further increases in power density have a marginal effect on the loss in mass of the Al2C>3/SiO2-based composite material. In contrast, Fig. 3.7 also shows that as the traverse speed is increased then the loss in mass experienced by the Al2O3/SiO2-based composite material decreases, again in a linear manner. After laser treatment the samples were stored in an uncontrolled environment (open laboratory) and weighed regularly every day for 12 days. The results of these tests are illustrated in Fig. 3.8, which shows clearly that the extent to which the Al2O3/SiO2based composite material regains mass is a function of the density of the energy deposited on its surface. This is perhaps not surprising since an increase in the energy density increases the likelihood of material ejection or porosity formation. Also, as the energy incident upon the A^CVSiCVbased composite material surface increases, the HAZ (Fig. 3.9) will consequently increase in size. The general mass regain experienced by the laser-treated Al2O3/SiO2-based composite material at the various laser power densities is thought to be the result of the rehydration through contact with the air of the HAZ, which comprises unslaked lime. This appears to be a reasonable assumption when one considers that in terms of absolute mass regain, the greatest mass regain occurs with the samples treated with the highest power density. For instance, the Al2O3/SiO2-based composite material sample treated at a power density of 1 kW/cm2 initially reduces in mass to 99.70 per cent of its

60


original mass, but after 12 days has regained 0.13 per cent to be finally 99.83 per cent of its original mass. In contrast, the Al2O3/SiO2-based composite material sample treated at a power density of 3 kW/cm2 initially reduces in mass to 98.46 per cent of its original mass, but after 12 days has regained some 0.51 per cent to be finally 98.97 per cent of its original mass.

Fig. 3.8 Mass regained by Al2O3/SiO2-based composite material after HPDL surface treatment with time (240 mm/min traverse speed)

Figure 3.9 shows the characteristic HAZ produced in the Al2O3/SiO2-based composite material as a result of HPDL interaction. The HAZ was observed to surround the semivitreous irradiated zone both on and below the surface. The HAZ was measured as being typically around 150 um thick both on and below the surface.

Fig. 3.9 Typical optical cross-sectional image of the HAZ produced on the Al2OySiO2-based composite material after HPDL surface treatment

61


Because the main component of the AlaCVSiCVbased composite material is chamotte, which contains a certain amount of CaCOs, on interaction with the HPDL beam the rapid localized surface heating will result in the decomposition of the CaCOs at temperatures between 825 and 950 °C in accordance with the recognized chemical reaction [141]

As can be seen, the breakdown results in unslaked lime (CaO) and CC>2 gas. The COi gas simply enters the atmosphere, while the CaCO rests around the semi-vitreous irradiated zone producing a mark. Indeed, this mark, which is the HAZ, can be observed by using a phenolphthalein indicator followed by water misting. This was possible because phenolphthalein is an indicator that is colourless when in contact with CaO, but appears violet-red in the presence of Ca(OH)2 due to the change in pH. Thus, by applying the phenolphthalein to the laser-irradiated zone and the surrounding area, then water misting the entire area to convert the CaO to Ca(OH)2, it was possible to ascertain the precise nature of the HAZ. According to Czernin [142], the breakdown of CaCOj occurs at a range of temperatures from 400 to 700 °C. This variation is said to be due to differing heating rates and exposure times [143]. The effect of increased heating rates is to 'telescope' reactions together, causing them to occur at higher temperatures. This is due to individual reactions not having sufficient time to reach completion, or equilibrium, before the rapidly rising temperature reaches the initiation temperature of neighbouring higher temperature reactions [144].

Fig. 3.10 Typical SEM surface images of the AljOj/SiCh-based composite material, (a) untreated and (b) HPDL treated From Fig. 3.10(a) it can be seen clearly that before laser treatment the surface of the Al2O3/SiO2-based composite material appears coarse, with individual crystals of the constituent components being clearly discernible. After HPDL treatment [Fig. 3.10(b)] there is more surface ordering, with the surface appearing cellular-dendritic, showing

62


that fusion of the individual particulates has occurred. As will be discussed in more detail later, such a solidification structure is indicative of rapidly solidified microstructures [145]. Figure 3.11 shows a typical cross-sectional view through the A^CVSiC^-based composite material (a) before and (b) after HPDL irradiation. As can be seen, an upper densified layer is clearly visible, with a gradual decrease in the degree of densification as the distance from the surface increases.

Fig. 3.11 Typical SEM cross-sectional view through the AljOs/SiOi-based composite material (a) before laser treatment and (b) after HPDL treatment An XRD analysis of the A^CVSiCVbased composite material surface before and after laser treatment (Fig. 3.12) revealed that, on the whole, the phases present within the laser-treated region were the same, however, their proportions were different. In particular, after HPDL treatment it was not possible to detect any SiC>2 while the AliOs was depleted. But, as the EDX analysis shows (Fig. 3.13), Si and Al were still present in similar proportions on the A^CVSiOi-based composite material surface before and after laser treatment. This indicates that partial laser vitrification of the AlOa/SiOabased composite material surface has occurred due to the fact that these materials are glass-forming elements, and as such, vitrified when irradiated.

63


Fig. 3.12 X-ray diffraction analysis of the AÔs/SiOi-based composite material, (a) before laser treatment and (b) after laser treatment

Fig. 3.13 Energy disperse X-ray analysis of the AUOs/SiOz-based composite material, (a) before HPDL treatment and (b) after HPDL treatment

Thermal analysis in the form of thermogravimetric analysis (TG) and DTA were performed in order to determine fully the thermal characteristics of the AhOa/SiOabased composite material. The analysis was carried out using a fully cured piece (10 x 10 x 4 mm3) of the AhOa/SiOi-based composite material which was crushed into manageable pieces and then fine ground using a pestle and mortar. The powder was subsequently sieved through a 75 ujn mesh and then thoroughly mixed to ensure an even mix. Exactly 30 mg of the homogeneous mix was placed in the platinum crucible within the micro-environment cup and heated in an atmosphere of air at a rate of 10 °C/min. The ground AlaOa/SiOa-based composite material samples were examined by simultaneous TG-DTA using a Rheometric Scientific STA 1500.

64


Table 3.4 Summary details of the TG-DTA findings on the A^CVSiCh-based composite material Observation Double endothermic peak

Event Loss in mass of 7.55% Removal of the surface water

267 °C

Endothermic peak

Loss in mass of 0.95% Driving out of the water hydrates

349 °C

Endothermic peak

366 °C

Endothermic peak

Region I

Temperature 80 °C 104 °C

II

III

IV 417 °C

Sharp endothermic peak

662 °C

Endothermic peak

1150°C

Irregular cluster of peaks

V

Loss in mass of 1.34% Decomposition of the formed Mg3(OH)2Si4010 Formation of the vitreous silica according to 2MgO3Si -» Mg2O4Si + SiO2 Loss in mass of 1 .07% Melting of Zn Loss in mass of 0.72% Fe2O3 a-/3 solid state phase transformation or, crystallization of the Na2O.2SiO2 within the waterglass Final fusion of the composite

The results obtained from the TG-DTA are summarized in Table 3.4. The AliCVSiChbased composite material displayed five overlapping, yet distinct regions of mass loss. The first stage (I) revealed that up to 236 °C, the AlaCVSiOa-based composite material experienced a relatively large mass loss of 7.55 per cent. This loss in mass was accompanied by a double endothermic peak at 80 and 104 °C. These effects are attributable to the removal of the surface water from the A^CVSiCVbased composite material. Similarly, within the second region (II) a relatively small mass loss of 0.95 per cent up to 342 °C occurred. This was associated with an endothermic peak at 267 °C which can be attributed to the driving out of the water hydrates from the various Al2O3/SiC>2-based composite material constituents. In the third region (EH), at temperatures up to 411 °C, the loss in mass experienced by the Al2CVSiO2-based composite material was 1.34 per cent. This mass loss was accompanied by a series of closely situated endothermic peaks. However, only a peak at 349 °C could be assigned with any certainty to a known occurrence; namely the decomposition of the formed Mg3(OH)2Si4Oio. Nevertheless, it is reasonable to attribute a peak observed at 366 °C to the formation of vitreous silica according to

A loss in mass of 1.07 per cent was experienced by the A^CVSiCVbased composite material up to 648 °C (IV), being accompanied by a single sharp endothermic peak at 417 °C. This peak can certainly be attributed to the melting of the Zn component. Since no peak is visible corresponding to the boiling point of Zn (907 °C), it is therefore reasonable to assume that the Zn remains in the elemental form and simply reacts with

65


some other constituents of the AbCVSiOa-based composite material. A relatively small mass loss of 0.72 per cent was seen to occur up to 1200 °C (V). This was accompanied by an endothermic peak at 662 °C which may possibly be associated with the FeiOs «-/? solid state phase transformation. Yet it is more probably due to the crystallization phenomena of the Na2O.2SiC>2 within the waterglass. Additionally, at around 1150 °C an irregular cluster of endothermic peaks can be seen. These can be reasonably attributed to the final fusion of the A^CVSiCVbased composite material, since on examination of the sample after the experiment, the sample appeared to be partially vitrified.

3.4.2 Wettability and surface energy characteristics Prior to laser treatment of the AhCVSiC^-based composite material surface it was not possible to fire the enamel on to the surface of the A^CVSiOi-based composite material. This was found to be due to the fact that the contact angle between the enamel and the untreated A^CVSiOi-based composite material surface was measured as being 118 degrees, consequently preventing the enamel from wetting the AÔa/SiCh-based composite material surface. An optical micrograph of a sessile drop of enamel (20 °C) placed on the surface of the AÔa/SiC^-based composite material before and after HPDL irradiation with the contact angle superimposed is shown in Fig. 3.14.

Fig. 3.14 Contact angles for the enamel on (a) the untreated surface of the AliOs/SiOi-based composite material and (b) HPDL-treated surface of the Al2O3/SiO2-based composite material Figure 3.15 shows the measured contact angles between the enamel and the surface of the AbOs/SiOi-based composite material before and after interaction with the selected industrial lasers. As is clearly evident from Fig. 3.15, under the experimental laser parameters, interaction with the CC>2 laser, the Nd:YAG laser, and the HPDL beams resulted in the contact angle between the enamel and the A^CVSiC^-based composite material reducing from 118 degrees to 31, 34, and 33 degrees respectively. In contrast, interaction of the AhOa/SiCVbased composite material with excimer laser radiation effected an increase in the contact angle to 121 degrees. Similarly, as Table 3.5 shows, with all the control liquids used the AhOa/SiC^-based composite material experienced a significant reduction in contact angle as a result of interaction with the CC>2 laser, the Nd:YAG laser, and the HPDL beams, while interaction with the excimer laser beam again resulted in an increase in the contact angle.

66


Fig. 3.15 Mean values of contact angles formed between the enamel and the AliOySiCh-based composite material before and after interaction with the selected lasers Table 3.5 Mean values of contact angles formed between the selected test liquids at 20 °C and the AÔs/SiOi-based composite material before and after interaction with the selected lasers

Liquid Human blood Human blood plasma Glycerol 4-Octanol

Untreated 61 64 34 29

Contact angle, 9 (deg) HPDL CO2 Nd:YAG 34 35 27 24

38 39 29 27

37 38 28 26

Excimer 74 84 57 44

3.4.2.1 Variations in surface roughness characteristics The changes in the surface roughness of the AÔs/SiOa-based composite material occasioned as a result of laser interaction are shown in Fig. 3.16. It can be seen clearly from Fig. 3.16 that considerable reductions in the surface roughness of the AhOa/SiC^based composite material were effected after interaction with the COi laser, the Nd:YAG laser, and the HPDL beams, reducing from an initial Ra value of 25.85 (im to 5.88, 6.56, and 6.27 |im respectively. In contrast, interaction of the AlzOa/SiC^-based composite material with excimer laser radiation resulted in a roughening of the AlaOs/SiCh-based composite material surface, causing the surface roughness to increase to an Ra value of 36.22 |im.

67


Fig. 3.16 Mean values of surface roughness on the Al2O3/SiO2-based composite material before and after interaction with the selected lasers

Fig. 3.17 Relationship between HPDL-treated AUOa/SiOi-based composite material contact angle and surface roughness (Ra)

Figure 3.17 shows the effect of the surface roughness of the AliCVSiCVbased composite material when treated with the HPDL on the contact angle. For experimental purposes the liquid used was glycerol. As can be seen from Fig. 3.17, variations in the surface roughness of the HPDL-treated AliCVSiCVbased composite material had a small but discernible effect on the measured contact angle, thus confirming that the surface roughness of the laser-treated A^CVSiOa-based composite material plays a significant role in the wettability performance of the Al2O3/SiO2-based composite material. Such results are in accord with those obtained by Feng et al. [146], who noted that contact angle was inversely proportional to surface roughness.

68


The smoothing effects of COi laser, Nd:YAG laser, and HPDL irradiation on the surface of the AÔs/SiCVbased composite material in comparison with the roughening effects of excimer laser irradiation are clearly discernible from Fig. 3.18.

Fig. 3.18 Typical SEM surface images of the Al2O3/SiO2-based composite material (a) before laser treatment and after laser interaction with (b) CO2 laser, (c) NdrYAG laser, (d) HPDL, and (e) excimer laser radiation

69


Figure 3.19 clearly shows that the surface condition of the HPDL-treated AlaC^/SiOibased composite material has a significant effect on the contact angle. In particular, Fig. 3.19 indicates that the melting, and therefore the partial vitrification of the glassforming elements (SiC>2 and AlaOa) within the AliCVSiOa-based composite material, is an essential prerequisite in order for significant reductions in the contact angle to be realized. As one can see, after the onset of melting at around 0.9 kW/cm2 the contact angle reduces sharply from 113 to 33 degrees, after which no further decrease (or increase) is discernible until the power density exceeds 2.1 kW/cm2. After this point a small increase in the contact angle from 33 to 39 degrees was observed. This is probably due to the fact that this power density level lies outside the optimum operating conditions for the traverse speed used in the experiments [112] and, as such, will cause an increase in the surface roughness. This increase, as was seen earlier, has a small but discernible effect on the contact angle. The mere reordering of the crystals that occur at power densities below 0.9 kW/cm2 appears to have only a slight effect on the contact angle, reducing it from 118 to 113 degrees. Indeed, work conducted by Zhang et al. [147] found that significant improvements in the bond strength of a Si3N4 ceramic could be realized only when excimer laser treatment of a structural alloy steel (SAE 4340) resulted in surface melting. Nevertheless, such a reduction in the contact angle reveals that laser interaction without the incidence of melting does affect slightly the wettability characteristics of the AÔs/SiCVbased composite material.

Fig. 3.19 Relationship between enamel contact angle on HPDL-treated Al2O3/SiO2-based composite material and power density

3.4.2.2 Variations in surface O2 content The O2 content of a material's surface is most certainly an influential factor affecting the wetting performance of the material [97, 148]. Wetting is governed by the first atomic layers of the surface of a material. So, in order to determine element content of 02 at the surface of the A^CVSiCVbased composite material, it was necessary to examine the surface using XPS.

70


As Fig. 3.20 shows, significant differences in the surface C>2 content of the AÔs/SiCVbased composite material after interaction with all the selected lasers were observed. Increases in the surface C>2 content were experienced by the A^CVSiCV based composite material after interaction with COa laser, the Nd:YAG laser, and the HPDL beams, increasing from an initial value of 37.6at% to 43.1at%, 41.4at% and 42.8at% respectively. Conversely, interaction of the A^CVSiCVbased composite material with excimer laser radiation resulted in the surface O2 content of the Al2O3/SiO2-based composite material decreasing slightly to 33.4at%.

Fig. 3.20 Al2O3/SiO2-based composite material surface O2 content before and after interaction with the selected lasers

3.4.2.3 Surface energy and the dispersive/polar characteristics As was shown in Chapter 2, it is possible to estimate reasonably accurately the dispersive component of the AbOa/SiCVbased composite material surface energy, y dsv, by plotting the graph of cos 6 against (yf v ) 1/2 /y/ v in accordance with equation (2.18). Figure 3.21 shows the best-fit plot of cos 9 against (ydlv)l/2/yiv for the untreated and laser-treated AliOs/SiOa-based composite material-experimental control liquids system. A comparison of the ordinate intercept points of the untreated and laser-treated Al2O3/SiO2-based composite material-liquid systems, shown in Fig. 3.21, reveals that for the untreated and excimer laser-treated AlaCVSiC^-based composite materialliquid systems, the best-fit straight line intercepts the ordinate relatively close to the origin. On the other hand, Fig. 3.21 shows that the best-fit straight line for the Al2O3/SiO2-based composite material-liquid systems of the COa, Nd:YAG, and HPDL-treated samples intercept the ordinate considerably higher above the origin. This is significant since interception of the ordinate close to the origin is characteristic of the dominance of dispersion forces acting at the Al2O3/SiO2-based composite material-liquid interfaces of the untreated and excimer laser-treated samples, resulting in poor adhesion [87, 135]. On the other hand, an interception of the ordinate well

71


above the origin is indicative of the action of polar forces across the interface, in addition to dispersion forces, hence improved wettability and adhesion is promoted [87, 135]. 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 A^CVSiCVbased composite material surface (untreated and lasertreated) did not occur [87, 135].

/ d j\lfl /y for the A^CVSiOi-based composite material in Fig. 3.21 Plot of cos 6 against (y* v /v contact with the wetting test control liquids, before and after laser treatment

As was shown in Chapter 2, in order to determine the polar component of the A^CVSiCVbased composite material surface energy, ypsv, it is necessary to calculate the dispersive component of the work of adhesion, W^, by using equation (2.19). Both Wad and W^d are related by the straight line relationship represented by equation (2.20). Thus, from the best-fit straight line plots of Wad against w£ for the Al2O3/SiO2-based composite material when it is both untreated and laser treated it is possible to determine the constants a and b for each separate condition of the Al2O3/SiO2-based composite material (Table 3.6).

72


Table 3.6 Values determined for the constants a and b from the plots of Wad against W*d for the AÔs/SiOi-based composite material, before and after laser treatment b (mj/m2)

a 1.22 1.94 1.86 1.71 1.12

Untreated CO2 laser Nd:YAG laser HPDL Excimer laser

11.3 -17.1 -21.5 -18.3 13.2

Since a linear relationship exists between the dispersive and polar components of the control test liquids' surface energies [112] which satisfies equation (2.20), then, as was shown previously, it is possible to calculate 7 psv directly for untreated and laser-treated AliCVSiOi-based composite material using equation (3.2) with the appropriate values of a given in Table 3.6. Table 3.7 gives the values determined for 7 dsv and 7 psv, as well as the total surface energy for both the untreated and laser-treated AÔs/SiC^-based composite material. Table 3.7 Determined surface energy values for the AUCVSiOi-based composite material before and after laser treatment

Surface energy component Dispersive component,

7 dsv

Untreated 84.2 (mJ/m )

Polar component, 7 psv (mJ/m2)

2

2.0

Condition CO2 Nd:YAG HPDL Excimer 90.7 76.9 86.6 89.0 36.8

36.2

25.9

0.3

As can be seen from Table 3.7, CC>2 laser, Nd:YAG laser, and HPDL treatment of the surface of the A^CVSiCVbased composite material resulted in an increase in the polar component of the surface energy, 7 psv. As was discussed earlier, such increases in the surface energy of the A^CVSiC^-based composite material, in particular the increase in 7 psv, have a positive effect upon the action of wetting and adhesion. Again, these changes in the surface energy of the AlaOs/SiOa-based composite material after treatment with these lasers are primarily due to the fact that the treatment of the AÔs/SiOa-based composite material surface results in the partial vitrification of the surface; a transition which is known to effect an increase in y£, [130], consequently causing a decrease in the contact angle. 3.4.2.4 Discussion of the laser effected wettability characteristics modification The results detailed previously show clearly that interaction of the AlaCVSiOi-based composite material with selected industrial lasers has resulted in the contact angle formed between the enamel and the control liquids altering to various degrees depending upon the laser type. Under the selected experimental laser operating parameters, interaction of the A^CVSiOi-based composite material typically with the

73


COi laser, the Nd:YAG laser, and the HPDL beams resulted in a decrease of similar proportion in the contact angle, while interaction of the A^CVSiCVbased composite material with excimer laser radiation resulted in an increase in the contact angle. Such changes in the value of the contact angle are influenced, depending upon the laser used, primarily by: 1. Surface melting and partial vitrification. Laser-induced melting and vitrification results in the occurrence of two main changes in the surface condition of the AliOs/SiC^-based composite material. These are: (i) Surface smoothing resulting from the laser melting of the A^CVSiC^-based composite material surface which consequently results in a reduction of the surface roughness, thus directly reducing the contact angle, 9. (ii) Increase in the polar component, ypsv, of the surface energy resulting from the melting and partial laser vitrification of the glass-forming elements within the AlaCVSiC^-based composite material composition, thus improving the action of wetting and adhesion by generating a surface with a more vitreous surface microstructure. 2. Surface roughening. An increase in the AliOs/SiCVbased composite material surface roughness resulting from laser ablation of the AliOa/SiCVbased composite material surface in turn results directly in an increase in the contact angle, 6. 3. Surface #2 content. An increase in the surface 62 content of the A^CVSiCVbased composite material resulting from laser treatment is an influential factor in the promotion of the action of wetting, since an increase in surface 62 content inherently effects a decrease in the contact angle, and vice versa. It is highly likely that the resultant contact angle between the AÔa/SiC^-based composite material-enamel systems and the A^CVSiCVbased composite materialliquid systems of the CO2, Nd:YAG, and the HPDL-treated samples are all similar in value due to the fact that interaction with these lasers caused surface melting (see Fig. 3.14), resulting in a significantly smoother surface. This, combined with the fact that vitrification of the A^CVSiCVbased composite material surface results in an increase in YsV as a result of the surface becoming less crystalline in nature, thus promoting wetting, would influence a reduction in the contact angle. From Table 3.7 it can be seen that interaction of CO2, Nd:YAG, and the HPDL radiation with the Al2O3/SiO2-based composite material resulted in similar increases in y Psv • However, absorption of CC>2 radiation by the Al2O3/SiO2-based composite material is higher than that of the Nd: YAG or the HPDL [112] and, since contact angle reduction is a function of surface melting and vitrification [112], then surface melting and vitrification will occur to a greater extent with the CC>2 laser, thus resulting in a marginally greater decrease in the contact angle. In contrast, as Fig. 3.18(e) shows, interaction of the Al2O3/SiO2-based composite material with excimer laser radiation did not cause melting of the surface, but instead induced surface ablation, which consequently resulted in a slightly rougher surface. Thus an increase in the contact angle was effected. Additionally, Kokai et al. [149] have concluded that, with excimer laser parameters which are conducive to the

74


production of plasma, as was the case with the AÔa/SiC^-based composite material, then the surface roughness is increased as a result of plasma-induced debris redepositing on the surface and excessive thermally induced surface fractures and porosities. Clearly, since plasma generation was observed, then surface roughening after excimer laser irradiation was perhaps to be expected. Yet both Liu et al. [150] and Nicolas et al. [151] have reported that irradiating ZrO2 with excimer laser radiation with energy densities in excess of 2.7 J/cm2, resulted in a reduction in surface roughness. Such reductions were attributed to the fact that at these levels of energy density, melting of the ZrO2 surface occurred. It was found that the depth of the laser melting, and in the case of the excimer laser, the ablation region, varied significantly according to laser type. Table 3.8 shows the differences in laser melt/ablation depth determined for each laser by means of crosssectional SEM analysis. Table 3.8 Determined laser melt/ablation depths for the Al2O3/SiO2-based composite material after laser irradiation Laser Laser melt/ablation depth

CO2

Nd:YAG

HPDL

Excimer

2 10 urn

50 urn

125 urn

30 urn

As is evident from Table 3.8, the differences in laser melt/ablation depth obtained with the Nd:YAG and excimer lasers was an order of magnitude smaller than those of the CO2 laser or the HPDL. The main reason for these large differences thought to be the pulsed nature of the beams of the Nd:YAG and excimer lasers, as opposed to the CW nature of the CC>2 laser or HPDL beams. Since the interaction time of a pulsed beam with a material is much shorter than that of a CW beam, the depth of the laser melt/ablation region will be much smaller due to the reduced time afforded for heat transfer. It is also very important to consider the surface 62 content of the A^CVSiC^-based composite material before and after treatment with the selected lasers. Increases in the surface C>2 content were experienced by the AÔs/SiCVbased composite material after interaction with the CO2 laser, the Nd:YAG laser, and the HPDL beams, while interaction of the AÔs/SiC^-based composite material with excimer laser radiation resulted in the surface 62 content of the AÔySiC^-based composite material decreasing. Such a result is in agreement with the findings of a number of workers [152, 153], who have noted that for many ceramic materials, irradiation with an excimer laser beam creates defective energy levels, in particular the formation of O2 vacancies. Since roughening of the surface does not necessarily create a surface with a more crystalline structure, it is reasonable to assume that the increase in the surface roughness after excimer laser irradiation, along with the associated reduction on the surface 62 content, are the principal reasons for the observed decrease in the contact angle.

