Modern Surface Technology 9783527315321, 3527315322

Citation preview

Modern Surface Technology Edited by Fr.-W. Bach, A. Laarmann, T. Wenz

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

Related Titles Watts, J. F., Wolstenholme, J.

An Introduction to Surface Analysis by XPS and AES 2003 ISBN 0-470-84712-3

Decher, G., Schlenoff, J. B. (eds.)

Multilayer Thin Films Sequential Assembly of Nanocomposite Materials New edition planned for 2007 ISBN 3-527-31648-5

Wicks, Z. W., Jones, F. N., Pappas, S. P.

Organic Coatings Science and Technology 1999 ISBN 0-471-24507-0

Modern Surface Technology Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, and Thomas Wenz

Translated from German by C. Benjamin Nakhosteen

The Editors Prof. Dr.-Ing. Friedrich-Wilhelm Bach Director University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

n All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Dr. Andreas Laarmann formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Dipl.-Ing. Thomas Wenz formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Translated from German by C. Benjamin Nakhosteen Original title: Moderne Beschichtungsverfahren © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany All rights reserved Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

Cover Picture Credits HVOF coating of shaft Institute for Materials Science, University of Hannover, Germany Top right Cooling element with Cu layer produced by cold gas spraying OBZ Dresel & Grasme GmbH, Bad Krozingen, Germany Centre left Sol-gel antireflection coating on glass Fraunhofer Institute for Solar Energy Systems (ISE), Freiburg, Germany Centre right Enhalpy probe diagnostics of the three cathode gun Triplex II Institute for Materials Science, University of Hannover, Germany Bottom left PN-PVD coating process Institute for Materials Science, University of Hannover, Germany Bottom right Fly cutters with build-up brazed wear protection coating (Brazecoat) Innobraze GmbH, Esslingen, Germany

Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at

© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Printed in the Federal Republic of Germany Printed on acid-free paper Composition K+V Fotosatz GmbH, Beerfelden Printing betz-druck GmbH, Darmstadt Bookbinding Litges & Dopf Buchbinderei GmbH, Heppenheim ISBN-13: 978-3-527-31532-1 ISBN-10: 3-527-31532-2

VII

Contents Preface

V

List of Contributors 1 1.1 1.2 1.3 1.4 1.5 1.6

2

2.1 2.2 2.2.1 2.2.1.1 2.2.2 2.2.2.1 2.2.2.2 2.3 2.3.1 2.3.1.1 2.3.1.2 2.3.1.3 2.3.1.4 2.3.1.5

XVII

Selecting Surface-treatment Technologies 1 W. Tillmann, E. Vogli Introduction 1 Requirements on Part Surfaces 1 Selecting Coating and Surface Technologies 4 Processes for Surface Modification and Coating 5 Economic Assessment of Surface-treatment Technologies 9 Summary and Conclusions 9 References 10 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance 11 M. Wägner Introduction 11 Fundamentals 11 Heat Treatment 11 Surface-hardening Processes 12 Stainless Steels 13 Classification of Stainless Steels 14 Stainless Austenitic Steels 15 Technologies for Surface Hardening of Austenitic Stainless Steels 19 Kolsterising 19 Influence on Microstructure 20 Influence on Chemical Composition 21 Influence on Mechanical Properties 21 Wear Resistance 21 Influence on Corrosion Resistance 23

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

VIII

Contents

2.3.2 2.3.2.1 2.3.2.2 2.3.2.3 2.4 2.4.1 2.5

Kolsterising plus PVD Coating Coating Adhesion 25 Wear Resistance 26 Fatigue Strength 26 Applications 28 Application Limitations 28 Outlook 29 References 29

3

Fundamentals of Thin-film Technology 31 M. Nicolaus, M. Schäpers Introduction 31 Classification of Thin-film Coating Processes 31 General Aspects of Gas-phase Coating Processes 32 PVD – Physical Vapour Deposition 32 Evaporation 32 Sputtering 33 Ion Plating 35 CVD – Chemical Vapour Deposition 35 Plasma Properties 36 Low-pressure Plasma 37 Coating Configuration 38 Coating Structure 38 Electrodeposition and Electroless Plating Processes 39 Introduction 39 Fundamental Terms 40 Electrolyte 40 Electrodes, Electrode Reactions, Electrode Potential 40 Electrolysis and Faraday’s Laws 42 Overpotential 44 Electroless Plating 44 Electrodeposition of Metal 45 Electrodeposition of Metal from Non-aqueous Solvents 47 Summary and Outlook 49 References 50

3.1 3.2 3.3 3.3.1 3.3.1.1 3.3.1.2 3.3.1.3 3.3.2 3.4 3.4.1 3.5 3.5.1 3.6 3.6.1 3.6.2 3.6.2.1 3.6.2.2 3.6.2.3 3.6.2.4 3.6.3 3.6.4 3.6.5 3.6.6

4

4.1 4.2 4.3 4.3.1 4.3.2 4.3.3

24

Innovations in PVD Technology for High-performance Applications 51 K. Bobzin, E. Lugscheider, M. Maes, P. Immich Introduction 51 Market Situation 52 Application Examples 53 Tool Coatings for Cutting 54 Tool Coatings for Forming 55 Coatings for Plastic Parts 57

Contents

4.3.4 4.3.5 4.4

Coatings for Machine Elements 58 Part Coating for High-temperature Applications Summary 61 References 62

5

Development and Status Quo of Thermal CVD Hard-material Coating 65 A. Szabo Introduction 65 Early CVD Hard-material Coating 66 Fundamentals of Deposition Processes 66 Chemical Mechanism 66 Interdisciplinary Fundamentals 67 CVD System and Reaction-chamber Techniques 67 Combination Coatings 70 Material and Coating Properties 73 Physical Properties of Coating Materials 74 Comparison of Coating Combinations 74 Classic TiC-TiN 74 Balanced TiN-TiC 74 Effects of Thermal Expansion 75 Effects of Hardness 77 Performance of Hard-material Coatings – Applications 77 Wear Resistance 79 Heat Treatment and Dimensional Accuracy 79 CVD Coating at Lower Temperatures 80 Moderate-temperature CVD, MTCVD 80 Plasma-activated CVD, PACVD 82 Summary and Conclusions 82 References 83

5.1 5.2 5.3 5.3.1 5.3.2 5.3.3 5.4 5.5 5.5.1 5.5.2 5.5.2.1 5.5.2.2 5.5.3 5.5.4 5.6 5.6.1 5.6.2 5.7 5.7.1 5.7.2 5.8

6 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10

60

Hot-filament CVD Diamond Thin Films 87 O. Lemmer, R. Cremer, D. Breidt, M. Frank, J. Müller Introduction 87 Differences of Diamond Tools 88 Substrate Pre-treatment 88 Production of CVD Diamond 89 Hot-filament Process 90 Controlling CVD Diamond Properties 92 Industrial Deposition of CVD Diamond 93 Post-treatment of CVD Diamond 93 Applications for Diamond-coated Tools 94 Summary and Conclusions 99 References 100

IX

X

Contents

7

7.1 7.2 7.2.1 7.2.2 7.2.3 7.2.4 7.3 7.3.1 7.3.2 7.3.3 7.3.4 7.3.5 7.4 7.4.1 7.4.2 7.4.3 7.4.4 7.4.5 7.5 7.6

8 8.1 8.2 8.2.1 8.2.2 8.3 8.3.1 8.3.2 8.4 8.4.1 8.4.2 8.4.3 8.5

An Introduction to Electrodeposition and Electroless Plating Processes 101 W. Olberding Introduction 101 Fundamentals of Electrodeposition (Considering Nickel Deposition as Example) 101 Structure of Electroplated Nickel Coatings 104 Deposition Mechanism 105 Current-density Distribution 106 Electroless Plating of Nickel 107 Overview of System Technologies 109 Barrel Plating 109 Rack Plating 111 Continuous Plating 112 Brush Plating 114 Tank Plating 114 Overview of Individual Process Steps in Electroplating Degreasing 114 Activating or Pickling 115 Carryover 115 Coating Passivating Materials such as Stainless Steel and Aluminium 116 Summary of Pre-treatment 116 Microstructuring and Electroforming 116 Summary 117 References 118

114

Fundamentals of Thermal Spraying, Flame and Arc Spraying Z. Babiak, T. Wenz, L. Engl Introduction 119 Fundamentals of Thermal Spraying 119 Structure of Thermal Spray Coatings 121 Adhesion of Thermal Spray Coatings 122 Flame Spraying 123 Flame Spraying Process 123 Materials and Applications 125 Arc Spraying 127 Arc Spraying Process 127 Special Arc Spraying Processes 131 Materials and Applications 131 Summary and Conclusions 134 References 134

119

Contents

9 9.1 9.2 9.2.1 9.3 9.3.1 9.3.2 9.3.3 9.3.4 10 10.1 10.2 10.2.1 10.2.2 10.2.3 10.3 10.4 10.5 10.6 10.6.1 10.6.2 10.6.3 10.6.4 10.7

11

11.1 11.2 11.3 11.4 11.5 11.6 11.6.1 11.6.2 11.6.3 11.6.4

Spray Materials 137 J. Beczkowiak Introduction 137 Spray Material Properties Determined by Production Issues 137 Powder-production Processes 138 Material Selection for Coating Applications 142 Materials for Wear Protection 143 Materials for Corrosion Protection 143 Materials for Biotechnology 144 Materials for Special Applications 144 High-velocity Oxygen Fuel Flame Spraying O. Brandt Introduction 145 Characteristics 146 HVOF Gun 146 Fuel Gases and Process Parameters 147 Spray Materials 148 Technical Considerations 150 Applications 151 Process Monitoring and Control 153 Development Trends 155 Application Technology 155 Coating Materials 155 Process Technology 156 Techniques and Methods 156 Summary 156 References 157

145

Triplex II – Development of an Economical High-performance Plasma Spray System for Highest-quality Demands even under Challenging Production Conditions 159 H. Zimmermann, H.-M. Höhle Introduction 159 Fundamentals of Plasma Spraying 161 Standard Plasma Gun Design 164 Development of the High-performance Three-cathode Plasma Gun Triplex 168 Triplex II – A New Era in Plasma Spraying Technology 171 Positive Feedback from Industry 174 Chromium Oxide Coating of Anilox Rollers for Printing Industry 174 Abradable Coatings 175 Thermal-barrier Coatings 177 Further Applications 178

XI

XII

Contents

11.7

Summary 178 References 178

12

System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying 179 W. Krömmer, P. Heinrich Introduction 179 System Design 179 Pressure Tank and Nozzle 179 Control Unit 180 Touch Screen 181 Main Mask Parameters 182 Gas Heater LINSPRAY® 183 Gas Supply for Cold Gas Spraying 184 Helium Recovery 185 Applications 186 Summary 188 References 189

12.1 12.2 12.2.1 12.2.2 12.2.3 12.2.3.1 12.2.4 12.2.5 12.2.6 12.3 12.4

13 13.1 13.2 13.3 13.3.1 13.3.2 13.3.3 13.3.4 13.3.5 13.4 13.4.1 13.5 13.5.1 13.6

14 14.1 14.1.1 14.1.2 14.2 14.2.1 14.2.2 14.3

Diagnostics in Thermal Spraying Processes 191 J. Prehm, K. Hartz Introduction 191 Classification of Diagnostic Methods 191 Methods for Particle Diagnostics 191 Laser Doppler Anemometry (LDA) 191 Phase Doppler Anemometry (PDA) 194 Laser Two-focus Method (L2F) 195 Particle Image Velocimetry (PIV) 195 In-flight Particle Diagnostics 197 Methods for Plasma and Hot Gas Diagnostics 198 Enthalpy Probe Diagnostics 198 Methods for Online Process Control 199 Particle-flux Imaging (PFI) 200 Summary and Conclusions 202 References 202 Sol-gel Coating Processes 205 M. Kursawe, V. Hilarius, G. Pfaff, R. Anselmann Introduction 205 Background and Origin of Sol-gel Chemistry 205 Material Fabrication by Means of Sol-gel Techniques 206 Sol-gel Coating Formation for SiO2 207 Coatings with SiO2 Sol from Salts of Silicic Acid 207 Coatings with SiO2 Sol from Si Alkoxides 208 Application Examples 210

Contents

14.3.1 14.3.2 14.3.2.1 14.3.2.2 14.3.3 14.3.4 14.4

15 15.1 15.2 15.3 15.4 15.4.1 15.4.2 15.4.3 15.5 15.5.1 15.5.2 15.6

16 16.1 16.2 16.2.1 16.2.2 16.2.3 16.2.4 16.2.5 16.3 16.3.1 16.3.2 16.4

Translating an Idea into a Product: Development of an Anti-reflection Coating for Glass 210 Application of Wet Chemical Coating Techniques for a Common Product Type: Pearlescent Pigments 214 Gloss and Colour 214 Production of Pearlescent Pigments with Interference Colours 215 Effect Pigments on SiO2 Flakes 217 Coating of SiO2 Spheres for Cosmetic Formulations 219 Conclusions 219 References 220 Hot-dip Coating 221 W. Bleck, D. Beste Mechanisms of Corrosion Protection 221 Phase Diagrams Fe-Zn, Fe-Al, Al-Zn, and Fe-Al-Zn 224 Metal Coatings 227 Systems Technology 229 Design of Hot-dip-coating Systems 229 Reacting Agents in Molten Zinc 231 Surface Post-treatment 233 Quality Control 234 Testing Mechanical Properties 234 Testing Corrosion Properties 234 Summary and Conclusions 236 References 237 Build-up Brazed Wear-protection Coatings 239 H. Krappitz Introduction 239 Brazing and Soldering 239 Fundamentals 239 Repair Brazing 241 Coating by Build-up Brazing of Sintered Hard Metals 242 Brazing of Ceramics 244 Brazing of Hard-material Particles 246 BrazeCoat Technology 248 Coating with Mats of Filler Metal and Hard Material (BrazeCoat M) 248 Coating with Suspensions of Filler Metal and Hard Material (BrazeCoat S) 250 Summary 252 References 252

XIII

XIV

Contents

17 17.1 17.2 17.3 17.4 17.5

18 18.1 18.2 18.3 18.4 18.4.1 18.4.2 18.4.3 18.4.3.1 18.4.3.2 18.4.4 18.4.4.1 18.4.4.2 18.4.4.3 18.4.4.4 18.4.5 18.4.6 18.5 18.5.1 18.5.1.1 18.5.1.2 18.5.2 18.5.2.1 18.5.2.2 18.5.2.3 18.5.2.4 18.6

19 19.1 19.2 19.2.1

Applications of Coating Processes in Brazing Technology 253 K. Möhwald, U. Holländer, A. Laarmann Introduction 253 Brazing Filler-metal Application by Thermal Spraying 254 Electroplating and Electroless Plating Methods for Brazing Filler-metal Application 257 Brazing Filler-metal Application by PVD 259 Summary and Conclusions 261 References 262 Surface Protection by Means of Build-up Welding 263 A. Gebert, B. Bouaifi Introduction 263 Process Variants 264 Characterisation of Build-up Welded Coatings 265 Build-up Welding Techniques 268 Distinguishing Features 268 Shop Welding (Manual Arc Welding, Gas Flame) 270 Processes with Protective Slag 271 Electroslag Build-up Welding (RES – Resistance Electroslag) 271 Submerged Arc Build-up Welding 271 Inert-gas-shielded Arc Welding 273 Tungsten Inert Gas Build-up Welding (TIG Process) 273 Gas-shielded Metal Arc Welding 274 Plasma-transferred Arc Process (PTA) 276 Plasma MIG Process 279 Resistance Roll Seam Technique 280 Laser Cladding 281 Coating Materials for Build-up Welding 283 Materials for Corrosion Protection 283 Corrosion-resistant Iron-based Materials 284 Nickel Alloys 285 Materials for Wear Protection 285 Nickel Hard Alloys 287 Iron Hard Alloys 288 Cobalt Hard Alloys 292 Aluminium Pseudo-alloys 294 Summary and Conclusions 295 References 296 Non-destructive Testing and Assessment of Coatings W. Reimche, R. Duhm Introduction 297 Coatings 297 Processes 297

297

Contents

19.2.2 19.2.3 19.3 19.3.1 19.3.2 19.3.3 19.3.4 19.3.5 19.3.6 19.3.7 19.3.8 19.4 19.4.1 19.5 19.5.1 19.5.2 19.5.3 19.5.4 19.5.5 19.6

Coating Properties 298 Test Planning 299 Thickness-measurement Techniques 299 Geometric Measurement of Parts 302 Differential Weight Analysis Before and After Coating 302 Coating-thickness Measurements Based on Magnetic Pull-off 302 Coating-thickness Measurements Based on Acoustic Principles 303 Coating-thickness Measurements with Magnetic-induction Techniques 305 Coating-thickness Measurements with Eddy-current Techniques 306 Coating-thickness Measurement by Means of X-ray Fluorescent Analysis 309 Coating-thickness Analysis by Means of Beta-backscatter Technique 310 Internal Stresses in Coatings 311 Roentgenographic Assessment of Internal Stresses – X-ray Diffractometry 311 Detecting Coating Defects 312 Detection of Open Defects in Coatings – Dye-penetration Test 313 Detection of Laminar Separation/Coating Delamination – Ultrasonic Testing 313 Detection of Laminar Separation/Coating Delamination – Lock-in Thermography 315 Detection of Internal Coating Defects – Eddy-current Testing 317 Assessment of Coating Adhesion by Means of Electromagnetic Testing 318 Summary and Conclusions 319 References 320

Subject Index

323

XV

V

Preface Technological developments in aerospace and other high-technology fields give rise to constantly increasing demands on part surfaces. Generally, surfaces that, e.g., withstand tribological loads, or show special properties in thermal and electrical conductivity or optical behaviour, require utilisation of coating processes. This book presents industrially implemented coating processes in the fields of build-up welding and brazing, plasma, arc, and flame spraying, sol-gel technology as well as the thin-film technologies, chemical vapour deposition and physical vapour deposition. Particular emphasis is placed on the combination of process and materials technology in terms of producing coatings that meet all necessary requirements. Alongside industrially relevant coating processes, newly developed technologies on the verge of industrial implementation are presented. Examples are processes for diamond synthesis, cold gas spraying or the processing of nano-sized particles. The aim of this book is to enable engineers and technicians working in development, design, and manufacturing to be able to estimate the potential of protective surface coatings and the associated processes in their fields of activity. The intention is that coating technologies serve as an integral part of development, design, and manufacturing. The Editors May 2006

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

XVII

List of Contributors Ralf Anselmann Degussa AG Creavis Technologies & Innovation Paul-Baumann-Strasse 1 45764 Marl Germany

Wolfgang Bleck RWTH Aachen University Department of Ferrous Metallurgy Intzestrasse 1 52072 Aachen Germany

Zenon Babiak formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

Kirsten Bobzin RWTH Aachen University Surface Engineering Institute Augustinerbach 4–22 52062 Aachen Germany

Joachim Beczkowiak H.C. Starck GmbH Ceramics and Surface Technology Am Kraftwerkweg 3 79725 Laufenburg Germany Dipl.-Ing. Daniel Beste RWTH Aachen University Department of Ferrous Metallurgy Intzestrasse 1 52072 Aachen Germany

Belkacem Bouaifi CeWOTec gGmbH Lassallestrasse 14 09117 Chemnitz Germany Oliver Brandt Becon Technologies GmbH Feuerwerkerstrasse 39 3602 Thun Switzerland Dirk Breidt CemeCon AG Research & Development Adenauerstrasse 20 B 1 52146 Würselen Germany

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

XVIII

List of Contributors

Rainer Cremer CemeCon AG Research & Development Adenauerstrasse 20 B 1 52146 Würselen Germany

Peter Heinrich Linde AG Business Segment Linde Gas Carl-von-Linde-Strasse 25 85716 Unterschleissheim Germany

Rainer Duhm University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

Volker Hilarius Merck KgaA Pigments R&D Frankfurter Strasse 250 64293 Darmstadt Germany

Lars Engl formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

Hans-Michael Höhle Sulzer Metco Europe GmbH Am Eisernen Steg 18 65795 Hattersheim Germany

Martin Frank CemeCon AG Research & Development Adenauerstrasse 20 B 1 52146 Würselen Germany Andreas Gebert CeWOTec gGmbH Lassallestrasse 14 09117 Chemnitz Germany Karsten Hartz University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

Ulrich Holländer University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Philipp Immich RWTH Aachen University Surface Engineering Institute Augustinerbach 4–22 52062 Aachen Germany Harald Krappitz Innobraze GmbH Fritz-Müller-Strasse 97 73730 Esslingen Germany

List of Contributors

Werner Krömmer Linde AG Business Segment Linde Gas Carl-von-Linde-Strasse 25 85716 Unterschleissheim Germany

Kai Möhwald University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

Monika Kursawe Merck KgaA Business Development Chemicals/ Advanced Materials Frankfurter Strasse 250 64293 Darmstadt Germany

Jürgen Müller CemeCon AG Research & Development Adenauerstrasse 20 B 1 52146 Würselen Germany

Andreas Laarmann formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Oliver Lemmer CemeCon AG Research & Development Adenauerstrasse 20 B 1 52146 Würselen Germany Erich Lugscheider RWTH Aachen University Surface Engineering Institute Augustinerbach 4–22 52062 Aachen Germany Michel Maes RWTH Aachen University Surface Engineering Institute Augustinerbach 4–22 52062 Aachen Germany

Martin Nicolaus University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Werner Olberding IGOS Institut für Galvano- und Oberflächentechnik GmbH Grünewalder Strasse 29–31 42657 Solingen Germany Gerhard Pfaff Merck KgaA Pigments PD Frankfurter Strasse 250 64293 Darmstadt Germany Jens Prehm formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany

XIX

XX

List of Contributors

Wilfried Reimche University of Hannover Institute for Materials Science Non-destructive Testing Lise-Meitner-Strasse 1 30823 Garbsen Germany Melanie Schäpers Draeger Medical AG & Co. KG Moislinger Allee 53–55 23542 Lübeck Germany Andreas Szabo Best-Surface Simanowizstrasse 12 71640 Ludwigsburg Germany Wolfgang Tillmann University of Dortmund Chair of Materials Technology Leonhard-Euler-Strasse 2 44227 Dortmund Germany

Evelina Vogli University of Dortmund Chair of Materials Technology Leonhard-Euler-Strasse 2 44227 Dortmund Germany Martina Wägner Bodycote Hardiff BV Paramariboweg 45 7333 PA Apeldoorn The Netherlands Thomas Wenz formerly: University of Hannover Institute for Materials Science Schönebecker Allee 2 30823 Garbsen Germany Harald K. Zimmermann Sulzer Metco AG (Switzerland) Rigackerstrasse 16 5610 Wohlen Switzerland

1

1 Selecting Surface-treatment Technologies W. Tillmann, E. Vogli, Chair of Materials Technology, University of Dortmund, Germany 1.1 Introduction

Nowadays, modern production processes require inherent state-of-the-art surface technologies. Furthermore, rising standards of technical products are creating the perception that surface technologies are often the central impetus needed for meeting product specifications. Design engineers thus face two essential tasks: On the one hand, part specifications need to be transformed into properties of materials and surfaces. On the other hand, selected materials technologies have to be integrated in corresponding process chains. Apart from the required part specifications, production costs and ecological aspects are important issues. Not only production standards but also economic conditions lead to increasing significance of surface technologies. Considering the two substantial domains of surface technology, tribology and corrosion, macroeconomics experts estimate that tribological damage causes a loss of approx. 1% of the German gross national product (GNP). The economic effect of corrosion damage is even higher, approaching approx. 3.5–4.2% of the GNP. Surface technologies therefore have to be considered as one of the key technology fields in production engineering. Here, one possible method for selecting surface-treatment processes that satisfy existing requirements of specific parts is introduced. In addition, a variety of surface-treatment processes are compared with respect to possible fields of application and characteristics specific to the individual processes.