75


3.4.3High-power diode laser-induced solidification microstructures and the resulting effects on wettability characteristics Variations in the HPDL operating parameters (power density and traverse speed) were seen to affect significantly the microstructures obtained within the laser-treated areas on the Al2O3/SiO2-based composite material. The product of power density and interaction time, which can be derived from the beam spot diameter and the traverse speed, yields the specific energy delivered to the surface. As such, it is possible to group the lasergenerated microstructures into those produced with high, medium, and low specific energies; since the microstructures observed within each region were reasonably typical across the group. Figure 3.22 shows the laser parameters that were found to produce microstructural changes in the AÔs/SiOa-based composite material, with the specific energy groups superimposed. Obviously laser-induced microstructures will be generated outside the selected energy density regions, but, for simplicity, these were not considered in the analysis since they do not lie within the optimum operating parameters.

Fig. 3.22 Schematic diagram of the melting and solidification region for HPDL surface treatment of the AUCVSiCh-based composite material Figure 3.23 shows a typical example of the microstructures obtained (a) in the centre and (b) on the edge of a HPDL-treated track on the AlaCVSiOa-based composite material surface, when using a relatively high specific energy (>600 J/cm2). As can be seen from both Fig. 3.23(a) and (b), the microstructures appear to be indicative of rapid solidification and are dendritic across the entire width of the laser-treated track. However, on the edge of the track the dendrites appear much finer and elongated. In both instances the structures appear ordered in orientation. Figure 3.24 shows a typical example of the microstructures obtained (a) in the centre and (b) on the edge of a HPDL-treated track on the AlaCVSiOi-based composite material surface under relatively medium specific energy conditions (500-600 J/cm2).

76


Figure 3.24(a) shows that a semi-ordered dendritic structure was observed in the centre of the laser track, while in contrast, Fig. 3.24(b) shows a semi-ordered and much finer elongated cell structure present on the edge of the solidification track. Moreover, both Fig. 3.24(a) and (b) display a microstructure typical of rapid solidification. The very light grey area visible in various parts across Fig. 3.24(b) is believed to be spherulites. Under lower magnification the semi-directional structure observed in Fig. 3.24(a) and (b) was seen typically to extend in a perpendicular direction from the edges of the laser melt track, tending inwards towards the centre of the track.

Fig. 3.23 Typical SEM surface images of the microstructures on the AI2O3/SiO2-based composite material with relatively high specific energy, (a) centre of the track (b) edge of the track

Fig. 3.24 Typical SEM surface images of the microstructures on the AlaOa/SiCh-based composite material with relatively medium specific energy, (a) centre of the track (b) edge of the track

77


From Figs 3.23 and 3.24 it can be seen that the solidification microstructures obtained differ not only with changes in specific energy, but even across the same track. Such differences in microstructure within the same track result from the fact that at the edge of the melt track the solidification rate, R, is low while the thermal gradient, G, is at its steepest, therefore GIR is high along the fusion line of the melt pool. Towards the centre of the melt zone the solidification rate is increased while the thermal gradient is reduced. Consequently GIR rapidly falls off as solidification proceeds towards the centre of the melt zone. This is the result of the high GIR ratio at the interface being just slightly less than that required for stability, that is the degree of constitutional supercooling is smaller, thus different microstructures at the edges of the solidification melt tracks can be formed. At the centre, however, the GIR ratio is comparatively smaller, which, in the instances where high and medium specific energies were used, allows dendritic structures to be formed within the AÔs/SiC^-based composite material. But, as Fig. 3.23 shows, a dendritic structure was observed in the high specific energy solidification track both in the centre (a) and at the edge (b), although the dendrites at the edge appeared much finer and elongated. It is proposed that under the conditions of high specific energy, the GIR ratio in both regions was small enough to ensure that a dendritic microstructure prevailed. In contrast, as Fig. 3.24 shows, with medium specific energy a dendritic structure was formed only in the centre (a) of the solidification track, while on the edge (b) a finer elongated, cell structure was observed. From this it would appear that the initial GIR ratio at the edge is high and rapidly diminishes towards the centre, thus, the much finer and elongated, oriented structure is formed at the edge, while dendrites are produced in the central area. Another possible reason for the formation of the fine, elongated structures at the edges of the solidification tracks seen in Figs 3.23(b) and 3.24(b), is the fact that, although the HPDL beam is not truly Gaussian in nature, the power intensity profile of the beam produces a temperature gradient perpendicular to the direction of traverse [154]. As such, the cooling rate, t (= GR), of the Al2O3/SiO2-based composite material will be much faster on the edge of the laser track than in the centre, and may therefore give rise to the much finer and elongated microstructures observed on the edges of the laser tracks in Figs 3.23(b) and 3.24(b). Indeed, such findings have been reported by a number of workers conducting research into the laser treatment of various ceramics and alloys. Pei et al. [155] noted that both equiaxed and dendritic microstructures were obtained in different regions of the same laser clad ZrOa layer, concluding that the differences were related to different cooling rates in the various regions of the laser clad ZrO2 layer. Similar results were obtained by Liu [156] after laser sealing Y2Oa-ZrO2 and MgO-ZrOa ceramic coatings. Shih et al. [157] observed that across a YBa2Cu3Ox and laser clad track different microstructures were found in different regions, as did Shieh et al. [158] across a SiC^-AÔs laser clad track. Furthermore, both workers noted that not only were cellular and dendritic microstructures visible, but also that the microstructures were much finer on the edge of the clad track than in the centre. Such differences in microstructure type and size were ascribed to the varying degrees of constitutional supercooling, which, according to McCallum et al. [159], are inherent in laser processes.

78


The directionally solidified nature of the microstructures observed in Figs 3.23(b) and 3.24(b), which were seen to extend in a perpendicular direction from the edges of the laser melt track, tending inwards towards the centre of the track, is thought to be due, again, to the fact that the HPDL beam intensity is at a maximum in the centre of the profile. Because of this, the highest temperatures occur in the centre of the track with the lowest temperatures being experienced on the edge of the laser track. Consequently solidification begins at the edges of the track and develops quickly inwards as the laser beam is traversed away. Such observations of directional solidification are in accord with those of Song and Netravali, and Bradley et al. [20] during CC>2 laser and HPDL treatment of AliOs-based refractory materials. It is clear that differences in the HPDL operating parameters produce very different microstructures in the A^CVSiC^-based composite material. Yet, despite this there was no discernible difference in the wetting characteristics of the Al2O3/SiO2-based composite material after laser treatment, both with high and medium energy densities. Indeed, as can be seen clearly from Fig. 3.19, the surface condition of the laser-treated Al2O3/SiO2-based composite material has a significant effect on the contact angle. In particular, Fig. 3.19 indicates that the melting and resolidification [i.e. the partial vitrification of the glass-forming elements (SiC>2 and A12O3) within the A^CVSiCV based composite material] is an essential prerequisite in order for significant reductions in the contact angle to be realized. 3.4.4 Bonding mechanisms between the high-power diode laser-treated AI2O3/SiO2-based composite material and enamel 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 (van der Waals). In chemical wetting adhesion is achieved as a result of reactions occurring between the mating surfaces, giving rise to chemical bonds [91]. In either case, the driving force for wetting is the reduction of the surface free energy of the solid AÔs/SiC^-based composite material by the liquid enamel (YSV - Ysi)- 1° practice, complex combinations of various bonding mechanisms actually occur, varying according to the types of materials used [91]. For the Al2O3/SiO2-based composite material and the enamel, the mechanisms involved in ceramic-glass bonding are reasonably applicable. These principally include physical bonding (van der Waals forces), chemical bonding (oxide transformation and 62 bridging) and, on a very small scale, electrochemical reactions such as the electrolytic effect due to the presence of ferric oxides within the Al2O3/SiO2-based composite material reacting with other oxides in the enamel [91]. In the particular case of the AlaCVSiC^-based composite material and the enamel, the bonding mechanism is principally the result of physical forces. This is because adhesion between many materials is assured by electron transfer and is therefore related to bandgap energy [100, 148]. Thus, for non-conducting materials, such as the A^CVSiCh-based composite material, with large bandgaps, there will be practically no free charges inside the ceramic crystals, even at elevated temperatures. In this case the electron transfer at the interface will not take place since the electron transfer depends

79


exclusively on the concentration of free charges in the ceramic crystal [100]. As a result, the chemical contribution to the work of adhesion is negligible since Wa2 laser, the Nd:YAG laser, and the HPDL, the wavelengths of which vary by more than an order of magnitude. Nonetheless, the work has shown that under the chosen experimental laser operating parameters, changes in the wettability characteristics of the A^CVSiOi-based composite material were seen to vary depending upon the laser type - in particular, whether the laser radiation had the propensity to cause surface melting.

82

Chapter 4

High-Power Diode Laser Modification of Selected Engineering Ceramic Materials for Improved Wettability Characteristics

This chapter examines the effects of HPDL radiation on a number of common engineering ceramics, with the aim of developing a possible generic approach to the laser modification of the wettability characteristics of materials. It was observed that HPDL radiation effected significant changes in the wettability characteristics of the selected engineering ceramic materials. The work detailed in this chapter reveals that the wettability characteristics of the engineering ceramic materials could be controlled and/or modified with HPDL surface treatment.

4.1 Introduction In terms of wettability characteristics modification, cognition of the influence that solid substrate surface conditions have on the wettability characteristics of a material are of great importance. In particular, factors such as surface morphology, surface microstructure, surface composition, and surface energy are pivotal in effecting changes to the wettability characteristics of a material. At present, very little published work exists regarding the effects of laser radiation, let alone HPDL radiation, on the wettability characteristics of ceramic materials. Indeed, the published work is predominantly concerned with the use of excimer laser radiation. Song and Netravali have carried out comprehensive research into the surface modification of ultra-high-strength polythene fibres using excimer laser radiation (308 nm) for enhanced adhesion, investigating the effects of laser operating parameters on the surface and adhesion characteristics [160] and the effects of the treatment environment [161]. Similarly, Kappel [162] has shown that the texturing of ceramics

83


with an excimer laser 248 nm) can improve the adhesion strength by up to 20 per cent. Such an improvement is said to be due to the formation of raised microscopic protrusions over the surface. This chapter examines the effects of HPDL radiation on a number of common engineering ceramics, with the aim of developing a possible generic approach to the laser modification of the wettability characteristics of materials.

4.2 Experimental procedures 4.2.1 Preparation of materials In order to investigate the HPDL modification of the wettability characteristics of a number of selected ceramic materials, the solid ceramic materials were cut into squares (10 x 10 mm2 with a thickness of 3 mm) for experimental purposes. The materials selected were common engineering ceramics: an SiOi/AÔs-based ceramic (ceramic tile), an SiO2/Al2O3/Fe2O3-based ceramic (clay quarry tile), AÔs and SiO2-TiO2 (crystalline). To ensure that the surfaces of the selected engineering ceramics were free from contamination the contact surfaces were cleaned in an ultrasonic bath. 4.2.2 Laser processing procedure The general laser processing experimental arrangement is very similar to that described previously in Chapter 3, in which the defocused beam of the HPDL was fired across the surfaces of the ceramic materials by traversing the samples beneath the laser beam using the x- and y-axis of the CNC table at speeds of 5-8 mm/s, while 3 1/min of coaxially blown C>2 assist gas was used to shield the laser optics.

4.3 High-power diode laser modification of the wettability characteristics of selected engineering ceramic materials 4.3.1 Contact angle and surface energy analysis procedure In order to investigate the effects of HPDL radiation on the wetting and surface energy characteristics of the selected engineering ceramic materials, control experiments were conducted using the sessile drop technique with a variety of test liquids with known surface energy properties. In this manner it was possible to quantify any surface energy changes in the ceramics resulting from HPDL interaction. The sessile drop control experiments were carried out as in Chapter 3, using the test liquids detailed in Table 3.2. The contact angles were subsequently determined using the same technique as described in Chapter 2. As in Chapter 2, the experimental results showed that throughout the period of the tests no discernible change in the magnitude of the contact angle occurred, indicating that thermodynamic equilibrium was established at the solid-liquid interface at the outset of the experiments. 4.3.2Effects of high-power diode laser radiation on contact angle characteristics As can be seen from Table 4.1, HPDL irradiation of the ceramic materials' surfaces resulted in all the materials displaying a reduction in the contact angle with all the control liquids used.

84

HPDL Modification of Engineering Ceramic Materials for Improved Wettability Characteristics

Table 4.1 Mean values of contact angles formed between the selected test liquids at 20 °C and the selected engineering ceramic materials before (UT) and after (LT) HPDL irradiation

Substrate Ceramic tile Clay tile A1a203 SiO2-TiO2 (cryst)

Blood UT LT 69 31 73 47 76 61 57 39

Contact angle, 6 (deg) Plasma Glycerol UT LT UT LT 32 41 18 71 33 76 49 57 55 51 78 63 36 28 61 41

4-octanol UT LT 35 16 54 29 51 49 31 26

4.3.2.1 Variations in surface roughness characteristics As has already been established, contact angle is greatly influenced by surface roughness, with reductions in the contact angle being realized as a result of the surface of a material being smoother. Indeed, as Fig. 4.1 shows, considerable reductions in the roughness of the surfaces of the selected engineering ceramics were obtained after laser treatment. Further, the typical resultant smoothing effects of HPDL irradiation on the surface of a clay quarry tile are discernible from Fig. 4.2.

Fig. 4.1 Mean values of surface roughness on the selected engineering ceramics before and after HPDL interaction

85


Fig. 4.2 Typical optical surface image of a clay quarry tile before and after HPDL treatment

4.3.2.2 Variations in surface O2 content Previously in Chapter 3, the 62 content of the AÔs/SiCVbased composite material surface was observed to play an influential role in the wetting performance of the Al2O3/SiO2-based composite material. Again, as in Chapter 2, the untreated and HPDL surface-treated selected engineering ceramic samples were examined on the surface using XPS. Figure 4.3 shows that differences in the surface C«2 content of the selected engineering ceramic materials resulted from interaction with all the selected lasers. As one can see from Fig. 4.3, increases in the surface C>2 content were experienced by all the selected engineering ceramics after interaction with the HPDL beam.

Fig.

86

4.3 Surface content of the selected engineering ceramics before and after interaction with the HPDL


Such observed increases in the surface O2 content of the selected engineering ceramic materials after HPDL irradiation are borne out somewhat by Fig. 4.2. As was discussed in Chapter 3, distinct changes in the colour of the surface of the clay quarry tile (which was typical of all the selected engineering ceramic materials), which were observed after HPDL treatment, can be largely attributed to the resultant phase transitions and to the presence in small concentrations of metal transition ions in various oxidation states within the ceramic composition. Of particular importance are ferric ions in the Fe3+ and Fe2+ oxidation state. In the case of the clay quarry tile it appears that both phases are present within the composition. The colour is therefore determined by the Fe3+/Fe2+ ion ratio, resulting in the black colour evident in Fig. 4.2. 4.3.3 Surface energy and the dispersive/polar characteristics As before, to estimate the dispersive component of the selected engineering ceramics' surface energy, yd, the graph of cos 6 against (ydlv )1/2/Y(v was plotted in accordance with equation (2.18), whereby the value of ydsv ^s estimated by the gradient [= 2(y f v ) 1/2 ] of the line which connects the origin (cos 9 = -1) with the intercept point of the straight line (cos 9 against (y^ v ) 1/2 /Y/v) correlating the data point with the abscissa at cos 6=1. Figure 4.4 shows the best-fit plot of cos 9 against (ydlv )1/2/Y(v for the untreated and laser-treated ceramic materials-experimental control liquids system. Contrasting the ordinate intercept points of the untreated and HPDL-treated selected engineering ceramic-liquid systems shown in Fig. 4.4 reveals that, in general, the bestfit straight lines for the untreated selected engineering ceramics intercept the ordinate relatively close to the origin. While conversely, the best-fit straight line for the HPDLtreated selected engineering ceramic-liquid systems intercept the ordinate considerably higher above the origin. Again, as was discussed previously, interception of the ordinate close to the origin is characteristic of the dominance of dispersion forces acting at the solid-liquid interfaces of the untreated selected engineering ceramics [87, 135]. Interception of the ordinate well above the origin for the HPDL-treated selected engineering ceramics indicates that polar forces across the interface, in addition to dispersion forces, are active, thus improved wettability and adhesion is promoted [87, 135]. Also, as in Chapter 3, since none of the best-fit straight lines for either the untreated or the HPDL-treated selected engineering ceramics intercept below the origin, then it can be said that the development of an equilibrium film pressure of adsorbed vapour on any of the selected engineering ceramic surfaces did not occur.

87


/

, \l/2

Fig. 4.4 Plot of cos 0 against \vlv ) /Yiv f°r the selected engineering ceramics in contact with the wetting test control liquids, before and after HPDL treatment In order to determine the polar component of the selected engineering ceramic materials' surface energy, ypsv, it is necessary to calculate the dispersive component of the work of adhesion, W^,, by using equation (2.19). Table 4.2 gives the values calculated for Wad and W^ for the selected engineering ceramic materials. Table 4.2 Calculated values of W^ and W^ for the selected engineering ceramics, before (UT) and after (LT) HPDL treatment Dispersive work of adhesion ( wd )

Work of adhesion (Wad) Engineering ceramic condition Ceramic tile (UT) Ceramic tile (LT)

Blood

Plasma

Glyc.

4-Oct.

Blood

Plasma

Glyc.

4-Oct.

64.5 88.2

66.9 93.3

112.2 123.7

50.0 53.9

58.2 59.2

59.8 58.7

106.5 108.7

47.6 48.6

Clay tile (UT) Clay tile (LT)

61.4 79.9

62.7 83.6

97.9 116.6

43.7 51.6

53.9 54.8

54.4 55.3

98.9 100.5

44.2 45.0

A12O3 (UT) A12O3 (LT)

59.0 70.5

60.9 79.5

99.8 103.3

44.8 45.5

51.0 52.2

51.5 52.7

93.6 95.8

41.9 42.8

SiO2-TiO2 (UT) SiO2-TiO2 (LT)

73.4 84.4

75.0 88.6

114.7 119.4

51.1 52.2

53.1 58.2

53.5 58.7

97.3 106.7

43.5 47.7

Both Wad and Wd are related by the straight line relationship represented by equation (2.20). Thus, from the best-fit straight line plots of Wad against W^ for the selected

88


engineering ceramic materials when they are both untreated and HPDL treated, it is possible to determine the constants a and b for each separate condition of the selected engineering ceramic material. The values determined for the constants a and b are given in Table 4.3. Table 4.3 Values determined for the constants a and b from the plots of Wad against Wd for the selected engineering ceramics, before (UT) and after (LT) HPDL treatment Engineering ceramic condition Ceramic tile (UT) Ceramic tile (LT)

a 1.44 1.89

b (mj/m2) -0.7 -0.3

Clay quarry tile (UT) Clay quarry tile (LT)

1.15 1.52

-2.9 -4.7

A12O3 (UT) A12O3 (LT)

0.74 1.26

14.82 -3.5

SiO2-TiO2 (cryst.) (UT) SiO2-TiO2 (cryst.) (LT)

1.37 1.72

-6.9 -5.2

By using the appropriate value of a given in Table 4.3 it is possible to calculate directly 7 fv for the untreated and HPDL-treated selected engineering ceramics using equation (3.2) (since the same control liquids are being used as in Chapter 3). Table 4.4 gives the values determined for ydsv for both the untreated and HPDL surfacetreated selected engineering ceramics. Table 4.4 Determined surface energy values for the selected engineering ceramics before (UT) and after (LT) HPDL surface treatment

Surface energy component

Selected engineering ceramic condition Ceramic Clay tile Al2O3 SiO2-TiO2 UT LT UT LT UT LT UT LT

Dispersive component (mJ/m2) Polar component (mJ/m2)

76.9 7.2

79.7 30.2

66.1 68.3 0.4 10.1

59.2 0.0

62.0 1.5

64.0 3.4

76.9 22.3

As can be seen from Table 4.3, HPDL surface treatment of the selected engineering ceramics has led to an overall increase in the total surface energy yiv. What is more, HPDL surface treatment of the selected engineering ceramics has also effected a significant increase in the polar component of the surface energy 7 psv. Again, these changes in the surface energy of the selected engineering ceramics after HPDL treatment are essentially the result of the laser-induced surface melting and partial vitrification of the surface; a transition that is known to effect an increase in 7 psv [130], consequently causing a decrease in the contact angle.

89


4.4 Discussion of high-power diode laser wettability characteristics modification The results presented in Section 4.3 are a clear indication that HPDL surface treatment of the selected engineering ceramics brought about a general reduction in the contact angle formed between the selected engineering ceramics and the control liquids, thereby resulting in a change in the wettability characteristics of the selected engineering ceramics. As was discussed in Chapter 3, such observed changes in the value of the contact angle were brought about primarily by: 1. Surface melting and partial vitrification. HPDL-induced melting and vitrification of the glass-forming and glass-modifying compounds in the selected engineering ceramics caused two main changes in the surface condition. These were: (i) Surface smoothing resulting from HPDL melting of the selected engineering ceramics' surfaces which consequently resulted in a generally significant reduction of the surface roughness, thus directly reducing the contact angle, 6. (ii) Increase in the polar component, ypsv, of the surface energy resulting from the melting and partial laser vitrification of the glass-forming elements within the selected engineering ceramics' composition, which in turn improves the action of wetting and adhesion by generating a surface with a more vitreous surface microstructure. 2. Surface O2 content. An increase in the surface 62 content of the selected engineering ceramics resulting from HPDL treatment due to the occurrence of surface oxidation during melting and vitrification. This was an influential factor in the promotion of the action of wetting, since an increase in surface O2 content inherently effects a decrease in the contact angle, and vice versa.

4.5 Summary The results presented in this chapter show that it is possible to alter the wetting characteristics of the selected ceramic materials using the HPDL. Moreover, the findings of this work are a further indication that the use of laser radiation to effect changes in the wetting characteristics of materials is a feasible technique.

90

Chapter 5

Laser Modification of Selected Metallic Materials for Improved Wettability Characteristics

Interaction of CO2, Nd:YAG, excimer, and HPDL radiation with the surface of a common mild steel (ENS) and the resulting effects on wettability characteristics of the steel are examined in this chapter. It was found that, in all instances, the laser radiation occasioned changes in the wettability characteristics of the steel, namely changes in the measured contact angle of certain liquids. In order to give the work a practical basis, the HPDL-induced changes in the wettability characteristics of the mild steel are quantified in terms of firing a vitreous enamel coating on to the HPDL-treated surfaces of the mild steel. The work described in this chapter has shown that the wettability characteristics of the selected mild steel could be controlled and/or modified with laser surface treatment. Moreover, the findings of this work indicate that a relationship between the change in the wetting properties of the mild steel and the laser wavelength may exist.

5.1 Introduction Comparisons of the differences in the beam interaction characteristics with various materials of the predominant materials processing lasers, the CO2, the Nd:YAG and the excimer laser, are limited. Previously the main fundamental differences resulting from wavelength variations of CO, CO2, Nd: YAG, and excimer lasers for a number of materials processing applications have been detailed [15-18]. Likewise, such practical comparisons between these traditional materials processing lasers and the more contemporary HPDL are even fewer in number. Previously Schmidt et al. [19] compared the performance of CO2, excimer, and HPDL in the removal of chlorinated rubber coatings from concrete surfaces, noting wavelength-dependent differences in the process performance. Additionally, Bradley

91


et al. [20] compared the CO2 and HPDL for the treatment of Al2O3-based refractory materials in terms of microstructure observing wavelength-dependent microstructural characteristics unique to each laser. In more comprehensive investigations, Lawrence et al. [21, 22] compared the effects of CO2, Nd:YAG, excimer, and HPDL radiation on the wettability characteristics of an Al2O3/SiO2-based ceramic, noting that changes in the wettability characteristics of the material varied depending upon the laser type. Both scientists and engineers alike have a great interest in understanding the interfacial phenomena between vitreous enamels and carbon steels, since in many practical applications where vitreous enamels are fired on to carbon steels, the performance of the article is directly linked to the nature of the enamel-steel interface. To date, very little work exists pertaining to the use of lasers for altering the surface properties of metallic materials in order to improve their wettability characteristics. Nevertheless, it is recognized within the currently published work that laser irradiation of material surfaces can affect their wettability characteristics. Previously Heitz et al. [106], Henari and Blau [107], and Olfert et al. [108] have found that excimer laser treatment of metals results in improved coating adhesion. The improvements in adhesion were attributed to the fact that the excimer laser treatment resulted in a smoother surface and as such enhanced the action of wetting. However, the reasons for these changes with regard to changes in the metals' surface morphology, surface composition, and surface energy are not reported. In contrast, work by Lawrence et al. on the laser modification of the wettability characteristics of a number of different composite [109-114] and ceramic materials [27] has shown that the wettability performance is affected by changes in the surface roughness, the surface O2 content, and the surface energy. This present work describes for the first time the beam interaction characteristics of a 1 kW CO2 laser, a 400 W Nd:YAG laser, a 5 W KrF excimer laser, and a 1.2 kW HPDL with a common mild steel, specifically in terms of their differing effects on the wettability characteristics. These incorporate chiefly: contact angle variations, the differences in morphological and microstructural features, the surface composition, and the surface energy changes.

5.2 Experimental procedures 5.2.1 Materials and laser processing procedures The solid materials used as substrates in the wetting experiments were rectangular billets (50 x 100 mm2 with a thickness of 3 mm) of common engineering low-carbon mild steel (ENS). The contact surfaces of the materials were used as-received in the experiments. The enamel used in the experiments consisted mainly of the following: SiO2, B2O3, Na2O, Mn, F, and small quantities of Ba, Al2O3, and Ni, while the powder size was less than 75 um. The enamel frit paste was allowed to cure at room temperature for 2 h and then irradiated immediately with the laser beams. The general laser processing experimental arrangement is as shown in Fig. 3.1 (see Chapter 3), wherein the defocused laser beams were fired back and forth across the surfaces of the mild steel by traversing the samples beneath the laser beam using the x- and y-axis of the CNC gantry table. The general operating characteristics of the lasers used in the study are detailed in Table 5.1. Both pulsed and CW lasers were used in the study, therefore both the average power and

92


the peak power of each laser will differ. So, in order reasonably to compare the effects of each laser on the wettability characteristics of the mild steel, the laser energy density (fluence) of each laser beam incident on the mild steel surface was set by manipulating the laser power density and traverse speed in the case of the CO2 laser and the HPDL, and the laser power density, pulse width, and frequency in the case of the Nd:YAG and excimer lasers, such that the energy density of each of the four industrial lasers incident upon the surface of the mild steel was around 159 J/cm2. Table 5.1 Details of the selected industrial lasers used

Lasant Wavelength Maximum average output Maximum pulse energy Pulse width Repetition rate Fibre core diameter Beam diameter/size Mode of operation

CO2 C02 gas 10.6 urn IkW ~ ~ ~ ~ 6 mm CW

Nd:YAG Nd:YAG crystal 1.06 um 400 W 70 J 0.3-10 ms 1-1000 Hz 600 um 6 mm Pulsed (rapid)

Laser HPDL GaAlAs 940 nm 1.2 kW ~ ~ ~ ~ 3 x 6 mm2 CW

Excimer KrF gas 248 nm 5W 35 J 20 ns 1-55 Hz ~ 1.8 x 1.8 mm2 Pulsed (multiple)

In order to analyse the laser-treated specimens, they were, in some instances, sectioned with a Struers cutting machine using a diamond-rimmed cutting blade, and then polished using cloths and diamond suspension pastes down to 3 um. Samples, both sectioned and unsectioned, were then examined using optical microscopy, SEM, EDX, XRD, and XPS techniques.

5.2.2 Wettability characteristics analysis procedure In order to characterize fully the effects of laser radiation on the wetting and surface energy characteristics of the mild steel, two sets of wetting experiments were conducted. The first set of experiments comprised control experiments carried out using the sessile drop technique with a variety of test liquids with known surface energy properties. Thus it was possible to quantify any surface energy changes in the mild steel resulting from laser interaction. The sessile drop control experiments were carried out using the same liquids and in the same manner as described in Chapter 3. The second set of experiments were conducted simply to determine the contact angle between the enamel and the mild steel before and after interaction with the HPDL. The enamel-mild steel wetting experiments were carried out in atmospheric conditions with molten droplets of the enamel (600 °C). The temperature of the enamel throughout the experiments was measured using a Cyclops infrared pyrometer. The droplets were released in a controlled manner on to the surface of the mild steel (HPDL treated and untreated) from the tip of a micropipette, with the resultant volume of the drops being approximately 15 x 10~3 cm3. Profile photographs of the sessile enamel drop were obtained for every 60 °C fall in temperature of the molten enamel drop, with the contact angle subsequently being measured, and a mean value being obtained. Again, the standard deviation due to experimental error was calculated as being ±0.2 degrees.

93


5.3 Effects of laser radiation on the wettability characteristics For demonstration purposes an optical micrograph of a sessile drop of enamel (20 °C) placed on the surface of the mild steel (a) before and (b) after HPDL irradiation, with the contact angle superimposed, is shown in Fig. 5.1. As can be seen, HPDL irradiation of the mild steel surface effected a discernible reduction in the enamel contact angle.

Fig. 5.1 Contact angles for the enamel on (a) the as-received mild steel surface and (b) the HPDL-treated mild steel surface (1500 mm/min traverse speed) As can be seen from Table 5.2, under the experimental laser parameters employed, laser irradiation of the surfaces of the mild steel samples resulted in changes in the contact angle. Table 5.2 shows clearly that such changes were dependent upon the laser used. In general, interaction of the mild steel with CO2 and the excimer laser radiation resulted in marginal increases in the contact angle between the mild steel and the control liquids. In contrast, interaction of the mild steel with the Nd:YAG and HPDL beams resulted in the contact angle between the mild steel and the control liquids reducing. Table 5.2 Mean values of contact angles formed between the selected test liquids at 20 °C and the mild steel before and after interaction with the selected lasers

Laser Untreated CO2 Nd:YAG HPDL Excimer

Blood 55 60 54 41 57

Contact angle, 0 (degrees) Plasma Glycerol 4-octanoI 44 40 59 67 51 49 43 37 48 39 32 30 41 64 46

5.3.1 Variations in surface roughness characteristics The various surface effects of the respective lasers on the mild steel are clearly discernible from Fig. 5.2. From the microstructures shown in Fig. 5.2 it would appear that in all instances of laser treatment, surface melting and resolidification to varying degrees was induced. Indeed, as Fig. 5.3 shows, reductions in the surface roughness of the mild steel were observed (using a Taylor-Hobson Surtronic 3+ profilometer) after interaction with the Nd:YAG and the HPDL beams, reducing from an initial Ra value of 1.46 um to 1.25 and 1.12 um respectively. However, interaction of the mild steel with the CO2 and excimer

94


laser radiation occasioned the roughening of the mild steel surface, causing the surface roughness to increase to the respective Ra values of 2.58 and 2.12 um.

Fig. 5.2 Typical SEM surface images of the mild steel (a) before laser treatment and after laser interaction with (b) CO2 laser, (c) Nd:YAG laser, (d) HPDL, and (e) excimer laser radiation

95


Fig. 5.3 Mean values of surface roughness on the mild steel before and after interaction with the selected lasers

5.3.2 Variations in surface 02 content It has already been established that the O2 content of a material's surface is an influential factor governing the wetting performance of the material [20, 21], with wetting being governed by the first atomic layers of the surface of a material. Thus, in order to determine element content of O2 at the surface of the mild steel, it was necessary to examine the surface using XPS.

Fig. 5.4 Mild steel surface O2 content before and after interaction with the selected lasers

Clear differences in the surface O2 content of the mild steel after interaction with all the selected lasers were observed. These differences are shown graphically in Fig. 5.4. Increases in the surface O2 content were experienced by the mild steel after interaction with CO2, the Nd:YAG, and the HPDL beams, increasing from an initial value of

96


34.2at% to 41.5, 35.7, and 40.1at% respectively. Conversely, interaction of the mild steel with excimer laser radiation resulted in the surface O2 content of the mild steel decreasing slightly to 32.8at%. 5.3.3 Surface energy and the dispersive/polar characteristics As has already been demonstrated in previous chapters, it is possible to estimate reasonably accurately the dispersive component of the mild steel surface energy, y dsv, by plotting the graph of cos 0 against (ydlv )1/2/y/v in accordance with equation (2.18). Figure 5.5 shows the best-fit plot of cos 9 against (y^ v ) 1/2 /7/ v for the untreated and laser-treated mild steel-experimental control liquids system. Figure 5.5 shows clearly that the best-fit straight lines for the mild steel-liquid systems of the Nd:YAG laser and especially the HPDL intercept the ordinate higher above the origin than those of the untreated, CO2 laser and excimer laser-treated mild steel samples. This is of great importance since interception of the ordinate close to the origin is characteristic of the dominance of dispersion forces acting at the mild steelliquid interfaces of the untreated and excimer laser-treated samples, resulting in poor adhesion [87, 135]. On the other hand, an interception of the ordinate well above the origin is indicative of the action of polar forces across the interface, in addition to dispersion forces, hence improved wettability and adhesion is promoted [87, 135]. Also, as was discussed in the preceding chapters, because none of the best-fit straight lines intercept below the origin, then one can assume that the development of an equilibrium film pressure of adsorbed vapour on the mild steel surface (untreated and laser treated) did not occur.

1/2

Fig. 5.5 Typical plot of cos 0 against IydA J\ /yjv for the untreated and laser-treated mild steel in contact with the test control liquids

97


As was shown previously in Chapter 2, in order to determine the polar component of the mild steel surface energy, y fv, it is necessary to calculate the dispersive component of the work of adhesion, W^, by using equation (2.19). Both Wad and W^ are related by the straight line relationship represented by equation (2.20). Thus, from the best-fit straight line plots of Wad against W^ for the mild steel when it is both untreated and laser treated it is possible to determine the constants a and b for each separate condition of the mild steel (Table 5.3). Table 5.3 Values determined for the constants a and b from the plots of Waa against W^ for the mild steel, before and after laser treatment

b (mj/m2) 20.8 -18.7 -25.6 -19.5 15.2

a 1.33 1.74 1.64 1.41 1.25

Untreated CO2 laser Nd:YAG laser HPDL Excimer laser

Since a linear relationship exists between the dispersive and polar components of the control test liquids' surface energies [112] which satisfies equation (2.20), then, as was shown previously, it is possible to calculate y fv directly for untreated and laser-treated mild steel using equation (3.2) with the appropriate values of a given in Table 5.3. Table 5.4 gives the values determined for yfv and /£,, as well as the total surface energy for both untreated and laser-treated mild steel. Table 5.4 Determined surface energy values for the mild steel before and after laser treatment

Untreated

C02

Condition Nd:YAG

HPDL

Excimer

Dispersive component, 7 sv (mJ/m )

66.00

65.9

65.4

64.6

65.5

Polar component, 7 psv (mJ/m2)

4.17

3.83

4.24

6.59

4.0

Surface energy component 2

As can be seen from Table 5.4, Nd:YAG laser, HPDL, and excimer laser treatment of the surface of the mild steel effected small increases in the polar component of the surface energy 7 psv. Such increases in the polar component of the surface energy of the mild steel have a positive effect upon the action of wetting and adhesion.

5.4 Discussion of laser-effected modification of wettability characteristics The results detailed previously show clearly that interaction of the mild steel with the selected industrial lasers has resulted in the contact angle formed between the control liquids altering to various degrees depending upon the laser type. Under the selected

98


experimental laser operating parameters, interaction of the mild steel typically with the Nd:YAG and HPDL beams effected a decrease in the contact angle, while interaction of the mild steel with COa and excimer laser radiation occasioned small increases in the contact angle. Such changes in the value of the contact angle are influenced, depending upon the laser used, primarily by: 1. Modifications to the surface roughness. Depending upon the laser used, in this instance the Nd:YAG laser and the HPDL, the particular experimental laser parameters employed resulted in the ideal degree of melting and resolidification of the mild steel surface being induced. This in turn resulted in a reduction of the surface roughness, thus directly reducing the contact angle, 9. 2. Surface roughening. Depending upon the laser used, in this case the CO2 laser, the particular experimental laser parameters employed resulted in a degree of melting and resolidification of the mild steel surface being induced which was not conducive to surface smoothing. Also, in the case of the excimer laser the resultant ablation of the mild steel surface resulted directly in an increase in the surface roughness. In both instances the increase in the mild steel surface roughness resulted inherently in an increase in the contact angle, 6. 3. Surface O2 content. An increase in the surface O2 content of the mild steel resulting from laser treatment is an influential factor in the promotion of the action of wetting, since an increase in surface O2 content inherently effects a decrease in the contact angle, and vice versa. 4. Increase in the polar component, y^, of the surface energy. This results from the melting and resolidification of the mild steel surface, thus creating a different microstructure that quite possibly improved the action of wetting and adhesion. From Fig. 5.2 it can be seen that microstructures of the CO2, Nd:YAG, and HPDLtreated samples appear to be indicative of melting and resolidification. Indeed, the microstructures of the Nd:YAG- and HPDL-treated samples are very similar in nature, yet the microstructure of the CO2 laser-treated sample differs greatly. It is surmised that in the instances of the Nd:YAG- and HPDL-treated samples, the optimum degree of surface melting is obtained, resulting in the minimum surface roughness. Similar laser-induced surface smoothing effects were obtained by Nicolas et al. [151] and Henari and Blau [107], who observed that excimer laser treatment of ceramics and metals could result in the generation of a smoother surface. In contrast, it is believed that in the instance of the CO2 laser-treated sample, where the surface of the mild steel has become roughened as a consequence of laser interaction, the roughening is occasioned as a result of excess energy being absorbed by the surface of the mild steel, leading therefore to a higher level of surface melting. This in turn causes micro-porosities and a generally rough surface profile. Indeed, this supposition is borne out somewhat by Fig. 5.6, which shows that the surface condition of the mild steel resulting from HPDL modification (with a number of different traverse speeds) greatly affected the measured contact angle between the mild steel and vitreous enamel.

99


Fig. 5.6 Relationship between surface roughness, contact angle (enamel), and traverse speed for the HPDL-treated mild steel

As can be seen from Fig. 5.6, at relatively low traverse speeds excess energy is deposited on the surface of the mild steel resulting in a high level of surface melting. This in turn causes porosities and a generally rough surface profile. As the traverse speed increases, however, the energy deposited on the surface of the mild steel reduces. Accordingly the degree of surface melting reduces ultimately to the optimum degree, resulting in the minimum surface roughness, and contact angle, at around 1500 mm/min. Beyond this point the surface roughness, and contact angle, can be seen to increase, indicating that insufficient melting, and consequently smoothing, was achieved. Again, such results are in accord with those obtained by Feng et al. [146], who noted that under certain surface conditions, contact angle reduction was inversely proportional to surface roughness. Moreover, Olfert et al. [108] found that excimer laser treatment of steel surfaces greatly improved the adhesion of a zinc coating. They asserted that laser treatment occasioned the smoothing of many of the high-frequency surface features, resulting in more complete wetting by the zinc. The CO2 laser operates in the CW mode and consequently the laser operating parameters used in the experiments were exactly the same as those of the HPDL in every way (power, beam diameter, traverse speed, etc.), the only difference being, however, the laser wavelength. There are many other factors that may quite possibly come into play, but since the microstructures obtained after CO2 [Fig. 5.2(b)] and HPDL [Fig. 5.2(d)] treatment were so different, then it is perhaps reasonable to propose that these changes are the result of wavelength-difference effected beam absorption. What is more, the definite similarity between the microstructures obtained after Nd:YAG [Fig. 5.2(c)] and HPDL [Fig. 5.2(d)] treatment indicate further the existence of a wavelength relationship. In contrast, as Fig. 5.2(e) shows, interaction of

100


the mild steel with excimer laser radiation did not cause melting of the surface, but instead induced surface ablation, which consequently resulted in a slightly rougher surface. Thus an increase in the contact angle was effected. Similar observations were made by Lawrence and Li [22] during surface treatment of a ceramic compound using an excimer laser. Additionally, Kokai et al. [149] have concluded that, with excimer laser parameters which are conducive to the production of plasma, as was the case with the mild steel, then the surface roughness is increased as a result of plasma-induced debris redepositing on the surface and excessive thermally induced surface fractures and porosities. Clearly, since plasma generation was observed, then surface roughening after excimer laser irradiation was perhaps to be expected. However, Liu et al. [150], Nicolas et al. [151], and Henari and Blau [107], have reported that irradiating ZrO2 with excimer laser radiation with energy densities in excess of 2.7 J/cm2, resulted in a reduction in surface roughness. Such reductions were attributed to the fact that at these levels of energy density, melting of the ZrO2 surface occurred. By means of cross-sectional SEM analysis it was possible to determine the laser melt/ablation depth for each laser. As Table 5.5 shows, the depth of the laser melting, and in the case of the excimer laser, the ablation region, varied significantly according to laser type. For the CO2, Nd:YAG, and HPDL the melt depths were measured as 170, 45, and 100 um respectively, while for the excimer laser the ablation depth was measured as 10 urn. Clearly, the differences in laser melt/ablation depth obtained with the Nd:YAG and excimer lasers was an order of magnitude smaller than those of the CO2 or HPDLs. The main reason for these large differences are thought to be due to the pulsed nature of the beams of the Nd:YAG and excimer lasers, as opposed to the CW nature of the CO2 or HPDL beams. Since the interaction time of a pulsed beam with a material is much shorter than that of a CW beam, consequently the depth of the laser melt/ablation region will be much smaller due to the reduced time afforded for heat transfer. Table 5.5 Determined laser melt/ablation depths for the mild steel after laser irradiation

Laser melt/ablation depth

CO2 170 um

Laser HPDL Nd:YAG 45 um 100 um

Excimer 10 um

It is also of great importance to consider the surface O2 content of the mild steel before and after treatment with the selected lasers. Increases in the surface O2 content were experienced by the mild steel after interaction with the CO2 laser, the Nd:YAG laser and the HPDL beams, while interaction of the mild steel with excimer laser radiation resulted in the surface O2 content of the mild steel decreasing. Such a result is in agreement with the findings of a number of workers [152, 153], who have noted that, for many materials, irradiation with an excimer laser beam creates defective energy levels, in particular the formation of 02 vacancies.