1.2 Requirements on Part Surfaces

Systematic selection of suitable surface treatments is always based on acquiring a complete set of requirements on the part surface with respect to intended operating conditions. According to Haefer [3], the surface is responsible for all meModern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

2

1 Selecting Surface-treatment Technologies

chanical, thermal, chemical, and electrochemical interactions with the environment. This leads to the main functions that need to be fulfilled by technical surfaces: · corrosion resistance · wear resistance · defined tribological behaviour · optical behaviour · decorative behaviour · matched interface behaviour (e.g. for joining purposes). In addition, especially highly specialised products may demand specific functions. Parts used in micro-technology for example can require special electromagnetic properties of surfaces. Ultimately, requirements on part surfaces are determined by the particular load conditions under which the final product operates. Figure 1.1 illustrates the main kinds of load conditions subdivided into volume and surface loads. Wear and corrosion are the main stresses that have to be controlled by surface technology in the realm of mechanical engineering. Incorrect materials selection as well as unsuitable or missing protective layers lead to manifold damages, some of which are shown in Figs. 1.2 and 1.3. In many cases, appropriate surface treatment can either prevent or at least delay such damages. However, adjusting part surface characteristics carefully is essential in order to handle overall operating conditions. Surface technology focuses on reacting adequately to the specific kinds of load and stress. For this, the materials properties of part surfaces are systematically modified or produced, particularly by means of:

Fig. 1.1 Main volume and surface loads on parts.

1.2 Requirements on Part Surfaces

Fig. 1.2 Wear phenomena.

Fig. 1.3 Corrosion phenomena.

· applying a protective coating to the workpiece · modifying the surface zone of the workpiece. Typical coating processes are chemical vapour deposition (CVD), physical vapour deposition (PVD), thermal spraying, build-up brazing and welding, as well as cladding and dip coating. Surface-modification processes, on the other hand,

3

4

1 Selecting Surface-treatment Technologies

include thermo-chemical diffusion processes, thermal surface hardening, implantation methods, and mechanical surface-hardening processes.

1.3 Selecting Coating and Surface Technologies

Designing a suitable surface treatment from a given combination of loads is challenging. Not only is it often difficult to precisely and thoroughly understand the operating conditions of a part, but very large variety of possible materials and materials technological processes have to be considered. Estimates indicate that the number of materials used in materials technology is in the range of 40 000–80 000. Moreover, including surface technologies, about 1000 different processes are used. In contrast, the mean vocabulary of a Central European spans approx. 5000 words. Quite obviously, the process of selecting an appropriate coating or surface treatment requires a systematic approach. The selection process needs to be implemented at an early stage of product development. It is necessary that developers already consider surface requirements during concept phases, directly after taking down customer and market demands. Based on the given operating conditions, four fundamental aspects should be clarified systematically [1, 6]. The following facets and questions need to be considered carefully: 1) Function: – What are the functional characteristics of the part surface? – What kind of requirements exist? 2) Purpose: – What needs to be maximised? – What needs to be minimised? 3) Limitations: – Which constraints and boundary conditions have to be met? e.g. – from a technical point of view – from an economic point of view – considering design-to-cost concepts – considering design for environment concepts – considering life-cycle costs 4) Options: – What options exist? This systematic approach basically represents the general framework of the requirement catalogue concluded from the set of loads and stresses. Subsequently, individual materials and surface technologies need to be analysed and assessed against this background. This search and evaluation should be performed in an equally systematic approach. Figure 1.4 illustrates an example of a systematic analysis sheet. Here, individual coating materials and processes can be rated with respect to selected properties, prerequisites, and restrictions. The listed se-

1.4 Processes for Surface Modification and Coating

Fig. 1.4 Example of a rating matrix for evaluating coating materials and/or processes against the background of a desired property catalogue.

lection of properties within the rating matrix as well as the corresponding prerequisites and restrictions originate from the formerly compiled catalogue of requirements. The approach presented here describes a workable method of correlating a catalogue of requirements with appropriate surface technologies. Certainly, the quality of results is determined and limited by the requirement catalogue developed in phase one. Furthermore, this methodology requires comprehensive knowledge of available materials and processes, a frequently limiting factor due to the already mentioned manifold process varieties and materials. 1.4 Processes for Surface Modification and Coating

Giving a detailed overview of the different surface modification and coating processes would go far beyond the scope of this chapter. Therefore, a general summary of the most important process classes is presented, along with their indi-

5

6

1 Selecting Surface-treatment Technologies

Fig. 1.5 Classification of surface-modification processes.

Fig. 1.6 Classification of coating technologies.

vidual assets and drawbacks. Figure 1.5 shows the systematic classification of surface-modification processes. Selected process technologies are presented in Table 1.1, including basic advantages and disadvantages [5]. Unlike surface-modification processes, coating involves covering the surface of a workpiece with a well-bonded layer of shapeless material. A possible classification of coating technologies is given in Fig. 1.6. Bond strength to the substrate material primarily determines the quality of a coating. This macroscopic property is controlled by: · materials combination · type of interface zone · microstructure and process conditions · substrate type and pre-treatment. A strong atomic bond in the contact zone is most favourable, provided that internal stresses within the coating are not too high and no long-term degradation occurs within the coating/substrate composite. Coating process and material combinations determine whether mechanical, chemical, or electrostatic bonds prevail, or whether diffusion occurs. Thus, preparation of the workpiece is a

1.4 Processes for Surface Modification and Coating

7

Table 1.1 Selected process technologies for surface modification. Advantages

Process technologies

Disadvantages

+ inexpensive + selective treatment possible + depth 1–10 mm

Hardening by means of induction flame laser, electron beam TIG (tungsten-inert gas)

– limited to steel, Co, 3–0.6% – distortion possible

+ applicable to many types of steel + well-controlled coating properties

Carburisation · diffusion of C (up to 0.8%) into surface including hardening · variety of different C-carriers

– distortion – cooling cracks

Carbonitriding + less distortion of surface compared to hardening and carburisa- · compare above, additional nitrogen tion · low-temperature process

– slow process

+ less distortion of surface + high elevated temperature hardness

Nitriding · N-diffusion, formation of surface nitrides

– slow process

+ good resistance against adhesive wear + allows oxidising for corrosion protection

Nitrocarburising – cf. nitriding

– modifies thin surface zone

+ high hardness

Boronising · boron diffusion for boride formation · also applicable for Co-, N-, Tialloys

– distortion (high process temperatures) – brittle – low corrosion resistance

+ inexpensive

Sherardising · Zn-diffusion with subsequent chromatising

– no wear protection

+ good corrosion protection + less vibration fatigue + increased resistance against stress-corrosion cracking and corrosion fatigue

Shot peening – modifies thin surface zone for plastic deformation of workpiece – low increase in hardness surface

see above

Deep rolling comparable with shot peening

– expensive

+ can create high surface hardness values + good wear and corrosion protection

Plating, metallising (e.g. Cr, V, Nb, Si-containing diffusion coatings) large variety of processes

– high process temperatures (distortion)

8

1 Selecting Surface-treatment Technologies

crucial factor in obtaining good coating adhesion. Apart from removing contamination, pre-treatments activate the substrate surface and therefore substantially influence the bond between coating and substrate. Typical mechanisms of surface activating are: · creating defects in the substrate · increasing surface energy · removing oxide layers. Table 1.2 summarises different coating technologies in common use and includes important process characteristics [4, 6]. Table 1.2 Selected coating technologies. Advantages

Technologies

Disadvantages

+ high hardness values + good corrosion resistance + reduces friction in contact with steel

Electrochemical deposition (e.g. Cr) up to 0.5 mm coating thickness

– coating of complex geometries is difficult – danger of hydrogen embrittlement – environmental problems

Chemical (electroless) deposition + low-temperature process from electrolyte solution (e.g. NiB, + very high corrosion protection + suitable for most metal substrates NiP) and many non-conducting materials + uniform coating thickness even on complex geometries

– expensive – additional heat treatment necessary

+ very high hardness values + good adhesion

CVD, chemical vapour deposition chemical vapour deposition at high temperatures

– distortion – coating of sharp-edged geometries is difficult – disposal of aggressive gaseous waste

+ dense coatings with high adhesion + low coating process temperature + allows deposition of pure elements, compounds and alloys

PVD, physical vapour deposition – evaporation – cathode sputtering

– low growth rate of coating – expensive vacuum process – restrictions in terms of part geometry

+ large variety of materials + good adhesion + properties well controllable by choice of materials and process

Thermal spray processes

– residual porosity – deposition efficiency of coating process (overspray)

+ + + + +

Build-up welding

– coating materials limited – impact on substrate material

Build-up brazing powdery hard material and brazing filler metal with binding agent protective gas process

– coating materials limited

very high adhesion large parts coatable inexpensive very high adhesion coating of complex geometries

1.6 Summary and Conclusions

Fig. 1.7 Procedure for determining costs per piece and customer benefit for surface-technology processes.

1.5 Economic Assessment of Surface-treatment Technologies

Next to selecting coatings from a technological point of view, the costs of available surface-treatment processes need to be taken into account. Considering all relevant cost elements associated with individual process steps by means of direct costing is necessary. As presented in Fig. 1.7, analysis yields individual costs per piece and thus allows comparing different surface-treatment processes. Furthermore, specific customer benefit of a surface technology can be an additional determining factor during economic assessment [2]. Inevitably, this task is more of a challenge than pure cost assessment because customer benefit is hard to quantify. The fundamental idea is to compare the two situations before and after optimisation using a quantitative approach. Once this method delivers quantified benefits associated to different surface treatments it represents an additional economic assessment tool. Process and material selection can thus use a supplementary criterion along with technical and economic ratings. The quantitative assessment of customer benefit can also be used to reduce complications of market launch for new surface-treatment processes.

1.6 Summary and Conclusions

Selecting an appropriate technology to produce a certain combination of surface characteristics is a very complex process. It involves systematic correlation of specifications with attainable surface properties. Usually, the selection process includes economic and ecological evaluations. Surface technologies are gaining importance as integral parts of manufacturing chains. While surface treatments nowadays are often carried out as separate

9

10

1 Selecting Surface-treatment Technologies

or post-processes, integration into process chains is on the advance. Aimed primarily at reduced production time, integration creates synergies as well. For instance, coating sheet metal with initially poor conductivity can ease subsequent electromagnetic forming steps. Also, part geometry close to final contour can be produced using enhanced process control during coating. These examples show that integration of surface treatment technologies and manufacturing process chains is essential along with developing new coating processes and materials.

References Charles, J. A., Crane, F. A. A.: Selection and Use of Engineering Materials, Butterworth, London, Boston, Singapore, Sydney, Toronto, Wellington, 1989 2 Grundmann, G., Blau, W.: Modellrechnungen anhand von Fallbeispielen, Series Wissenstransfer Oberflächentechnik, VDI-TZ, Düsseldorf, 2002 3 Haefer, R.: Oberflächen- und Dünnschicht-Technologie, Springer, Ber1

lin, Heidelberg, New York, London, Paris, Tokyo, 1987 4 Kienel et al.: Vakuumbeschichtung 2–5, VDI, Düsseldorf, 1994 5 Kohtz, D.: Wärmebehandlung metallischer Werkstoffe, VDI, Düsseldorf, 1994 6 Tillmann, W.: Anforderungen an heutige Bauteiloberflächen, DGM training seminar, Witten, September 2003

11

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance M. Wägner, Bodycote Hardiff BV, Apeldoorn, Netherlands 2.1 Introduction

In everyday life we are surrounded with applications that depend on stainless steel. Typical examples are found in food processing, medical technology, electronics, and the automotive industry. However, commonly used highly corrosion resistant austenitic steels show low hardness with overall unfavourable wear resistance. Therefore, a wear-protected surface on top of the retained corrosion resistant substrate is the ideal combination. This chapter presents technologies for surface hardening using diffusion processes – known as Kolsterising – in order to improve the wear resistance of austenitic steel, as well as a newly developed combination of this diffusion process with PVD technology.

2.2 Fundamentals 2.2.1 Heat Treatment

The following hardening mechanisms are used to increase the strength of metallic materials [1]: · work hardening · refinement of grain size · solid-solution strengthening · precipitation hardening · dispersion strengthening · martensitic transformation.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

12

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

In hardened materials, movement of dislocations is reduced or inhibited, and thus, tension levels necessary for deformation increase. Heat treatment is the most important means for controlled modification of metallic materials properties, characterised by a precise sequence of heating, holding at temperature, and cooling in controlled atmospheres. According to [2], the term heat treatment is referred to as: “a collective term for a single process, or a combination of production processes, used to optimise or obtain specific processing characteristics or usability by changing materials properties. It is used for solid metallic materials and includes thermal, thermochemical, or thermo-mechanical operations.” Heat treatment focuses either on improving processing characteristics such as machining or forming properties, or on improving usability features including hardness, toughness, strength, or wear resistance. Classifications of heat-treatment processes distinguish: · thermal processes · thermo-chemical processes · thermo-mechanical processes.

2.2.1.1 Surface-hardening Processes Surface-hardening processes yield cost-efficient solutions and controlled optimisation of metallic surface properties. This group of processes has been gaining importance continuously for years. Surface hardening is focused mainly on increasing wear resistance and static as well as dynamic fatigue strength. However, corrosion resistance and optical properties are of interest as well. In particular, hybrid processes that combine a selection of different technologies, such as plasma nitriding and PVD coating, have been established and are important for many branches of industry. Thermo-chemical technologies are used to modify chemical composition and microstructure mainly within the surface zone by thermal and chemical processes. Surface hardening essentially relies on the same basic processes, however, innovations and developments have been rapid in recent years. Examples are the development of laser heat treatment succeeding flame hardening, low-pressure carburisation instead of gas carburising or powder carburisation, and plasma nitriding as an alternative to salt bath nitriding. Non-metal Diffusion Processes Most surface hardening depends on non-metal diffusion processes. In diffusion processes, the concentration of selected elements such as carbon and/or oxygen is increased within the surface zone. Carburisation, case hardening, nitriding, and nitrocarburising are commonly used. Increased hardness values in the surface zone result from martensitic transformation in case hardening and solidsolution strengthening as well as precipitation hardening in nitriding.

2.2 Fundamentals

Correct thermo-chemical treatment includes careful control of process parameters such as temperature, pressure, time, and also the chemical activity of the reacting agents. The complete reaction involves three stages: · atmospheric transport of reacting agents · phase-boundary reaction between atmosphere and substrate material · diffusion of the component into the material’s surface. Usually, process performance is determined by the impurity diffusion rate within the substrate material [2]. The Kolsterising technology for surface hardening of austenitic steels presented here is also influenced substantially by the laws of carbon diffusion at low temperatures. Diffusion [3] Generally, diffusion is a macroscopic mass transport based on atomic motion exceeding the distance between single atoms. The driving forces of diffusion include local concentration gradients. The atomic motion via vacancies or along interstitial paths (as observed in carburisation and nitriding) leads to a concentration equilibrium. The kinetics of diffusion are defined in Fick’s laws:   @c Fick’s first law: ‰cm 2 s 1 Š jˆ D @x

Mass flux j represents the number of atoms passing through the perpendicular surface A within the time interval dt. Mass flux is directly proportional to the diffusion coefficient D. The diffusion coefficient (in cm2/s) is a temperature-dependent material parameter. Diffusion coefficient:

D ˆ D0
exp… Q=R
T†

Fick’s second law describes the change in concentration with time at a given point:  2  @c @ c Fick’s second law: ˆD @t @x2 The diffusion depth x with time is approximated by the following parabolic law: x % (D · t)–1/2, D = temperature-dependent diffusion coefficient. This chapter focuses on the non-metal diffusion process Kolsterising® used for surface hardening of austenitic stainless steels. 2.2.2 Stainless Steels

Corrosion resistance of stainless steels requires a chromium proportion of more than 12 per cent in weight as an alloy component of a substitutional solid solution within the iron lattice. The chromium content is responsible for the spon-

13

14

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

taneous formation of an extremely strong and chemically stable chromium oxide layer. This layer is commonly referred to as the passivation layer, and is the main reason for the chemical resistance of stainless steels. In 1912, the Krupp company patented the first stainless steel (18/8 CrNi steel), known as V2A-steel.

2.2.2.1 Classification of Stainless Steels A large number of today’s stainless steels are specified in the EN 10088 standard [4]. Chromium and nickel are the primary alloying elements. Depending on alloy composition, crystalline structures of stainless steels vary and thus influence usability and processing properties. In classifications of stainless steels, steel grades are named according to structural characteristics: · ferritic steels · martensitic steels · austenitic steels · ferritic-austenitic steels (duplex steels).

The amount of ferrite and austenite formers in the alloy determine the crystalline structure [5]. Ferrite-stabilising Alloying Elements Chromium – molybdenum – silicon – vanadium – tungsten – titanium – niobium – aluminium Being the most important alloying element, the ferrite former chromium promotes formation of a ferritic crystal structure. At room temperature, steels with 13–30 per cent by weight chromium and 0.1 per cent by weight carbon show ferrite structure. Carbide- and nitride-stabilising elements (V, W, Ti, Nb) support the ferrite solid solution as ferrite formers, and draw the austenite stabilising elements carbon and nitrogen from the lattice by forming precipitations. Austenite-stabilising Alloying Elements Nickel – carbon – nitrogen – manganese Nickel is a strong austenite former. The austenite field can be enlarged considerably provided that concentration of austenite formers, above all nickel between 8–30 per cent by weight, is sufficiently high. The A3 transition is shifted to lower temperatures, and thus, the austenitic structure is stable even at room temperature. The most common structure diagram for this type of steel is the Schaeffler diagram. Initially developed for welding of chromium-nickel-steels, it characterises the crystal structure after cooling from very high temperatures. Alloying elements are summarised to chromium and nickel equivalents, and determine the proportions of structural constituents.

2.2 Fundamentals

Examples of Stainless Steels See Table 1.1, p. 16.

2.2.2.2 Stainless Austenitic Steels Modern technology fields such as food processing and medical technology depend on austenitic chromium-nickel steels. They are the most commonly used stainless steels. Covering a range from X5 CrNi 1810 to X1 NiCrMoCu 2520, these steels withstand many corrosion chemical stresses. The alloying element molybdenum yields pitting corrosion resistance, particularly in chloride containing aqueous solutions. Characteristics of Austenitic Steels · high corrosion resistance · high toughness with breaking elongation values of approx. 50% · good low-temperature toughness (impact value)

Fig. 2.1 Schematic classification of heat-treatment techniques [1].

15

1.4319 1.4310 1.4449 1.4439 1.4420 1.4523 1.4583 1.4505 1.4506 1.4577 1.4465 1.4427

Austenitic steels X 5 CrNi 18 7 X 12 CrNi 17 7 X 5 CrNiMo 17 13 X 3 CrNiMoN 17 13 5 X 5 CrNiMo 18 11 X 10 CrNiMoTi 18 12 X 10 CrNiMoNb 18 12 X 5 NiCrMoCuNb 20 18 X 5 NiCrMoCuTi 20 18 X 5 CrNiMoTi 25 25 X 2 CrNiMoN 25 25 X 12 CrNiMoS 18 11

a) b)

£ 0.10 £ 0.05

1.4460 1.4582

Austenitic-ferritic steels X 8 CrNiMo 27 5 X 4 CrNiIdnNb 25 7

0.22 0.35 0.43 0.95 1.20

£ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0 £ 2.0

£ 2.0 £ 2.0

£ 1.0 £ 1.0 £ 1.0 £ 1.0 £ 1.0 £ 1.0 £ 1.0

to to to to to to to

12.5 18.5 14.0 14.0 17.5 19.0 18.0

16.0 16.0 16.0 16.5 16.5 16.5 16.5 16.5 16.5 24.0 24.0 16.5

to to to to to to to to to to to to

18.0 18.0 18.0 18.5 18.5 18.5 18.5 18.5 18.5 26.0 26.0 18.5

26.0 to 28.0 24.0 to 26.0

10.5 16.5 12.0 12.0 15.5 17.0 16.0

% Cr

to 1.3 to 1.3 to 0.8

to 2.0 to 1.3

– – 4.0 4.0 1.2 2.5 2.5 2.0 2.0 2.0 2.0 2.0

to to to to to to to to to to

5.0 5.0 1.7 3.0 9.0 2.5 2.5 2.5 2.5 2.5

1.3 to 2.0 1.3 to 2.0

– 1.5 0.9 – 0.9 0.9 0.4

% Mo

7.0 to 8.0 7.0 to 9.0 12.5 to 14.5 12.5 to 14.5 9.0 to 12.0 12.0 to 14.5 12.0 to 14.5 19.0 to 21.0 19.0 to 21.0 24.0 to 26.0 22.0 to 25.0 10.5 to 13.5

4.0 to 5.0 6.5 to 7.5

£ 0.5 £ 1.0 £ 1.0 £ 1.0 £ 1.0 – –

% Ni

Unless otherwise noted, silicon concentration is limited to 1.0 %, phosphorus to 0.045 %, and sulfur to 0.030 %. Portion of the niobium may be replaced by twice as much tantalum.

£ 0.07 £ 0.12 £ 0.07 £ 0.04 £ 0.07 £ 0.10 £ 0.10 £ 0.07 £ 0.07 £ 0.07 £ 0.03 £ 0.12

£ 0.08 £ 0.10 0.17 to 0.25 to 0.33 to 0.85 to 0.95 to

Ferritic and martensitic steels X 5 CrTi 12 1.4512 X B CrMoTi 19 1.4523 X 20 CrMo 13 1.4120 X 30 Cr 13 1.4028 X 35 CrMo 17 1.4122 X 90 CrMoV 18 1.4112 X 105 CrMo 17 1.4125

% Idn

%C

Designation

Material no.

Chemical composition

Steel grade

Table 2.1 Guaranteed chemical composition of steels (cast analysis).

– – – N 0.10 to 0.20 – Ti ³ 5 ´ % C Nb ³ 8 ´ % Cb) Cu 1.8 to 2.2; Nb ³ B ´ % Cb) Cu 1.8 to 2.20; Ti ³ 7 ´ % C Ti ³ 10 ´ % C N 0.08 to 0.16 P £ 0.06; S 0.15 to 0.95

– Nb ³ 10 ´ % C b)

Ti ³ 6 ´% C Ti ³ 7 ´ C – – – V 0.07 to 0.12 –

% other a)

16

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

2.2 Fundamentals

· · · ·

paramagnetic low 0.2% offset yield stress of 200–300 N/mm2 low wear resistance, low surface hardness values of 200 to 300 HV0.05 high adhesion tendency (cold welding, fretting).

Corrosion Resistance of Austenitic Stainless Steels Corrosion resistance is not a fixed material characteristic. In fact, boundary conditions such as chemical composition, material properties, and conditions of use greatly influence corrosion resistance. Corrosion resistance requires a stable passivation layer at the surface. A reducing environment, e.g. corrosive media such as hydrochloric, sulfuric, or phosphoric acid, can weaken or prevent formation of the passivation layer, and thereby evenly corrode the surface. Predominantly influencing parameters are temperature and concentration of the corrosive agent. In practice, steel with a weight loss of less than 0.3 g/m2 h is considered resistant. Additional influencing factors are the steel’s chemical composition and microstructure. Figure 2.2 illustrates the corrosion resistance of

Cr-Ni

Fig. 2.2 Corrosion resistance in sulphuric acid [5].