101


From the previous discussion it is unclear whether after laser surface treatment the surface roughness, the microstructural changes, or the O2 content alone, or a combination of these, are the principal factors influencing the observed changes in the wettability characteristics of the mild steel. But, by grinding the surfaces of the untreated and laser-treated mild steel samples down to 1 um, while still retaining a laser-treated surface, it was possible to isolate the effects of surface roughness by rendering them non-effective, and to investigate at least the effects of the microstructural changes and possibly those of the O2 content. An examination of the contact angle characteristics of the ground mild steel samples using only glycerol revealed that the contact angle was consistently around 30 degrees across the range of samples. In addition, from an XPS analysis of the O2 content of the mild steel samples, it was found that the O2 content of the untreated sample remained around the original value at 33.8at%. For the CO2, the Nd:YAG, and the HPDL-treated samples, however, the O2 content was found to have reduced to a level similar to that of the untreated sample (34.3, 32.8, and 33.2at% respectively), while the excimer laser-treated sample increased to a level similar to that of the untreated sample (34.0at%). Since the measured contact angles of the ground samples were consistently similar, despite the presence of the laser-induced microstructures, then it is reasonable to conclude that microstructure does not influence the wettability characteristics of the mild steel. Indeed, it would appear that the wettability characteristics of the mild steel are influenced predominantly by the surface roughness and, to some extent, by the surface O2 content.

5.5 Bonding mechanisms between the high-power diode lasertreated mild steel and enamel In practical enamelling operations complex combinations of the various bonding mechanisms (detailed in Chapter 2) actually occur, varying according to the types of materials used [91]. For the mild steel and the enamel, the mechanisms involved in the bonding are principally: physical bonding (van der Waals forces), mechanical bonding, chemical bonding (oxide transformation and O2 bridging), and on a very small scale, electrochemical reactions such as the electrolytic (redox) effect due to the presence of ferric oxides within the mild steel reacting with other oxides in the enamel [91]. Invariably, the preponderant bonding mechanisms between mild steels and enamels are physical and mechanical bonding [95]. However, an EDX analysis conducted at the interface between the mild steel and the enamel revealed the presence of a small diffusion region which contained elements unique to both the mild steel and the enamel. This is perhaps to be expected since enamel glazes on steels are typically bonded as a result of some of the base material dissolving into the glaze [91], with wetting characteristics often being achieved or enhanced by a reaction at the interface at an elevated temperature [72]. Indeed, when an oxide layer is present on the surface of a metal, as was the case for the HPDL-treated mild steel, then typically the fired enamel dissolves this oxide layer. Subsequently a redox reaction has to occur at the interface to form more metal oxide [72]. In the case of the HPDL-treated mild steel and the enamel, which is an Fe/CoO-sodium silicate glass system, then because the oxidation potential for Fe is higher than that for Co,

102


the redox reaction forms FeO at the interface. Thus a layer of FeO continually remains at the interface and acts as the compatible oxide phase that provides the chemical bonding between the HPDL-treated mild steel and the enamel [72]. Additionally, intrinsic within the redox reaction formation of FeO is the alloying of the Fe with the reduced Co to form dendrites in the interfacial zone by means of a microgalvanic cell mechanism [72, 95]. Furthermore, this alloying has a negative free energy which therefore contributes to the driving force for the net redox reaction formation of FeO [72].

Fig. 5.7 Typical cross-section SEM image of the bond region between the enamel and the HPDL-treated mild steel

As can be seen from Fig. 5.7, it is not possible to discern any dendritic growth within the enamel glaze in the bond region, which is characteristic of enamels fired on to substrates containing Fe, Si, and in particular Co [95]. However, it can be seen that enamel is held firm in the surface irregularities, thus ensuring sound adhesion. Again, such mechanical bonding is typical of enamel glazes on metals [95].

5.6 Summary Interaction of CO2 laser, Nd:YAG laser, HPDL, and excimer laser radiation with the surface of the mild steel was found to effect changes in the wettability characteristics of the material. It was observed that interaction of the mild steel with Nd:YAG and HPDL radiation effected reductions in the contact angle. In contrast, interaction of the mild steel with CO2 and excimer laser radiation resulted in a slight increase in the contact angle. Such changes were identified as being primarily due to: 1. The generation of a smoother surface after Nd:YAG and HPDL treatment due to optimum surface melting and resolidification. 2. The surface roughness of the mild steel increasing after interaction with CO2 and excimer laser radiation due to excess surface melting and ablation respectively, which in turn resulted directly in an increase in the contact angle.

103


3. Changes in the surface 02 content of the mild steel; increasing after interaction with CO2, Nd:YAG, and HPDL radiation due to surface melting, and decreasing after interaction with the excimer laser due to the creation of defective energy levels. 4. Increases in the polar component of the surface energy resulting from the melting and resolidification of the mild steel surface, thus creating a different microstructure that quite possibly improved the action of wetting and adhesion. However, it was found that changes in the wettability characteristics of the mild steel appeared to be predominantly influenced by the surface roughness, while the microstructure appeared not to have any effect on the wetting properties of the mild steel. Additionally, surface O2 content is also thought to play a minor role. It is a distinct possibility that a wavelength dependence of the change in the wetting properties can be deduced from the findings of this work. This is clear from the very similar properties of the surfaces irradiated with the Nd:YAG laser and the HPDL, the wavelengths of which vary by little more than 66 nm. It is important to note, however, the high degree to which beam mode (temporal and spatial) will influence the laser process. Nonetheless, the work has shown that under the chosen experimental laser operating parameters, changes in the wettability characteristics of the mild steel were seen to vary depending upon the laser type.

104

Chapter 6

Laser Modification of Selected Polymer Materials for Improved Wettability Characteristics

The effects of CO2, Nd:YAG, excimer, and HPDL radiation on the wettability characteristics of a number of bio-compatible polymer materials, namely polyethylene (PE) and polymethyl methacrylate (PMMA) are investigated in this chapter. Interaction of the laser radiation with the surface of the polymers was seen to effect varying degrees of change to the wettability characteristics of the materials depending upon the laser used. The work presented in this chapter has demonstrated that the wettability characteristics of the PE and PMMA could be controlled and/or modified with laser surface treatment. However, a wavelength dependence of the change in the wetting properties could not be deduced from the findings of this work.

6.1 Introduction Much work has been carried out to study the effects of laser wavelength variation for medical and surgical applications, revealing clear differences in the performance and effectiveness of many different lasers for such applications. In contrast, comparisons of the differences in the beam interaction characteristics with various materials of the predominant materials processing lasers, the CO2, the Nd:YAG, and the excimer laser, are limited. Previously, the main fundamental differences resulting from wavelength variations of CO, CO2, Nd:YAG, and excimer lasers for a number of materials processing applications have been detailed [15-18]. Likewise, such practical comparisons between these traditional materials processing lasers and the more contemporary HPDL are even fewer in number. Previously, Lawrence et al. compared the effects of CO2, Nd:YAG, excimer, and HPDL radiation on the wettability characteristics of an Al2O3/SiO2-based ceramic [21, 22], and a mild steel [23, 24], noting that changes in the wettability characteristics of the ceramic and the steel varied depending upon the laser type.

105


In general, polymer materials posses poor adhesion characteristics. Consequently it is very difficult to wet, and therefore bond, almost all polymer materials to adhesives without modifying their surface in order to improve their wettability characteristics. Traditionally a number of surface modification techniques have been employed to achieve such ends. Previously Fourche [163] investigated a number of mechanical and flame methods which essentially roughened the polymer surface thus promoting improved mechanical bonding. A number of workers have developed chemical treatments primarily to increase the surface O2 content of various polymer materials. These include oxidation in chromic acid [164] and sulphoration in chlorosulphonic acid [165]. However, these chemical treatments require prolonged immersion of the polymer materials in the acids in order to obtain any significant improvement in interfacial bond strength. What is more, these treatments are often accompanied by an undesirable loss in mechanical strength. Kaplan et al. [166] used the corona discharge treatment to modify the surface of ultra-high strength polyethene (UHSPE) fibres. It was noted that an increase in interlaminar shear strength was effected, indicating that better interfacial bonding of the fibres had been achieved. Plasma treatments have also shown a good deal of promise for polymer material surface modification. Holmes and Schwartz [167] studied the effect of ammonia plasma on UHSPE fibres and concluded that the adhesion of the fibres to epoxy resin was improved without the usual detrimental reduction in fibre strength. Li and Netravali [168-170] also investigated the effects of plasma (allylamine and ammonia) on UHSPE fibres and found that huge increases of over 300 per cent in the interfacial shear strength were occasioned after some plasma treatments due to increases in the surface roughness. Similar large increases in the interfacial shear strength of UHSPE fibres after plasma treatment were also observed by Nguyen et al. [171]. Despite the very beneficial improvements in interfacial shear strength that can be obtained from such plasma treatments, the actual process itself is extremely 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. Lawrence and Li have amply demonstrated the practicability of employing different types of lasers to effect changes in the wettability characteristics of ceramics [21, 22, 27] and metals [23, 24] for improved adhesion and bonding. In the case of polymer materials, excimer laser radiation has been shown to be a very effective means of enhancing their wettability characteristics. First demonstrated in 1982 by Srinivasan and MayneBanton [172], the excimer laser is typically used to remove thin surface layers of polymers. In recent years this technique has been applied to control precisely and alter the surface characteristics of a number of polymer materials. Much research has been carried out to study the effects of excimer laser radiation on the wettability characteristics of polyethylene terephthalate (PET) in film [106], fibre [173], and sheet [174] form. The work on PET sheet by Andrew et al. [174] revealed that excimer laser treatment resulted in surface roughening. It was suggested that this was probably due to the differential etching of crystalline and amorphous regions in the material. Surface roughening was also obtained on polyparaphenylene terephthalamide (PPTA) fibres after excimer laser treatment by Watanabe and Takata [175]. A generally more undulating surface morphology, as opposed to a finely roughened surface, was

106


observed by Heitz et al. [106] and Watanabe et al. [173] after excimer laser treatment of PET films and fibres respectively. Both studies reported an increase in adhesion, citing the reason as being the occurrence of photochemical reactions which were accompanied by slight surface ablation. Similar improvements in adhesion strength of excimer laser-treated polypropylene (PP) films were reported by Breuer et al. [176] when using certain optimum laser parameters. It was proposed that such improvements were the result of the laser generating a more polar surface. Laurens et al. [177, 178] also concluded that a more polar surface resulted from the excimer laser treatment of polyether-etherketone (PEEK). Furthermore, these workers found that the choice of laser wavelength has a crucial influence on the resultant wettability characteristics of the PEEK. Comprehensive and detailed investigations by Song and Netravali [160, 161, 179] into the effects of excimer laser radiation on the interfacial characteristics of UHSPE fibres and epoxy resin revealed a considerable increase in the interfacial shear strength resulting after laser treatment. This enhancement in the interfacial shear strength was attributed to the increase in surface roughness, the increase in surface O2 content, and the increased polar nature of the fibres after excimer laser treatment. Yet despite the large amount of work conducted with excimer lasers, no published literature to date exists pertaining to the use of other industrial lasers to modify the wettability characteristics of polymer materials. The work presented in this chapter describes for the first time the beam interaction characteristics of the bio-compatible polymer materials polyethylene (PE) and polymethyl methacrylate (PMMA) with a 1 kW CO2 laser, a 400 W Nd:YAG laser, a 5 W KrF excimer laser, and a 120 W HPDL, focusing particularly on the differing effects thereof on the wettability characteristics. These chiefly incorporate contact angle variations, the differences in morphological and microstructural features, the chemical composition of the surface, and the surface energy changes. Despite the fact that PE is widely used in surgery for total joint replacement and PMMA is used extensively in surgery as bone cement, no published literature exists pertaining to the enhancement of wettability characteristics using lasers or other means.

6.2 Experimental procedures 6.2.1 Materials and laser processing procedure The solid materials used as substrates in the wetting experiments were cubes (25 x 25 mm2 with a thickness of 5 mm) of PE and PMMA. The contact surfaces of the materials were used as-received in the experiments. The general laser processing experimental arrangement comprised of the defocused laser beams being fired back and forth across the surfaces of the PE by traversing the samples beneath the laser beam using the x- and y-axis of the CNC gantry table (see Fig. 3.1). The general operating characteristics of the lasers used in the study are detailed in Table 6.1. Both pulsed and CW lasers were used in the study, therefore, both the average power and the peak power of each laser differed. Consequently, in order reasonably to compare the effects of each laser on the wettability characteristics of the polymers, the laser energy density (fluence) of each laser beam incident on the surface of the polymers was set by manipulating the laser power density and traverse speed in the case of the CO2 laser and the HPDL, and the laser power density, pulse width, and frequency in

107


the case of the Nd:YAG and excimer lasers, such that the energy density of each of the four lasers incident upon the PE surface was around 300 mj/cm2. Table 6.1 Details of the selected industrial lasers used

Lasant Wavelength Maximum average output Maximum pulse energy Pulse width Repetition rate Fibre core diameter Beam diameter/size Mode of operation

CO2 CO2 gas 10.6 um 1 kW ~ ~ ~ ~ 6 mm CW

Laser HPDL Nd:YAG Nd:YAG crystal GaAlAs 810±20 nm 1.06 um 400 W 60 W ~ 70 J ~ 0.3-10 ms ~ 1-1000 Hz 1 mm 600 um 6 mm 6 mm CW Pulsed (rapid)

Excimer KrF gas 248 nm 5W 35 J 20 ns 1-55 Hz ~ 1.8 x 1.8 mm2 Pulsed

In order to analyse the laser-treated specimens, they were, in some instances, sectioned. Samples, both sectioned and unsectioned, were then examined using optical microscopy, SEM, EDX, XRD, and XPS techniques. 6.2.2 Wettability characteristics analysis procedures To investigate the effects of laser radiation on the wetting and surface energy characteristics of the polymers, wetting experiments were conducted. The experiments comprised control experiments carried out using the sessile drop technique with a variety of test liquids with known surface energy properties. Thus it was possible to quantify any surface energy changes in the polymers resulting from laser interaction. Table 6.2 Total surface energy (%,) and the dispersive (yfv) and polar (/^) components for the selected test liquids [180]

Liquid Glycerol Formamide Etheneglycol Dimethylsulphoxide

Y (mj/m2) 63.4 58.2 48.0 44.3

ri (mj/m2) 37.0 39.5 29.0 36.1

y;

(mj/m2) 26.4 18.7 19.0 8.2

The sessile drop control experiments were carried out using glycerol, formamide, etheneglycol and dimethylsulphoxide. Details of the test liquids, along with their total surface energy ( y 2 ) , dispersive ( y f v ) and polar ( y f v ) component values are given in Table 6.2. The experiments were conducted in atmospheric conditions at a temperature of 20 °C. The droplets were released in a controlled manner on to the surface of the test substrate materials (laser treated and untreated) from the tip of a micropipette, with the resultant volume of the drops being approximately 6 x 10~3 cm3. Each experiment

108


lasted for three minutes with profile photographs of the sessile drops being obtained every minute. The contact angles were then measured with a mean value being subsequently determined. The standard deviation due to experimental error was calculated as being ±0.2 degrees. As described in previous chapters, it was observed during the wetting experiments that, throughout the period of the experiments, no discernible change in the magnitude of the contact angle was observed, indicating that thermodynamic equilibrium was established at the solid-liquid interface at the outset of the experiments.

6.3 Effects of laser radiation on wettability characteristics As can be seen from Tables 6.3 and 6.4, under the experimental laser parameters employed, laser irradiation of the surfaces of the polymer samples resulted in changes in the contact angle to varying degrees. Both these tables show clearly that such changes were dependent upon the laser used. It can be seen that, in general, interaction of the polymers with CO2 radiation resulted in an increase in the contact angle between the polymers and the control liquids, while interaction of the polymers with Nd:YAG and HPDL radiation resulted in a slight reduction. In contrast, interaction of the polymers with the excimer laser beam resulted in a considerable reduction in the contact angle between the polymers and the control liquids. Table 6.3 Mean values of contact angles formed between the selected test liquids at 20 °C and the PE before and after laser treatment


Glycerol 66 68 59 67 39

Contact angle, 6 (degrees) Formamide Etheneglycol Dimethylsulphoxide 61 58 46 57 41 55 54 52 39 55 52 41 37 33 27

Table 6.4 Mean values of contact angles formed between the selected test liquids at 20 °C and the PMMA before and after laser treatment


Glycerol 70 65 67 64 42

Contact angle, 0 (degrees) Formamide Etheneglycol Dimethylsulphoxide 62 41 59 59 55 38 60 39 56 57 53 37 38 24 35

109


6.3.1 Variations in surface roughness characteristics An analysis of the polymers' surface roughness (using a Taylor-Hobson Surtronic 3+ profilometer) after treatment with these lasers revealed that the surface roughness had increased, as Fig. 6.1 shows.

Fig. 6.1 Mean values of surface roughness on the polymers before and after interaction with the selected lasers As has already been seen previously, according to Feng et al. [146], under certain surface conditions, contact angle is inversely proportional to surface roughness. Clearly this proposal is borne out somewhat by the CO2, Nd:YAG, and HPDL, but does not hold in the case of the excimer laser. It is believed that the increase in surface roughness of the polymers after excimer laser treatment is counteracted by increases in the surface O2 content and the surface energy changes (as discussed later). The various surface effects of the respective lasers on the PE and the PMMA are clearly discernible from Figs 6.2 and 6.3 respectively. As can be seen from these figures, the PE and PMMA surfaces resulting from exposure to the various laser beams appear to display similar characteristics. From the micrographs shown in Figs 6.2 and 6.3 it would appear that in the instances where CO2, Nd:YAG, and HPDL radiation was incident with the polymer surface, surface melting and resolidification to varying degrees was induced. On the other hand, these figures show that interaction of the polymers with excimer laser radiation resulted in surface ablation.

110


Fig. 6.2 Typical SEM surface images of the PE (a) before laser treatment and after interaction with (b) CO2 laser, (c) NdrYAG laser, (d) HPDL, and (e) excimer laser radiation

111


Fig. 6.3 Typical SEM surface images of the PMMA (a) before laser treatment and after interaction with (b) CO2 laser, (c) Nd:YAG laser, (d) HPDL, and (e) excimer laser radiation

112


6.3.2 Variations in surface 02 content It has already been established that the O2 content of a material's surface is an influential factor governing the wetting performance of the material [97, 148], with wetting being governed by the first atomic layers of the surface of a material. Thus, in order to determine element content of O2 at the surface of the mild steel, it was necessary to examine the surface using XPS. As is evident from Fig. 6.4, only very marginal differences in the surface O2 content of the polymers after interaction with the CO2, Nd:YAG, and HPDL beams were observed. Conversely, interaction of the PE with excimer laser radiation resulted in the surface 02 content of the polymers increasing markedly.

Fig. 6.4 Surface O2 content of the selected polymers before and after interaction with the selected lasers 6.3.3 Surface energy and the dispersive/polar characteristics As has already been demonstrated in previous chapters, it is possible to estimate reasonably accurately the dispersive component of the PE and PMMA surface energy, yl> by plotting the graph of cos 0 against (y^ v ) 1/2 /Y;v in accordance with equation (2.18). Figures 6.5 and 6.6 show the best-fit plot of cos 9 against (y fv )1/2/y/v for the untreated and laser-treated PE-experimental control liquids system and the PMMAexperimental control liquids system respectively.

113


/

d \lf2

Fig. 6.5 Typical plot of cos d against (y /v J /y/v for the untreated and laser-treated PE in contact with the test control liquids

/

. \l/2

Fig. 6.6 Typical plot of cos 6 against (y (v J ///v for the untreated and laser-treated PMMA in contact with the test control liquids

As in previous chapters, a comparison of the ordinate intercept points of the untreated and laser-treated PE-liquid systems and PMMA-liquid systems given in Figs 6.5 and 6.6 respectively, reveals that the best-fit straight lines for the polymer-liquid systems of the CO2, Nd:YAG, and HPDL intercept the ordinate at similar points. Indeed,

114


the intercept points of the polymer-liquid systems for these lasers are close to the ordinate intercept point of the untreated polymers. Conversely, the ordinate intercept points for the excimer laser polymer-liquid systems are considerably higher above the origin than those of the untreated, CO2, Nd:YAG, and HPDL polymer-liquid systems. This is of great importance since interception of the ordinate well above the origin is indicative of the action of polar forces across the interface, in addition to dispersion forces, hence improved wettability and adhesion is promoted. On the other hand, as the ordinate interception point approaches the origin, the influence of polar forces diminishes and dispersion forces begin to dominate [87, 135]. Thus, from Figs 6.5 and 6.6 it can be asserted that polar forces are in attendance more in the excimer lasertreated PE and PMMA samples than those either untreated or treated with the other lasers. Furthermore, because none of the best-fit straight lines intercept below the origin, then as seen previously, it can be said that the development of an equilibrium film pressure of adsorbed vapour on either the PE or the PMMA surface (untreated and laser treated) did not occur. As was shown previously in Chapter 2, in order to determine the polar component of the PE and PMMA surface energy, /£,, it is necessary to calculate the dispersive component of the work of adhesion, W^, by using equation (2.19). Both Wad and W^ are related by the straight line relationship represented by equation (2.20). Thus, from the best-fit straight line plots of Wad against W^ for the PE and PMMA when it is both untreated and laser treated it is possible to determine the constants a and b for each separate condition of the PE and the PMMA. The values of a and b for the PE and PMMA after surface treatment with the various industrial lasers are given in Table 6.5. Table 6.5 Values determined for the constants a and b from the plots of Wad against W^ for the PE and PMMA, before and after laser treatment

PMMA

PE

Untreated CO2 laser Nd:YAG laser HPDL

Excimer laser

a 1.14 1.16 1.15 1.17 1.24

b (mj/m2) -6.3 -4.8 -13.7 -1.7 -16.7

a 1.12 1.14 1.15 1.13 1.28

b (mj/m2) -2.9 -4.5 -2.1 -6.1 -20.9

Since a linear relationship exists between the dispersive and polar components of the control test liquids' surface energies [179] which satisfies equation (2.20), then, as was shown previously, it is possible to calculate 7 psv directly for untreated and laser-treated PE and PMMA by using the appropriate values of a given in Table 6.5. But, because the test liquids used in the experiments differ from those used in the previous chapters, equation (2.24) now becomes

115


Table 6.6 gives the values determined for y dsv and y fv, as well as the total surface energy for both the untreated and laser-treated mild steel. Table 6.6 Determined surface energy values for the PE and PMMA before and after laser treatment PE

Condition Untreated CO2 laser Nd:YAG laser HPDL Excimer laser

d

2

2

Y sv (mj/m )

?'„ (mJ/m )

33.7 34.2 35.0 34.6 35.9

3.4 4.3 3.9 4.5 10.2

d

Y sv

PMMA (mj/m ) Yda (mJ/m2) 2

37.2 38.6 38.1 39.1 40.1

2.7 3.7 4.2 3.3 15.5

6.4 Discussion of laser-effected wettability characteristics modification The results detailed previously show clearly that interaction of both the PE and the PMMA with the selected industrial lasers has resulted in the contact angle formed between the control liquids altering to various degrees depending upon the laser type. Under the selected experimental laser operating parameters, interaction of the polymers typically with the CO2, Nd:YAG, and HPDL beams effected very little change in the measured contact angles, while interaction of the polymers with the excimer laser radiation occasioned considerable decreases in the measured contact angles. Such changes in the value of the contact angle are influenced, depending upon the laser used, primarily by changes in the surface morphology, changes in the surface O2 content, and changes in the surface energy. From Figs 6.2 and 6.3 it can be seen that morphologies of the CO2-, Nd:YAG-, and HPDL-treated samples appear to be indicative of melting and resolidification. As is evident from the analysis of the surface roughness, the melting and subsequent resolidification after treatment with these lasers appears to have had very little effect on the surface roughness. In contrast, as can be seen from Fig. 6.2(e) and Fig. 6.3(e), interaction of the PE and the PMMA with excimer laser radiation under the chosen operating parameters did not cause melting of the surface, but instead induced surface ablation which consequently resulted in a slightly rougher surface. Similar observations were made by Lawrence et al. [21-23] during surface treatment of ceramics and metals with excimer laser radiation. Additionally, Kokai et al. [149] have concluded that, with excimer laser parameters which are conducive to the production of plumes, as was the case with both the PE and the PMMA, then the surface roughness is increased as a result of plume-induced debris redepositing on the surface and excessive thermally induced surface fractures and porosities. Since plume

116


generation was observed, then surface roughening after excimer laser irradiation was perhaps to be expected. However, Liu et al. [150], Nicolas et al. [151] and Henari and Blau [107], have reported that irradiating ZrC>2 with excimer laser radiation with energy densities in excess of 2.7 J/cm2, resulted in a reduction in surface roughness. Such reductions were attributed to the fact that at these levels of energy density, melting of the ZrO2 surface occurred. According to equation (2.14) and the work of Feng et al. [146], an increase in the contact angle would be expected after interaction of the polymers with the excimer laser. However, this is clearly not the case as is apparent from Tables 6.3 and 6.4, which show that significant reductions in the contact angle were observed after excimer laser treatment. It is postulated that the increase in surface roughness of the polymers after excimer laser treatment is counteracted to a large degree by the increases in the surface 02 content and the surface energy changes. As was discussed earlier, the surface O2 content of the polymers before and after treatment with the CO2, the Nd:YAG, and the HPDL altered very little. But, after interaction with the excimer laser beam the surface O2 content of the polymers increased markedly. Such a finding is contrary to those of a number of workers [152, 153] who have noted that, for ceramics, irradiation with an excimer laser beam creates defective energy levels, in particular the formation of O2 vacancies. Nevertheless, the increase in surface O2 content of polymer materials after excimer laser interaction is well documented [176— 179]. Such an occurrence is due to the fact that polymers undergo main-chain scission when subjected to concentrated levels of UV radiation in air [181]. Thus, when exposed to the excimer laser radiation under the selected laser operating parameters, both the PE and the PMMA can easily generate free radicals as transient species. The O2 in the air can then react readily with the free radicals under the high temperatures generated on the surface of the polymers. It is suggested that oxidation in this manner is the reason for the observed increase in the surface O2 content of the polymers after excimer laser treatment. Furthermore, the exhibited increase in the polar nature of the polymers after excimer laser treatment can also be attributed to this photo-oxidation on the polymers' surface. As such, it is reasonable to assume that this surface oxidation will result in the generation of some surface O2-containing polar functional groups [179]. By means of cross-sectional SEM analysis it was possible to determine the laser melt/ablation depth in the PE and PMMA samples for each laser. As is evident from Table 6.7, the depth of the laser melting, and in the case of the excimer laser, the ablation region, varied significantly according to laser type. As one can see, the differences in laser melt/ablation depth obtained with the Nd:YAG laser and particularly the excimer laser, were smaller than those of the CO2 or HPDL. The main reason for these differences is thought to be due to the pulsed nature of the beams of the Nd:YAG and excimer lasers, as opposed to the CW nature of the CO2 or HPDL beams. Since the interaction time of a pulsed beam with a material is much shorter than that of a CW beam, consequently the depth of the laser melt/ablation region will be much smaller due to the reduced time afforded for heat transfer.

117


Table 6.7 Determined laser melt/ablation depths for the PE and PMMA after laser irradiation

Laser melt/ablation depth PE PMMA

CO2 75 um 56 um

Laser Nd:YAG HPDL 31 um 67 um 24 um 53 um

Excimer 8 um 12 um

The excimer laser aside, the morphology of the PE and the PMMA generated after treatment with the CO2, the Nd:YAG, and the HPDL was very similar. It is therefore possible to conclude that a wavelength dependence of the change in the wetting properties cannot be deduced from the findings of this work. This is evident from the very similar properties of the surfaces irradiated with the CO2, the Nd:YAG, and the HPDL, the wavelengths of which vary by more than an order of magnitude. Nonetheless, the work has shown that under the chosen experimental laser operating parameters, the wettability characteristics of the polymers were seen to alter to various degrees depending upon the laser type, and in particular, whether the laser radiation had the propensity to cause melting or ablation of the surface. Consequently, it is distinctly possible that pulse width may have played an important role.

6.5 Summary Interaction of CO2 laser, Nd:YAG laser, HPDL, and excimer laser radiation with the surface of PE and PMMA sheet was found to effect varying degrees of change to the wettability characteristics of the materials. Overall, the changes in the wettability characteristics of the two polymers were found to be similar in magnitude. It was observed that interaction of the polymers with CO2, Nd:YAG, and HPDL radiation resulted in very little change in the contact angle, and therefore the wettability characteristics, with the control liquids used. In contrast, interaction of the polymers with excimer laser radiation occasioned a marked decrease in the contact angle, and therefore the wettability characteristics, with the control liquids used. Such changes after excimer laser treatment were identified as being primarily due to the increase in the surface C>2 content of the polymers and the increase in the polar component of the surface energy. These occurrences are believed to be the result of photo-oxidation on the polymers' surface which, in turn, is assumed to have resulted in the generation of some surface (^-containing polar functional groups, thus effecting the observed changes in the wettability characteristics of the polymers after excimer laser treatment. A wavelength dependence of the change in the wetting properties cannot be deduced from the findings of this work. This is clear from the very similar properties of the surfaces irradiated with the CC>2 laser, the Nd:YAG laser, and the HPDL, the wavelengths of which vary by more than an order of magnitude. Nonetheless, the work has shown that under the chosen experimental laser operating parameters, changes in the wettability characteristics of the polymers were seen to vary somewhat depending upon the laser type. In particular, this was seen to depend up on whether the laser radiation had the propensity to cause surface melting or ablation. As such, the influence of pulse width cannot be discounted.

118

Chapter 7

Practical Applications of Lasers for the Modification of Wettability Characteristics

To date, a number of practical applications of the use of lasers for modifying the wettability characteristics of the materials detailed in the previous chapters have been identified. These are the sealing of ceramic tiles, the enamelling of OPC, and the enamelling of mild steel. The findings of this work indicate that the laser is not only an ideal tool, but arguably a unique tool for altering the wetting characteristics of engineering materials.

7.1 Introduction This chapter discusses the various practical applications for the laser modification of the wettability characteristics of the materials detailed in the preceding chapters. Firstly, the Al2O3/SiO2-based composite material described in Chapter 3, along with a vitreous enamel, has been employed to create a new grout material for the sealing of ceramic tiles. Secondly, to prolong the service life of OPC and also to generate a unique surface not demonstrated previously, HPDL radiation was used to alter the wettability characteristics of the OPC such that a vitreous enamel layer could be fired on the surface. Finally, laser radiation has been used to alter the wettability characteristics of the common engineering carbon steel described in Chapter 5 in order to facilitate the hitherto impossible task of enamelling the carbon steel in normal atmospheric conditions without pre-treatment of the steel. In addition, it is believed that the work detailed in Chapter 6, which described the use of laser radiation to enhance the wettability characteristics of a number of bio-compatible polymer materials (PE and PMMA), will prove useful in the medical field. Here it is believed that the laser-induced changes in the wettability characteristics of the polymers will aid

119


patient recovery and improve the effectiveness of implant surgery. Consequently this aspect of the use of lasers for the modification of the wettability characteristics of materials is a field of ongoing research for the authors. Similarly, further research is also being conducted to study a variety of practical applications for the laser enhancement of the wettability characteristics of not only the ceramic materials detailed in Chapter 4, but many other types of ceramics.

7.2 A two-stage ceramic tile grout sealing process 7.2.1 Application background Ceramic tiles are applied to the walls and floors in a multitude of places, from hospital operating theatres to industrial clean-rooms. Currently, tiles are applied to surfaces using either tile grouts or adhesive, with tile grout (typically epoxy based) or silicon resin being used to fill the void between adjoining tiles. A major difficulty with tiled surfaces is that contaminants can enter into, and exit a space via a tiled surface, through the tile grouts used to fill the void between adjoining tiles. In the case of silicon resin sealants, contamination can pass around the edges of the seal because the sealant is not absolutely bonded to the ceramic tiles due the presence of small voids that increase in size over time along the sealant/tile interface. The problem is compounded, as a result of the tile grout's porosity, by water, germs, and other harmful agents, which have the ability to permeate into cavities behind the ceramic tiles, corroding the bonding agent used to fix the tile to the substrate, the substrate itself, or even the ceramic tile. Moreover, the predominant problem with commercially available tile grouts is that over time they become contaminated, and have to be either removed physically or mechanically, or directly tiled over; an arduous, ineffective and costly undertaking that is not only time consuming, but often results in damage to the ceramic tiles. A possible solution to these problems would be to use a laser to produce an impervious surface glaze on a vitrifiable substance or compound placed in the void between adjoining vitrified ceramic tiles, thereby sealing the ceramic tiles without damaging them through exposure to heat. The propensity of a material to be readily laser vitrified and glazed is dependent upon the material possessing all of the following characteristics: (a) enough vitrifiable elements within its composition to become vitrified when irradiated; (b) high thermal conductivity to cope with the rapid heating/cooling cycle; and (c) similar thermal properties to those of the vitrified ceramic tiles. To ensure that any seals produced are completely functional, the post-process glazed surface layer of the seal must satisfy the following criteria: 1. The glazed surface layer of the seal must bond completely with the edges of the vitrified ceramic tiles without cracking, spalling or pitting. 2. The glazed surface layer of the seal must be free of cracks, blistering and porosities. 3. It must be capable of preventing water, germs, and/or other harmful agents from permeating through its surface. That is, it must be amorphous.

120


4. It must be chemically resistant to any harmful agents likely to be present within the working environment. That is, it must be amorphous. 5. It must be resistant to any thermal and/or physical forces likely to be present within the working environment. 6. The glazed surface layer of the seal and the substrate beneath must be long lived and must not deteriorate over a short time. As one can see from the above list of criteria, many immediately exclude all polymeric and polycrystalline materials. Metals are also eliminated because of the incompatibility of their thermal characteristics. As such, the only possible alternatives are a range of inorganic ceramic materials. Consequently, a new ceramic tile grout was created from an amalgamation of a number of selected oxides, mainly A12O3 and SiO2, which was termed the amalgamated oxide compound grout (AOCG). However, as was shown in Chapter 3, the seals generated using the AOCG were of a polycrystalline nature, even after HPDL treatment. As such, they would be unable to function correctly and therefore would not satisfy the criteria detailed above. In particular, they would allow water, germs, or other harmful agents to permeate through their surface. Previous research by other workers has indicated that certain glasses, in particular vitreous enamels, would be most suitable for use as a laser-glazed coating material. Indeed, the deposition of thin silica coatings on to metallic substrates using a CO2 laser has been shown to be feasible [182], while, again using a CO2 laser, vitreous enamel frits have been fired successfully on to steel [183, 184] and glass [185-187] substrates. Yet in all these previous studies, pre- and post-heating of the enamels in a furnace to temperatures in the range of the enamel melting temperature was necessary in order to relieve thermal stresses and avoid microcracking. Table 7.1 Thermal characteristics of selected glasses and a typical vitreous enamel

Thermal/mechanical property Softening point (°C) Thermal expansion coefficient (K~1) Viscosity (P) Poisson's ratio Young's modulus (MPa)

Glass/enamel type Pyrex Enamel Soda-LimeSilica 780 510 700 33 x 10~7 94 x 10-7 92 x 10-7 109 1010 108 0.176 0.189 0.220 5.21 x 10-4 6.42 x 10-4 7.70 x 10-4

As Table 7.1 shows, the thermal characteristics of many vitreous enamels make them inherently more suited to laser glazing. This is due in particular to the relatively low softening temperature and viscosity as well as the relatively high Young's modulus. However, the difficulties experienced previously [112] during the laser glazing of aluminoborosilicate glass are still apparent when employing a vitreous enamel. Namely, that a complete seal cannot be generated from the enamel alone, since the thermal stresses generated are too great and complete glazing throughout the section does not occur due to the relatively high speed of the laser-glazing process.

121


So as to generate satisfactory seals using a vitreous enamel, it is essential that the vitreous enamel possesses the following characteristics: (a) a relatively low softening temperature and high thermal conductivity to enable it to cope with the rapid heating/cooling cycle; (b) low viscosity when melted to allow easy flow; and (c) similar thermal properties, e.g. thermal expansion coefficient, to those of the vitrified ceramic tiles. To ensure that any seals produced with the vitreous enamel are completely functional, the post-process laser-glazed amorphous surface must satisfy, as well as the criteria detailed previously, the following: 1. The glazed surface layer of the seal must bond completely with the edges of the vitrified ceramic tiles without cracking, spalling, or pitting. 2. The glazed surface layer of the seal must be free of cracks, blistering, and porosities. 7.2.2 Sealing process development To overcome the inherent problems of the polycrystalline nature of the AOCG seal, an amorphous surface layer on top of the AOCG was produced by applying a commercially available enamel frit on to the AOCG surface, and then laser firing the enamel frit in order to vitrify it and thus seal the surface. Figure 7.1 illustrates the four distinct steps involved in the two-stage ceramic tile grout seal development.

Fig. 7.1 Schematic diagram of the four steps involved in the two-stage ceramic tile grout sealing process

122


As Fig. 7.1 shows, the sealing process is achieved in two stages, thus the seal itself consisted of two distinct components: an AOCG substrate and a glazed enamel surface, with the aim being that the AOCG substrate provides a tough bulk substrate which simply acts as a filler for the void between adjoining tiles, while the enamel provides an impervious surface glaze. In Step 1, UK standard 150 x 150 x 5 mm3 vitrified ceramic tiles of various colours: white, navy blue, leaf green and jet black, were cut into smaller pieces, 20 x 20 mm2, for experimental purposes and applied in pairs to an OPC substrate using standard epoxy tile grout (Vallance Limited). The spacing between the vitrified edges of each tile pair was the industry recommended 1.5 mm. The fixed ceramic tile pieces were then allowed to set for the standard setting time of 12 h. The tiles were then sealed with the first stage of the sealing technique by applying the AOCG into the void between adjoining ceramic tiles, flush to the surface of the tiles. The AOCG was then allowed to cure for 8 h. In Step 2 the set compound was then irradiated using the HPDL as shown in Fig. 7.1, and overlaid directly with a 250 um layer of the commercially available enamel frit (Ferro Limited) which, in order to form a manageable paste, was mixed with 20wt% white spirit (Step 3). The composition of the enamel is as described in Chapter 3. The main mechanical and thermal properties of the enamel and the vitrified surface of the ceramic tiles used are detailed in Table 7.2. The enamel frit paste was allowed to cure for 1-2 h and then irradiated immediately with the defocused HPDL beam (Step 4). Table 7.2 Principal mechanical and thermal properties of the seal enamel and the vitrified surface of the ceramic tiles [112]

Mechanical/thermal property Density (g/ml) Hardness (Hv) Softening point (°C) Coefficient of thermal expansion

Thermal conductivity (W/mK) Specific heat (J/kg °C, @ 0 °C)

Seal enamel 2.5-2.6 590 510 94 x 10-7 0.99-1.30 840

Vitrified ceramic tile 2.2-2.6 570 1100 89 x 10-* 1.14-1.36 880

7.2.3 Seal characterization and performance testing To fully characterize the two-stage enamel seals it was necessary to conduct a series of detailed tests and analyses. This testing and analysis focused chiefly upon the general effects of the laser interaction, the resultant physical and mechanical properties of the seals, the corrosion resistance characteristics, and the resultant microstructures. The mechanical and chemical testing of the laser-generated seals was carried out in conjunction with testing on existing epoxy tile grouts so as to determine, and ultimately compare, the physical properties of both materials. Current British and International standards, in relation to tile grout, are concerned only with water absorption and compressive strength, while for actual ceramic tiles standard tests exist for the determination of water absorption and chemical resistance only. Consequently, it was not possible to test the laser-generated seals according to, and strictly adhering to, established tests. As such, wherever possible tests based on current standards were developed to investigate specific aspects of particular relevance to the

123


laser-generated seals, namely the pull-off strength, the surface roughness, the rupture strength, the wear resistance, the permeability characteristics (water) and the corrosion resistance. 7.2.3.1 General effects and observations of high-power diode laser interaction Effects of power density variations Power density variations were observed to have a significant effect upon the surface morphology of the enamel laser glazes. A series of experiments were conducted with an O2 shroud gas for a range of power densities and traverse speeds. A minimum power density level of around 1 kW/cm2 was observed, below which incomplete glazing of the enamel occurred, regardless of the traverse speed. A typical example of a relatively low power density (1 kW/cm2) laser glaze is the top track shown in Fig. 7.2. As can be seen, laser interaction at this level resulted in a seal which appears only partially vitrified. At a relatively medium power density (2 kW/cm2), however, the quality of the enamel surface glaze on the enamel was much improved. A typical example of a medium power density enamel surface glaze is the centre track shown in Fig. 7.2. Here complete vitrification of the enamel has occurred, with the surface displaying very few microcracks and no porosities. The bottom track shown in Fig. 7.2 shows an example characteristic of glazing within the relatively high power density range (3 kW/cm2). As can be seen, the quality of the enamel surface glaze is extremely poor, displaying many large microcracks and porosities. However, both the microcracks and the porosities appear to be regular in both periodicity and intensity.

Fig. 7.2 Effects of laser interaction on enamel surface glazes at various laser power densities

124


Effects of traverse speed variations As with power density variations, changes to the traverse speed had a significant effect upon the surface morphology of the enamel glaze. A series of experiments were conducted with an 02 shroud gas for a wide range of traverse speeds and power densities. Typical results obtained at 2 kW/cm2 for relatively low (480 mm/min) traverse speeds are shown in Fig. 7.3.

Fig. 7.3 Effects of laser interaction on enamel surface glazes at various traverse speeds

From the experiments it was observed that typically, at relatively low and high traverse speeds, the surface condition of the enamel glazes was unacceptable. As the tracks on the top and bottom of Fig. 7.3 show, the glazed surfaces displayed many large porosities and microcracks. At traverse speeds in the range of 180-420 mm/min, however, good quality surface glazes on the enamel could be generated which displayed neither porosities nor microcracks (see Fig. 7.3, centre track). Operating window The energy density incident upon the enamel surface at any moment is a function of both the power density and the traverse speed. Therefore, if either one is altered individually, then such a change can be compensated for through an alteration of the other. As such, it is possible by means of precisely structured and accurate experimentation to determine the exact operating window for the laser glazing of the enamel (Fig. 7.4). Within these conditions good-quality surface glazes displaying neither microcracks nor porosities could be generated.

125


Fig. 7.4 Schematic diagram of the operating window for laser surface treatment of the enamel

7.2.3.2 Pull-off strength testing Experimental procedure In order to test the strength of the bond between the laser-generated enamel seal and the laser-treated AOCG and ceramic tile surface, as well as the strength of the lasertreated AOCG, pull-off tests were conducted as shown in Fig. 7.5.

Fig. 7.5 Schematic diagram of the experimental set-up for the pull-off tests

For the tests the AOCG, enamel/AOCG and the enamel/ceramic tile were prepared as large-area samples (10 x 10 mm2). For the tests on the complete seal the ceramic tiles were cut into smaller pieces (20 x 20 mm2) and applied in pairs using an Araldite epoxy to an OPC substrate and allowed to cure for 24 h. The spacing between the

126


vitrified edges of each tile pair was the industry recommended 1.5 mm. The samples were then laser sealed as described earlier. High-tensile aluminium test dollies of 5 mm diameter were then attached to the AOCG and enamel surfaces and to the axially opposite substrate surfaces using Araldite epoxy; they were then left to cure for 24 h. In order to ensure axial accuracy, which is essential for true results, the test dollies were set in position using identical V-blocks. As shown in Fig. 7.6, the samples were placed into an Instron 4507 tensile/compressive test rig by mounting the test dollies into the jaws of the rig. A tensile force was then applied until failure, with the energy being recorded. Results As Fig. 7.6 shows, the results obtained for the AOCG varied markedly with changes in the laser operating parameters. A post-test analysis of the samples showed that the material failed approximately 150 um below the laser-treated surface within the HAZ of the AOCG. A similar observation was made by Sugimoto et al. [122] during laserglazing work on concrete. An optical analysis of the detached surfaces showed that the enamel had detached cleanly and completely from both the AOCG and the ceramic tile at the interface.