17

18

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

Fig. 2.3 Schematic diagram of the most common corrosion types [5].

different alloys exposed to sulfuric acid. Sensitisation of steel can be caused by chromium depletion in the matrix due to chromium carbide precipitation, grain-boundary segregation, delta ferrite regions within the microstructure, and strain-induced martensite, all generating potential weak points. Appropriate materials selection and heat treatment can minimise these influences. Chemical corrosion can occur as an even surface attack, or appear localised. While even surface corrosion is controllable and predictable, localised corrosion phenomena result in limited durability and incalculable working life. Selective corrosion such as pitting or crevice corrosion, which can be caused by chloride containing solutions, locally destroys the passivation layer, forms local elements, and leads to anodic dissolution of the active area. Stress corrosion cracking can appear when the part is subject to additional tensile stress. Figure 2.3 gives an overview of the most common corrosion types. Heat Treatment of Austenitic Stainless Steels Austenitic stainless steels do not show allotropic phase transformations. Therefore, classic hardening techniques are inapplicable, i.e. quenching does not lead to martensitic transformation. Heat treatment is applied after forming, e.g. when strain-induced martensite is present due to mechanical pre-treatment, or in case of steel sensitisation, e.g. caused by chromium precipitation at grain boundaries. Austenitic steels are commonly used in solution annealed and quenched condition. Solution anneal-

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels

ing prevents steel sensitisation caused by precipitation and chromium depletion in the matrix and reduces deformation stresses and strain-induced martensite. Temperatures for solution annealing range from 1000 to 1150 8C, depending on alloy composition. At this temperature, chromium carbides dissolve in the matrix. Subsequent quenching to room temperature avoids re-precipitation of carbides or intermetallic phases that could reduce corrosion resistance and cause embrittlement. At room temperature, crystal structure usually is austenitic. Heat treatment in the temperature range of 500 to 900 8C (depending on the alloy) should be avoided because chromium carbides precipitate or intermetallic phases form, e.g. tetragonal r-phase. As a result, corrosion resistance drops and the material embrittles.

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels

Conventional technologies for surface hardening that increase carbon and/or nitrogen concentration can not be applied to stainless steels unless accepting a loss of corrosion resistance. At common nitriding temperatures, chromium precipitation, and therefore chromium depletion in the matrix, occurs. 2.3.1 Kolsterising

The Kolsterising® process, developed by Prof. Kolster, is a surface hardening technique for stainless austenitic and ferritic-austenitic steels. It has been in industrial use since the late 1980s. The objectives of the process are [6]: · surface hardness values between 1000 and 1200 HV0.05 · no loss of substrate corrosion resistance · increased wear protection, especially against adhesion (cold welding) · no change in shape, dimensional accuracy, or colour · uniform hardening effect, even in borings and cracks. The technology is used for near-surface carburisation at temperatures below 500 8C. Due to low process temperatures, carbon atoms remain in solid interstitial solution. Chromium carbide precipitation does not occur, and thus, corrosion protection persists. The solid solution of carbon in the fcc lattice produces internal compressive stresses, increasing hardness values to 1200 HV. The obtainable diffusion depth of carbon depends on process time. Industrial suppliers offer diffusion depths of 20 to 40 lm. Figure 2.4 illustrates the diffusion zone versus hardness HV0.05. The results presented here refer to solution annealed steel, 1.4435 type, Kolsterised to a depth of 35 lm.

19

20

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

Fig. 2.4 Hardness distribution HV 0.05 of Kolsterised coating.

2.3.1.1 Influence on Microstructure Figure 2.5 shows scanning electron microscope images of a polished surface before and after Kolsterising® [6]. The internal compressive stress, caused by forced solid solution of carbon, locally activates slip planes for strain relief. Slip lines on the surface confirm this micro-deformation. The influence of the process on surface roughness is shown in Fig. 2.5. In practice, roughness values Ra of more than 0.2 lm do not increase significantly. Light-optical microscopic examination reveals the carburised near-surface zone, visible as a white seam, thickness approx. 35 lm, known as the S-phase, Fig. 2.6 [7]. The microstructure near the surface is unchanged and this again indicates that the surface hardening method is a diffusion process. Furthermore, even grain boundaries in the hardened zone are visible, even though the layer is less sensitive to etching. The microstructure shows that the composition of the hardened surface zone is homogeneous and precipitation-free.

a

Fig. 2.5 Scanning electron microscopic image of an austenitic CrNiMo steel surface before and after Kolsterising. (a) 1.4435, polished with Al2O3, (b) 1.4435, polished with Al2O3, Kolsterised.

b

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels Fig. 2.6 Optical microscope image of microstructure in the near-surface zone of a Kolsterised austenitic CrNiMo steel.

Radiographic analyses of Kolsterised and untreated specimens indicate that austenite is the only phase present, and reveals that the lattice is distinctly deformed. Internal compressive stresses determined by the sin2w method reach –1390 MPa.

2.3.1.2 Influence on Chemical Composition Auger analysis of a Kolsterised specimen is presented in Fig. 2.7. The point analysis in Fig. 2.7 b verifies that the steel’s qualitative composition is unchanged. Only the quantitative proportion of carbon is increased considerably in the near-surface zone compared to the substrate (Fig. 2.7 c). The fact that no foreign alloying elements are added makes the technology particularly promising for applications in food processing, medical, and pharmaceutical industry.

2.3.1.3 Influence on Mechanical Properties Investigations of treated and untreated specimens showed no significant change of mechanical properties such as tensile and impact strength. Due to the internal compressive stresses in the near-surface zone, the dynamic fatigue strength of austenitic stainless steels can be increased considerably using this surface hardening technology (Fig. 2.8).

2.3.1.4 Wear Resistance Surface hardening is usually used to increase wear resistance, and particularly, to provide anti-galling characteristics in austenite/austenite material combinations. The influence of Kolsterising on friction coefficients and mass loss of the friction pairing has been investigated with pin-on-disc tests. In this test a pin is pressed onto a rotating disc with defined perpendicular force. In the test, the frictional force is measured and, after a certain test period, the mass loss of the pin and the disc is determined [9].

21

22

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance Fig. 2.7 Auger analysis of Kolsterised near-surface zone.

Tests were carried out under dry wear conditions and with 0.9% physiological saline solution. Figure 2.9 shows that surface hardened specimens are subject to considerably less mass loss. This effect is particularly noticeable under corrosive strain (0.9% NaCl solution). Summarising the results, wear resistance under dry conditions is increased 80-fold and in 0.9% saline solution 180-fold.

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels

Fig. 2.8 Influence of near-surface modification on fatigue strength.

Fig. 2.9 Influence of Kolsterising on wear behaviour in pin-ondisc test. Load 10 N, rotational speed 46 revs/min, disc diameter 33 mm, duration 2.5 h.

2.3.1.5 Influence on Corrosion Resistance The influence of surface hardening on corrosion resistance of austenitic stainless steel is crucial for process applications in many fields, e.g. food processing and medical technology. Investigations of pitting corrosion resistance according to ASTM G48 in 10% FeCl3 solution reveal the influence of Kolsterising with different diffusion depths on the critical pitting corrosion temperature (Fig. 2.10). No negative influence of the process on the critical pitting corrosion temperature was found,

23

24

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

b Fig. 2.10 (a) Microstructure after corrosion test. (b) Influence of Kolsterising on critical pitting corrosion temperature of an austenitic CrNiMo steel in 10% FeCl3 – test according to ASTM.

furthermore, values increased slightly. Edges that are usually particularly affected are better protected as well. Examinations of susceptibility to stress corrosion cracking using 10% FeCl3 solution and 80% Rp0.2 tensile strain also show distinct improvement compared to untreated samples [10]. Apart from the positive results in chloride-containing media, tests under other conditions such as NaOH and H2SO4 are successful as well. However, it is important to note that Kolsterising is limited to applications where the material is in a stable passive condition, i.e. under certain conditions corrosion resistance may decrease. Boiling acetic acid (30–50%) produces noticeable corrosion by destroying the passivation layer, and also corrodes the Kolsterised layer. The substrate material should also be free of potentially weakening areas with negative influence, e.g. delta ferrite and strain-induced martensite. 2.3.2 Kolsterising plus PVD Coating

Since plasma nitriding combined with CrN coatings was first used some years ago to protect forming tools, a number of technologies combining diffusion processes and hard-material coatings have evolved. Plasma nitriding produces a dif-

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels

Fig. 2.11 Hardness distribution of Kolsterised + DLC-coated specimen.

fusion zone with internal compressive stress forming a supporting layer for the subsequently deposited CrN layer. The mechanical properties such as coating adhesion improve considerably. This established a basis for combining Kolsterising with friction-reducing coatings applied by PVD technology (unbalanced magnetron sputtering) for lubricant-free applications that require minimal friction and/or high wear resistance [11]. Solution annealed 1.4404 with defined roughness values of Ra = 0.15 lm and Rz = 0.82 lm was used as substrate material. Investigations covered and compared: · untreated substrate · Kolsterised substrate · substrate coated with W-C : H and a-C : H (DLC coating) · substrate Kolsterised and coated with W-C : H as well as a-C : H (DLC coating). After duplex treatment, the hardness of the 3-lm DLC coating at the surface reaches 3000 HV0.05, followed by a diffusion zone of approx. 25 lm. In the interface between DLC coating and diffusion zone, hardness values drop to approx. 1000 HV0.05. Within the diffusion zone, hardness gradually decreases and finally reaches the initial substrate hardness of 280 HV0.05 (Fig. 2.11).

2.3.2.1 Coating Adhesion Scratch tests were performed in order to assess adhesion of the DLC coating, comparing duplex treatment and purely DLC-coated samples. In this test a hard metal tip is pressed to, and drawn across the surface with progressive loads. The penetration depth into the surface is then measured. The increase in coat-

25

26

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

Fig. 2.12 Scratch test – adhesion measurement.

ing adhesion for duplex treatment by a factor of 40 is shown in Fig. 2.12. This effect is attributed to the diffusion zone that mechanically supports the top layer [11].

2.3.2.2 Wear Resistance Wear resistance is examined by pin-on-disc tests. In order to simulate true operating conditions as closely as possible, the pinon-disc tests are carried out using a 100Cr6 (AISI 52100) ball, Æ = 6 mm, with a perpendicular load of 5 N, and speed of 0.1 m/s without any lubricant added. Friction coefficients of DLC-coated and duplex-coated samples (l = 0.12) are considerably lower than for untreated material (l = 0.75). The wear coefficients of samples and balls after the test are shown in Fig. 2.13. All three surface treatments produce positive results compared with untreated substrate material. Best results are achieved using duplex surface treatment, reducing wear coefficients by a factor of 5000. Wear of the counterpart is also reduced considerably by a factor of 350 [11].

2.3.2.3 Fatigue Strength In order to assess the fatigue strength of DLC and duplex-coated samples, a hard metal ball (Æ = 5 mm) is loaded onto the surface with a force of 100 N in an impact test with 1 million cycles at a frequency of 50 Hz. This test particularly demonstrates the advantages of the duplex treatment. The hard metal ball produces a distinct mark on the surface with a diameter of 500 lm on samples coated with DLC (Fig. 2.14). Within the mark, the coating suffers concentric cracks.

2.3 Technologies for Surface Hardening of Austenitic Stainless Steels

Fig. 2.13 Pin-on-disc test: wear measurement.

a)

b)

Fig. 2.14 Fatigue test: surface image after continuous test.

Marks produced on duplex-treated samples are considerably smaller with diameters of only 250 lm. Here, no cracks develop within the coating. The coating is thus crack-free. Transfer of the results presented here to other coatings is limited. Wear and corrosion values are typical system properties and results were obtained under laboratory conditions. Therefore, these values are not universally valid for real tribological systems [11].

27

28

2 Stainless Austenitic Steel – Surface Hardening for Increased Wear Resistance

Fig. 2.15 Application examples.

2.4 Applications

Combinations of increased wear resistance, fatigue strength, and retained corrosion resistance are beneficial for many applications (Fig. 2.15). Applications are found in miscellaneous industrial areas. The pump industry is one of the most important industrial fields. The service life of dosing pumps for food processing was increased substantially. Furthermore, soft packings became dispensable due to the anti-galling properties of metal pump components operating in direct contact. Fretting was prevented and the relative motion of metal parts was trouble-free. Hygiene was improved and maintenance effort reduced. One of the first applications was found in cutting rings for tube fittings. When the swivel nut is screwed tight, cutting rings penetrate stainless steel tubes to a depth of approx. 0.1 mm with their cutting edges in order to obtain a secure joint. The hardness of the cutting ring needs to be higher than that of the tube, and at the same time, corrosion resistance must be maintained. 2.4.1 Application Limitations

Available chamber sizes limit applications of the Kolsterising® process. Chambers are limited to a diameter of 480 mm and height of 560 mm. The substrate material affects the results of surface hardening. Delta ferrite portions or strain-induced martensite impair results. However, appropriate surface machining or heat treatment prior to Kolsterising® can avoid these negative influences.

References

2.5 Outlook

The stainless steel market has been growing continuously for years. Stainless austenitic steel with its excellent properties can be utilised for a new range of applications when combined with Kolsterising®. In cases where other materials or constructions were necessary due to reduced wear resistance and/or cold welding, austenitic steel with high corrosion resistance and Kolsterised surface can now be used. In addition, new component properties are achieved by combining surface hardening and PVD technology. Although their application potential is still under investigation, results are very promising.

References 1 Vollertsen, F., Vogler, S.: Werkstoffei-

2

3

4 5

6

7

genschaften und Mikrostruktur. Hanser, Munich, 1989 Eckstein, H. J.: Technologie der Wärmebehandlung von Stahl. VEB Deutscher Verlag für Grundstoffindustrie, Leipzig, 1977 Böhm, H.: Einführung in die Metallkunde. BI Wissenschaftsverlag, Mannheim, 1987 EU Standard EN 10088 Jäniche, W., et al.: Werkstoffkunde Stahl, Vol. 2: Anwendungen, Chapter D13 Nichtrostende Stähle. Verein der Eisenhüttenleute. Springer, Berlin, 1985 Gümpel, P., et al.: Nichtrostende Stähle. Expert, Kontakt & Studium, Vol. 493, 1996 Gümpel, P., Waegner, M.: Randschichthärten von austenitischen und ferritisch-

8 9

10

11

austenitischen Stählen. Stahl, December 2002, issue 6, pp. 28–32 Bell, T.: Surface hardening of austenitic stainless steel Gümpel, P., v. d. Jagt, R.: Kolsterizing: A Novel Surface Hardening Treatment for Austenitic and Duplex Stainless Steels. Proceedings, 4th ASM Heat Treatment and Surface Engineering Conference, Vol. 3, October 19th–21st, 1998, Florence, Italy Rey, O., Jacquot, P.: Kolsterising: the hardening of austenitic stainless steel. Proceedings, 4th European Stainless Steel Conference, pp. 251–253, Paris, 2002 Hurkmans, T., et al.: Duplex treatment of austenitic stainless steel for improved corrosion and wear resistance. ASM Conference on Surface Technology, San Diego, April 2004

29

31

3 Fundamentals of Thin-film Technology M. Nicolaus, Institute for Materials Science, University of Hannover, Germany M. Schäpers, Draeger Medical, Lübeck, Germany 3.1 Introduction

The numerous applications of modern surface and thin-film technology appear in all aspects of everyday life. They are necessary for products such as compact disks, solar cells, heat-protection glass, or in modern production systems, e.g. for machining. Although their process technology is costly, physical vacuum coating systems are common for producing functional thin films on an industrial scale. They allow application-specific deposition of high quality coatings on many different surfaces.

3.2 Classification of Thin-film Coating Processes

Coating technologies are commonly classified according to coating thickness. Usually, those up to 10 lm are considered thin-film coatings. The main aspect, however, is that thin films can have completely different properties compared to bulk material and thicker coatings, in part due to their different microstructure. Common processes in recent thin-film technology include PVD and CVD technology as well as chemical and electrochemical plating (Fig. 3.1). The fact that coating material is applied to the substrate on an atomic scale, in the form of individual atoms, ions, clusters, or molecules, is common to nearly all processes. PVD and CVD technologies deposit coating material via the gas phase, whereas chemical and electrochemical processes operate in aqueous media.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

32

3 Fundamentals of Thin-film Technology Fig. 3.1 Classification of thin-film techniques.

3.3 General Aspects of Gas-phase Coating Processes 3.3.1 PVD – Physical Vapour Deposition

In physical vapour deposition (PVD), coating material is deposited from the gas phase without any chemical reactions being involved. Developed in the 1960s, the technology allows thin-film deposition of metalloid and ceramic coating materials on virtually any technically relevant substrate. Apart from alloys and metals, ceramic, as well as glass substrates, even plastics can be PVD coated using pulsing in order to reduce process temperature. PVD coatings are used in various applications. Not only applicable for conventional wear-resistance coatings, PVD can produce decorative coatings, thermal-barrier coatings, and optical surfaces, e.g. for microelectronics. Usually, PVD involves high-vacuum technology. Three main process variants are common: evaporation, ion plating, and sputtering (Fig. 3.2).

3.3.1.1 Evaporation The coating material for evaporation is heated in a crucible, e.g. resistance heated, until it evaporates. Provided that vacuum is sufficiently high (10–3 to 10–6 Pa), the mean path length of vapour atoms is high enough for negligible particle interaction. Particles therefore move towards the substrate and the walls of the chamber on a straight trajectory, and are deposited, thus producing the coating layer. Vacuum evaporation features relatively high evaporation rates (e.g. 10–3 g/cm2 s) and evenly distributed coating thickness with sufficient adhesion. The throwing

3.3 General Aspects of Gas-phase Coating Processes

Fig. 3.2 Basic variants of PVD.

power, i.e. the ability to evenly coat parts with complex shapes, is relatively low due to the straight particle trajectories inherent to the process. Molecular beam epitaxy (MBE) is one of the most important evaporation methods for production of well-defined thin films. This technique allows oriented growth of the coating on the crystalline substrate surface. The method is used primarily to produce high quality coating structures for semiconductors (Fig. 3.3 [1]). One advantage of MBE is the ability to yield extremely sharp interfaces between coating layers. 3.3.1.2 Sputtering Unlike evaporation in high vacuum, sputtering requires an inert process gas such as argon and a high-voltage source. The mean path length of argon at a pressure of 0.1 to 1 Pa is several millimetres. In DC sputtering, a constant potential of –1 to –5 kV is applied to the coating material and a glow-discharge plasma is ignited. The coating material is then the target of a constant bombardment by ionised gas particles with high kinetic energy. Due to this effect, coating materials in sputtering processes are referred to as sputter targets, or simply targets. The plasma burns between the target cathode and the grounded substrate as well as the walls of the high-vacuum chamber (recipient), both working as anodes. Compared to evaporation, selection of coating material for sputtering is far less limited because the target material is not released thermally but by impact processes. The method is independent of the coating material’s vapour pressure. Sputtering yields slow deposition rates (< 10–4 g/cm2 s), good coating adhesion, but relatively uneven coating thickness. Due to higher process gas pressure, the particle interaction of vapour and process gas is higher, resulting in better throwing power.

33

34

3 Fundamentals of Thin-film Technology

Fig. 3.3 Schematic representation of a system for molecular beam epitaxy. Transmission electron microscopic image of a Si/SiGe superlattice cross section.

Fig. 3.4 Schematic representation of a magnetron cathode with symmetric field lines (balanced magnetron). (a) Magnetic field (top view). (b) Cross section of field lines between poles.

DC sputtering has been developed further, resulting in high-frequency sputtering (RF sputtering) that has paved the way for additional applications. Here, an HF source produces a high frequency. All surfaces are negatively charged compared to the plasma and acquire a self-bias due to electrons with high mobility that arrive at the surfaces much quicker than the slower plasma ions. Layers of charged particles that can be assigned to different capacities then appear between the cathodes and anodes.

3.3 General Aspects of Gas-phase Coating Processes

In widespread magnetron sputtering, permanent magnets are arranged so that the magnetic flux lines penetrate the target vertically (Fig. 3.4). The superposition of electric and magnetic fields causes the electrons to follow a spiral motion in front of the cathode. The increased probability density of the particles thus leads to increased ionisation and sputtering rate.

3.3.1.3 Ion Plating In ion plating, the properties of thin surface zones are modified selectively by ion implanting. The substrate and coating are subject to a continuous bombardment of high-energy particles, prior to and during coating growth, resulting in increased coating adhesion and optimised microstructure within the coating. The bombardment is created by additional anodes placed within the glow-discharge volume, as well as electromagnetic fields (permanent magnets) behind and in front of the target. Deposition rates (speed of coating growth) are high in ion plating, while substrate temperature remains relatively low (T < 300 8C). Ion plating produces coatings with good to very good adhesion and high deposition rates reaching 10–2 g/cm2 s, however, coating thickness is rather unevenly distributed. With the PVD processes presented here, hard coatings of nitrides, carbides, or carbonitrides can be produced in high-vacuum processes, when reactive gases such as nitrogen (N2) or acetylene (C2H2) are employed. Many PVD coatings used nowadays, especially for wear resistance, are produced by reactive ion plating. These processes generally differ in the type of evaporation used for the metal coating component. For evaporation either an electron beam (anode target), arc (arc target), or magnetron sputtering is applied. Table 3.1 summarises characteristic properties of typical PVD processes.

Table 3.1 Summary of typical PVD processes. Process

Deposition rates (lm s–1)

Pressure (Pa)

Particle energy (eV)

Process temperature

Adhesion

Evaporation Sputtering Ion plating

0.05–25 0.0001–0.7 0.01–25

10–3 10–1–1 10–1–1

Mz+ (phase II) + ze–

(1)

If the forward reaction (oxidation) in Eq. (1) is predominant, i.e. metal dissolution forms cations (positive ions), the electrode becomes negatively charged due to the accumulation of electrons at the surface. In contrast, a predominant reverse reaction will cause the electrode to be charged positively because electrons are drawn from the surface. A positive electrode surface attracts hydrated anions and forms an electrolytic double layer (Fig. 3.8), described in the Helmholtz model of the double layer, and also referred to as the Helmholtz plane. Due to heat-induced movement of the molecules, the double layer is followed by a zone of diffuse charge distribution, increasing with distance to the electrode. This diffuse double layer is described in the Gouy-Chapman model. The equilibrium reaction is influenced by the resulting electrical potential difference between electrode and electrolyte, and this variation must be considered when formulating the conditions of equilibrium. The energy necessary for an ion i with charge zie to enter, from infinity, the inside of a mixed phase with the electrical potential ', is not only the chemical potential – the partial molar free enthalpy – but in addition zie' or ziF' (per mole i). This value is known as the electrochemical potential: li ˆ li ‡ zi  e
NA
' ˆ li ‡ zi
F
' li : li : zi: e: F: ':

electrochemical potential of i chemical potential of i charge number of i elementary charge Faraday’s constant electrical potential

Fig. 3.8 Rigid (Helmholtz) and diffuse (Gouy-Chapman) electrolytic double layer.

…2†

41

42

3 Fundamentals of Thin-film Technology

The conditions of equilibrium require every particle type to have the same electrochemical potential in every phase. Considering the stoichiometric coefficients of reactants and products, we find for the phase boundary metal/metal ions: lM ˆ lMz‡ ‡ z
le

…3†

Ultimately, the above considerations yield the equilibrium Galvani potential difference: D' ˆ '…II† '…a† : R: T: aMz+:

'…I† ˆ '0 ‡

R
T
ln aMz‡ z
F

…4†

Galvani potential of phase a gas constant absolute temperature activity of metal ions

Equation (4) describes that an increase in activity of the metal ions, e.g. by an increase in metal ion concentration, causes the potential difference of the metal electrode to rise, and therefore, the metal electrode is more positively charged than the solution. As the potential at the phase boundary metal/metal ions (Galvani potential) defies experimental measurement, a reference point must be defined. The standard hydrogen electrode (SHE) can serve as a reference electrode, when assigned a potential of zero. In the SHE, a platinum plate, additionally platinised, is immersed in an acid aqueous solution with hydrogen (H+) ion activity of 1. The platinum plate is constantly kept saturated with hydrogen by a continuous hydrogen gas stream at a pressure of 1013.25 mbar (1 atmosphere). The equilibrium potential of the metal electrode is given by the Nernst equation: e ˆ e0 ‡

R
T
ln aMz‡ z
F

…5†

e, e0: electrode and standard electrode potential of the metal electrode The SHE serves as a reference electrode and fixed reference point, and it allows different electrode systems to be compared. The resulting table of electrode potential [14] is known as the electrochemical series (Table 3.3).