Fig. 7.6 Relationship between pull-off strength of laser-treated AOCG layer and laser operating parameters

For the enamel/AOCG and the enamel/ceramic tile the average bond strength was recorded as 60 and 45 MPa respectively. The results obtained from the tests showed little variation within the optimum laser operating parameters, indicating that neither the power density nor the traverse speed influenced the bond strength of the enamel.

127


7.2.3.3 Rupture strength testing Experimental procedure Tests were conducted to determine the rupture strength of the enamel glaze when comprising a complete two-stage seal. Additionally, for comparison purposes, rupture strength tests were carried out on the vitreous glaze on the surface of a ceramic tile. The two-stage enamel seal test samples were prepared as described above. The samples were placed on to the sample stage of an Instron 4507 tensile/compressive test rig, as shown in Fig. 7.7, and then subjected to a compressive rupture force until the enamel seal failed (cracked), with the energy being automatically recorded. The rupture force was applied by means of a high-tensile steel indentor with a 1 mm radius point.

Fig. 7.7 Schematic diagram of the experimental set-up for the rupture tests

Results The results of the tests are summarized in Table 7.3. As can be seen there was little variation between the average rupture strength of the enamel seal and the vitreous tile glaze, 2.6 and 3.0 J respectively. However, the recorded rupture strengths of enamel seal ranged from 2.3 to 2.7 J, while those obtained for the vitreous tile glaze ranged from only 2.9 to 3.0 J. This relatively large difference in the range of the recorded rupture strengths is probably due mainly to the shape of the tile edges; since the enamel seal when laser fired naturally assumes a concave surface geometry. Thus, the strains within the enamel layer are higher, therefore reducing the strength (compared with an enamel seal with a flat surface profile) by some 40-50 per cent [95]. Furthermore, it is well established that substrate thickness has a significant effect upon the rupture strength of an enamel coating [95]. Consequently, because the thickness of the AOCG was not controlled as accurately as the thickness of the bulk ceramic tile material, slight variations in the recorded rupture strength are perhaps to be expected.

128

Practical Applications of Lasers for the Modification of Wettability Characteristics Table 7.3 Rupture strength measurements for ceramic tile and laser-generated enamel seal

Material Ceramic tile (vitreous glaze) Two-stage seal (enamel glaze)

Rupture strength (J) Range Average 2.9-3.0 3.0 2.3-2.7 2.6

In any system, such as the two-stage ceramic tile grout seal, where an enamel coating is placed on to a solid substrate, the rupture strength of the coating is, among other things, directly influenced by the compressive strength of the substrate. A series of experiments was therefore conducted to determine the effects of HPDL interaction with the AOCG, in particular with regard to temperature. This was achieved by heating a number of AOCG samples in regular steps up to 800 °C. The cold compressive strength of the samples was than obtained as depicted in Fig. 7.7. Additionally, to give a practical reflection of the AOCG thermal characteristics in terms of compressive strength, a number of commercially available epoxy tile grout samples were also tested in the same manner. The cold compressive strength values obtained for both the AOCG and the epoxy tile grout are shown in Fig. 7.8.

Fig. 7.8 Cold compressive strength behaviour with increased temperature of the AOCG and conventional epoxy tile grout As can be seen from Fig. 7.8, as the temperature is increased up to 150 °C the compressive strength of the AOCG increases to a maximum of 12.4 MPa. Once the temperature exceeds 150 °C, however, the compressive strength of the AOCG decreases to 8.8 MPa. After this point the compressive strength decreases uniformly to a minimum of 8.0 MPa at 600 °C. However, at around 700 °C an interim maximum in compressive strength of 8.9 MPa was observed. The maximum compressive strength of 12.4 MPa for the AOCG compares favourably with 10.5 MPa for the conventionally

129


available, waterproof, ready-mixed, coloured tile grout at room temperature [188]. However, the general thermal resistance properties of the AOCG were found to be far superior to those of the waterproof, ready-mixed, coloured tile grout. Even at a temperature of 800 °C no breakdown of the AOCG was observed, while it is known that waterproof, ready-mixed, coloured tile grout loses its integrity at temperatures above 200 °C [189]. Moreover, at temperatures above approximately 100 °C the complete irreversibility of the water glass reaction, that is, the rehydration of the water glass when exposed to water, was achieved. 7.2.3.4 Wear analysis Experimental procedure The wear resistance of a material in general is determined primarily by the hardness of the material in comparison with the hardness of other materials with which it conies into contact [190]. However, wear resistance does not always increase with hardness [191]. Tests were therefore conducted in accordance with Lawrence and Li [25] in the manner shown in Fig. 7.9 to determine the exact difference in wear resistance between a conventional epoxy tile grout, the vitreous glaze on a ceramic tile surface, and the enamel surface of a two-stage seal itself. In the case of the epoxy tile grout and the enamel seal the test samples were prepared as described above, while for the ceramic tile the test piece was simply a 25 x 25 mm2 section. The test samples, in all cases, were then attached using a Loctite adhesive to an OPC substrate. The samples were then weighed and subjected to a friction force for 8 h, being removed from the machine and weighed at 2 h intervals.

Fig. 7.9 Schematic diagram of the experimental set-up for the wear tests

Results Usually the greater the hardness of a material, the higher its wear resistance. So, the greater hardness of the compositional components of the enamels used on ceramic tiles and for the sealing of ceramic tiles (principally SiO2, B2O3, Na2O, and Mn) in comparison with epoxy tile grouts (principally CaCO3 and dolomite) may account for improved wear resistance. Yet it is known that wear resistance does not always increase with hardness [191].

130


Fig. 7.10 Relationship between weight loss and friction time for the conventional epoxy tile grout, the vitreous glaze on a ceramic tile surface and an enamel seal

Figure 7.10 shows the relationship between weight loss and the friction time for all three test pieces. As can be seen, the wear resistance of the enamel seal is fractionally greater than that of the vitreous glaze on the ceramic tile surface. However, the enamel seal shows a considerable increase in wear resistance over the epoxy tile grout, with the weight loss being four times lower than for the epoxy tile grout after 4 h, and eight times lower after 8 h. 7.2.3.5 Water permeability testing Experimental procedure Perhaps the most important function of the enamel seal is its capability of preventing harmful agents from permeating through it. In order to test the permeability of the enamel, comparison experiments with conventional epoxy tile grout were conducted in terms of water permeability. The tests were based on BS 6906 [192]. For the experiment, the ceramic tiles were cut into smaller pieces (20 x 20 mm2), applied in pairs using a commercial tile grout to a ceramic tile substrate, and allowed to cure for 24 h. The spacing between the vitrified edges of each tile pair was the industry recommended 1.5 mm. The samples were then filled in the conventional way with tile grout or laser sealed as described above. A scaled glass tube 1.2 m long with an outside diameter of 12 mm and an inside diameter of 10 mm was luted and sealed completely on to the samples using a silicone sealer and a Loctite adhesive. The tubes were then filled with water to a height of 1 m so as to give a reasonable pressure head, and bunged to prevent any evaporation. Figure 7.11 details the experimental set-up. The whole set-up was then weighed. The experiments were carried out in a temperature-controlled room held at 15 °C ± 3 °C for 72 h.

131


Fig. 7.11 Schematic diagram of the experimental set-up for the water permeability tests

Results On completion of the experiments, measuring and weighing of the test equipment revealed that the conventional epoxy tile grout had an average water permeability of approximately 4.18 x 10~3 mg/h cm2, while the enamel seal displayed no measurable permeability. Such a result confirms that, not only is the enamel seal fully amorphous, but that since there are no cracks or porosities in the enamel glaze or the interface between the enamel glaze and the borosilicate glass tile surface, it is reasonable to assume that a continuous impervious surface has therefore been created across the surface of the tiles and the seal. 7.2.3.6 Surface roughness/cleanability analysis An important feature of the enamel seal is its surface roughness; since it is this that determines the cleanability of the tile surface as a whole. Using the Taylor-Hobson Surtonic 3+ surface texture measuring instrument, a series of measurements were taken on the surface of a layer of an epoxy grout (polished to ensure a measurement could be taken), the vitrified surface of a ceramic tile and the surface of a lasergenerated enamel seal. On each sample four measurements were taken in different positions and in different directions on the surface, with an average being taken. Table 7.4 summarizes the surface roughness (Ra) measurement results. Table 7.4 Surface roughness (Ra) measurements for epoxy tile grout, ceramic tile and laser-generated enamel seal

Surface Epoxy tile grout (polished) Ceramic tile (glazed surface) Enamel seal

Surface roughness (Ra) Range Average 3.83 urn 2.36-5.72 um 0.06 um 0.06 um 0.20-0.64 um 0.36 um

As Table 7.4 shows, the surface roughness of the enamel seal is many times less than that of a conventional epoxy grout even when polished. In ordinary operating conditions where the surface roughness of the epoxy grout is not polished, but is

132


determined by the means of application, the surface roughness was measured in excess of 30 um. Clearly, in this situation the surface roughness, and therefore the cleanability, of the enamel seal will be considerably better. In order to analyse the effects of laser treatment on the surface roughness of the enamel glazes, a series of experiments was conducted with an O2 shroud gas using a wide range of power densities and traverse speeds, thus allowing the effects of laser treatment to be determined in terms of both power density and traverse speed. The surface roughness values (Ra) are shown in Fig. 7.12.

Fig. 7.12 Relationship between enamel glaze surface roughness (Ra) and power density for various traverse speeds

7.2.3.7 Corrosion resistance characteristics Experimental procedure Tiled surfaces are often subjected to corrosive substances, either as part of the normal service environment and/or as a result of routine cleaning. Therefore corrosion resistance tests based upon BS 6431 [193] were conducted using nitric acid, sodium hydroxide and a detergent cleaner (Premier Products MP9). The experiments were carried out by releasing small droplets of the corrosive agents in the concentration ratios of 80, 60, 40, 20, and 10 per cent, respectively, on to the surface of the epoxy tile grout, the AOCG, and the enamel glaze at hourly intervals for 4 h. The droplets were released in a controlled manner on to the material surfaces from the tip of a micropipette, with the resultant volume of the drops being approximately 6 x 10~3 cm3. The samples were then examined optically and mechanically tested to determine the compressive strength and wear life characteristics. High concentrations of the various corrosive agents were used principally to accelerate the tests. However, in practice 60 per cent nitric acid is used within the nuclear processing industry as a solvent for

133


nuclear fuels [194], while within the food processing and brewing industries, tiled surfaces are washed repeatedly many times a day with detergent cleaners [195]. Effects on morphology and microstructure Figure 7.13 shows the surface condition of the epoxy tile grout before and after exposure to all three reagents at 80 per cent concentration.

Fig. 7.13 Surface condition of the epoxy tile grout (a) untreated and after exposure to (b) 80 per cent concentration nitric acid, (c) 80 per cent concentration sodium hydroxide alkali, and (d) 80 per cent concentration detergent

All three substances in the concentrations 80, 60, and 40 per cent were seen to attack immediately the epoxy tile grout surface, with the nitric acid and sodium hydroxide attacking with greater severity than the detergent, while the enamel glaze displayed no discernible microstructural changes or signs of devitrification due to corrosion. Moreover, it was seen that the AOCG displayed a similar resistance to reagent exposure as the enamel glaze. This similar corrosion resistance of the AOCG with the

134


enamel glaze indicates that a seal comprising the AOCG with a laser-glazed enamel coating could function correctly in situ, even if the enamel coating became damaged. Effects on compressive strength Tests conducted according to ASTM C579-91 [196] revealed that exposure of the epoxy tile grout to the reagents had a significant effect upon the compressive strength and the wear resistance of the epoxy grout. As Fig. 7.14 shows, exposure of the epoxy grout to nitric acid and sodium hydroxide in the concentrations 40-80 per cent resulted in an average loss of compressive strength of approximately 35-71 per cent. In the case of the detergent a significant loss in compressive strength only occurred with concentrations above 40 per cent. Here the average loss in compressive strength for concentrations in the range 60-80 per cent was approximately 15-30 per cent. This compares with no discernible difference in the compressive strength of the AOCG, and for the enamel seal, the rupture strength.

Fig. 7.14 Variation in compressive strength of epoxy tile grout with reagent type and reagent concentration

Effects on wear life To determine the effects of corrosion on the epoxy tile grout, the AOCG and the enamel glaze, wear tests were conducted as described above. For the AOCG and the enamel glaze it was observed that the wear life characteristics were not affected. However, the wear life characteristics of the epoxy tile grout were altered considerably. Figure 7.15 shows the variation in wear resistance of the epoxy tile grout when exposed to the reagents with an 80 per cent concentration.

135


Fig. 7.15 Relationship between weight loss and friction time for the epoxy tile grout with different reagent types at the maximum concentration (80 per cent)

As can be seen from Fig. 7.15, the wear resistance of the epoxy tile grout is particularly affected through interaction with the nitric acid and the sodium hydroxide. Here the weight loss was approximately five times higher than for the unexposed epoxy tile grout after 4 h, and approximately ten times higher after 8 h for both acids. In the case of the detergent the weight loss was twice as high as that recorded for the unexposed epoxy tile grout after 4 h, and five times higher after 8 h. 7.2.4 Discussion of results 7.2.4.1 Cracking and the formation mechanism The formation of cracks can be attributed mainly to thermal stresses generated during laser irradiation. This is due to the fact that the enamel has a low thermal conductivity, and as such, during laser heating a large thermal gradient between the melt zone and the substrate exists, which results in thermal stresses. Additionally, despite the fact that the process of laser firing the enamel frit results from a high specific rate of energy which in turn facilitates localized melting of the enamel frit, the fact that a certain amount of the heat will be conducted to sections of the seal where the enamel is already glazed, combined with the existence of a relatively cold AOCG substrate, means that thermal stresses will be generated. During the heating phase the stresses will be compressive and relieved by plastic deformation, thus precluding crack formation. At high temperatures (T> Tm) the stresses can also be relieved [197-199]. However, during cooling when the temperature falls below Tm, then stresses will accumulate. If the fracture strength of the material is exceeded, then cracking within the melted layer will occur. Within the optimum operating conditions the glazes produced with the enamel fired using the HPDL displayed no porosities and, depending upon the shroud gas used (see Fig. 7.4), few to no microcracks. The principal reason for the reduction, and even

136


elimination of microcracks is the reduction in the thermal gradient, AT, during laser irradiation. This is due partly to the fact that the softening point of the enamel frit is around 510 °C, much lower than many of the other materials tested. Therefore, the tensile stresses that result from the unrelieved elastic stresses that occur due to the contraction of the material between the softening point (510 °C) and ambient temperature (20 °C) are much reduced to a value below that of the fracture strength. The thermal stress, (7, induced by a thermal gradient can be calculated using the Kingery equation

where, Ey is Young's modulus, AT is the temperature change, B is the coefficient of thermal expansion and vp is Poisson's ratio. Aris the difference between the critical temperature (below which stresses can no longer be relieved) and ambient temperature. For Pyrex glass this is the difference between the softening point, 780 °C, and ambient temperature, 20 °C. Since the softening point of Pyrex glass is 780 °C [200], that is Pyrex glass can be plastically deformed at temperatures above 780 °C, thermal stresses arising during cooling from above 780 °C are relieved by plastic deformation. However, unrelieved elastic stresses result due to contraction occurring between 780 °C and ambient temperature, 20 °C. Thus for Pyrex glass A7 = 760 °C. So, by using the following values for Pyrex glass: Ey = 6.42 x 104 MN/m2, ft = 33 x 10"7 K"1, Ar = 760 °C and v = 0.176, when the Pyrex glass powder was irradiated by the laser beam the thermal stress produced according to equation (7.1) is 185 MN/m2. Since this is in excess of the fracture strength of Pyrex glass, 120 MN/m2 [200], cracking will occur and can only be avoided by severe distortion or through the reduction of Ar by pre-heating. For the enamel, however, the softening temperature is 510 °C, therefore A! is only 480 °C. Also, Ey = 6.25 x 104 MN/m2, /3 = 33 x 10"7 K"1 and v = 0.162 [201]. Again, using equation (7.1) the thermal stress induced in the enamel during HPDL irradiation is calculated to be 118 MN/m2, which is below the fracture strength of the enamel, 135 MN/m2 [201]. As such, cracking will not occur, making any preheating completely unnecessary. As can be seen from Fig. 7.16, under the extreme laser processing conditions of relatively high power density (3.25 kW/cm2) and slow traverse speed (120 mm/min) with an Ar shroud gas, microcracking within the enamel glaze occurred both parallel and perpendicular to the surface. Such findings are in accord with the nature of the tensile stresses produced during cooling of the solidifying layer in terms of the glazed layer section thickness. During cooling of a relatively thin glazed layer, the temperature is almost uniform across the section, therefore it only experiences a twodimensional stress at the surface. However, for relatively thick section glazes, as is the case in question, the temperature gradient across the depth is present along with the gradients at the surface. Consequently the three-dimensional nature of the stresses produces microcracks that are both perpendicular and parallel to the surface [199, 202].

137


Fig. 7.16 SEM cross-section view of enamel glaze on laser-treated AOCG showing mic roc racks parallel and perpendicular to the surface

7.2A.2 Surface rippling and the formation mechanism Within the laser-generated meltpool of the AOCG there are a great many forces acting, as Fig. 7.17 shows.

Fig. 7.17 Schematic representation of the forces acting on the melt pool

One of the largest forces acting on the melt pool is the surface shear force, which results from the variation in surface tension due to the steep thermal gradients within the melt pool and is given by [35]

Also, as a consequence of the differences in temperature between different regions of the melt pool, and hence differences in surface tension, the melt pool is subject to very strong thermocapillary convection or Marangoni mixing forces [203] that promote compositional homogeneity. Indeed, Chan et al. [204] have modelled the flow within the melt pool by solving the Navier-Stokes equation together with the heat flow equations and speculate that the melt pool rotates approximately five times before

138


solidification. Moreover, thermocapillary convection is the single most important factor influencing the geometry and surface characteristics of the meltpool [205-207]. Indeed, as a result of such forces acting within the meltpool, the undulating surface ripples observed on the surface of the enamel seal are generated. 7.2.4.3 Porosities and the formation mechanism Within the optimum processing conditions no porosities were observed on the glazed enamel surface. Yet porosities were a common feature when processing outside the optimum conditions, as Fig. 7.18 shows, with the porosities varying in size from microscopic pits to large craters depending upon the laser operating parameters.

Fig. 7.18 Optical view of porosities formed on the surface of the enamel glaze with (a) relatively low laser energy density and (b) relatively high laser energy density

In the case of the enamel, for all instances of porosity formation the mechanism behind their development is the outcome of gas escaping from within the melt and disrupting the surface [208]. With regard to the enamel, the gas is likely to be O2, formed and released during the vitrification reaction [209]. If the laser energy density incident on the enamel is too low, then the generated O2 cannot escape from the molten enamel surface easily because of the high viscosity of the melt. As such, when the O2 eventually does penetrate the melt surface, the resultant porosity is not filled by the flow of the melt; since the insufficient energy density is unable to maintain a high enough temperature for an adequate length of time and thus decrease the overall viscosity of the melt [202]. In this case the porosities formed are typically small and shallow, being regular in both periodicity and intensity [Fig. 7.18 (a)]. On the other hand, if the laser energy density incident on the enamel is too high, then boiling of the surface may occur. At the same time, an increase in 02 formation may occur within the melt. These individual pockets of O2 formation may combine and rise to the surface of the melt. Once the energy density decreases (as the laser traverses away), then the additional 02 will attempt to escape from the molten surface [Fig. 7.18(b)]. However, as Fig. 7.19 shows schematically, the solidifying melt will

139


prevent this, causing bubbles to form (a). The excessive 02 gas pressure will firstly cause the bubbles to expand (b) and ultimately rupture the walls of the bubbles (c), creating a sharp 'knife-edge' porosity [117, 123]. These types of porosity are usually large, deep, and randomly spaced.

Fig. 7.19 Illustration of the large 'knife-edge' porosity formation

7.2.4.4 The effects of different shroud gases on the enamel glaze condition Because of its open structure, it is possible for gases to dissolve molecularly in glass, and, if the gas molecule is small enough, it can diffuse rapidly in a simple glass such as fused silica [200]. Indeed, gases such as H2 and O2 are known to dissolve molecularly in glass and can also react with the glass network [200]. As Fig. 7.20 shows, the quality of the enamel glazes produced, in terms of smoothness, porosity, and microcracks, was greatly influenced by the type of shroud gas employed. When using both (a) Ar and (b) 62 shroud gases the surfaces appear undulated and display very few porosities. However, the crack density in the Ar shroud gas sample was found to be higher than that of the sample treated with an 02 shield gas.

Fig. 7.20 Typical SEM surface images of the enamel glaze, (a) Ar shroud gas and (b) O2 shroud gas

140


Fig. 7.21 Best-fit cooling curves for the laser-glazed enamel when Ar and O2 shroud gases are employed

The use of O2 as the shroud gas significantly reduced the number of microcracks and porosities within the enamel glaze as well as producing a much smoother surface. This indicates that the O2 interacted with the glass network increasing the heat generation and subsequently the fluidity of the melt. In contrast, the Ar did not interact with the glass network and was consequently trapped within the more viscous melt in the form of bubbles. As such, when employing O2, inherent gas bubbles generated during vitrification of the enamel escaped from the melt more readily due to its lower viscosity, thus reducing the number of porosities and microcracks. Indeed, as the cooling curves for the enamel shown in Fig. 7.21 indicate, when the laser beam has been removed, the cooling of the enamel is faster when an Ar shroud gas rather than an O2 shroud gas is employed. Therefore, the length of time that the enamel is of a sufficient fluidity to allow the generated gas bubbles to escape the surface easily is much reduced, resulting in porosities and microcracks. 7.2.4.5 Stability to devitrification of the enamel glaze HPDL interaction with the enamel basically fires the enamel frit, vitrifying it and thus generating, in effect, a glass coating. As Fig. 7.22 shows, a cross-sectional SEM analysis of the enamel glaze revealed there to be no crystals within the glaze, while the XRD analysis of the enamel glaze shown in Fig. 7.23 confirmed it to be completely amorphous.