3.6.2.3 Electrolysis and Faraday’s Laws Electrolysis is a process where a substance is dissociated electrochemically due to an external source of direct current that passes through the substance, and alters the equilibrium at the phase boundary metal/metal ions. In the electrolysis cell, anions move towards the positively charged anode (oxidation) whereas cations are attracted by the negative cathode (reduction). Michael Faraday formulated the fundamental laws of electrolysis in 1833. Faraday’s law describes

3.6 Electrodeposition and Electroless Plating Processes Table 3.3 Standard electrode potential of selected metals at 25 8C [14]. Electrode reaction

EMF (V)

Electrode reaction

EMF (V)

Na+ + e– ?Na Mg2+ + 2e– ? Mg Al3+ + 3e– ? Al Ti2+ + 2e– ? Ti V2+ + 2e– ? V Cr2+ + 2e– ? Cr Zn2+ + 2e– ? Zn Cr3+ + 3e– ? Cr Fe2+ + 2e– ? Fe In3+ + 3e– ? In Ni2+ + 2e– ? Ni 2 H+ + 2e– ? H2

–2.71 –2.36 –1.66 –1.63 –1.19 –0.91 –0.76 –0.74 –0.44 –0.34 –0.23 ±0.00

Cu2+ + 2e– ? Cu Cu+ + e– ? Cu Ag+ + e– ? Ag Hg2+ + 2e– ? Hg Pt2+ + 2e– ? Pt Au2+ + 2e– ? Au Au3+ + 3e– ? Au

+0.34 +0.52 +0.80 +0.80 +1.20 +1.36 +1.69

how the amount of electric charge flowing through the electrolyte influences the rate of dissociation. The rate of dissociation per unit time and area is: 1 dni 1 dQ 1 

j ˆ
ˆ A dt zi
F A
dt zi
F

…6†

The dissociated mass amounts to: 1 dmi 1 dQ 1
ˆ ˆ
Mi
Mi
j A dt zi
F A
dt zi
F

…7†

The deposition rate is an additional value that is of particular interest for metal plating. The following equation takes into account the density of the deposited metal: dx 1 dmi 1
ˆ
dt A dt qi A: ni: zi: F: Q: j: mi: Mi: qi: x:

surface area of electrodes amount of substance i charge number of i Faraday constant quantity of charge electric current density mass of i molar mass of i density of i thickness of coating

…8†

43

44

3 Fundamentals of Thin-film Technology

Faraday’s law refers only to the Faraday current, which is caused directly by the electrode reaction. The capacitive current (occasionally called non-Faraday current) originates from charge building up in the double layer and is not taken into consideration in the above equation. In this context, a further value is introduced – the current efficiency: fˆ

mobs: mtheor:

…9†

The equation describes the ratio of observed metal dissolution to theoretical dissolution as predicted by Faraday’s law. Usually, current efficiency is below 1, which means that additional reactions take place, e.g. hydrogen development occurring at the cathode.

3.6.2.4 Overpotential The overpotential g is the difference in the electrode potential of an electrode between its equilibrium potential and its operating potential when a current is flowing. At low current density, limited speed of charge transfer through the phase boundary electrode/electrolyte alters the electrode potential. This type of overpotential is known as the charge-transfer overpotential gT. The charge-transfer overpotential depends on the species involved in transfer as well as the type of electrolyte and electrode. Transfer limitation usually occurs along with other limitations due to slow mass transport from the electrode surface to the electrolyte and vice versa (diffusion limitation). Consequently, this type of overpotential is called the diffusion overpotential gD. Furthermore, insufficient reaction processes of accompanying reactions, e.g. adsorption and desorption processes as well as pre- and post-reactions, add to the reaction overpotential gR. 3.6.3 Electroless Plating

Electroless or chemical plating takes place without any external source of current. Metal ions in the electrolyte solution are reduced and deposited on the workpiece. This type of coating process usually yields coating thicknesses of up to 10 lm. Metal ions can be reduced using two different approaches. When the standard electrode potential of the deposition metal is more positive than the workpiece, a spontaneous redox reaction develops. Considering the reaction Fe…s† ‡ Cu2‡ …l† ! Fe2‡ …l† ‡ Cu…s† the corresponding part reactions and standard electrode potentials are: I II

Cu2+ + 2e– ? Cu Fe2+ + 2e– ? Fe

e0 = +0.34 V e0 = –0.44 V

3.6 Electrodeposition and Electroless Plating Processes

The total electromotive force (EMF) for this reaction is De = 0.78 V. The electromotive force is related to the standard reaction Gibbs energy: Dr G ˆ

z
F
De

…10†

The standard reaction Gibbs energy amounts to –151 kJ/mol. Therefore, the forward reaction is spontaneous. The second approach for electroless plating uses a reducing agent that is added to the electrolyte and allows metal deposition. The reactions involved are catalytic. The chemical reactions occur, due to the catalytic nature of the substrate surface in coating of metal substrates. For non-metal parts to be processed, activation of the surface with an appropriate catalyst is inevitable. Surface activation is a multi-step process: 1) surface cleaning (brushing, blasting, degreasing) 2) etching (acid or alkaline) 3) neutralising 4) applying the catalyst (for non-metal workpieces) 5) adding accelerator. Etching not only cleans the surface but increases surface roughness of the workpiece, providing an anchor effect for subsequent metal deposition. Usually, palladium is used as a catalyst. In a solution of PdCl2 and SnCl2, colloidal palladium is formed, and bound to excess tin(II) chloride: 2‡

0

Pd Cl2 ‡ SnCl2 ! Pd
SnCl2 ‡ additional tin(IV) complexes Accelerators, e.g. citric, hydrochloric, oxalic, or fluoroborate acid, are added to remove the tin(II) chloride and additional tin(IV) complexes bound to the palladium. After removing the tin(II) chloride, palladium deposits on the workpiece surface and provides nucleation sites (Fig. 3.9). 3.6.4 Electrodeposition of Metal

Coating a substrate with metal or an alloy using an external source of direct current, is referred to as electrodeposition or galvanic deposition. The deposition metal is provided as an electrolyte solution with metal ions. The substrate to be coated serves as cathode (negative pole) where metal ions from the electrolyte are reduced and deposited elementally at the surface. Corresponding to the charges involved, oxidation takes place at the anode (positive pole). Two types of anodes are used: 1) Soluble anode: The anode is made from the same metal as the coating so that the electrolyte solution does not suffer depletion of metal ions (Fig. 3.10). Consequentially, spent anodes need to be replaced from time to time.

45

46

3 Fundamentals of Thin-film Technology

Fig. 3.9 Principle of electroless metal plating on electrically non-conductive substrates.

2) Insoluble (inert) anode: Inert anodes are made from lead alloys or precious-metal-coated titanium plates. When using this type of anode, monitoring of the electrolyte is necessary in order to determine whether adding of additional metal ions is advisable. Electrodeposition is applicable only when the substrate material is an electrical conductor. Therefore, when coating of insulators, e.g. plastics, is desired, the surface of the material affords pre-coating with a thin conducting film. For precoating, the surface is activated and subsequently coated using electroless plating, PVD, or CVD processes. The surface is then ready for electroplating. In aqueous electrolytes though, electroplating is limited by the standard electrode potential (see above) of the coating material. Under standard conditions, thermodynamics require the standard electrode potential to be more positive than that of the standard hydrogen electrode for deposition to occur. However, as practice shows, deposition of metals like nickel, chromium, or indium from

3.6 Electrodeposition and Electroless Plating Processes

Fig. 3.10 Principle of metal electroplating (soluble anode).

aqueous electrolytes is possible although their standard electrode potential is more negative than that of the SHE (Table 3.3). This is due to inhibited hydrogen production in such systems. Deposition potential versus equilibrium potential therefore shifts to the cathodic side. 3.6.5 Electrodeposition of Metal from Non-aqueous Solvents

For many technically important metals, e.g. magnesium, aluminium, titanium, vanadium, or tungsten, the deposition potential is strongly negative. Only hydrogen would develop in aqueous electrolytes. Therefore, these metals can only be deposited from non-aqueous electrolytes. The most common example of this is aluminium produced from a liquid solution of aluminium oxide and cryolite (Al2O3/NaAlF4). Due to high temperature (900–1000 8C), liquid aluminium is formed, qualifying this method not only for galvanising of parts and electroforming. Bunsen was the first to investigate the electrolytic deposition of aluminium from an aluminium(III) chloride/sodium chloride molten salt in 1854. The first patent involving aluminium deposition from molten fluoride salt was granted in 1924. All commonly used inorganic molten salt is based on a mixture of aluminium(III) chloride and alkali metal halogenide. The main advantage, high conductivity in the range of 5 to 10 X–1 cm–1, originates from complete dissociation of the alkali metal halogenide. Furthermore, metal halogenides, e.g. AlCl3, dissolve well in liquid alkali metal halogenides [15]. The main disadvantages of the technique are high process temperature (> 350 8C), corro-

47

48

3 Fundamentals of Thin-film Technology

Fig. 3.11 Aluminium coating produced by means of electroplating.

siveness of the chemicals involved, and the sensitivity of salt to atmospheric humidity [16]. Current investigations therefore focus on complex fluoroaluminates serving as electrolytes (Fig. 3.11) with high application potential due to preferable chemical properties (hydrolysis immunity, low corrosiveness) [17]. Apart from aluminium, other metals have been deposited from inorganic molten salt, e.g. titanium, chromium, molybdenum, tungsten, niobium, and tantalum. Molten salt here is generally a KX-LiX (X = Cl, F) type mixture where the deposition metal is a complex anion, e.g. potassium heptafluorotantalate (K2TaF7) [18, 19] for tantalum deposition. In 1948, Hurley and Wier were the first to use organic ionic solvents for production of non-aqueous electrolytes in aluminium deposition. They investigated mixtures of N-ethyl pyridinium chloride/aluminium(III) chloride and N-ethyl pyridinium bromide/aluminium(III) chloride [20, 21]. Ionic liquids, also referred to as room temperature molten salts (RTMS), are molten salts with melting points below 100 8C, and represent a new class of non-molecular, ionic liquids. Generally, the term refers to organic salt containing an organic cation and a significantly smaller inorganic anion [22]. Due to reduced ion mobility, the conductivity of these systems is considerably lower (approx. 10–1 X–1 cm–1) than in inorganic molten salt [23]. The most prominent example of such ionic liquids is 1-ethyl-3-methyl-imidazolium chloride/aluminium chloride (EMICl/AlCl3) (Figs. 3.12 and 3.13). In this electrolyte, formation of [AlCl4]– (1 : 1 composition) and [Al2Cl7]– (high-AlCl3 side) complexes is observed. Cathodic deposition of aluminium, here, occurs due to reduction of the [Al2Cl7]– complex [24, 25]: 4‰Al2 Cl7 Š ‡ 3e ! Al ‡ 7‰AlCl4 Š Experimental investigations yield that aluminium deposition on the low-AlCl3 side, up to 1 : 1 composition, was unfeasible because the organic cation reduced more easily than the [AlCl4]– complex. Development of customised system and process technology was necessary due to the disadvantages related to air and humidity sensitivity of the aluminium electrolyte. Examples are airtight cham-

3.6 Electrodeposition and Electroless Plating Processes

Fig. 3.12 1-ethyl-3-methyl-imidazolium chloride/aluminium chloride.

Fig. 3.13 Liquidus curve of EMICl/AlCl3.

bers and sluice systems. Consequently, operating costs for such systems are higher than for conventional galvanotechnics. 3.6.6 Summary and Outlook

The fundamentals of galvanotechnics are based on the laws introduced by Michael Faraday in 1833 and Walter Nernst in 1898. Since then, metal and alloy deposition have developed continuously. Galvanotechnics provide decorative and corrosion-resistant coatings and are used in electronics as well as the printed circuit board industry. Electrolyte baths and compositions are subject to continuous development and optimising in order to meet rising demands from industry. For this, experimental electrochemical procedures are necessary, e.g. current density/potential measurements. The results yield possible reaction mechanisms in metal deposition processes, and ultimately allow optimisation of electrolyte compositions.

49

50

3 Fundamentals of Thin-film Technology

References 1 Paul Scherrer Institute, Switzerland,

2 3

4

5

6

7

8 9

10

11

12

13

14

15

website http://lmn.web.psi.ch/shine/sigec. htm Frey, H., Kienel, G.: Dünnschichttechnologie. VDI, Düsseldorf 1987 Eckert, G., Riemann, K.-U.: Schuhmacher, A.: Zum Technologiepotential der Plasmaphysik, Vol. 5. Series Technologie Aktuell, V. Bödecker Ed., VDI, Düsseldorf 1985 Huddlestone, R. H., Leonard, S. L.: Plasma Diagnostic Techniques. Academic Press, New York, 1965 Franz, G.: Oberflächentechnologie mit Niederdruckplasmen. Beschichten und Strukturieren in der Mikrotechnik. Springer, 1994 Grill, A.: Cold Plasma in Materials Fabrication. From Fundamentals to Applications. IEEE Press, New York, 1994 Musil, J., Kadlec, S., Vyskocil, J., Poulek, V.: Reactive deposition of hard coatings. Surface and Coating Technology, 39/40 (1989) pp. 301–314 Movchan, A., Demchishin, A. V.: Phys. Met. Metallogr. 28 (1969) p. 83 Thornton, J. A.: High rate thick film growth. Annual Review of Materials Science, 7 (1977) pp. 239–260 Messier, R., Giri, A. P., Roy, R. A.: Revised structure zone model for thin film physical structure. Journal of Vacuum Science & Technology A2(2) (1984) pp. 500–503 Atkins, P. W.: Physikalische Chemie. VCH Verlagsgesellschaft mbH, Weinheim, 1st Ed., 1988 Hamann, C. H., Vielstich, W.: Elektrochemie. Wiley-VCH, Weinheim, 3rd completely revised edition, 1998 Bockris, J. O’M., Reddy, A. K. N.: Modern Electrochemistry. Plenum Press, New York, 1970 Antelman, M. S.: The Encyclopedia of Chemical Electrode Potentials. Plenum Press, New York, 1982 Øye, H. A.: The Power of Thermodynamic Modelling – Examples from Molten Halide Mixtures. In: Metallurgical and Materials transactions B, 31, 2000, No. 4, pp. 641–650

16 Blander, M.: Molten Salt Chemistry. In-

terscience, New York, 1964 17 Holländer, U.: Unpublished research re-

18

19

20

21

22

23

24

25

sults, Institute for Materials Science. University of Hanover White, S. H., Twardoch, U. M.: The chemistry and electrochemistry associated with electroplating of group VIA metals. In: Journal of Applied Electrochemistry, 17, 1987, pp. 225–242 Lantelme, F., Barhoun, A., Li, G.: Electrodeposition of Tantalum in NaCl-KClK2TaF7 Melts, in: Journal of the Electrochemical Society, 139, 1992, No. 8, pp. 1249–1255 Hurley, F. H., Wier, P. T.: Electrodeposition of Metals from Fused Quaternary Ammonium Salts, in: Journal of the Electrochemical Society, 98, 1951, No. 5, pp. 203–206 Hurley, F. H., Wier, P. T.: The Electrodeposition of Aluminium from Nonaqueous Solutions at Room Temperature. In: Journal of the Electrochemical Society, 98, 1951, No. 5, pp. 207–212 Wasserscheid, P., Keim, W.: Ionische Flüssigkeiten – neue „Lösungen“ für die Übergangsmetallkatalyse, in: Angewandte Chemie, 2,000, 112, pp. 3926– 3945 Hussey, Ch. L.: Room Temperature Molten Salt Systems, in: Mamantov, G. Ed.: Advances in Molten Salt Chemistry, Vol. 6. Plenum Press, New York, 1987, pp. 185–230 Liao, Q., Pitner, W. R., Stewart, G., Hussey, Ch. L.: Electrodeposition of Aluminium Chloride-1-Methyl-3-ethylimidazolium Chloride Room Temperature Molten Salt + Benzene. In: Journal of the Electrochemical Society, 144, 1997, No. 3, pp. 936–943 Pitner, W. R., Hussey, C. L.: Electrodeposition of Zinc from the Lewis Acidic Aluminium Chloride-1-Methyl-3-ethylimidazolium Chloride Room Temperature Molten Salt. In: Journal of the Electrochemical Society, 144, 1997, No. 9, pp. 3095–3103

51

4 Innovations in PVD Technology for High-performance Applications K. Bobzin, E. Lugscheider, M. Maes, P. Immich Surface Engineering Institute (IOT), RWTH Aachen University, Germany 4.1 Introduction

Surface and thin-film coating technology is substantially gaining importance for industrial applications. Applications of thin films are based on the interrelationship between material properties and surface conditions. Nowadays, important manufacturing branches, e.g. automotive, aerospace, and power engineering, depend strongly on modern surface technologies. Emerging, new industries such as medical and microtechnology as well as microelectronics, owe most of their growth to the processes developed in surface technology. Some examples of applications in optics are: thin films for reduced or increased reflection and optical absorption as well as optical filters. Electronics and microelectronics use thin-film contacts, resistors, and capacitors. Thin coatings are also applied in chemical process technology and mechanical engineering for wear reduction, defined frictional behaviour, corrosion resistance, high-temperature thermal resistance, and last but not least, for decorative reasons. Additional applications for surface refinement are found in tool coatings with wear-resistant materials or biocompatible implants for medical use. In the realm of modern high technology, the ability to optimise substrate and surface properties independently by means of surface technology, has proven equally beneficial and necessary in many areas of engineering science. From a tribological point of view, surface properties and their defined production is particularly important. In particular, modern PVD (physical vapour deposition) and CVD (chemical vapour deposition) technologies provide an excellent means of altering chemical composition as well as physical and mechanical surface properties.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

52

4 Innovations in PVD Technology for High-performance Applications

4.2 Market Situation

A programme evaluation conducted in 1993 by the ifo Institute (Institute for Economic Research) on behalf of the BMBF (Federal Ministry of Education and Research) revealed that prognoses of large world-wide growth rates for thin-film coatings existed as early as 1990. Depending on the investigated industry branch, 5–20% annual growth is expected (Fig. 4.1). The prognoses have proven correct, as shown in a 2001 study on the surface technology market situation of the VDMA (German Engineering Federation) on behalf of the BMBF. VDMA-organised suppliers of industrial plasma surface technology registered an average increase of 10% in orders, more than half, 60%, were inland orders. Service work dominated inland orders, while international partners mostly requested systems and installations. The substantial growth of the market is attributed to continuous expansion into new fields of application. The Federal Ministry of Education and Research (BMBF) has encouraged and promoted the development of broad know-how in the field of surface technology during the past 15 years. Until 1993, the BMBF research funding programme

Fig. 4.1 Average annual growth prognoses and market shares of thin film systems [source: ifo Institut für Wirtschaftsforschung – Institute for Economic Research, 1993].

4.3 Application Examples

for thin-film technology raised approx. 150 million DM (approx. 77 million euros) for research and development in industry and academic sciences. Research continued subsequently within the scope of the BMBF programme for surface and coating technology (OSTec) from 1993 to 1997. Concrete steps are needed to take advantage of the benefits of surface and coating technology and the present predominant technology position of Germany compared to international competitors. In the realm of an OSTec evaluation conducted by the Munich-based ifo-Institute, knowledge deficits, insufficient standardisation measures, as well as inadequate education and training were identified as the main restraints. Similar, at times considerable, qualification deficits were found in a VDMA study (2001) in the branches of potential applications. Results suggest that further market penetration is inhibited by the observed qualification deficits and due to missing capacities to untangle the situation. The ifo-Institute’s report on the OSTec funds reads as follows (1993): ”Interviews with suppliers of systems and installations confirmed the relevance of personnel and qualification. A large number of experts pointed out that potential users, in part, do not even know about capabilities and possibilities of surface technology, especially thin-film coatings. [. . .] Even worse, many enterprises do not have any OSTec-specific product and process knowledge. In these cases, suppliers need contacts to competent experts. . .” (translated from the German)

4.3 Application Examples

Surface treatment focuses on increasing the value of parts and tools. Frequently, increased service life or efficiency is desired due to rising application demands. However, higher profits may also involve using less-expensive substrate materials, or equipping parts with decorative surfaces. Development of coated parts always requires system development, considering the substrate, the coating material, the interface, as well as boundaries to the surrounding environment. A system-wide, holistic approach is necessary to develop a unit that is able to withstand applied loads and surrounding conditions. In many fields of manufacturing, modern surface technology has led to great improvements in performance and service life. Especially in cutting technology, coated tools are indispensable, e.g. for turning, milling, and drilling operations. PVD processes provide coatings spanning from simple binary hard materials such as TiN with hardness values of approx. 2200 HV to crystalline diamond, the hardest natural material. Coatings referred to as hard/soft combination layers integrate low friction coefficients with high wear resistance. Process developments nowadays even allow coatings on temperature-sensitive parts

53

54

4 Innovations in PVD Technology for High-performance Applications

made of low-alloy steel or thermoplastic, therefore yielding new fields of application. 4.3.1 Tool Coatings for Cutting

Titanium nitride (TiN) was the first coating successfully used for industrial steel machining. TiN has been a well-established coating in different tribological applications for decades. Prominent features include high hardness values and chemical resistance to many media common in tribological contact. However, growing demands in many fields of cutting technology limited the use of this coating. But it has served as a basis for subsequent coating developments, e.g. Ti(C,N) or (Ti,Al)N coatings. Due to higher oxidation, heat, and wear resistance (Ti,Al)N coatings allowed higher cutting speed, were preferable, and met the production demands of modern cutting industry. A number of coatings have been developed in coating technology to meet the various demands. Examples are hard, multi-layer, and nano-structured wear-resistance coatings, e.g. TiAlN+MoS2, TiAlN+Al2O3, TiAlN+WC/C [4]. The aim in developing these systems is to combine the properties of hard and wear-resistant surfaces with soft, friction-reducing features. These coatings have yielded excellent results particularly in dry machining [9]. Another combination uses alternate deposition of hard materials, producing so-called nano-laminated coatings. Very thin film layers (2–40 nm) in these coatings limit dislocation movement and therefore yield extreme mechanical properties, e.g. very high coating hardness. Development needs in modern cutting applications are clearly focused on complex coatings (Fig. 4.2). Apart from socalled nano-laminated structures, nano-composites have high potential due to embedded hard phases in an amorphous matrix. Two different materials are deposited simultaneously, a crystalline and an amorphous phase. The nano-composite structure is produced by phase separation. In contrast to multi-layer coatings, where any desirable material combination can be deposited, nano-composite production is limited to certain material combinations. Thermal resistance of the coating is a key feature for tool service life in high-temperature applications. In several investigations, dual-phase nano-composites have proven to exhibit very high thermal endurance and temperature stability [6, 19, 20]. A combination of new coating materials is favourable in applications where hardness is less relevant but lubrication properties dominate, as in dry machining or cutting of materials that tend to adhere, e.g. austenite, nickel-base, and in particular aluminium alloys. Examples are graded ZrC and MQT (minimum quantity lubrication), used in many cutting applications [9]. Good results for adhesion resistance have also been obtained using Al2O3. Current research work focuses on high-temperature, oxidation-stable material combinations as well as new multi-layer and composite layers [10, 11]. Industrial and scientific interest is also directed at oxidic solid lubricants based on the formation of understoichiometric compounds and corresponding crystallo-

4.3 Application Examples

Fig. 4.2 Schematic representations and images of microstructures in complex coatings.

graphic structure (Magnéli phases). New process technology allows deposition of nano-composites and electrically insulating coatings, e.g. c-Al2O3. 4.3.2 Tool Coatings for Forming

The use of PVD coatings to protect the surface of forming tools is increasing. Surface treatment is applied to injection moulds for plastics, diecasting moulds and tools for semi-solid forming of metal (thixoforming). These tools suffer from strong mechanic and thermal stresses occurring in the thixoforming process of aluminium. In addition to the corrosive properties of the processed materials, stress and strain combinations include erosion, thermal fatigue with cracking and adhesion occuring. Figure 4.3 shows a typical forming tool for aluminium diecasting. Thixoforming is developing towards processing of steel that again will increase process temperatures considerably (1200–1400 8C). Forming tools for such conditions require highly heat resistant, oxidation-stable, and thermal-fatigue-resistant coatings. The development of Cr-based coatings allowed PVD coatings to be used under corrosive conditions [3, 5, 23]. The advantage of these coatings is that they form a stable, thin oxide layer at the surface on top of the dense coating structure. Compared to Ti-based coatings, surface energy is lowered, and therefore, adhesion resistance improves notably when Cr-based coatings are applied [3]. Also, less releasing agent is needed to securely run the processes. (Cr,Al)N coatings are successors to CrN coatings, and demanding with respect to process technol-

55

56

4 Innovations in PVD Technology for High-performance Applications Fig. 4.3 Thixoforming for steel forming.

ogy. Compared to CrN, coating hardness is increased due to solid-solution formation. Furthermore, high residual compressive stresses in PVD coatings of forming tools reduce the formation of heat checkings (fine cracks), resulting in smoother surfaces of cast parts [12]. (Cr,Al)N coatings combine corrosion resistance to releasing agents as well as molten aluminium, for aluminium diecasting and thixoforming [5, 7]. For steel forming, oxide ceramic coatings are applicable due to their mechanical and thermal properties, e.g. chemical inertness and high mechanical stability. As far as current investigations have revealed, Al2O3 and t-ZrO2 coatings produced in reactive magnetron sputter ion plating (MSIP) (Fig. 4.4) show good results for thixoforming of steel [7, 13, 18]. When insulating coatings are reactively deposited from metallic targets, target poisoning must be prevented. Target poisoning, or the formation of a reaction layer, cuts plasma power density due to dielectric coating formation. Pulse-sputter processes have been developed in order to stabilise reactive sputtering. For reactive deposition of oxide coatings in sputter processes, the pulse source switches target polarity from negative to positive for short periods of time in socalled bipolar pulsed processes. During the positive pulse, electrons are drawn to the positive surface and the oxide layer is discharged. In unipolar pulse processes, in contrast, the energy source operates at low power and generates stronger pulses only during short periods of time. This causes an instantaneous, considerable, but short rise in plasma density, and prevents the target from encountering undesired thermal stress [22].

4.3 Application Examples

Fig. 4.4 SEM images of ZrO2 (left) and Al2O3 coatings (right) deposited by means of MSIP pulse technology.

PVD coatings are also used successfully in plastics processing. Here, coatings prevent corrosion, abrasion, and adhesion, depending on composition of the processed plastics and possible filler materials. When handling elastomers, gas evolution during crosslinking can cause strong acids to develop, and thus, parts and/or coatings involved are subject to corrosive attack. On the other hand, particle-reinforced plastics primarily produce abrasive damage. The majority of coated components are part of plastifying units, e.g. barrel extruders, extrusion screws and cavities. The length of barrel extruders can reach several metres, presenting a challenge for the vacuum-system technology. 4.3.3 Coatings for Plastic Parts

For years, the plastics market has been experiencing high growth rates. Modern plastic developments continuously establish new market segments. In many applications, plastics nowadays are an alternative to glass or ceramic materials. For automotive (Fig. 4.5), and traffic engineering in general, lightweight applications using plastics are increasing. The main challenge in coating plastics is the low substrate tolerance to thermal stress. About 15 years ago, plastics were developed and put into operation that showed higher mechanical stability and allowed temperatures in the region of 200 8C, e.g. polyphenylene sulfide (PPS) and polyetheretherketone (PEEK). On controlling process parameters appropriately, PVD can operate below these maximum temperatures [2, 15]. Initially, coatings were applied to plastics mostly for scratch resistance or decorative purposes. Today, the market for PVD coatings is dominated by optical applications as well as coatings for the packaging industry. Many different optical coating functions, e.g. anti-glare and anti-reflection, as well as permeation barrier effects have been realised.

57

58

4 Innovations in PVD Technology for High-performance Applications

Fig. 4.5 Coated plastic parts for automotive industry – today and tomorrow [source: Adam Opel AG].

4.3.4 Coatings for Machine Elements

Wear protection and reduction of friction are prominent factors influencing parts and machine components, and therefore, they affect service life and maintenance programmes of entire technical installations. The temperature sensitivity of the substrate material during deposition processes of wear protective coatings is the main challenge. The breakthrough of thin-film technology for parts was promoted by the development of modern diesel engines. Extreme operating conditions in these sophisticated engines required new material concepts. In order to increase the specific output, the engines operate at high injection and ignition pressures (> 2000 bar). The achieved high power density yields low pollution and low fuel consumption. In addition, there is a demand for long service intervals and a trend towards lightweight constructions, e.g. aluminium crankcases. The first successful use of PVD technology in mass production was on a plain bearing in the rocker arm of a modern direct injection pump injector for diesel engine (Fig. 4.6). Annual coating production has already reached approx. 3.5 million sets and predictions suggest a further rise towards 6 million sets per year until 2006. Nowadays, the use of PVD technology is profitable when processing simple shaped parts with a geometry well suitable for coating. Valves, pump systems, and plain bearings offer high potential for PVD applications. Tribological systems involving high Hertzian stresses, often with superimposed translatory and boring motion, represent a much greater challenge. In addition, complex part geometries make coating processes difficult, and part handling tasks frequently represent the main concern.

4.3 Application Examples

Fig. 4.6 Rocker-arm plain bearing for pump injector unit in modern diesel engines.

Fig. 4.7 Spindle bearing, inner ring coated with CrAlN and CrAlN+C.

In successfully coated thrust bearings from a standardised testing facility for roller bearings (FAG Kugelfischer, FE8), where Hertzian stresses of up to 1900 N/mm2 occur, metal containing carbon coatings reduced cumulative wear by 90%. The main focus in these applications is on emergency running properties, use in challenging environments, as well as optimised friction coefficients and service life [3, 16]. Spindle bearings require particularly sophisticated coating technology (Fig. 4.7). The performance of cutting materials is improving continuously, in part due to coatings, and cutting speeds. Thus, requirements on machine tool spindles are rising. Optimal results require high-speed spindle bearings. Apart from optimising lubrication composition and supply, the most promising means of enabling high-revolution operation are new material combinations for rolling elements and bearing tracks [17]. A number of research programmes focus on coatings in hybrid bearings that contain ceramic rollers in contact with the coated inner

59

60

4 Innovations in PVD Technology for High-performance Applications

and outer rings. In future, this bearing type may well be an alternative to full ceramic bearings, nowadays used for dry running applications. Substitution of the expensive, and difficult to mount, full ceramic bearings is desired. Furthermore, coated steel bearing rings with optimised friction coefficients can reduce shear and tensile stress within the surface. Compared to standard bearings, hybrid bearings (steel bearings with ceramic balls) allow an increase in rotational speed of 15–30%. Practice shows that service life of hybrid bearings is at least twice as high as for steel bearings [21]. Coated bearing rings additionally improve performance. 4.3.5 Part Coating for High-temperature Applications

Ecologically and economically electric energy supply requires the implementation of modern, up-to-date power plant technology. Examples are gas turbine power plants and combined facilities with gas and steam turbines. High burning gas temperatures within the gas turbine allow best energy efficiency with respect to energy carriers (usually fossil fuels) (Fig. 4.8). Components exposed to such extreme operating conditions inside gas turbines require a sophisticated combination of protective coating materials and coating technology. Manifold loads and strains on these materials include corrosion, oxidation, and thermal fatigue. Ceramic, thermally insulating coatings can effectively protect parts and, in particular, metallic gas turbine blades. Zirconium dioxide (ZrO2) has a number of ideal material properties. Apart from the ability to serve as a thermal insulation material due to low thermal conductivity, the thermal expansion is similar to that of metallic turbine blades,

Fig. 4.8 Performance increase for turbines due to EB-PVD coated blades [source: turbine blade – MTU, diagram – Rolls Royce].

4.4 Summary

Fig. 4.9 Elongation-tolerant thermal barrier EB-PVD coatings for turbine blades.

in spite of the ceramic microstructure. Further developments in thermal-barrier coatings also investigate pyrochlore compounds. Here, lanthanum zirconate has proven to shift the sintering tendency of thermal-barrier coatings to even higher burning gas temperatures. Nowadays, and for this application, EB-PVD (electron beam PVD) is well established, particularly due to very high deposition rates and the possibility to carefully control surface temperature of geometrically complex substrates. The thickness of thermal-barrier coatings here is in the range of 250 lm or more. Thermal-barrier coatings produced by EB-PVD show a characteristic, columnar, crystalline structure, yielding high elongation tolerance (Fig. 4.9). Appropriate process control can also result in zig-zag microstructure of EB-PVD coatings. This fishbone (or herringbone) structure is expected to additionally decrease the thermal conductivity of the coatings.

4.4 Summary

Surface and thin-film technology have become important parts of industrial applications. Their main purpose is to provide surface treatment of materials for increased part and tool value. Today, PVD is an economical process technology for all parts with simple, well-coatable geometry. PVD provides coatings spanning from simple binary hard materials such as TiN to crystalline diamond. In cutting applications, a wide range of state-of-the-art coatings are established, including multi-layer and nano-structured coatings for increased wear resistance with and without integrated, friction-reducing solid lubricant layers, as well as complex oxidation-resistant coatings, and metastable systems with or without solid lubricant top coats. Since developments in pulse technology advanced,

61

62

4 Innovations in PVD Technology for High-performance Applications

Al2O3 and ZrO2 PVD coatings are a promising, economical alternative for steelforming applications. Today, research demands for new materials in part coatings are developing towards complex coatings, rendered possible only by continuously developing systems and process technology.

References 1 Bärwulf, St.: Entwicklung und Qualifi-

2

3

4

5

6

7

8

zierung von PVD-Trockenschmierstoffschichten auf der Basis oxydischer Materialien. Doctoral Thesis, RWTH Aachen (2001), Shaker, ISBN 3-8265-8314-0 Beitinger, G., Kolbeck, G., Feldmann, K.: Metallisieren von Thermoplasten. Metalloberfläche 55 (2001) 3, pp. 40–45 Bobzin, K.: Benetzung und Korrosionsverhalten von PVD-beschichteten Werkstoffen für den Einsatz in umweltverträglichen Tribosystemen. Doctoral Thesis RWTH Aachen (2000), Shaker, ISBN 3-8265-7414-1 Erkens, G., Cremer, R., Hamoudi, T., Rambadt, S., Wirth, I.: Supernitride. Eine neue Generation von Schichtwerkstoffen; EFDS Workshop: Beschichtete Werkzeuge – höhere Wirtschaftlichkeit in der Ur- und Umformtechnik; Frankfurt/Main, December 5th, 2002, proceedings Guerreiro, S.: Qualität- und Standzeitverbesserung von Aluminiumdruckguß durch Einsatz moderner PVD-Schichten. Doctoral Thesis RWTH Aachen (1998), VDI, ISBN 3-18-353305-7 Holubar, P., Jilek, M., Sima, M.: Nanocomposite nc-TiAlSiN and nc-TiN-BN coatings: their application on substrates made of cermented carbide and results of cutting tests. Surface and Coatings Technology 120–121 (1999) pp. 184–188 Hornig, Th.: Entwicklung von Werkstoffverbunden für den Einsatz in Thixoformingwerkzeugen für die Aluminiumund Stahlverarbeitung. Doctoral Thesis RWTH Aachen (2002), Mainz, ISBN 3-89653-935-3 Knotek, O., Lugscheider, E., Bärwulf, St., Barimani, C.: Process Windows and Properties of Tungsten- and VanadiumOxides deposited by MSIP-PVD-Process.

9

10

11

12

13

14

15

16

Materials Research Society, Symposium Proceedings Vol. 555 (1999) Lugscheider, E., Leyendecker, T., et al.: PVD hard coated reamers in lubricantfree cutting. Surface and Coatings Technology 112 (1999), pp. 146–151 Lugscheider, E., Knotek, O., Zimmermann, H.: Investigation of the mechanical and structural properties of Ti-Hf-CN arc PVD coatings. Surface and Coatings Technology 116–119 (1999) pp. 139– 243 Lugscheider, E., Bobzin, K., Klocke, F., et al.: Weiterentwicklung umweltverträglicher Tribosysteme in der Kaltumformung mittels PVD-Technologie, Proceedings GfT’2000, Göttingen, September (2000) Lugscheider, E., Bobzin, K., Hornig, Th., et al.: PVD-Hardcoatings Protecting the Surface of Tools for Thixoforming. Advanced Engineering Materials (2000) Lugscheider, E., Bobzin, K., Bärwulf, St., Hornig, Th.: Oxidation characteristics and technical properties of Chromium based PVD-hardcoating for use in semisolid forming tools. Surface & Coatings Technology 133–134 (2000) 540–547 Lugscheider, E., Knotek, O., Bobzin, K., Bärwulf, St.: Characteristic Curves of Voltage and Current, Phase Generation and Properties of Tungsten- and Vanadium-Oxides Deposited by Reactive DC-MSIP-PVD-Process for Self-Lubricant Applications. Surface and Coatings Technology 142–144 (2001) 137–142 Lugscheider, E., Bobzin, K., Hornig, Th., Beckers, M.: PVD-Beschichtungen lassen Kunststoffe kalt. Metalloberfläche 55 (2001) 10, pp. 34–37 Lugscheider, E., Bobzin, K.: Übertragung tribologischer Funktionen der Schmierstoffe auf die Werkstoffoberfläche mittels PVD-Technologie. Tribologie und

References Schmierungstechnik 49 (2002) 1, pp. 16–22 17 Lugscheider, E., Bobzin, K., Colmenares, C., Weck, M., Krell, M., Boutemy, F.: Untersuchung von PVD-beschichteten Wälzlagern für umweltverträgliche, schnelldrehende Werkzeugmaschinenspindeln. VDI Symposium: Gleit- und Wälzlagerungen, September 9th–11th, 2002, Fulda, VDI Report 1706, Düsseldorf VDI (2002), ISBN 3-18-091706-7, pp. 77–96 18 Lugscheider, E., Bobzin, K., Maes, M., Abdel-Samad, A.: PVD oxide coatings for the Use in thixoforming of steel. S2P, Cyprus, Sept. 21st–23rd, 2004 19 Männling, H.-D., Patil, D. S., Moto, K., Jilek, M., Veprek, S.: Thermal stability of

20

21 22

23

superhard Nanocomposite coatings consisting of immiscible nitrides; Surface and Coatings Technology 146–147 (2001) Sept./Oct. 2001, 263–267 Mitterer, C., Mayrhofer, P. H., Musil, J.: Thermal stability of PVD hard coatings; Vacuum 71 (2003) 279–284 N. N.: Präzisions-Schrägkugellager. SNFA Main Catalogue, 6th edn (1998) Scholl, R. A.: Asymmetric bipolar pulsed power: a new power technology. Surface and Coatings Technology 98 (1998) 823– 827 Schrey, A.: Dünne Hartstoffschichten zum Korrosionsschutz. Doctoral Thesis RWTH Aachen (1993), Mainz, ISBN 3-930085-45-3

63

65

5 Development and Status Quo of Thermal CVD Hard-material Coating A. Szabo, Best-Surface, Ludwigsburg, Germany 5.1 Introduction

Chemical reactions involving the deposition of material from a gas or vapour phase are generally referred to as CVD (chemical vapour deposition). Here, the deposited product is the result of chemical reactions between precursor materials. Activation energy for the chemical reactions is supplied by thermal energy at elevated temperatures. Temperatures range from several 100 8C to far above 1000 8C. HTCVD (high-temperature CVD) hard coatings have been developed empirically during the past 4–5 decades, following application-oriented demands. Development of systems and materials demonstrate the requirements of industrial applications on the one hand, and the pragmatic approach on the other. Thermal CVD coating technology has evolved to an industrially usable technology for many diverse applications. Several disadvantages of high-temperature processes have been eliminated successfully by partially modifying the chemical reactions and by selecting appropriate precursors. MTCVD (moderate-temperature CVD) though, is usually limited to deposition of titanium carbonitrides. In certain cases, low-pressure glow-discharge plasma can supply the necessary activation energy to the molecules and atoms. PACVD (plasma-activated CVD) allows deposition at low temperatures, between room temperature and several 100 8C. However, it is limited to special applications. This outline summarises the historical development and common variants of industrial CVD hard-material coating. Analysis and comparative assessment of properties will serve as guidelines to the interested reader and user.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

66

5 Development and Status Quo of Thermal CVD Hard-material Coating

5.2 Early CVD Hard-material Coating

High-temperature CVD technology (HTCVD) for titanium-based hard-material coatings resulted, by pure coincidence, from empirical studies during the 1950s. Thermo-chemical reduction of titanium tetrachloride (TiCl4) in a hydrogen and nitrogen atmosphere produced a yellow, hard material, titanium nitride, which formed a hard surface coating [40]. On the surface of carbon steel, titanium carbide (TiC) developed in a nitrogen-containing atmosphere. These coatings were silver-grey and even harder. The hard material (TiC) can also be produced in a carbon-containing atmosphere from TiCl4 [41]. Deposition variants and properties of titanium nitride and titanium carbide have been thoroughly investigated [39, 42, 62]. Relatively few publications, though, discuss basic principles of the formation mechanisms and chemical reaction kinetics in TiN and TiC production [14, 49, 62], although certain studies feature theoretical and process-analytical considerations [23, 36, 54, 59].

5.3 Fundamentals of Deposition Processes 5.3.1 Chemical Mechanism

The reaction that produces hard-material coatings from titanium tetrachloride, TiCl4, is thermally activated. Chlorine atoms are removed as HCl (hydrogen chloride) in multi-step reduction reactions. Remaining intermediates containing titanium react during transition with possible reacting agents such as nitrogen as well as activated carbon provided by methane, and form nitride and carbide, respectively. Carbonitrides are formed when nitrogen and carbon donors appear simultaneously, with the C : N ratio depending on the C : N supply and availability. The following gross reaction equations summarise the formation of the produced hard materials: TiCl4 ‡ H2 ‡ N2 ! TiN ‡ HCl TiCl4 ‡ H2 ‡ CH4 ! TiC ‡ HCl TiCl4 ‡ H2 ‡ N2 ‡ CH4 ! TiCN ‡ HCl (Stoichiometric coefficients were omitted purposely as components react incompletely and additional by-products are formed, respectively.) The coating composition was generally limited to titanium-base (TiN and TiC) because the metal donor TiCl4 is relatively easy to evaporate and is a com-

5.3 Fundamentals of Deposition Processes

monly available, inexpensive liquid. Although one mole of TiCl4 usually produces up to four moles of hydrogen chloride, large hydrogen surplus is necessary to promote the deposition reaction, according to the law of mass action (Guldberg and Waage’s law). Further details of chemical reactions, fundamental processes, and by-products are not covered in this chapter. 5.3.2 Interdisciplinary Fundamentals

The art of CVD hard-material coating has become an interdisciplinary field of expertise. This includes optimal collaboration of chemistry and physics, hardmaterial science, systems engineering, steel technology and heat treatment, as well as fundamental knowledge of functions in tool design and mechanical engineering. The resulting properties of coated parts depend on a large number of parameters, not discussed in any more detail here [15, 44]. Extensive investigations show that considerable effort has been devoted to determining chemical and physical processes as a theoretical base for better understanding of the involved processes [9–11, 35, 48]. Research and development results, as well as better analytical understanding of fundamental processes, led to early, considerable insight, which, however, has not thoroughly found its way into industrial practice [46]. 5.3.3 CVD System and Reaction-chamber Techniques

During the early stages of development, CVD systems were built and used with movable retorts for shaft furnaces, e.g. by Metallgesellschaft in Frankfurt, Germany. This type of CVD coating system dominated the early days in Germany and was copied by other industrialised countries (Fig. 5.1). During the following years, deposition processes and system technology were empirically improved. Later, bell-type reactors with vertical retorts were developed and operated. Bernex of Switzerland most notably promoted this type of system design (Fig. 5.2). Process parameters were adjusted differently for TiN and TiC, due to the dissimilarity of the involved chemical formation reactions, deposition mechanisms, and thus, growth characteristics of the two coating materials [2, 10]. In the direction of gas flow, consumption of TiCl4 leads to depletion, and therefore reduced concentration of TiCl4 within the reactor. As a result of reduced titanium supply, the deposition rate continuously decreases. Therefore, keeping uniform coating deposition, i.e. coating thickness, within tolerable limits has always been difficult in larger and longer reactors [21, 36]. Many system designs were introduced in order to yield better uniformity of coating deposition in larger-sized reactor volumes. The required parameter sets varied considerably. Tuning and matching all the essential parameters, and to

67

68

5 Development and Status Quo of Thermal CVD Hard-material Coating Fig. 5.1 CVD system, shaft furnace type (Dörrenberg).

Fig. 5.2 CVD system, tandem, bell-type reactor (IonBond).

guarantee that they would remain constant under industrial production conditions, was challenging. Industry requires simple and robust technologies [60]. One way of balancing non-uniform deposition rates due to TiCl4 depletion was to increase deposition temperature, and thus, deposition rates. For this, heating furnaces with several heating zones were designed, featuring up to six individually controllable heating zones. However, this method is applicable and reliable only for standard deposition of monolayers [1].

5.3 Fundamentals of Deposition Processes

For material combinations, each coating material requires different parameter sets and therefore different temperature gradients. In addition, these also depend on the specific pressure and composition conditions. To date, this type of design outlasted only for simple standard processes in industrial routine. In part, extensive automation has been introduced to these processes [24, 38]. Different CVD systems feature different types of gas flow, e.g. rising or descending gas flow [55]. An effective means of compensation involves a design where the direction of reactive gas flow alternates between upward and downward flow within the reactor, even during the process [6, 45]. Radial gas flow is an additional way to yield better uniformity by controlled gas flow. Applications feature, for instance, a centre tube with lateral holes, suitable also for rotating operation [2, 21]. This technique has proven its value primarily for series coating of hard alloy indexable inserts. A CVD technique developed in the former German Democratic Republic for industrial production utilises a revolving principle. All process stages are integrated within this remarkable design, and the change-over takes place in a protective gas atmosphere [9]. The demand for large CVD coated parts has led to the design of large-volume reactors. Diameters have grown from initially 20–30 cm up to 75 cm, today. However, growing reactor size requires deposition parameters to be adjusted constantly, and involves considering the relevant chemical and physical boundary conditions (Fig. 5.3). Detailed solutions were developed mostly on an experimental, empirical basis [1, 51, 53, 64]. Rapidly growing interest gave rise to a research project, investigating production and interaction of hard material combinations with TiC and TiN, during the late 1970s.

Fig. 5.3 Large CVD system, tandem (Dörrenberg).

69

70

5 Development and Status Quo of Thermal CVD Hard-material Coating

Paterok described in detail combinations of the two important hard materials TiN and TiC in varying functional coatings [45, 46]. However, industrial practice did not adequately incorporate the analysed results and described insights. Remarkably enough, Paterok’s early insights precisely reflect the optimal coating combination that is being used today. Then, it was extremely difficult to produce the described optimal coating combination of steel-TiN-TiCN-TiC, considering the available thermal CVD technology.