141


Fig. 7.22 Typical SEM cross-section view of the enamel glaze/AOCG interface showing the absence of any crystallization in the enamel glaze

Fig. 7.23 XRD analysis of the laser-glazed enamel

Notwithstanding this, it is possible for the enamel glaze to become crystalline through the destruction of the glassy state by means of a process known as devitrification. Devitrification can occur in two ways. The first is the breakdown of the glass surface by corrosion or weathering. But, as was seen earlier, exposure of the enamel glaze to extremely harsh reagents did not cause the glaze to devitrify. The second, considered by many as the only true form of devitrification, is when the overall composition remains unchanged but crystals separate in the glassy medium, destroying the glassy state. The morphology of the crystals formed varies from spherical to highly dendritic [210]. Glass can be regarded as a randomly oriented system of silicon-oxygen tetrahedra lacking periodicity and symmetry, and having an energy content a little greater than the corresponding crystal form. Therefore, if certain favourable conditions are in place, then this metastable glassy phase may re-orient its network to the conditions of lowest energy and crystals will be deposited. These favourable conditions can be generalized as being dependent upon temperature and composition.

142


A sufficiently high temperature is one of the fundamental conditions for crystallization, since it is essential in order for the movement of atoms to allow orientation and the presence of crystallization centres. Such centres occur usually at the glass/air boundary, around a porosity or at the surface of an entrapped bubble and, although not all instances of devitrification are due to the presence of bubbles or porosities, in nearly all cases of devitrification such features will be found [190]. Clearly, as was seen earlier, within the optimum operating conditions neither porosities or bubbles were formed within the enamel glaze. Thus the possibility of devitrification of the enamel glaze due to temperature-induced exploitation of crystallization centres is significantly reduced. The composition of the enamel also played an important part in the stability to devitrification of the glaze. In particular it is known that Pb lends glasses a high resistance to devitrification, even in long periods at the most favourable temperatures [211], while compounds such as A12O3 and MgO are also known to be very useful in assuaging devitrification problems [190]. This is because the inclusion of these elements within the enamel composition creates an enamel without a high liquidus temperature and therefore a reduced tendency towards devitrification [190]. In contrast, high softening point enamels containing elements such as CaO and Mo tend to devitrify rather easily [190]. Indeed Mo has been seen to be active in the devitrification process, playing an important role in phase selection during the crystallization [212]. Of great importance in providing stability towards devitrification is the composition of the substrate on to which the enamel is fired. This is because when most enamels are fired, the enamel used to generate the two-stage seals included, some degree of diffusion across the interfaces occurs. As such, the surface elements within a substrate come into play significantly and even if an enamel is not given easily to devitrification, the substrate can cause some degree of devitrification at the interface [213]. 7.2.4.6 Wear life characteristics The mechanical and chemical tests conducted showed clearly that the two-stage enamel seal significantly out-performed the conventional epoxy tile grout in all the test areas. Indeed, in many instances the performance of the two-stage enamel seal approached, and occasionally surpassed, that of the ceramic tiles themselves. Moreover, the superior mechanical and chemical performance of the two-stage enamel seal over conventional epoxy grout suggests that the life characteristics of the seal are also superior to those of conventional epoxy grout. This was especially true in the case of chemical resistance and water permeability, where the enamel seal proved to be resistant to both. This marked variation in corrosion resistance is due to the difference in composition of the epoxy tile grout and the AOCG and enamel. The epoxy tile grout consists largely of limestone, dolomite, and organic epoxy, which are readily attacked by acids. In contrast, the inherently relatively high contents of SiO2 and B2O3 in the AOCG and enamel composition ensure an increase in acid resistance [95]. The difference in permeability performance is due to the contrasts in structure of the epoxy tile grout and

143


the enamel seal of the two-stage ceramic tile grout seal; with the epoxy tile grout comprising a porous polycrystalline structure, while the structure of the enamel seal is of a dense amorphous nature. However, the in situ relative thickness of the epoxy tile grout and the enamel coating must be considered in order to give a true interpretation of the actual wear life characteristics, particularly when considering the wear resistance (with and without exposure to corrosive chemical agents). The increase in wear life of the two-stage enamel seal over the epoxy tile grout is defined as

where the wear life is determined from

Table 7.5 summarizes the wear rate details and the nominal life increase of the enamel seal over the conventional epoxy tile grout. As is evident from the table, the enamel surface of the two-stage ceramic tile grout seal gave an increase in actual life over the epoxy tile grout regardless of the environment. However, as can be seen, the increase in actual life of the enamel seal over the epoxy tile grout varies markedly depending upon the working environment. Nonetheless, the most common working environment for the two-stage enamel seal would involve contact with detergent acids, therefore yielding significant economic savings since such a tiled surface lasts around ten times longer than one sealed with conventional epoxy tile grout. Table 7.5 Wear rate details and the nominal life increase of the two-stage enamel seal over conventional epoxy tile grout in various corrosive environments

Epoxy tile grout Two-stage enamel seal Wear life (x100%)

Density 2180 kg/m3 2650 kg/m3 -

Wear rate (mg/cm2/h) Thickness Normal Detergent NaOH 53.8 96.3 2000 um 12.3 1.3 1.3 500 um 1.3 2.9

13.1

23.4

HN03 125 1.3 30.4

7.2.5 Summary To determine the possibility of producing a laser-induced impervious surface glaze on a vitreous material placed in the void between adjoining ceramic tiles, thus sealing and joining the tiles together permanently, a two-stage process has been developed using a new grout material which consists of two distinct components: (1) an AOCG substrate and (2) a HPDL-glazed enamel surface. The AOCG provides a tough, heat resistant bulk substrate, while the enamel provides an impervious and cleanable surface. Tiles were successfully sealed with power densities as low as 500 W/cm2 and at rates of up to 600 mm/min.

144


The HPDL two-stage ceramic tile grout sealing technique was shown to result in the generation of crack- and porosity-free seals, produced without the need for special atmospheric conditions or pre-heating, owing to: 1. The elimination of microcracks as a result of the enamel possessing thermal properties such that the ensuing thermal gradient from HPDL firing was reduced to such a level that thermal stresses in excess of the fracture strength were avoided. 2. The elimination of porosities resulting from an understanding of the basic porosity formation mechanisms, and accordingly establishing the appropriate laser operating parameters so as to avoid conditions favourable to porosity formation. From the results of the detailed experimental work presented in this chapter, it was seen that the generation of the enamel surface glaze resulted in a seal with improved mechanical and chemical properties over conventional epoxy tile grouts. Both epoxy tile grout and HPDL-generated enamel seals were tested for: (i) bond strength; (ii) rupture/impact strength; (iii) surface roughness; (iv) water permeability; (v) acid/alkali resistance and (vi) wear/life characteristics. The enamel seal showed clear improvements in strength, surface roughness, and wear/life characteristics, while being impermeable to water, and being resistant (up to 80 per cent concentration) to nitric acid, sodium hydroxide, and detergent acids. The mechanical and chemical testing of the seals revealed: 1. For the enamel/AOCG and the enamel/ceramic tile the average bond strengths were recorded as 60 and 45 MPa respectively. 2. The rupture strength of the enamel glaze was found to be comparable with that of the ceramic tiles themselves, with the average rupture strength of the enamel glaze being recorded as 2.6 J, while the average rupture strength of the ceramic tile surface was 3.0 J. 3. The average surface roughness of the seals and the tiles was 0.36 and 0.06 um respectively, while for the conventional epoxy tile grout the average surface roughness when polished was 3.83 um, and in excess of 30 um without polishing. 4. Optical inspection and water permeability tests revealed that a complete bond between the amorphous enamel crack- and porosity-free glaze and the tiles, as well as the laser-treated AOCG substrate, was achieved. 5. Both the ceramic tiles and the enamel glaze showed complete resistance to 80 per cent concentration nitric acid, sodium hydroxide, and detergent acids. In contrast, conventional epoxy tile grout displayed little resistance to any of the reagents. 6. Life assessment testing revealed that enamel seals had an increase in actual wear life of 2.9 to 30.4 times over conventional epoxy tile grout, depending upon the corrosive environment. Clearly, the economic and material benefits to be gained from the deployment of such an effective and efficient method of tile sealing could be significant.

145


7.3 The enamelling of ordinary Portland cement by means of high-power diode laser radiation The unique characteristics of lasers furnishes them with the ability to be employed for the non-contact processing of materials which are otherwise difficult to process. Concrete is one such material since it is a composite, consisting of an array of fine and coarse aggregate pieces embedded within an OPC matrix. Consequently, the processing and surface treatment of concrete can be extremely laborious. This section describes the utilization of HPDL to produce for the first time an enamelled glaze on the 'as-cast' OPC surface of concrete and the effects thereof on the OPC's mechanical, chemical, and physical properties. Details of the OPC and enamel composition can be found in Chapter 3. The concrete sample blocks and the enamel frit were irradiated using the defocused high-order mode HPDL beam with a beam spot diameter of 2-5 mm and laser powers (measured at the workpiece using a Power Wizard power meter) of 20-100 W. The defocused laser beam was fired across the surfaces of the concrete samples by traversing the samples beneath the beam using the x- and y-axis of the CNC gantry table at speeds ranging from 60-600 mm/min (see Fig. 3.1). The laser optics were protected by means of a coaxially blown Oi shield gas jet at a rate of 5 1/min. To determine the characteristics of the HPDL-generated glazes on the treated OPC surface of concrete and the HPDLfired enamel, the samples were examined using optical microscopy, SEM, EDX, and XRD techniques. 7.3.1 General results of high-power diode laser enamelling 7.3.1.1 Operating window Figure 7.24 schematically illustrates the HPDL OPC surface of concrete glazing and concrete enamelling operating windows in terms of traverse speed and power density. Within the optimum operating conditions good-quality OPC and enamel glazes displaying few porosities and microcracks could be produced. Furthermore, from Fig. 7.24 it is possible to ascertain the maximum enamelling rate that it may be possible to achieve using the HPDL. This was calculated as being 0.34 m2/h for a circular beam of 5 mm diameter with a laser power of 100 W and a traverse speed of 720 mm/min.

146


Fig. 7.24 Schematic representation of the operating window for the concrete glazing and enamelling process using the 120 W HPDL

7.3.1.2 High-power diode laser-fired enamel glaze It was observed that, prior to laser irradiation, it was not possible to fire the enamel on to the OPC surface of concrete. Indeed, HPDL interaction with the enamel when placed on the untreated OPC surface simply resulted in the 'balling' of the enamel (the formation of small spheres approximately the diameter of the laser beam itself). Such observations are in accord with those of Bourell et al. [214] and Agarwala et al. [215], who noted the balling phenomena during laser sintering work of silica-based materials. After HPDL surface treatment of the OPC surface, however, it was possible to fire the enamel directly on to the OPC surface. The mechanism of this phenomenon is based entirely on the wettability characteristics of the OPC surface (see Chapter 3).

Fig. 7.25 Typical optical surface morphology of the HPDL-fired enamel glaze on HPDL-treated OPC (1.75 kW/cm2 power density, 360 mm/min traverse speed)

147


Fig. 7.26 Typical SEM micrograph of the cross-section of the HPDL-fired enamel glaze on HPDL-treated OPC (1.75 kW/cm2 power density, 360 nim/min traverse speed)

The typical surface morphology of the HPDL-fired enamel glaze produced on the HPDL-treated OPC surface of concrete is shown in Fig. 7.25. As can be seen, neither crack nor porosity formation were discernible on the enamel glaze. This is brought about as a result of the thermal characteristics of the enamel itself being conducive to firing by means of HPDL radiation (as discussed earlier). An XRD analysis of the HPDL-fired enamel glaze revealed it to be fully amorphous. The typical crosssectional view of the HPDL-fired enamel glaze produced on the HPDL-treated OPC surface of concrete is shown in Fig. 7.26. In this micrograph, the enamel glaze appears to be well bonded to the HPDL-treated OPC surface of the concrete. Surface roughness measurements of the surfaces of the untreated OPC, the HPDLtreated OPC, and the HPDL-fired enamel glaze were carried out. The investigation revealed that the average surface roughness of the HPDL-fired enamel glaze was 0.21 um. This compares with an average of 21.91 um for the untreated OPC surface and 2.88 um for the HPDL-treated OPC surface. Clearly, this considerable improvement in the surface roughness of the OPC occasioned by HPDL enamelling makes the surface that much easier to clean and maintain. 7.3.2 Mechanical, physical, and chemical properties 7.3.2.1 Bond strength In order to ascertain the strength of the bond between the HPDL-fired enamel glaze on the HPDL-treated OPC surface, pull-off tests were conducted. For the tests the concrete was prepared as relatively small-area samples (25 x 25 mm2). High-tensile aluminium test dollies were then attached to the glazed surface and to the axially opposite concrete substrate surface using Araldite epoxy and left to cure for 24 h. In order to ensure axial accuracy (essential for true results), the test dollies were set in position using identical V-blocks. The samples were placed into an Instron 4507 tensile/compressive test rig by mounting the test dollies into the jaws of the rig as

148


shown in Fig. 7.5. A tensile force was then applied until failure, with the energy being simultaneously recorded. A post-test analysis of the samples showed that the material failed below the HPDLtreated surface of the OPC in the HAZ. Within the optimum laser operating parameters the average bond strength of the glaze was recorded as 2.4 MPa. This compares with 6.3 MPa for the untreated OPC surface of concrete. Moreover, in all of the samples tested, not one failed at the interface between the HPDL-fired enamel glaze and the HPDL-treated OPC surface. It is therefore reasonable to assert that the bond strength of the HPDL-fired enamel glaze and the HPDL-treated OPC surface is somewhat greater that 2.4 MPa. 7.3.2.2 Rupture strength Tests were conducted to determine the rupture strength of the OPC glaze. Test samples were prepared as described above. The samples were placed on to the sample stage of the Instron 4507 tensile/compressive test rig and then subjected to a compressive rupture force until the OPC glaze failed (cracked), with the energy being simultaneously recorded (see Fig. 7.7). The rupture force was applied by means of a high-tensile steel indentor with a 1 mm radius point. The results of the tests revealed that the average rupture strength of the HPDL-fired enamel glaze was 2.8 J while the rupture strength of the untreated OPC surface was some 4.3 J. The rupture strength of the HPDL-treated OPC surface was measured as being only 0.8 J. 7.3.2.3 Wear resistance Tests were conducted in accordance with Fig. 7.9 to determine the exact difference in wear resistance characteristics of the HPDL-fired enamel glaze, the HPDL-treated OPC surface, and the untreated OPC surface. For experimental purposes the concrete was cut into smaller pieces (25 x 25 mm2). Half of the samples were then laser treated, with half of the laser-treated samples being enamelled. All the samples were then weighed and subjected to a friction force for 8 h, being removed from the machine and weighed at intervals of 2 h. Figure 7.27 shows the relationship between weight loss and the friction time for the OPC glaze and the untreated OPC. As can be seen, the HPDL-treated OPC surface shows a significant increase in wear resistance over the untreated OPC surface, with the weight loss being two times lower after 4 h, and three times lower after 8 h. What is more, the HPDL-fired enamel glaze exhibits not only a much greater wear resistance than the untreated OPC surface, but a marked increase in wear resistance over the HPDL-treated OPC surface.

149


Fig. 7.27 Relationship between weight loss and friction time for the untreated OPC surface, the HPDL-treated OPC surface, and the HPDL-generated glaze

7.3.2.4 Water sorptivity testing In order to test the water sorption properties of the HPDL-fired enamel glaze, or in other words, the effect of the HPDL-fired enamel glaze on the water absorption characteristics of the concrete, comparison experiments with both the untreated OPC surface and the HPDL-treated OPC surface were conducted by measuring the water sorptivity. For the experiment the HPDL-treated and untreated OPC samples were cut into smaller pieces (25 x 25 mm2). The tests were conducted in accordance with the standard procedure as used by Hall and Tse [216]. The samples were dried to a constant weight in an air oven at 65 °C to ensure all pores were free of water. The surfaces of the samples were then immersed in water and weighed at regular intervals. The side faces of the three samples were shielded from the water by means of an Araldite coating. In order to determine the sorptivity, i was plotted against the square root of time so as to give a straight line, as shown in Fig. 7.28. i is defined as

where Am is the cumulative change in mass with time and A is the absorbing surface area. The sorptivity, S, of the sample surfaces is simply the gradient of this line. As is evident from Fig. 7.28, the sorptivity of the untreated OPC surface was a typical 0.096 mm/mm1/2 [216], compared with 0.043 mm/min1/2 for the HPDL-treated OPC surface. It is therefore reasonable to conclude that since the HPDL-treated OPC surface has half the sorption of the untreated OPC surface, then the HPDL-treated OPC surface afforded the concrete twice as much resistance to water absorption than the untreated OPC surface. In addition, because the best-fit straight line for the HPDL-

150


treated OPC surface is below that of the untreated OPC surface, as well as intercepting the axis at a point below the untreated OPC surface, then it can be concluded that the rate of absorption of the laser surface-glazed concrete is much less than that of the untreated concrete. Furthermore, it is a distinct possibility that the HAZ, which was identified as CaO resulting from the dehydration of the Ca(OH)2, may, once rehydrated, act as a barrier towards liquids such as water, therefore augmenting the resistance of the HPDL-treated OPC surface to water absorption. Moreover, as Fig. 7.28 shows, the sorptivity of the HPDL-fired enamel glaze was practically zero. Thus, it is possible to assert that since the sorption of the HPDL-fired enamel glaze is negligible, then the enamel glaze provides the untreated OPC surface of the concrete with almost complete resistance to water absorption.

Fig. 7.28 Water absorption for the untreated OPC surface, the HPDL-treated OPC surface, and the HPDL-fired enamel glaze

7.3.2.5 Corrosion resistance Concrete surfaces are often subjected to corrosive substances, either as part of the normal service environment and/or as a result of routine cleaning. Therefore corrosion resistance tests based upon BS 6425 [217] were conducted using nitric acid, sodium hydroxide, and detergent cleaner (Premier Products MP9). The experiments were carried out by dropping small amounts of the corrosive agents in the concentration ratios of 80, 60, 40, 20, and 10 per cent on to the surface of the untreated and HPDLglazed OPC surface of concrete at hourly intervals for 4 h. The samples were then examined optically and mechanically tested in terms of compressive strength and wear. High concentrations of the various corrosive agents were used principally to accelerate the tests. However, in practice 60 per cent nitric acid is used within the nuclear processing industry as a solvent for nuclear fuels [194].

151


All three substances in the concentrations 80, 60, and 40 per cent were seen to attack immediately the untreated OPC surface, with the nitric acid and sodium hydroxide attacking with greater severity than the detergent, while the HPDL-glazed surface displayed no discernible microstructural changes or signs of devitrification due to corrosion. Tests conducted according to ASTM C579-91 [196] revealed that exposure of the untreated OPC surface to the reagents had a significant effect on the compressive strength and the wear resistance of the OPC. Exposure of the OPC to nitric acid and sodium hydroxide in the concentrations 40-80 per cent resulted in an average loss of compressive strength of approximately 19-37 per cent. In the case of the detergent a discernible loss in compressive strength only occurred with concentrations above 40 per cent. Here the average loss in compressive strength for concentrations in the range 60-80 per cent was approximately 17 per cent. This compares with no discernible difference in either the wear resistance or the compressive strength of both the HPDLfired enamel glaze or the HPDL-treated OPC surface. Chemical attack also accounted for a large reduction in the wear resistance of the untreated OPC surface when exposed to the reagents with an 80 per cent concentration. The wear resistance was significantly affected when the OPC was exposed to nitric acid and sodium hydroxide. Here the weight loss was approximately five times higher than for the unexposed OPC after 4 h, and approximately 11 times higher after 8 h for the nitric acid. In the case of the detergent, the weight loss was marginal after both 4 h and 8 h. No discernible change in the wear resistance of either the HPDL-treated OPC surface or the HPDL-fired enamel was observed. 7.3.2.6 Wear life characteristics As the results of the mechanical and chemical tests show, the HPDL-fired enamel glaze out-performed the untreated OPC surface in almost all the test areas. Moreover, the generally superior mechanical and chemical performance of the HPDL-fired enamel glaze over the untreated OPC suggests that the life characteristics of the HPDL-fired enamel glaze may also be superior to those of untreated OPC. This was especially true in the case of chemical resistance and water absorptivity, where the HPDL-fired enamel glaze proved to be resistant to both. This marked variation in corrosion resistance and absorptivity performance is due to the difference in structure of the HPDL-fired enamel glaze and the untreated OPC. Whereas the HPDL-fired enamel glaze is of an amorphous nature, the untreated OPC comprises a porous polycrystalline structure, thus the untreated OPC is readily attacked by acids while the amorphous structure of the HPDL-fired enamel glaze ensures an increase in acid resistance [95]. Yet, as has already been seen, in any analysis of the wear life of the two materials the in situ relative thickness of the HPDLfired enamel glaze and the untreated OPC layer on concrete must be considered in order to give a true interpretation of the actual life characteristics, particularly when considering the wear resistance (with and without exposure to corrosive chemical agents). Using equations (7.3) and (7.4) it was possible to calculate the wear rate

152


details and the nominal life increase of the HPDL-fired enamel glaze over the untreated OPC surface in a variety of environments (Table 7.6). To simulate service in a number of environments the untreated, laser-treated, and enamelled OPC surfaces were exposed prior to wear testing to a detergent, NaOH and HNOs. As Table 7.6 shows, the HPDL-fired enamel glaze gives an increase in actual life over the untreated OPC surface regardless of the environment. However, as can be seen, the increase in actual life of the HPDL-fired enamel glaze over the untreated OPC surface varies considerably depending upon the working environment. But, notwithstanding this, arguably the most common working environment for an OPC surface would involve some contact with at least detergent acids, therefore yielding significant economic savings since a HPDL-fired enamel glaze surface lasts around four times longer than one which is unglazed. Table 7.6 Wear rate details and the nominal life increase of the HPDL-fired enamel glaze over untreated OPC in various corrosive environments

Untreated OPC HPDL-fired enamel glaze Wear life (x100%)

Density 2220 kg/m3 2650 kg/m3 ~

Wear rate (mg/cm2/h) Thickness Normal Detergent NaOH 18.5 73.8 1500 nm 9.8 1.3 1.3 750 um 1.3 4.5

8.5

33.9

HN03 114.8 1.3 52.7

7.3.3 Summary Using a 120 W HPDL, the firing of a vitreous enamel frit to produce an enamel glaze on the ordinary Portland cement (OPC) surface of concrete was successfully demonstrated with power densities as low as 1 kW/cm2 and at rates up to 780 mm/min. The glazes produced were typically 750 urn in thickness and displayed no discernible microcracks or porosities. A maximum coverage rate of 0.34 m2/h was calculated. Mechanical testing of the HPDL-fired enamel glazes revealed that the average rupture strength of the HPDL-fired enamel glaze was 2.8 J, while the rupture strength of the untreated OPC surface was some 4.3 J. The average bond strength of the glaze was recorded as 2.4 MPa. This compares with 6.3 MPa for the untreated surface of the OPC. The HPDL-fired enamel glaze exhibited exceptional resistance to chemical attack and water absorption, while the untreated OPC surface was highly susceptible to both. Life assessment testing revealed that the HPDL-fired enamel glaze effected an increase in wear life of 4.5 to 52.7 times that of an untreated OPC surface, depending on the corrosive environment. Clearly, the economic and material benefits to be gained from the deployment of such an effective and efficient coating on OPC could be significant.