5.4 Combination Coatings

Initially, TiN and TiC hard-material coatings competed against each other. But the formation mechanisms and several properties of the two hard materials differ slightly [42], so TiN and TiC can complement one another, depending on the demands of a specific application. It was not until the 1970s that combination coatings, composed of several hard materials, were suggested, and introduced to industrial applications [4, 22, 23, 43, 47]. Due to high technical demands of graded steel-TiN-TiCN-TiC coating combinations developed by Paterok, industrial coating suppliers preferred a simple, pragmatic approach involving steel-TiC-TiCN-TiN coating combinations that grow spontaneously on carbon steels. This type, known as TiC-TiN or sandwich coating, has become a multi-purpose industrial product that, today, shows sufficient overall performance for many applications. In practice, however, this coating is inadequate for extreme applications [17, 27–30, 33] (Figs. 5.4 and 5.5). The mechanical and physical characteristics of coated steels influence the overall performance of coatings considerably. Composite carbide zones form on the steel surface at the interface due to alloying, depending, amongst others, on the hard material forming metals (Ti, Cr, V, etc.) [11]. Promising results were obtained with a CrC-TiC coating, although production is significantly more challenging than for TiC-TiN [16]. The combination steel-CrC-TiC basically serves the same purpose as TiN-TiC, i.e. produces a stress-reduced, well-balanced coating. Titanium carbide (TiC) is formed in a chemical reaction from thermally activated, hydrogen-reduced titanium and active carbon. Titanium is fed to the surface via gas phase; carbon can be supplied – also from gas phase – by decomposition of organic molecules, e.g. methane, CH4, but also from the carbon-containing steel itself. The deposition and growth rate of titanium carbide in CVD technology therefore shows non-linear characteristics. The process of TiC deposition depends on carbon diffusion. Decarburisation of TiC-coated steel is less critical due to subsequent carbon diffusion, which compensates for decarburisation in the near-surface zone [47, 59]. In practice, precise control of carbon supply via gas phase is advisable at all times.

5.4 Combination Coatings

Fig. 5.4 TiN growth characteristic.

Fig. 5.5 TiC growth characteristic.

Titanium nitride (TiN) develops in a heterogeneous chemical reaction at the surface, where coating components are supplied via gas phase. It crystallises at the surface and forms a sealed uniform coating around the part. Growth and deposition rate are directly (linear) proportional to process time. Titanium nitride coatings prevent carbon from diffusing from the substrate into the coating; TiN serves as a diffusion barrier. Therefore, TiC coatings can grow directly on carbon steels, but not sufficiently on TiN, or similar materials, without an additional carbon supply. Due to this property, TiN can be used as a barrier layer preventing carbon diffusion within CVD processes. Thus, embrittlement due to decarburisation of carbides and g phase formation in the interface beneath the coating is prevented in CVD processing of hard-metal substrates [4, 47, 56, 58] (Fig. 5.6). The first successful results included coating of hard-metal tools and hard alloy indexable inserts for cutting applications. In contrast to tool steel, hard metal (tungsten carbide with cobalt binder) does not require any special heat treatment after coating at high temperature. Most hard metal suppliers offer highquality indexable inserts of their own production (Fig. 5.7). Thermal CVD technology allows analogous deposition of further hard materials and substances by decomposition of metal halogenides and related volatile compounds.

71

72

5 Development and Status Quo of Thermal CVD Hard-material Coating

Fig. 5.6 TiN barrier layer against decarburisation/carbon diffusion (IonBond).

Fig. 5.7 Hard-metal indexable inserts, CVD coated.

Additional hard materials, apart from titanium-based materials, particularly carbides and nitrides of zirconium, chromium, tungsten, vanadium, tantalum, as well as other refractory metals, have been found, produced, and investigated. Particular attention has been paid to chromium carbide (CrC) because highperformance tool steels are mostly chromium alloyed and the thermal expansion of chromium carbide is almost equal to that of chromium alloyed steel [8, 12, 13].

5.5 Material and Coating Properties

5.5 Material and Coating Properties

In a broader definition, nitrides and carbides can be considered ceramic materials [50]. These materials are known to be hard and brittle. Ceramic coating material deposited at high temperature, stabilises during cooling by micro-cracking, which leads to moderate stress relief. This type of coating usually builds up compressive stress. However, hard-material coatings are relatively elastic and flexible, provided that the coating thickness is not too high. Such thin CVD coatings initially follow the elastic deformation of steel under strain without developing cracks. Practical limits are reached only during plastic deformation of the steel. Then, permanent cracks develop that gradually cause the coating to spall. A comparative test method of thin-coating adhesion uses this effect. The indent caused by a Rockwell hardness test represents, at the same time, a defined deformation of the steel. The deposited hard-material coating follows the deformation to a certain degree, then cracks, and possibly spalls partially. The degree of flaking is an empirical measure of coating adhesion on steel substrate. This test method has been standardised for adhesion measurements of thin coatings deposited by PVD technology [7] (Fig. 5.8).

Fig. 5.8 DIN testing method, thin-film adhesion.

73

74

5 Development and Status Quo of Thermal CVD Hard-material Coating

In practice, the method is applicable for estimating CVD coating adhesion as well. Due to high deposition temperatures, CVD coatings typically develop higher adhesion and better properties. 5.5.1 Physical Properties of Coating Materials

The overall properties of the combined steel substrate and hard-material coating are determined mainly by two hard material parameters: one, the intrinsic stresses [34, 63] caused by thermal expansion of the materials involved, and two, the hardness of the coating material itself [5, 20, 32]. 5.5.2 Comparison of Coating Combinations 5.5.2.1 Classic TiC-TiN Titanium carbide (TiC) hard material typically grows on carbon-steel substrates. The kinetics and chemical mechanisms of TiC deposition have been thoroughly investigated. In particular, the role of carbon has been revealed [11]. The growth characteristics show that a predominant proportion of carbon is supplied by diffusion from the solid phase, i.e. the substrate [36, 49, 59]. Initially, the TiC growth rate is high, and gradually decreases during coating growth due to reduced carbon diffusion through the coating. Additional carbon is provided via gas phase by thermal decomposition of methane (CH4). Within the transition zone, where a titanium carbonitride mixed phase (TiCN) is formed, nitrogen and methane are supplied simultaneously. The particular C : N proportion in the transition zone is more or less directly proportional to the ratio of C and N activity in the relevant chemical reactions during deposition. Formation of titanium nitride is promoted by adding nitrogen as N2, or by decomposing NH3 [22]. Growth of TiN shows a linear characteristic in practice too, because the entire active nitrogen originates from the gas phase. When the amount of methane as carbon donor decreases, the C : N ratio within the carbonitride changes correspondingly. Without any added methane, pure titanium nitride (TiN) is formed. The sequence within the coating is: steel-TiC-TiCN-TiN (Fig. 5.9).

5.5.2.2 Balanced TiN-TiC The hard material combination of steel-TiN-TiCN-TiC is a well-balanced system. Physical properties, such as thermal expansion and hardness, gradually increase across the coating. TiN can be applied to all types of steel or other substrates, because the entire coating material required is supplied via gas phase. Since titanium nitride prevents carbon diffusion from the steel, TiN serves as a barrier layer within this

5.5 Material and Coating Properties

Fig. 5.9 Coating combination: steel-TiC-TiCN-TiN, classic.

Fig. 5.10 Coating combination: steel-TiN-TiCN-TiC, balanced.

coating. TiC can only be deposited on TiN when all the required carbon is provided constantly via gas phase from an adequate source. A low-stress, elastic transition zone between TiN and TiC layers requires deposition of a graded TiCN intermediate layer with opposed C and N supply. In practice, the gas composition is constantly adjusted. The TiCN intermediate layer gradually changes to TiC due to a continuously reduced nitrogen supply. Further growth of TiC is promoted only by carbon from the gas phase. Adequate carbon donors ensure a continuous supply of carbon within the complete reactor volume. This is one of the characteristic features of the process. The sequence within the coating combination is: steel-TiN-TiCN-TiC (Fig. 5.10). 5.5.3 Effects of Thermal Expansion

The combination coating steel-TiC-TiCN-TiN faces highest stresses within the interface between steel and TiC (cf. Table 5.2, left).

75

76

5 Development and Status Quo of Thermal CVD Hard-material Coating Table 5.1 Influencing material properties (rounded values). Material

Hardened steel TiN

Thermal expansion in 10–6/K 12 9 Hardness in HV (Vickers) 800 (64 HRC) 2500

TiCN

TiC

8–7 3000

6 3600

Individual structures of presented coatings yield the following differences:

Table 5.2 Effects of variations in thermal expansion. Coating type TiC-TiN Sequence

Steel

TiN-TiC TiC

Values

12 6 shows minimum Differences +6

TiN

Steel

9

12 9 smooth transition +3

–3

TiN

TiC 6 +3

(+ = compressive stress, – = tensile stress, for succeeding layer) These differences create stresses in the interfaces between individual layers:

Table 5.3 Effects of variations in hardness. Coating type TiC-TiN Sequence Values

Steel

TiN-TiC TiC

TiN

800 3600 2500 shows maximum Differences +2800 –1100

Steel

TiN

TiC

800 2500 3600 smooth transition +1700 +1100

From the very start, the top coat is responsible for the functional coating characteristics.

Here, the highest difference in expansion occurs, and therefore, tensile stresses in the steel and corresponding compressive stresses in the TiC layer reach maximum values. Furthermore, tensile stresses are induced in the subsequent TiN layer. Under extreme strain, the added external forces can accumulate and cause the coating to spall [27–30]. Also, the difference in thermal expansion produces the highest shearing force in this area. This effect leads to increased swelling at corners and edges [28]. Stress differences in the transition zones of balanced steel-TiN-TiCN-TiC coatings are only approx. half as large as the difference in stress between successive layers (cf. Table 5.2, right). Stresses and therefore forces are always lower, and additionally, they decrease gradually within the coating.

5.6 Performance of Hard-material Coatings – Applications

5.5.4 Effects of Hardness

The functional top layers of steel-TiC-TiCN-TiN coatings feature hardness values of approx. 2500 HV (cf. Table 5.3, left). This value represents the actual functional hardness of the surface towards the environment. The hardest material (TiC) is hidden underneath, in the centre of the coating. TiC, if at all, does not reveal its real function until later during its service life when wear and abrasion have progressed. Steel-TiN-TiCN-TiC coatings show maximum hardness values of approx. 3600 HV at the functional surface, therefore, they are particularly suitable to provide wear resistance (cf. Table 5.3, right). When abrasion progresses and finally removes the TiC top layer, the lilac-yellow TiCN/TiN intermediate layer appears. However, the coating still retains its functional properties. The complete TiN layer remains fully functional and serves as a wear reserve. The user, though, can easily detect the beginning of wear early enough due to the apparent change in surface colour, well before the coating is completely worn and the steel substrate suffers possible damage. Therefore, reasonable precautions, including coating removal and recoating, can be well timed. Thus, using the described coating reduces possible risks even further. A well-balanced, intelligent coating yields higher safety and therefore increased performance.

5.6 Performance of Hard-material Coatings – Applications

The newly developed TiN-TiC combination layers broaden the mechanical strain resistance of coated steel. This type of balanced coating combination on tool steels was designed for industrial applications. Introduced in 1997 in Germany, it has since been made commercially available [18]. The domain of the described coatings in industrial applications is metal forming. Investigations of obtained tool life, e.g. in deep drawing of high-strength sheet steel, showed that combined TiN-TiCN-TiC hard-material coatings yield enhanced wear resistance under high pressure and increased wear conditions. Thus, production-relevant performance (tool life) is again improved. The coating combination adds to the variety of existing dedicated CVD hard-material coatings, and provides an alternative technology applicable to high-performance tools and parts subject to high mechanical abrasive wear. In industrial use of coated forming tools that are subject to extreme strain, the new TiN-TiC coatings give a competitive edge. Practically no coating failures in terms of spalling were observed. Classic TiC-TiN coatings suffered greater coating failure, particularly at high-stressed edges, the main functional areas. In most cases failure was observed at the interface between the steel and the TiC layer, in rare cases the intermediate TiCN layer failed [28]. Comparison suggests

77

78

5 Development and Status Quo of Thermal CVD Hard-material Coating

Fig. 5.11 Drawing die, functional TiC coating (Dörrenberg).

Fig. 5.12 Large segmented tool (Dörrenberg).

that the well-balanced, stress-reduced steel-TiN-TiCN-TiC coating copes with higher loads and yields better resistance in practice. The results reveal increased overall performance of intelligently coated parts and tools (Figs. 5.11 and 5.12). In these applications, where abrasion occurs due to friction, surface hardness and chemical resistance of the functional surface predominantly determine coating resistance [25]. Many additional examples of applications are found in other branches of engineering [24]. In plastics processing, a number of critical cases involve wear due to a combination of abrasion and/or corrosion, which lead to severe damages [3].

5.6 Performance of Hard-material Coatings – Applications

Glass-fibre-reinforced plastics and plastics with mineral filler cause abrasion and/or corrosion of steel machine part surfaces, e.g. in extruders and surfaces of injection moulding dies. Hard and ceramic protective coatings, mainly applied by CVD technology and therefore featuring excellent adhesion, can, to a large extent, prevent such damage. CVD TiN-TiCN-TiC coatings increase the service life of parts manifold [19]. CVD hard-material coatings yield considerably higher resistance to abrasive wear than comparable PVD hard-material coatings. 5.6.1 Wear Resistance

The basic principle of wear-protecting, hard-material coatings is derived from the tribological rule that harder materials generally suffer less wear. In addition, hardness characteristics and strength of the steel substrate material, which supports the hard-material coating, also affects the overall system performance. Specific steel properties such as carbon content and alloying elements strongly contribute to hardenability and therefore determine the results [15]. Due to the inert and ceramic nature of hard materials, typical material interactions, frequently observed in metal forming, are prevented, e.g. metallic cold welding (galling). Friction coefficients are, and remain, relatively low on the smooth ceramic surface [52]. Considering tribology, a lubricant also promotes hydraulic separation of the contacting surfaces that are in relative motion. Dissimilar materials exposed to friction experience considerably less interaction. Ceramic hard coatings thus prevent cold welding as well as abrasive wear. The amount of lubricant otherwise necessary under routine conditions, e.g. uncoated systems, is effectively reduced. However, a minimum amount of lubricant is indispensable. Hard-materialcoated surfaces often require only little lubrication. In certain, though rare cases, emulsions or even water are sufficient. 5.6.2 Heat Treatment and Dimensional Accuracy

The biggest problem and disadvantage of CVD steel coating technology is the unavoidable high temperature of 850 8C to over 1000 8C. Being well above the tempering temperature, subsequent heat treatment of the coated steel parts or tools is obligatory. Heat treatment though always bears the risk of dimensional changes by distorting, shrinking, or twisting. This disadvantage, to date, limits certain applications of CVD coatings. Steel quality as well as process characteristics of the heat treatment influence dimensional changes. For certain steel qualities, geometrical changes can be corrected successfully; steel quality, therefore, is of fundamental importance. Despite certain technical disadvantages, CVD coatings outperform hard-material coatings deposited by PVD technology, particularly due to wear resistance and adhesion.

79

80

5 Development and Status Quo of Thermal CVD Hard-material Coating

The problem of dimensional changes in steel during heat treatment has been subject to thorough investigation in recent years [15, 26, 31]. Today, careful choice of steel quality in combination with an adequate CVD coating technique and subsequent accurate heat treatment allows CVD coating and geometrically controlled heat treatment of parts and tools with close tolerances.

5.7 CVD Coating at Lower Temperatures 5.7.1 Moderate-temperature CVD, MTCVD

The basic disadvantages and negative effects of classic HTCVD are a result of the high temperatures necessary for the process. In particular, formation of g phases within the cemented carbides/hard metals and dimensional changes such as distortion cause problems. The energy required for activating H2, N2, and CH4 is provided thermally at temperatures around 1000 8C in classic thermal HTCVD processes. Due to the demand for hard-material coating at reduced temperature, MTCVD (moderate-temperature CVD) was developed. Using reacting agents that require less activation energy, the CVD process can operate at lower temperatures. Limited selection of adequate reactants serving as nitrogen and/or carbon donors restricts applications of MTCVD technology. Most common in practice are organic nitrile compounds (R-CN) [37]. Thermal reactions of TiCl4 with organic nitrile compounds inevitably produce carbonitrides. Parameter variations can be used to control the carbon to nitrogen ratio within the carbonitride, to a certain extent. TiCl4 ‡ H2 ‡ R-CN ! TiCN ‡ HCl ‡ RH=RCl (Stoichiometric coefficients were omitted purposely as components react incompletely and additional by-products are formed, respectively.) MTCVD operates at temperatures between 800 and 900 8C. At these temperatures, g phases within the hard material are avoided, stresses induced in the interface between the TiCN layer and the substrate are reduced, and the anticipated geometrical changes in steel parts decrease (Fig. 5.13). Furthermore, the coating growth rate of TiCN produced from organic nitrile compounds is considerably higher, allowing higher coating thickness within reasonable process time. Combining HT and MTCVD in single hybrid processes yields coatings benefiting from the particular advantages of both technologies (Fig. 5.14).

5.7 CVD Coating at Lower Temperatures

Fig. 5.13 Edge of a hard-metal indexable insert, MTCVD coated. No (eta) g phases (IonBond).

Fig. 5.14 Combination of MT and HTCVD coatings (Bernex).

81

82

5 Development and Status Quo of Thermal CVD Hard-material Coating

5.7.2 Plasma-activated CVD, PACVD

An additional means of providing necessary activation energy for reactive components in CVD processes is silent electric discharge. In cold discharge plasma, the active species such as ions, radicals, and excited states are the energy carriers. The energy content of such excited species can correspond to temperatures of several 1000 8C. In spite of this, the macroscopic temperature of low-pressure discharge plasma remains low, with a maximum of several 100 8C. Developers have tried to activate thermal TiCl4 CVD processes by means of plasma discharge. However, these and other analogous approaches failed due to lack of chemical and physical process knowledge. In contrast, certain plasma-assisted deposition processes are feasible, when elementary atomic processes are utilised to deposit micro- and nano-crystalline structured material. A typical example are carbon coatings, which have been subject to ongoing development since the late 1970s [57, 61]. DLC (diamond-like carbon) coatings are state-of-the-art highly cross-linked, amorphous carbon coatings. It is unfortunate that many carbon coating materials are referred to as DLC although only a few meet the necessary criteria precisely. An accurate classification is currently being established. Amorphous, carbon-based coatings show very low friction coefficients, high inertness, and particularly low external adhesion. Therefore, the coatings are used in applications that require low lubricant-free friction, or when adhesion to the surrounding media is extremely undesirable. Examples are sliding surfaces and complex-shaped surfaces for producing plastic, elastomer, aluminium, and copper parts, as well as several special applications. Coating and material development is progressing constantly. On a laboratoryproject scale, plasma deposited coatings have proven that considerably improved coating properties are obtainable. Amongst these are amorphous, high-temperature resistant coatings with particularly high adhesion to the substrate, and micro- or nano-crystalline diamond coatings with true diamond structure. These materials have already been deposited with relatively high coating thickness.

5.8 Summary and Conclusions

Prospects are good for the coming years of thermal CVD technology development. The full potential of thermal CVD technology has not yet been tapped. Technological development during the past 10 to 20 years has, in part, been neglected. Improved development and optimisation of system and process technology can lead to higher capacity, larger reactor volume, and increased growth rates.

References

Deposition reactions can be optimised and growth rates accelerated with added knowledge of the determining elementary reactions. Now, depositable new materials and coating combinations, including further refractory metals apart from titanium, promote additional applications, allow use of better hard materials in existing applications, and generally assist problem solving. New technologies, currently under development, reduce dimensional changes in steel materials due to heat input. Additional coating materials yield precisely matched high-performance coatings. Medium-term predictions suggest considerably improved CVD technology with performance, though already high, increasing further. Thermal CVD technology is comparatively straightforward. The process is smooth and relatively simple to stabilise from a technical point of view. Coatings produced by means of HT- or MTCVD show increased service life as well as higher strength, performance, and stability than comparable PVD coatings. Thus, a favourable cost-performance ratio reduces overall product costs, and combined with new application prospects, CVD technology will be even more competitive.

References 1 Benninghoff, H., Zickler, H.: Funktio-

2

3

4

5

6

7

nelle verschleißfeste Oberflächen durch CVD-Verfahren. Feinwerktechnik & Messtechnik, 86 (1978) 8, pp. 389–394 Bernex Seminar “CVD-CVD/MT-Verfahren”, October 24th, 1984, Bernex GmbH, Langenfeld Bonetti, R.: Verschleißfeste CVD-Schichten. Schweizer Maschinenmarkt, 52, (1982) Bräuning, H.: Eine besondere Titankarbid-Beschichtung von Hartmetall-Wendeschneidplatten für Fräswerkzeuge. VDI-Z, 116 (1974) 11, pp. 878–882 Campbell, I. E., Sherwood, E. M. (Eds.): High Temperature Materials & Technology, John Wiley, 1967 Dannöhl, H. D., Güttinger, U.: Apparaturen und Produktionsanlagen zur Beschichtung. Haus der Technik – Lecture Notes 338, Essen, 1974, pp. 39–44 DIN Technical Report 39, Jehn, Hermann: Charakterisierung dünner Schichten, p. 210. Deutsches Institut für Normung e.V., 1st edn, Beuth, 1993

8 Ebersbach, G., Mey, E.: Metallkarbid-

9

10

11

12

behandlung von Werkzeugen und Bauteilen. Neue Hütte, 16, No. 3, 1971, pp. 145–148 Ebersbach, G., Mey, E.: Herstellung, Gebrauchseigenschaften und Anwendung verschleißbeständiger Metallkarbidschichten aus der Gasphase. Schmierungstechnik, Berlin, 14 (1983), 4, pp. 113–117 Ebersbach, G., Mey, E.: Thermodynamik und Kinetik der Bildung und des Wachstums von Titankarbidschichten auf metallischen Substraten. lecture III/6, Inmetherm ’75, Karl-Marx-Stadt, 1975 Ebersbach, G., Mey, E., Pursche, G.: Verbesserung des Verschleißverhaltens von Stählen durch thermochemisches Abscheiden von Metallkarbiden. Schmierungstechnik, 8 (1977) 1, pp. 14 et seqq. Ebersbach, G., Mey, E., Ullrich, G.: Vanadinkarbidbehandlung von Umformwerkzeugen und Bauteilen. Die Technik, 29 (1971) 4, pp. 273–275

83

84

5 Development and Status Quo of Thermal CVD Hard-material Coating 13 Ebersbach, G., Mey, E., Ullrich, G.:

14

15

16

17

18

19

20

21

22

23

24

25

Chromkarbidbehandlung von Werkzeugen und Bauteilen, Die Technik, 24 (1969) 8, pp. 526–530 Ebersbach, G., Mey, E., Ullrich, G.: Titankarbidbehandlung von Werkzeugen und Bauteilen. Die Technik, 24 (1969), 7, pp. 460–464 Ebersbach, G., Mey, E., Ullrich, G.: Thermische Nachbehandlung metallkarbidbehandelter Werkstoffe. Die Technik, 29, No. 1, 1974, pp. 44–47 Ebersbach, G., Mey, E., Ullrich, G.: Oberflächenveredlung von Umformwerkzeugen durch Dampfphasenprozesse. Fertigungstechnik und Betrieb, 29, (1979) 5, pp. 281–282 Eversberg, K.-R.: Umformwerkzeuge leben länger. Maschinenmarkt, 92 (1986) 12, pp. 48–53 Formplast Verschleißschutz GmbH & Co KG: Besigheim, Company Notes, Brochure, 1997 Formplast Verschleißschutz GmbH & Co KG: Beschichtungen für verschleißund korrosionsbeanspruchte Bauteile und Formen in der Kunststoffverarbeitung – Hartstoffbeschichtung mit CVDTechniken. Final Report 1999 Friedrich, C., Berg, G., Broszeit, E., Berger, C.: Datensammlung zu Hartstoffeigenschaften, Materialwissenschaft und Werkstofftechnik, 28 (1997) pp. 59–76 Hänni, W., Hintermann, H. E.: CVD of TiN at the tempering temperature of alloy steels. Euro-CVD 4, Eindhoven, 1983, pp. 468–478 Hertz, D., Gindraux, G., Tannenberger, H.: Protective CVD coatings on long pieces. Euro-CVD 4, Eindhoven, 1983, pp. 350–356 Hintermann, H. E., Gass, H.: Die chemische Abscheidung aus der Gasphase – CVD. Oberfläche – Surface, 12 (1971) 10, pp. 177–180 Horvath, E., Zimmermann, H.: CVDBeschichtungstechnik im Werkzeugbau. Fertigung, Technische Rundschau, Bern, 1979 Inzenhofer, A.: Verschleißschutz durch CVD-Behandlung. Wt – Zeitschrift für industrielle Fertigung, 75 (1985) 4, pp. 251–256

26 Jönsson, R.: Richten wärmebehandelter

27

28

29

30

31

32 33

34

35

36

37

38

Werkstücke. Härterei Technische Mitteilungen, 53 (1998) 1, pp. 5–8 Keller, K.: Voraussetzungen für den erfolgreichen Einsatz hartstoffbeschichteter Werkzeuge in der Kaltmassivumformung. Werkstatttechnik, 77 (1987) pp. 629–634 Keller, K.: Beschichten von Umformwerkzeugen, Voraussetzungen, Ausführungen. Werkstatt und Betrieb, 121 (1988) 11, pp. 913–917 Keller, K.: Beschichten von Umformwerkzeugen, Trägerwerkstoffe. Werkstatt und Betrieb, 122 (1989) 1, pp. 85–87 Keller, K.: Beschichten von Umformwerkzeugen, Anwendererfahrungen. Werkstatt und Betrieb, 122 (1989) 5 Keßler, O., Hoffmann, F., Mayr, P.: Maßund Formänderungen bei der Gasabschreckung des unbeschichteten und CVD-beschichteten X155CrVMo12-1. Härterei Technische Mitteilungen, 50 (1995) 6, pp. 337–343 Kieffer, R., Benesovsky, F.: Hartstoffe. Springer, Vienna, 1963. König, U., Dreyer, K., Reiter, N., Kolaska, J., Grewe, H.: Stand und Perspektive bei der chemischen und physikalischen Abscheidung von Hartstoffschichten auf Hartmetallen. Technische Mitteilungen, Krupp, Vol. 39 (1981) H 1, pp. 13–22 Krause, H., Jühe, H. H.: Röntgenographische Eigenspannungsmessungen an aus der Gasphase abgeschiedenen Titancarbidschichten. Härterei Technische Mitteilungen, 32 (1977) 6, pp. 302–307 Leonhardt, A., Schläfer, D., Seidler, M., Selbmann, D., Schönherr, M.: Microhardness and texture of TiCx layers on cemented carbides. Journal of the Less Common Metals, 87 (1982), pp. 63–69 Lindström, J. N., Stjernberg, K. G.: Rate determining steps at CVD of Al2O3, TiC and TiN. Euro-CVD 5, Uppsala, 1985, pp. 169–182 Mathesius, H. A., Bonetti, R.: CVDBeschichtung bei Mitteltemperaturen. Industrie Anzeiger, no. 31, 1985 Mohn, E., Bonetti, R., Wiprächtiger, H.: Protective CVD coatings for the tool industry: Requirements for process control and equipment. Journal de Physique, 50 (1989) C5, pp. 811–819

References 39 Münster, A.: Eigenschaften und Anwen-

40

41

42

43

44

45

46

47

48

dungen von Titannitrid und Titancarbid, Angewandte Chemie, 69 (1957) 9, pp. 281–290 Münster, A., Ruppert, W.: Oberflächenschichten aus Titannitrid. Die Naturwissenschaften, 39 (1952) 15, pp. 349–350 Münster, A., Ruppert, W.: Oberflächenschichten aus hochschmelzenden Titanverbindungen, Zeitschrift für Elektrochemie. Ber. Bunsenges. f. phys. Chem., 57 (1953) 7, pp. 564–571 Münster, A., Ruppert, W.: Zur Thermodynamik einiger Reaktionen des Titantetrachlorids. Zeitschrift für Elektrochemie. Ber. Bunsenges. f. phys. Chem., 57 (1953) 7, pp. 558–564 Oldewurtel, A., Eversberg, K. R.: CVDBeschichtung von Werkzeugen. Härterei Technische Mitteilungen, HTM, 44 (1989) 1, pp. 25–31 Oldewurtel, A., Eversberg, K. R.: Mehrleistung von CVD-beschichteten Umformwerkzeugen. Wt – Zeitschrift für industrielle Fertigung, 75 (1985) 5, pp. 299–304 Paterok, L. J.: Hartstoffbeschichtung durch thermochemische Abscheidung aus der Gasphase, Bernex GmbH, Final Report. In: Tribologie (Reibung, Verschleiß, Schmierung) Vol. 1. Springer, 1984, pp. 251–328, ISBN 3-540-10800-9 and 0-387-10800-9 Paterok, L. J.: Beitrag zur Herstellung und dem Verhalten von CVD-VerschleißSchutzschichten an Hochleistungsschnellarbeitsstahl. Doctoral Thesis, RWTH, Aachen, 1984 Perry, A. J., Horvath, E.: Interlayers formed in the carbide coating of steel by chemical vapour deposition. Thin Solid Films, 62 (1979) pp. 133–143 Pursche, G., Rosert, R., Resch, H., Steinhäuser, S., Fabian, D., Ebersbach, G., Richter, E., Zenker, R., Zenker, U., Büge, B.: Verschleißschutz durch Oberflächenschichten – eine Betrachtung zum Entwicklungsstand und zu den Entwicklungstendenzen. Schmierungstechnik, Berlin, 17 (1986) 4, pp. 111–115. Schmierungstechnik, Berlin, 17 (1986) 5, pp. 142–148. Schmierungstechnik, Berlin, 17 (1986) 6, pp. 168–173

49 Ramaekers, P. P. J., van Loo, F. J. J.: The

50 51

52

53

54

55

56

57

58

59

60

61

growth of TiC on iron-carbon substrates by CVD. Euro-CVD 4, Eindhoven, 1983, pp. 546–552 Römpp: Chemie Lexikon, 9th edn, 1995, Thieme, pp. 2190 et seqq. Keramik Ruppel, W., Schlamp, G.: Werkzeuge mit Titankarbid-Überzug. Bänder Bleche Rohre, 1964, no. 10, pp. 553–559 Ruppert, W.: Nitrier-, Chromkarbid- und Titankarbidschichten bei Werkzeugen. Bänder Bleche Rohre, 11 (1970) 4, pp. 216–222 Ruppert, W.: Die Abscheidung von Titankarbidüberzügen auf Stahloberflächen. Metalloberfläche, 14 (1960) 7, pp. 193– 198 Ruppert. W.: Abscheidung, Eigenschaften und Verhalten von Titancarbidschichten auf Stahloberflächen. Doctoral Thesis, Technical University of Darmstadt, 1959 Ruppert, W.: Gasplattieren von Eisenbasiswerkstoffen mit Hartstoffen, dargestellt am Beispiel von Titancarbidüberzügen. Haus der Technik – Lecture Notes 338, Essen, 1974, pp. 4–18 Schedler, W.: Für die meisten Anwendungsgruppen nur zwei Hartmetallsorten. Maschinenmarkt, 80 (1974) 68, pp. 1320–1322 Schürer, C.: Diamantähnliche Kohlenstoffschichten: Herstellung, Eigenschaften und Anwendungen. Vakuum in Forschung und Praxis, 3 (1996) pp. 163– 169 Sproul, W. D., Richman, M. H.: Effect of the eta layer on TiC-coated cementedcarbide tool life. Journal of Vacuum Science Technology, 12 (1975) 4, pp. 842–844 van der Straten, P. J. M., Michorius, M. M., Verspui, G.: The decarburisation of steel during the CVD of TiC. EuroCVD 4, Eindhoven, 1983, pp. 553–567 van der Straten, P. J. M., Verspui, G.: Aufbringen verschleißfester Schichten auf Werkzeugstahl nach dem CVD-Verfahren. VDI-Z, 124 (1982) 18, pp. 693– 698 Weissmantel, C.: Proceedings 7th Internat. Vac. Congr. & 3rd Intern. Conf. Solid Surfaces, Vienna, 1977, p. 1533

85

86

5 Development and Status Quo of Thermal CVD Hard-material Coating 62 Wiegand, H., Ruppert, W.: Einige Eigen-

schaften der Werkstoffkombination Stahl mit Titankarbidüberzug. Metalloberfläche, 14 (1960) 8, pp. 229–235 63 Wolfstieg, U., Macherauch, E.: Zur Definition von Eigenspannungen. Härterei

Technische Mitteilungen, 31 (1976) 1/2, pp. 2–3 64 Zimmermann, H., Bartknecht, W.: Precautionary design of CVD systems. EuroCVD 3, Neuchatel, 1980, pp. 62–70

87

6 Hot-filament CVD Diamond Thin Films O. Lemmer, R. Cremer, D. Breidt, M. Frank, J. Müller, CemeCon AG, Würselen, Germany 6.1 Introduction

Increasing requirements for high-performance cutting tools have led to a number of parallel developments, especially in high-speed, hard, and dry cutting of alloys and composite materials that are difficult to machine, e.g. highly abrasive materials or materials that show a tendency to form built-up edges. On the one hand they provoke development of new, market-driven protective coatings. On the other hand demand promotes an intensified integral approach to the overall cutting process. This includes sophisticated matching procedures in order to adapt tool/coating systems to workpiece/machine systems [16]. This chapter focuses on deposition of diamond thin films on hard-metal tools by means of hot filament CVD technology, with respect to substrate material, intended applications, and the problems mentioned above. Table 6.1 Properties of CVD diamond thin films [14] Property

Value

Hardness

8000–10000 HV0.05

Thermal conductivity

W m–1 K–1 approx. 1000 GPa 1.1 10–6 K–1

Modulus of elasticity Coefficient of thermal expansion

Unit

500–2000

Friction coefficient

0.02–0.1

–

Poisson’s ratio Roughness Ra

0.1–0.2 0.1–0.4

– lm

Comments

Reference value

texture dependent, as in natural diamond depends on density of grain boundaries extremely high between 0–100 8C, increases as temperature rises depending on friction pairing and running-in period diverse in literature depends on structure and thickness of coating

CBN 4500

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

Cu

400

TiN

9.6

Teflon 0.05

88

6 Hot-filament CVD Diamond Thin Films

Diamond has been deposited with reference to commercial CVD processes for approx. 15 years. Diamond features outstanding properties (Table 6.1), and therefore, is applicable to many different types of application. Most important for cutting applications are the extreme hardness and the excellent tribological properties of diamond. Compared to conventional coatings, they can yield an increase in service life by an order of magnitude (tenfold), reduce cutting force, and allow dry machining of workpieces. Machining of many materials is possible and economic only due to diamond-coated tools that can be used to machine Al-Si alloys, graphite, hard metal and ceramic green compacts [9–11], as well as fibre-reinforced composite materials and sandwich structures.

6.2 Differences of Diamond Tools

Today, diamond cutting materials for hard metal indexable inserts are available in different types: as sintered material (PCD), CVD diamond thick coatings, or CVD diamond thin films. The two bulk material type tools with PCD and CVD diamond thick coatings yield high cutting performance, i.e. high volume of workpiece material removed, however, they cannot be formed, e.g. for applications where chip breakers are needed at the cutting edge [18]. The main advantage of CVD diamond thin films is that they allow deposition on complexshaped substrates, e.g. drills or milling cutters [1, 5, 7].

6.3 Substrate Pre-treatment

In many cases, pre- and post-treatment of coatings and tools is necessary in addition to pure coating. Different cleaning, blasting, and polishing procedures for treating tool surfaces and cutting edges have considerable impact on the behaviour of the entire system. This applies to CVD coating of hard metal as well. Today, commercially available diamond-coated tools usually are made from hard metal, i.e. sintered tungsten carbide embedded in a matrix with different contents of cobalt, because this base material supports the hard diamond coating well. However, the cobalt binder used in the substrate catalyses the undesired formation of graphite during coating. Avoiding this effect is crucial during the early coating phases because, otherwise, the diamond coating does not adhere sufficiently to the substrate. Cobalt removal involves sophisticated processes that have to be matched carefully to each hard-metal type [13, 15]. One way of avoiding graphite formation between substrate surface and diamond coating is by applying a diffusion barrier. However, this technique is rare in tool coating due to the high costs involved. Mostly, chemical or thermal treatment is used to remove or inactivate [4] the cobalt in the near-surface zone of the sub-

6.4 Production of CVD Diamond

Fig. 6.1 Pre-treatment of hard metals [3].

strate. The chemical etching procedure for cobalt as well as subsequent phases of pre-treatment are illustrated in Fig. 6.1. Pre-treatment techniques strongly influence the resulting interface with respect to adhesion and roughness, but also affect profitability and reliability of the coating process. A rough surface creates mechanical bond between substrate and diamond coating, and is essential for the bond strength due to the extremely different properties of substrate and coating material (modulus of elasticity, hardness). When, at the same time, the coating process initially promotes growth of pure, crystalline diamond into the rough surface, very high bond strength is possible even for thicker coatings (> 20 lm). In addition, the rough and therefore large surface area in the interface reduces crack growth within this critical zone. Using coarse-grained hard-metal grades is economical, however, fine-grain types are pretreatable as well. Figure 6.2 shows the cutting edge of a diamond-coated hard metal that is subject to considerable strain during operation and requires optimal pre-treatment.

6.4 Production of CVD Diamond

Several techniques have been established for commercial use of the low-pressure synthesis of CVD diamond. Combustion, microwave plasma, DC arc jet, and hot-filament processes are predominant. Apart from equivalent operating pressure, the common principle of these techniques involves decomposition of carbon-containing gases, e.g. methane, acetylene, or carbon monoxide, with subsequent carbon deposition at the substrate surface. Undesired graphite formation can be avoided by setting appropriate process parameters and by hydrogen etching. Atomic hydrogen therefore plays an important role in the nucleation

89

90

6 Hot-filament CVD Diamond Thin Films

Fig. 6.2 Morphology of etched hard metal surface [3]

and growth of diamond coatings. The described techniques, however, differ significantly in their ability to coat complex-shaped geometries. Hot-filament processing is most common for coating complex shaped tools, e.g. drills or milling cutters.

6.5 Hot-filament Process

The CVD or hot-filament process is the best known and most suitable process technology for tool coating with CVD diamond thin films. In this technique, the required atomic hydrogen is produced with electrically heated glowing wires made from refractory metals (Ta, W, or Re). An appropriate spatial arrangement provides for an even distribution of the required species, even when processing substrates with complex geometry. Therefore, the process is easy to adapt for different batch sizes, and economical diamond coating of tools is feasible [12]. Figure 6.3 shows the basic principle. Atomic hydrogen is produced in the activation zone (Fig. 6.4 a), diffuses across the recipient, recombines at the CH4 molecule and thus forms the methyl radical CH3·, representing the actual growth species (Fig. 6.4 b) for CVD diamond deposition. Further recombination of atomic hydrogen at the substrate surface (Fig. 6.4 c) produces free spots at the growth front (Fig. 6.4 d) and transports sp2 hybridised (graphitic) carbon back to the gas phase. Growth of the diamond film initially starts with crystallisation at nucleation sites, the finally

6.5 Hot-filament Process

Fig. 6.3 Diamond-coated cutting edge of optimally pre-treated finest grain hard-metal substrate [3].

Fig. 6.4 Principle of hot-filament technique for CVD diamond-coating deposition [19].

closed film forms when small nuclei islands intergrow (Figs. 6.4–6.6). Crystals with beneficial orientation to the substrate gradually overgrow adjacent crystals. Conventional deposition of CVD diamond therefore forms characteristic columnar structures with rough surfaces (Fig. 6.5, left).

91

92

6 Hot-filament CVD Diamond Thin Films

6.6 Controlling CVD Diamond Properties

Economic and reliable coating of high-performance tools with complex geometry requires careful control of further thin-film properties such as film morphology. Under certain boundary conditions, production of smooth-surface diamond thin films is possible. Here, other than in conventional deposition, process parameters are adjusted in order to promote continuous re-nucleation (new formation of nuclei, Fig. 6.5, right). Such nano-crystalline diamond coatings can actually level uneven substrate areas. Due to the smooth surface, they are particularly adequate for machining of soft non-ferrous metals because formation of built-up edges is prevented. In Fig. 6.6, the structure of conventional micro- and smooth nano-crystalline diamond coatings is compared. Beneficial properties of micro- and nano-crystalline diamond coatings can be combined in multi-layer structures (Fig. 6.7). This concept increases fracture toughness and the growth of cracks is prevented. Cracks, therefore, cannot easily progress to the interface between the substrate and the first diamond layer, and thus, do not cause failure of the entire coating. Formation of compressive stresses due to different coefficients of thermal expansion between substrate

Fig. 6.5 Schematic representation of micro- and nano-crystalline CVD diamond coatings [17].

Fig. 6.6 Surface structure of micro- and nano-crystalline CVD diamond coatings on hard-metal substrates [3].

6.8 Post-treatment of CVD Diamond

Fig. 6.7 Micro- and nano-crystalline CVD diamond coating multi-layer on hard-metal substrate [3].

and coating, and between micro- and nano-crystalline diamond, has the positive effect that vertical cracks change direction and rather grow parallel to the substrate at multi-layer interfaces [8].

6.7 Industrial Deposition of CVD Diamond

CemeCon uses fully automated coating systems (CC 800 D type, Fig. 6.8) in order to carry out industrial hot-filament deposition of CVD diamond thin films. Reproducibility of CVD processes and uniform, homogeneous coating of tools and parts in large batches is guaranteed by acquiring, documenting, and controlling all relevant process parameters.

6.8 Post-treatment of CVD Diamond

The coating thickness and grain size of diamond thin films as well as the interface to the substrate affect the inevitable formation of rounded cutting edges on hard-metal cutting tools coated with CVD diamond. Post-treatment in terms of

93

94

6 Hot-filament CVD Diamond Thin Films

Fig. 6.8 CC 800 D type CVD diamond-coating systems, based on hot-filament technique [3].

sharpening the cutting edge is advisable when a particular application requires extra high surface quality of workpieces or reduced cutting force. GfD (Ulm, Germany), CemeCon, and WMtech (Ulm, Germany) collaborated in order to develop a sharpening process for cutting-edge radii of less than 0.5 lm on coated tools [6]. The noticeable geometric difference of the sharpened edge (Fig. 6.9, right) compared to an ideal geometry is intended and increases cutting-edge stability. Such sharpened cutting edges allow fine machining of alloys that are abrasive and tend to form built-up edges, and they allow clean cutting of fibres in wood and fibre-reinforced materials. 6.9 Applications for Diamond-coated Tools

Conventional diamond tools are already being used in several fields of applications (Fig. 6.10). These tools (with PCD and CVD diamond thick coatings) allow

Fig. 6.9 Plasma sharpened cutting edge of nano-crystalline CVD diamond coating on hard-metal substrate [6].

6.9 Applications for Diamond-coated Tools

Fig. 6.10 Applications of diamond-coated cutting tools [3].

cutting of highly abrasive materials, e.g. Al-Si alloys or MMCs (metal matrix composites). Non-ferrous metals can be machined with PCD tools when very high surface quality is required. Here, PCD directly competes with common uncoated or PVD as well as CVD-coated hard-metal cutting tools. Prospects are good for increased use of CVD diamond thin films as their properties can be matched to processing tasks in a wide range. The following examples illustrate successful applications. The first example is a crystalline diamond coating (CCDia®08) for machining graphite, hard metal, and ceramic green compacts. The coating is extremely resistant to abrasive wear. The cost : performance ratio is good, however, the surface is rough and therefore machining is limited to materials that do not tend to stick and adhere to the surface. Compared with uncoated tools, service life was improved 13-fold for graphite milling at cutting speeds of 600 m/min (Fig. 6.11). The following application introduces another crystalline diamond coating (CCDia®HiCo) for machining graphite, hard metal, and ceramic green compacts, yet this coating is optimised for tools with high cobalt content. Just as CCDia®08, it is limited to materials that do not tend to stick to the surface. Here, economic advantages are a result of reduced stock keeping since coating of many different hard-metal types is possible. Compared with uncoated tools, service life was improved 19-fold for graphite HSC (high-speed cutting) milling at cutting speeds of 1800 m/min (Fig. 6.12). Another example presents a multi-layer diamond coating (CCDia®FiberSpeed) with nano-crystalline top coat for processing fibre-reinforced plastics, and composites as well as sandwich materials such as Al-Ti and Ti-CFRP (carbon-fibre-

95

96

6 Hot-filament CVD Diamond Thin Films

Fig. 6.11 Application example: milling of graphite (V 1466) with CCDia®08 [2].

Fig. 6.12 Application example: milling of graphite (EK 85) with CCDia®HiCo [2].

reinforced plastics). Due to the multi-layer concept, the coating is extremely resistant to abrasive wear, and toughness values are high. Coating adhesion to the substrate is very good for all types of hard metals with medium cobalt content and the smooth surface allows machining of materials with a high tendency to stick to the surface. Compared with uncoated tools, service life was 1,000% higher for carbon-fibre-reinforced plastic (CFRP) milling, and 680% compared to a TiAlN coating, at cutting speeds of 600 m/min (Fig. 6.13). The same diamond coating (CCDia®FiberSpeed) qualifies for milling Al-TiCFRP composite material at cutting speeds of 250 m/min also. In this application, service life was 900% higher (Fig. 6.14).

6.9 Applications for Diamond-coated Tools

Fig. 6.13 Application example: milling of carbon-fibre-reinforced plastic (CFRP) ® with CCDia FiberSpeed [2].

Fig. 6.14 Application example: milling of composite material (Al-Ti-CFRP) ® with CCDia FiberSpeed [2].

A thick multi-layer diamond coating (CCDia®Tiger) is recommendable for the most challenging applications, including every example mentioned above. The coating is extremely resistant to abrasive wear, very tough, and adhesion to the hard-metal substrate is remarkable due to a strong mechanical bond. It is particularly adequate for machining material with high tendency to stick, and overall performance is outstanding. Compared to uncoated tools, service life increases by 1300% for abrasive AlSi17Mg milling, and 1100% compared to TiAlN-coated tools, at cutting speeds of 750 m/min (Fig. 6.15). ® The same diamond multi-layer (CCDia Tiger) is superior also in milling of MMCs with 30% SiC content. At cutting speeds of 450 m/min, service life increased 20-fold (Fig. 6.16).

97

98

6 Hot-filament CVD Diamond Thin Films

Fig. 6.15 Application example: milling of AlSi17Mg with CCDia®Tiger [2].

Fig. 6.16 Application example: milling of MMC with CCDia®Tiger [2].

The last example features a sharpened diamond coating, designed specifically for precision machining of Al-Si and Ti-Al alloys, MMCs, and plastics with mineral fillers. Sharpening produces extremely sharp edges and is applicable to all planar geometries. In cutting plastic foil with embedded abrasive colour pigments (titanium oxide), ceramic blades increased service life by 133%. A sharpened diamond coating, though, yielded an increase in service life of 667% (Fig. 6.17).

6.10 Summary and Conclusions

Fig. 6.17 Application example: cutting of plastic foil with sharpened diamond coating [2].

6.10 Summary and Conclusions

Development of hard-metal tools with CVD diamond thin films requires a systematic approach that considers and matches all components involved, i.e. tool substrate, protective coating, workpiece, and machines. This optimisation, along with selective adjustment of custom coating properties, is a prerequisite for the successful production and launch of new products. Close co-operation of tool producers, coating suppliers, and end users is essential for the development process. Examples show the elaborate approach needed to tap the full potential of the outstanding coating material, CVD diamond. Sophisticated applications are possible due to optimal substrate pre-treatment, adjusted coating thickness, modification of coating morphology, use of multi-layer coatings, and, last but not least, sharpening of tool edges. CVD diamond coatings are particularly useful for new, difficult-to-machine high-tech materials. They are less expensive than tools with PCD and CVD diamond thick coatings, and can be produced with complex geometries. Diamond-coated cutting tools are being used successfully in many different fields of application, and their economic potential is high. As coating suppliers’ and tool producers’ knowledge is growing constantly, an increase of applications and a market expansion for CVD diamond thin films is most likely.