153


7.4 The enamelling of carbon steel by means of high-power diode laser radiation The work presented in this section describes the use of the novel 1.2 kW HPDL to alter the wettability characteristics of the common engineering carbon steel (ENS) detailed in Chapter 5, thus facilitating the hitherto impossible task of enamelling carbon steel in normal atmospheric conditions without pre-treatment chemical cleaning of the steel. Details of the OPC and enamel composition can be found in Chapter 3. The laser used in the study was a 1.2 kW HPDL (Rofm-Sinar, DL-012), emitting at 940 nm wavelength. The laser beam was focused directly on to the samples with a 6 x 20 mm2 rectangular beam at a fixed power of 500 W. The laser was operated in the CW mode. The beam was traversed across the samples by means of mounting the assembly head on to the z-axis of a three-axis CNC table (see Fig. 3.1). The focused laser beam was thus fired across the surface of the mild steel, and then subsequently the enamel frit, by traversing the samples beneath the laser beam using the x- and y-axis of the CNC table at speeds of 250-2000 mm/min, while O2 gas was pumped into a gas box in order to assist the surface treatment process. 7.4.1 Enamel glaze characteristics As was discussed in Chapter 5, it was not possible to fire the enamel on to the asreceived surface of the mild steel. After HPDL surface treatment of the mild steel, however, it was possible to fire the enamel directly on to the mild steel. Figure 7.29 shows the typical surface morphology of the HPDL-fired enamel glaze on the mild steel. The glaze was typically slightly undulating, with the undulations being regular in both periodicity and intensity. The thickness of the glaze was regular across the surface, being typically around 450 um. Also, owing to the thermal characteristics of the enamel itself (as discussed earlier) the glaze displayed no cracks or porosities. Such glazes could be generated within a small laser parameters operating window of 1.752.5 kW/cm2 power density and 480-600 mm/min traverse speed.

Fig. 7.29 Typical surface morphology of the HPDL-fired enamel glaze on the mild steel (2 kW/cm2 power density, 480 mm/min traverse speed)

154


7.4.2 Mechanical, physical, and chemical characteristics 7.4.2.1 Bond strength To determine the strength of the bond between the conventionally fired and HPDLfired enamel glazes, pull-off tests were conducted as shown in Fig. 7.7. For the tests, small samples of the steel were prepared (20 x 20 mm2). High-tensile aluminium test dollies were then attached to both the glazed surface and the axially opposite steel substrate surface using Araldite epoxy and left to cure for 24 h. The mean maximum bond strength of the conventionally fired enamel glaze was recorded as 77 MPa, while the HPDL-generated glaze was 70 MPa. In addition, no discernible difference in the results obtained was observed within the optimum laser operating parameters, indicating that neither the power density nor the traverse speed influenced the bond strength of the enamel glaze. Furthermore, a post-test analysis of both the conventionally fired and HPDL-fired enamel glazes revealed that the glazes failed mainly around the enamel/steel interface. However, as is typical of well-bonded enamel glazes on steel substrates, small amounts of the enamel glaze remained bonded to the steel. Notwithstanding this, according to Vargin [95] it is still reasonable to assume that bond strength values obtained are valid, since most of the glaze failed at the interface rather than within the glaze itself. It is proposed that the close agreement of the recorded mean maximum bond strength of the conventionally fired and the HPDL-generated enamel glazes can be attributed to the fact that the HPDL-generated glaze displayed bonding mechanisms similar to those of a conventionally fired glaze. 7.4.2.2 Rupture/impact strength Tests were conducted to determine the rupture/impact strength of the conventionally fired and HPDL-fired enamel glazes. The tests were carried out as shown in Fig. 7.7 while the test samples were prepared as described above. The results of the tests revealed there to be little difference in the recorded values of average rupture strength for the conventionally fired enamel glaze and the HPDLgenerated glaze, 4.8 and 4.3 J respectively. Again, no discernible difference in the results obtained was observed within the optimum laser operating parameters, indicating that neither the power density nor the traverse speed influenced the rupture/impact strength of the enamel glaze. It is suggested that such similar values of rupture/impact strength were the result of both the conventionally fired and the HPDL-generated glazes possessing very similarly shaped cross-sections; that is a flat interface with the steel substrate and an almost flat upper surface. Indeed, the relatively small difference between the recorded rupture/impact strengths of the conventionally fired and HPDL-generated glazes is probably due mainly to the fact that the upper surface of the HPDL-fired glaze exhibited a very slight concave surface geometry. Thus, the strains within the enamel layer were marginally higher and reduced the strength in comparison to an enamel seal

155


with a flat upper surface profile, as was the case with the conventionally fired enamel glaze [95]. 7.4.2.3 Surface roughness An important feature of any enamel glaze is its surface roughness; since it is this property that invariably determines the functionality and cleanability of the surface. Using a Taylor-Hobson Surtronic 3+ surface texture measuring instrument, a series of measurements were taken on the conventionally fired and HPDL-fired enamel glazes. On each sample four measurements were made in different positions and in different directions on the surface, with an average being taken. The surface roughness of the conventionally fired enamel glaze was significantly less than that of the HPDL-fired enamel glaze, 0.55 (im compared with 1.88 îm respectively. Clearly, in this situation the surface roughness, and therefore the cleanability, of the conventionally fired enamel glaze will be considerably better. It is believed that this variation in surface roughness is due to the vastly different firing times of the two processes (hours in the conventional manner and seconds for the HPDL technique) and the particle size of the enamel frit itself. Since the firing times are somewhat fixed, it is postulated that the problem of a rougher surface resulting from the HPDL technique could be overcome by using smaller sized enamel particles, an approach that would not greatly improve the result when firing in the conventional manner. 7.4.2.4 Wear resistance Tests were conducted in accordance with Fig. 7.30 to determine the exact difference in the wear resistance characteristics of the conventionally fired and the HPDL-generated enamel glazes. For experimental purposes the glazed samples were cut into smaller pieces (25 x 25 mm2). All the samples were then weighed and subjected to a friction force for 8 h, being removed from the machine and weighed at two-hourly intervals. Figure 7.30 shows the relationship between weight loss and the friction time for the conventionally fired enamel glazes and the HPDL-generated enamel glazes. As is evident from this figure, both glazes displayed an exceedingly high degree of wear resistance. Further, practically no difference in the wear resistance of the two glazes was observed. Arguably, this indicates that the two glazes possess the same mechanical properties and similar amorphous structure, despite their very different firing techniques. This is, perhaps, a valid argument when one considers the previous findings of this work.

156


Fig. 7.30 Relationship between weight loss and friction time for the conventionally fired and HPDL-generated enamel glazes

7.4.2.5 Corrosion resistance Enamelled surfaces are often subjected to corrosive substances, either as part of the normal service environment and/or as a result of routine cleaning. Corrosion resistance tests based upon BS 6428 [218] were therefore conducted using nitric acid, sodium hydroxide and detergent cleaner (Premier Products MP9). The experiments were carried out by dropping small amounts of the corrosive agents, in the concentration ratios of 80, 60, 40, 20, and 10 per cent, on to the surfaces of the conventionally fired and HPDL-generated enamel glazes at hourly intervals for 4 h. The samples were then examined optically and mechanically tested in terms of bond strength, rupture/impact strength, and wear life. High concentrations of the various corrosive agents were used principally to accelerate the tests. Both the conventionally fired and the HPDL-generated enamel glazes displayed complete resistance to all three substances, even at the highest concentrations. But, perhaps more importantly, even when exposed to the highest concentrations, no discernible effect on the bond strength, rupture/impact strength, and wear life of either enamel glaze was observed. Furthermore, no discernible microstructural changes or signs of devitrification due to corrosion were seen in either glaze. Again, this is perhaps a strong indication that the two glazes possess a similar amorphous structure, despite their very different firing techniques. 7.4.3 Summary HPDL surface treatment made it possible, for the first time, to fire an enamel glaze directly on to the ENS carbon steel substrate in normal atmospheric conditions after HPDL surface pre-treatment, but without pre-treatment chemical cleaning of the steel. Furthermore, through the employment of the HPDL, the whole enamel firing process was reduced from one that takes hours to complete, to one that was complete in minutes.

157


Table 7.7 Recorded values for selected mechanical and physical properties of a conventional and HPDL-fired enamel glaze

Mechanical/physical property Mean bond strength Mean rupture/impact strength Mean surface roughness (Ra) Wear (after 8 h)

Enamel glaze firing technique Conventional HPDL 77 MPa 70 MPa 4.3 J 4.8 J 0.55 um 1.88 um 93 mg/cm2 85 mg/cm2

As Table 7.7 shows, mechanical, physical, and chemical testing of the HPDL-fired enamel glaze revealed the glaze to possess similar properties to those of a conventionally fired enamel glaze. The mean average bond strength of the HPDL-fired enamel glaze was recorded as 70 MPa, this compares with 77 MPa for a conventionally fired enamel glaze. The average rupture strengths for the conventionally fired enamel glaze and the HPDL-generated glaze were recorded as 4.8 and 4.3 J respectively. Exceptional wear and corrosion resistance was displayed by both the HPDL- and conventionally fired enamel glazes, with both only wearing by around 2.8 mg/cm2 after 8 h and both showing complete resistance to acid, alkali, and detergent attack. It is believed that such results are due to the two glazes possessing the same mechanical properties and similar amorphous structure, despite their very different firing techniques. The surface roughness, however, of the conventionally fired enamel glaze was significantly less than that of the HPDL-fired enamel glaze, 0.55 ujn compared with 1.88 ujn respectively. It is believed that this variation in surface roughness is due to the vastly different firing times of the two process (hours in the conventional manner and seconds for the HPDL technique) and the particle size of the enamel frit itself. It is postulated that the problem of a rougher surface resulting from the HPDL technique could be overcome by using smaller sized enamel particles.

158

Chapter 8

Conclusions

Comparative studies of the differences in the beam interaction characteristics with various materials of the predominant materials processing lasers: the CO2, the Nd:YAG, and the excimer laser, are limited. This work not only identified these differences in performance, but categorized and quantified them in terms of changes in the wettability characteristics of a number of engineering materials, leading to the stage where generic assumptions can be made. Similarly, very little work has been conducted to study the possible employment of lasers for altering the surface properties of materials in order to improve their wettability characteristics. This work, on the other hand, elucidated not only the reasons for changes in the wettability characteristics of a number of engineering materials after laser surface treatment, but also identified and investigated a number of possible areas of practical application. The work presented herein revealed that interaction of CO2, Nd:YAG, excimer, and high-power diode laser (HPDL) radiation with the surfaces of the selected engineering materials effected significant changes in the wettability characteristics of the materials depending upon both the material itself and the laser type. In general, changes in the wettability characteristics of the materials as a result of laser treatment were found to be influenced primarily by the following: 1. Surface melting and partial vitrification. In the instances where laser irradiation induced surface melting and vitrification, the two main changes in the surface condition of the material were seen to be: (i) surface smoothing resulting from the laser melting of the material's surface which therefore reduced the surface roughness of the material. So, according to the theory of wettability, this in turn will directly occasion a reduction in the contact angle, 9.

159


(ii) increases in the polar component, y psv, of the surface energy resulting from the melting and partial laser vitrification of the materials, thereby improving the action of wetting and adhesion by generating a surface with a more polar microstructure. 2. Surface roughening. On the occasions when laser interaction with the surfaces of the selected engineering materials resulted in an increase in the surface roughness of the material, either due to ablation or excessive melting, then according to the theory of wettability, an increase in the contact angle, 6, would be directly brought about. 3. Surface O2 content. Increases, or decreases, in the surface O2 content of the materials due to surface oxidation resulting from laser treatment was identified as further promoting the action of wetting, since an increase in surface O2 content inherently effects a decrease in the contact angle. Specifically in terms of the materials studied in this work, the following can be asserted with regards to the observed effects of laser radiation on their wettability characteristics: 1. Composite materials. For both the ordinary Portland cement (OPC) and the Al2O3/SiO2-based composite material, the change in phase from crystalline to semior fully amorphous resulting from treatment with certain lasers was found to be essential in order for the wettability characteristics of the materials to be altered in a beneficial manner. Such phase changes effected a reduction in the surface roughness, an increase in ypsv> and an increase in the surface O2 content. It was observed that surface melting, and hence the phase changes, of both the OPC and the Al2O3/SiO2-based composite material, could only be brought about when using the CO2, the Nd:YAG, and the HPDL. In the instances when the excimer laser was employed to treat the Al2O3/SiO2-based composite material, the surface roughness of the material was seen to increase after treatment due to the ablation of the surface, while the 02 content of the Al2O3/SiO2-based composite material decreased after interaction with the excimer laser owing to the creation of defective energy levels. Further, excimer laser treatment of the surface of the Al2O3/SiO2-based composite material was found to effect very little change in yfv of the material. From this study it was possible to conclude that under the chosen experimental laser operating parameters, changes in the wettability characteristics of the OPC and the Al2O3/SiO2-based composite material were seen to vary depending upon the laser type. In particular, whether the laser radiation had the propensity to cause surface melting. Consequently the influence of pulse width cannot be discounted. 2. Engineering ceramic materials. HPDL treatment of the surfaces of all the selected ceramic materials (the SiO2/Al2O3-based ceramic (ceramic tile), the SiO2/Al2O3/Fe2O3-based ceramic (clay quarry tile), and the Al2O3 and SiO2-TiO2 (crystalline)) resulted in an improvement in the wettability characteristics of the materials for similar reasons to those discussed previously. Firstly, HPDL interaction with the surfaces of the materials effectively sintering of the ceramic materials' surfaces thereby reducing the surface roughness. Secondly, partial HPDL

160

Conclusions

vitrification was found to have occasioned an increase in y psv. Finally, the HPDLinduced surface melting resulted in an increase in the surface O2 content. 3. Metallic materials. Interaction of COi, Nd:YAG, HPDL, and excimer laser radiation with the surface of the mild steel studied was found to effect changes in the wettability characteristics of the material. It was observed that interaction of the mild steel with Nd:YAG and HPDL 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. Such changes were identified as being primarily due to: (i) the generation of a smoother surface after Nd:YAG and HPDL treatment due to optimum surface melting and resolidification; (ii) the surface roughness of the mild steel increasing after interaction with CO2 and excimer laser radiation due to excess surface melting and ablation respectively; (iii) changes in the surface O2 content of the mild steel - increasing after interaction with CO2, Nd:YAG and HPDL radiation due to surface melting, and decreased after interaction with the excimer laser due to the creation of defective energy levels; and (iv) increases in ypsv resulting from the melting and resolidification of the mild steel surface which thus created a different microstructure that quite possibly improved the action of wetting and adhesion. However, it was found that changes in the wettability characteristics of the mild steel appeared to be predominantly influenced by the surface roughness, while the microstructure appeared to have very little effect on the mild steel's wetting properties. Additionally, surface O2 content is also thought to play a minor role. This in an area of ongoing research by the authors. 4. Polymer materials. Interaction of CO2, Nd:YAG, HPDL, and excimer laser radiation with the surface of polyethylene (PE) and polymethyl methacrylate (PMMA) sheet was found to effect varying degrees of change to the wettability characteristics of the materials. It was observed that interaction of the PE and PMMA with CO2, Nd:YAG, and HPDL radiation resulted in very little change in the wettability characteristics of the materials. On the other hand, interaction of the PE and PMMA with excimer laser radiation occasioned a marked improvement in the wettability characteristics of the materials. Such changes after excimer laser treatment were identified as being primarily due to the increase in the surface O2 content of the PE and the increase in 7 psv. These occurrences are believed to be the result of photo-oxidation on the PE and PMMA surfaces which, in turn, is assumed to result in the generation of some surface O2-containing polar functional groups, thus effecting the observed changes in the wettability characteristics of the PE and PMMA after excimer laser treatment. This study showed that under the chosen experimental laser operating parameters, changes in the wettability characteristics of the PE and PMMA were seen to vary somewhat depending upon the laser type. In particular, whether the laser radiation had the propensity to cause surface melting or ablation. As such, the influence of pulse width cannot be discounted. The studies of the effects of laser radiation on the wettability characteristics of engineering materials detailed in this work have lead to the development of a number of possible practical applications, most of which are currently being actively

161


investigated by the authors. At present these include: the sealing of ceramic tiles; the enamelling of OPC, and the enamelling of mild steel. The findings of this work already indicate that the laser is not only an ideal tool, but arguably a unique tool for altering the wetting characteristics of engineering materials. A major focus of this work was the employment of the relatively new HPDL for the processing of the selected engineering materials. The relatively new research that has been conducted firmly establishes the potential of HPDLs for widespread materials processing. With superior material coupling capabilities (notably many metallic materials), high operating efficiencies, high levels of reliability, and power levels low enough to enable a standard mains supply, it is inevitable that HPDL materials processing applications will evolve. Such applications will either be completely original in nature, or simply supersede technologically mature devices in current applications. What is more, HPDLs have features unlike any other lasers, they are compact, portable, and the nature of their construction means that they can be operated free of any maintenance or adjustment. Also, the intrinsic output modulation of HPDLs makes possible the on-line feedback of process-relevant signals to control the beam parameters. Such features will undoubtedly shape future materials processing applications. Indeed, HPDLs are currently being used for soldering telephone connectors, and, on a smaller scale, for the transformation hardening of a limited range of materials, and the cutting of foils. The areas of HPDLs technology and fabrication are active fields of research, ensuring that the output power of HPDLs continues to increase by a factor of around two each year. But, perhaps more importantly, advances in the area of coherent, single-mode output HPDLs are increasing the power and the intensity of these devices at such a rate, that by the early years of this century it is expected that they will be capable of accomplishing the vast majority of materials processing tasks. The HPDL is possibly the only potential manufacturing tool that offers the prospect of a truly flexible and portable manufacturing system. It can be moved from one location to another very easily, has exceedingly low running costs, and could be adapted to an extremely wide variety of applications by simply changing a computer operating program; essential attributes of a manufacturing system in an era of world class manufacturing (WCM) and just in time (JIT) manufacturing techniques. At present, the total laser technology market has a value of around £700 million per annum, with HPDLs and low power diode laser (LPLD) accounting for about £20 million (3 per cent) and £180 million per annum (26 per cent) respectively. The prime factor influencing the proliferation of HPDL applications is cost. Because of their mass production capability, it is envisaged that extremely low-cost HPDL output (

Laser Modification of the Wettability Characteristics of Engineering Materials

Chemical Modification of Lignocellulosic Materials

Chemical Modification of Lignocellulosic Materials

Principles of laser materials processing

Physics of Solid-State Laser Materials

Laser Fabrication and Machining of Materials

Laser materials processing

Micromachining of Engineering Materials (Mechanical Engineering)

Handbook of Engineering Polymeric Materials

Handbook of Engineering Polymeric Materials

The science and engineering of materials

Solid-State Laser Engineering

Cell Colonization Control Physical and Chemical Modification of Materials

Semiconductor-Laser Fundamentals: Physics of the Gain Materials

Engineering materials

Engineering materials

Semiconductor-Laser Fundamentals : Physics of the Gain Materials

Modification of Proteins

Surface Modification of Textiles

Engineering Mechanics 2: Mechanics of Materials

Engineering Aspects of Self-Organizing Materials

Plasma-surface modification of biomaterials

Wettability at High Temperatures

Physics and Engineering of New Materials

Chemical Processing of Ceramics (Materials Engineering, 8)

Handbook of Imaging Materials (Optical Engineering)

Deformation and Fracture Mechanics of Engineering Materials

Handbook of Imaging Materials (Optical Engineering)

Essentials of materials science and engineering

Fundamentals of Laser Dynamics

Laser Modification of the Wettability Characteristics of Engineering Materials

Chemical Modification of Lignocellulosic Materials

Chemical Modification of Lignocellulosic Materials

Principles of laser materials processing

Physics of Solid-State Laser Materials

Laser Fabrication and Machining of Materials

Laser materials processing

Micromachining of Engineering Materials (Mechanical Engineering)

Handbook of Engineering Polymeric Materials

Handbook of Engineering Polymeric Materials

The science and engineering of materials

Solid-State Laser Engineering

Cell Colonization Control Physical and Chemical Modification of Materials

Semiconductor-Laser Fundamentals: Physics of the Gain Materials

Engineering materials

Engineering materials

Semiconductor-Laser Fundamentals : Physics of the Gain Materials

Modification of Proteins

Surface Modification of Textiles

Engineering Mechanics 2: Mechanics of Materials

Engineering Aspects of Self-Organizing Materials

Plasma-surface modification of biomaterials

Wettability at High Temperatures

Physics and Engineering of New Materials

Chemical Processing of Ceramics (Materials Engineering, 8)

Handbook of Imaging Materials (Optical Engineering)

Deformation and Fracture Mechanics of Engineering Materials

Handbook of Imaging Materials (Optical Engineering)

Essentials of materials science and engineering

Fundamentals of Laser Dynamics

Recommend Documents