99

100

6 Hot-filament CVD Diamond Thin Films

References 1 Berky, E., et al.: Fräsen von Flugzeugin-

2 3 4

5

6

7

8

9

10

tegralbauteilen aus Alu-Knetlegierungen mit Minimalmengenschmierung im Serieneinsatz bei der DASA, VDI Report 1339, VDI, Düsseldorf (1997) pp. 123– 135 CemeCon AG, Würselen: Product Catalogue CemeCon AG, Würselen: Company Image Cremer, R., Müller, J., Neuschütz, D., Leyendecker, T., Lemmer, O., Frank, M., Gussone, J.: Improved Chemical Vapor Deposition of Diamond Layers on Cemented Carbides by Formation of Tungsten-Cobalt-Boron Phases in the Interface Region. Electrochemical Society Proceedings 13 (2001) pp. 365–372 Fernandes, A. J. S., Silva, V. A., Carrapichano, J. M., Dias, G. R., Silva, R. F., Costa, F. M.: MPCVD diamond tool cutting-edge coverage: dependence on the side wedge angle. In: Diamond and Related Materials 10 (2001) pp. 803–808 Flöter, A., Gluche, P.: Verbesserung der Schärfe diamantbeschichteter Hartmetallklingen, IDR Industrie Diamanten Rundschau 38 (2004) pp. 110–112 Gäbler, J., et al.: Chemical vapour deposition diamond microtools for grinding, milling and drilling. In: Diamond and Related Materials 9 (2000) pp. 921– 924 Gåhlin, R., Alahelisten, A., Jacobson, S.: The effects of compressive stresses on the abrasion of diamond coatings. Wear 196 1–2 (1996) pp. 226–233 König, W., Klocke, F.: Fertigungsverfahren 1 – Drehen, Fräsen, Bohren, 5th revised edn. Springer (1997) p. 156 König, W., Klocke, F., König, M.: Hochleistungszerspanung von Graphit, wt-Produktion und Management 85 (1995) pp. 503–509

11 Lemmer, O.: Revolution im Formenbau

12

13

14

15

16

17

18

19

– Graphit kontra Kupfer – HSC-Bearbeitung mit CVD-diamantbeschichteten Werkzeugen statt konventionellem Fräsen. Stahl 59 (1997) Leyendecker, T., Lemmer, O., Esser, S.: Ein neuer Beschichtungsprozeß in einem innovativen Anlagenkonzept zur Produktion von haftfesten Diamantbeschichtungen. Proceedings of the 14th International Plansee Seminar, 4 C9A, Reutte, Austria (1997) Lux, B., Haubner, R.: CVD Diamond for Cutting Tools. In: Dischler, B., Wild, C. (Eds.) Low-Pressure Synthetic Diamond, Springer Series in Materials Processing, Springer, Berlin, Heidelberg (1998) pp. 223–242 Spear, K. E., Dismukes, J. P.: Synthetic Diamond: Emerging CVD Science and Technology, John Wiley & Sons (1994) Sun, F. H., Zhang, Z. M., Chen, M., Shen, H. S.: Fabrication and application of high quality diamond-coated tools. Journal of Materials Processing Technology 129 (2002) pp. 435–440 Tönshof, H. K., Mohlfeld, A., Spengler, Chr.: Pre-Treatment of Coated Tools for Cutting Applications, Proceedings of the 2nd International Conference on Coatings in Manufacturing Engineering, Mai 9th–10th, 2001, Hannover (2001) P2–1 VDI Guideline 2840: Kohlenstoffschichten, Übersicht über Schichttypen und Eigenschaften. Carbon films – Basic knowledge, coating types and properties. VDI-Gesellschaft Produktionstechnik (ADB), Düsseldorf (2003) Wertheim, R., Satran, A.: Neue Werkzeugkonzepte zur Bearbeitung von Formen und Gehäusen. VDI Report 1399, VDI Düsseldorf (1998) pp. 351–378 http://www.cvd-diamantwerkzeuge.de/

101

7 An Introduction to Electrodeposition and Electroless Plating Processes W. Olberding, Institut für Galvano- und Oberflächentechnik Solingen GmbH (Institute for Electroplating and Surface Technology), Solingen, Germany 7.1 Introduction

From a historic point of view, electrodeposition is one of the oldest coating processes in industrial use. The basic principles were developed and applied in the early nineteenth century. Batteries supplied the electrical energy required for electrochemical conversion of dissolved metals, however, this source of energy was far from sufficient. By the mid-nineteenth century, development of the dynamo allowed enough electrical energy to be produced for electrochemical deposition from electrolytes on an industrial scale (see, for instance, [3] and references there). Electrodeposition is one of the most common techniques of coating many different materials for aesthetic or technical reasons, since.

7.2 Fundamentals of Electrodeposition (Considering Nickel Deposition as Example)

Considering nickel as example, fundamentals of electroplating are simple to explain. Most of the mechanisms described basically apply to other systems as well. Brugger [1] explains the electrodeposition of nickel in detail. This chapter focuses on providing only necessary fundamentals of mechanisms and electrode processes for electroplating. Nickel salt is well dissolvable in water, nickel then exists as bivalent ions within the solution. When a source of direct current is applied the electrons cannot simply pass through the solution. The electrons are rather transferred to a species at one electrode, and released at the other. Transport of electrons within a solution is always coupled to a chemical reaction. The electrode with excess electrons takes on a negative charge and is referred to as the cathode, the other, with a lack of electrons, is the anode. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

102

7 An Introduction to Electrodeposition and Electroless Plating Processes

During nickel electrolysis, two electrons from the cathode are released to a positive nickel ion, which is transformed to metallic nickel: Ni2‡ ‡ 2e

! Ni0

…1†

The metallic nickel is deposited at the surface of the electrode that serves as an electron donor, and forms a coating. In nickel electrolysis, the anode is usually made of metallic nickel. Here, two electrons are taken from the solid, metallic nickel atoms, transforming these into bivalent ions that leave the metal matrix and dissolve in the solution: Ni0 ! Ni2‡ ‡ 2e

…2†

As, in this process, electrons flow from one side to the other across the external circuit, the overall process would represent a separation of charges. Consequently, the difference in potential between the electrodes would have to rise constantly with time in order to sustain electron flow. The flow of ions from one electrode to the other balances the separation of charges. Ion mobility can either be realised by connecting the two electrolyte volumes with an ion-permeable membrane, or by immersing both electrodes in the same solution volume. In nickel electroplating, both the anode and the cathode are usually within a single solution. Figure 7.1 shows a schematic representation of a galvanic cell for nickel deposition. Ideally, every ion deposited from the solution is replaced by an additional ion entering the solution, and thus, the composition of the solution, which is referred to as the electrolyte, is unchanged. Furthermore, all nickel ions can be considered as identical species. A simplified view of the complete system suggests that the nickel that dissolves at one electrode is deposited at the other. Unfortunately, nature is more complicated. In addition to nickel deposition, other side reactions occur at the electrodes, as in every chemical process. In this case, additionally, water decomposes at the cathode: 2H2 O „ …H3 O†‡ ‡ OH 2…H3 O†‡ ‡ 2e

! H2 ‡ 2H2 O

…3†

Fig. 7.1 Schematic representation of an electrolysis cell. Usually, anodic and cathodic electrolytes are not separated.

7.2 Fundamentals of Electrodeposition (Considering Nickel Deposition as Example)

At the cathode, apart from nickel deposition, formation of hydrogen gas is observed. Depending on the type of electrolysis process used, the proportion of electrical energy lost by the hydrogen formation is usually in the range of 2 to 6% of the total energy consumed, but can reach up to 95% in certain processes. In contrast, electrical energy lost in anode-side reactions amounts to less than 1%. The obvious consequence of this for the process is that nickel ion concentration increases so that, ideally, regular electrolyte dilution is necessary. In practice, however, a certain electrolyte proportion is lost during the process. The loss is compensated by adding fully demineralised water, so that the added nickel due to unequal current efficiency can even be beneficial. Furthermore, the electrolyte’s pH, simply representing the reciprocal H3O+ concentration, rises due to the loss of H3O+ ions at the cathode. Consequently, concentration of these particular ions is balanced by adding acid. The produced hydrogen, additionally, settles onto the surface of the deposited material and causes coating defects. Gas bubbles act as insulators and nickel deposition ceases in the immediate vicinity. As a result, pores are formed within the coating (Fig. 7.2). Pores can cause tremendous problems. For example, when a nickel coating protects steel from corrosion, a pore can be the origin of increased corrosion, and the end product is likely to be useless. Consumption of H3O+ ions leads to increased pH around the electrode. This increase, however, promotes formation of nickel hydroxide, which precipitates from the solution directly onto the electrode surface: Ni2‡ ‡ 2…OH†

! Ni…OH†2

…4†

Fig. 7.2 Steel sheet electroplated with several hundred micrometres of nickel. Indentations in the coating caused by hydrogen blisters are clearly visible. Increased current density at edges leads to increased build-up on edges and dull appearance due to exceeding of electrolyte operating range.

103

104

7 An Introduction to Electrodeposition and Electroless Plating Processes

This effect produces a dull, unaesthetic precipitate. A buffer, added to the electrolyte, counteracts and binds protons when pH decreases, and releases protons when pH increases. In electrolytic nickel deposition, boric acid is commonly used to suppress nickel hydroxide precipitation. In order to prevent localised, reduced or increased ion concentration, sufficient electrolyte exchange is necessary at the electrode surface. This is guaranteed either by continuously running circulation pumps, or by air injection. Furthermore, parts to be coated are moved within the electrolyte to ensure relative motion between parts and electrolyte. Electrolytes should show good electrical conductivity. Therefore, ions are added that conduct electrical energy particularly well. For nickel plating, e.g. chloride ions are introduced to the electrolyte. Undisturbed ion exchange requires careful control of electrolyte temperature, and wetting of the parts to be coated is guaranteed by adding a wetting agent. 7.2.1 Structure of Electroplated Nickel Coatings

The structure of deposited nickel coatings can be controlled by changing the electrolyte composition. Certain organic substances, added to the electrolyte, inhibit crystal growth in certain directions. This leads to crystal growth yielding laminar coating microstructure rather than columnar structure, which would form when the coating grows parallel to the lines of electric flux (Fig. 7.3). Therefore, the surface is smoother, and thus, appears slick and shiny. The additives are referred to as brighteners, the corresponding electrolytes are called bright electrolytes. A different way of changing the appearance of the coatings is to add different organic additives that smooth the rough surface. The suggested mechanism of this effect involves organic substances covering the surface in the direction of electric flux, which prevent the coating from growing in this direction. The result is a considerably smoother surface. The organic substances added to the electrolyte decompose at the electrodes, and the decomposition products are partially integrated into the coating.

Fig. 7.3 Complex coating structure of an automotive trim. Lamellar structure of bright nickel coating is clearly visible.

7.2 Fundamentals of Electrodeposition (Considering Nickel Deposition as Example)

Varying amounts of carbon and sulfur are integrated into the coating, depending on the added substances, applied current density, temperature, and pH. The sulfur content of the coating is in the range of 300 to 500 ppm, depending on deposition parameters and electrolyte type. The amount of carbon is usually lower, between approx. 150 to 300 ppm (in both cases, referred to mass). Carbon and, in particular, sulfur have a considerable effect on the corrosion behaviour of the coating. Coatings with high concentration of sulfur are generally far less noble than coatings with low sulfur content. The direction of corrosion (surface corrosion or pitting corrosion) can be adjusted by combining different nickel coatings with varying sulfur content. Terms used for such coatings are, for instance, duplex or triplex nickel coatings. 7.2.2 Deposition Mechanism

Fundamentals of deposition mechanisms are essential in order to understand another principle of controlling the results of nickel coating. During deposition, a nickel ion is, initially, discharged and deposited onto the surface as a not yet tightly bound metal atom. In a second, subsequent phase the atom finds its final position on the surface (Fig. 7.4). The very first atom deposited onto the surface, has to form a completely new nickel nucleus, as no other nucleation sites are present (process of nucleation). A second nickel atom, deposited in the immediate vicinity of the first, has the choice of either forming an additional nucleus, or attaching itself to the first nucleus, and therefore, increasing its volume (nucleus growth). By controlling the relative rates of nucleation and nuclei growth, coatings with entirely different properties are producible. While a combination of high nuclei growth rate with reduced nucleation rate yields a few large crystallites, high rates of nucleation combined with low nucleation growth rates lead to the formation of many small crystals. Coatings with many small, rather than a few large, crystallites typically show considerably less porosity, resulting in increased corrosion protection. The coating is easier to deform and has a brighter surface.

Fig. 7.4 Schematic representation of nickel deposition: initially, the nickel discharges at the surface (left) and subsequently diffuses to the final position (right).

105

106

7 An Introduction to Electrodeposition and Electroless Plating Processes

Relative rates of nucleation and nuclei growth are controllable by, for example, applying differently shaped electrical currents. At constant DC current, newly deposited atoms tend to attach to the active surface of existing crystallites. Whereas, when the electrical current is interrupted continuously (up to several hundred times per second), new crystals keep forming in random places and considerably more and smaller nuclei are produced. When, in addition to interrupting, the polarity of the current is changed continuously, peaks in the coating dissolve, yielding even smoother deposition and better levelling of indentations. This technique is commonly used for copper deposition on circuit boards. Adequately shaped pulses allow controlled deposition, so that bores in the circuit boards of less than 1 mm in diameter are completely filled with copper, and thus, both surfaces are connected electrically. In conclusion, even for the comparatively simple deposition of nickel, coating results are determined strongly by the chosen electrolyte, i.e. whether acid, alkaline, or sulfamate electrolyte is used, and depend on additives to the basic electrolyte, as well as the adjusted electrical current shape. Coatings with other chemical elements are produced in similar processes. Certain techniques involve additional means, e.g. chemical deposition of nonautocatalytic coatings or cementation. However, it would go beyond the scope of this chapter to describe these processes as they are covered, in detail, in other published work.

7.2.3 Current-density Distribution

The thickness of deposition is a crucial criterion for deposition results. The amount of nickel deposited is simple to calculate: An electric current of two electrons yields a single deposited nickel atom. Faraday’s law gives the deposition equivalent for nickel, stating that per amperehour (Ah), a mass of 1.0947 grams of nickel is deposited. This is only true in the case that the applied electrical energy is used entirely for the nickel deposition. As described above, a proportion of electrical current is used in side reactions, e.g. decomposition of water, so that an additional correction value is needed for the equation. Considering the total amount of metal deposited on the substrate, this calculation is correct. However, measuring the produced thickness of deposition often yields a considerable difference between theoretically calculated and obtained values. These discrepancies are usually due to extremely irregular distribution of electrical current across the substrate. Current density, i.e. the amount of electrons flowing per unit area, is significantly higher at workpiece edges and corners compared to, for instance, the centre of a flat surface area. Therefore, the amount of metal deposited at edges and corners is much higher than at the centre of planar surfaces (Fig. 7.2).

7.2 Fundamentals of Electrodeposition (Considering Nickel Deposition as Example)

Differences in the distance from the cathode to the anode, often found, e.g., in processing of unevenly shaped workpieces, has a direct effect on the current density in the observed area. The electrolyte acts as a resistor, thus, when doubling the distance between the electrodes and maintaining constant potential, the electrical current is halved. This effect, though undesired in practical coating processes, can be used in order to test electrolyte quality: A cathode is fixed at an angle to an anode so that a descending gradient of current density forms across the width of the sheet metal cathode. A single deposition process within the so-called Hull cell yields results that characterise the coating quality obtained with the electrolyte within a wide range of current density. For the specialist, results reveal possible over- or underdosage within the electrolyte, and allow correct adjustment of the electrolyte bath. Also, process parameters can be adjusted to match the modified electrolyte properties. 7.2.4 Electroless Plating of Nickel

Whereas the methods described above feature electrons for metal ion reduction that are applied by an external source of electrical energy, the electrons in electroless plating originate from within the system. The process is therefore also referred to as autocatalytic or chemical plating [6].

Fig. 7.5 Schematic comparison of deposition distribution of electroless deposited coatings (left) and electroplated coatings (right); taken from [5].

107

108

7 An Introduction to Electrodeposition and Electroless Plating Processes

As no electrical current is fed to the system, no lines of electric flux develop. Consequently, in contrast to the above methods, the coating thickness is evenly distributed and edge effects are not observed (Fig. 7.5). A chemical agent that releases electrons, and therefore allows reduction of metal ions in the solution, is added to the electrolyte. For nickel, use of hypophosphite is common: 3Na…H2 PO2 † ‡ 3H2 O ‡ Ni…SO†4 ! 3NaH2 PO3 ‡ Ni ‡ 2H2

…5†

Deposition is successful only due to the fact that the electrons are not transferred to nickel directly but via the metal surface of the coated workpiece (Fig. 7.6). Were this not the case, the two substances would react directly within the solution and produce a fine nickel powder. Part of the added hypophosphite decomposes to phosphorous acid and phosphorus in a side reaction: 3Na…H2 PO2 † ! NaH2 PO3 ‡ H2 O ‡ 2NaOH ‡ 2P

…6†

Reaction (6) reveals that also, in addition to elemental nickel, elemental phosphorus develops. The phosphorus is integrated into the coating. Phosphorus contents are in the range of 1 to approx. 14 per cent by mass, depending on pH, temperature, and electrolyte composition. As a result, autocatalytically deposited nickel coatings feature rather different properties compared to coatings produced by means of electrodeposition. At phosphorus contents of, for example, 14%, coatings are no longer metallic but show partially ceramic properties. The ferromagnetism of nickel thus disappears. Additional physical properties are also affected considerably. Vickers hardness (HV) of electroplated nickel without additives is approx. 260 HV, of electroplated nickel with organic additives approx. 400 to 500 HV, and electroless deposited nickel coatings reach hardness values of approx. 600 to 800 HV. Furthermore, phosphorus and nickel form a solid solution, and 1 hour of heat treatment at 400 8C causes nickel phosphide precipitation, producing even higher hardness values in the coating [5]. Using this, hardness values of up to 1100 HV are obtainable that correspond to the hardness of the coatings known as hard chrome (Fig. 7.7).

Fig. 7.6 Schematic representation of reaction mechanism in electroless nickel plating with hypophosphite layer. Electron transfer progresses from the hypophosphite layer to nickel via the coating substrate surface. Thus, nickel is also deposited onto the substrate and does not precipitate in the solution.

7.3 Overview of System Technologies

Fig. 7.7 Change in hardness of nickel coatings during heat treatment, taken from [5].

An additional benefit is that the phosphorus content reduces the melting point of nickel coatings. Nickel coatings with a phosphorus content of 11% melt at temperatures as low as 875 8C (melting point of pure nickel: 1453 8C), qualifying the coated material for brazing. This is used, for instance, in industrial production of heat exchangers.

7.3 Overview of System Technologies

This section introduces and describes differences of the three most important processes in industrial use: barrel plating, jig or rack plating, and continuous plating. The two more exotic technologies brush plating and tank plating will only be discussed briefly. 7.3.1 Barrel Plating

In barrel plating, the substrate bulk material is fed to a barrel. The barrel opening is closed with an appropriate cover. The walls of the barrel are perforated with holes or slots. The barrels are driven and rotated by toothed gearings, mounted laterally.

109

110

7 An Introduction to Electrodeposition and Electroless Plating Processes

Fig. 7.8 Small barrel for laboratory-scale electroplating. The left side shows the toothed actuation gearing. The electrical contact is visible on the inside.

The barrels are fixed to frames, and dipped into the different processing solutions, either by automatically driven units, or manually, with or without the aid of pulley blocks. Contacts on the barrel walls facing towards the inside, or insulated wires fed through openings at the side of the barrels provide electrical contact of the bulk material (Fig. 7.8). One third of the barrel volume should be filled with bulk material. The bulk material itself conducts the electricity within the barrel. Counter electrodes are usually mounted outside the barrel. The electric field lines penetrate the barrel through the holes and slots. Therefore, only the outer approx. 2 to 3 cm of bulk material are exposed to the electric field. Workpieces further inside the bulk material are shielded by the outside workpieces and therefore do not experience any electrical current. Even custom-built systems where anodes are mounted inside the barrels do not guarantee that electrical current flows through all workpieces simultaneously. At worst, this can lead to passivation of parts not exposed to the initial electrical current. For all parts to be coated evenly on average, and to guarantee exchange of electrolyte inside the barrel, it is rotated within the processing solution. Hence, all workpieces revolve continuously inside the barrel so that every single part changes position from top to bottom and is evenly exposed to the electric field. This movement, however, creates relative motion and impact between workpieces, and has a negative effect on surface quality. Therefore, barrel plating is only used to process workpieces that are either so small and lightweight that the impact occurring does not influence surface quality (e.g. buttons), or where

7.3 Overview of System Technologies

surface quality issues are less crucial (e.g. screws, electrical contacts). However, barrel plating does allow cost-efficient plating of bulk material. 7.3.2 Rack Plating

Rack plating is used for parts where the appearance of surfaces is critical, e.g. decorative strips, faucets, or shower heads. Here, workpieces are mounted on racks or jigs and dipped into the processing solutions. Electrical current is fed through the complete racks when they are immersed in the processing bath. In order to guarantee that only the workpieces and not the rack itself is coated, racks are resistant to chemicals and coated with non-conductive coatings. The only uncoated parts of the rack are contacts where the workpieces are mounted, in order to ensure proper flow of electrical power to the substrate workpieces (Fig. 7.9). In this process, racks and workpieces move within the electrolyte solution to guarantee sufficient electrolyte exchange around workpiece surfaces. Proper arrangement of substrate workpieces on the rack is crucial: An idealised process model suggests that the complete rack appears as a single surface that is coated according to the geometry of the developing electric field. Consequently, workpieces at the outside show higher coating thickness compared to those inside the workpiece jig. One possibility to work around this problem is to reduce the surface in the outside area, e.g. by mounting fewer parts here than inside the cluster. Although this reduces the observed difference in coating thickness, it does not compensate the effect completely.

Fig. 7.9 Pilot-plant electroplating system at IGOS GmbH with rack.

111

112

7 An Introduction to Electrodeposition and Electroless Plating Processes

The complete system is far more complicated. Workpieces are usually not components with a single planar surface but three-dimensional bodies. The surfaces of these bodies and the counter electrodes are not necessarily in a planeparallel configuration. Surfaces may be arranged perpendicular to, facing away from, or hidden from the counter electrode by other parts of the workpiece. Usually the only solutions for hidden surfaces involve either applying secondary electrodes in the immediate vicinity of the workpiece to purposely increase current density in certain regions, or shielding certain areas to selectively reduce current density. Naturally, both methods increase product prices considerably. Generally, it is advisable to consider which surfaces are relevant, or absolutely irrelevant, for the final product and then build racks accordingly that primarily yield optimal coating results on relevant surfaces. Ideally, it is advisable to consider coating processes in design and construction of the final product. Minor design changes can reduce severe problems in electroplating drastically, e.g. extending coverings by a few millimetres or avoiding pocket holes.

7.3.3 Continuous Plating

Strictly speaking, continuous plating is a special kind of rack plating. However, it is introduced here as an independent technology in brief due to certain special process characteristics. Rather than mounting the coating workpieces to a rack and coating these individually, continuous plating features a conveyer belt-type arrangement where the coating substrate is guided through the system and coated directly (Fig. 7.10). The belt passes through several subsequent tanks containing different processing solutions. Consequently, as substrate speed is constant in all tanks, the substrate material is exposed to every bath for the same period of time. The only means of realising the desired treatment time in different process steps is to adjust the length of the tanks or the number of successive tanks containing the same solution. Systems can feature more than 50 treatment tanks. In order to simplify substrate exchange, and to avoid having to feed a substrate through the whole system, new substrate is bonded or welded to the previous belt. After plating, the joint is cut out and disposed of. As the complete system layout is matched to a particular substrate and product, changing to a new product usually involves entirely rebuilding the system. An advantage of this technology is that very wide belts, and therefore, very large surfaces can be coated with constant quality. Plane flat belts allow very low cathode/anode distance, yielding relatively constant coating results across the width of the belt. Due to constant speed and treatment time in process steps along the complete length of the belt, coating results are well reproducible and constant.

7.3 Overview of System Technologies

Continuous plating allows very high throughput. This reduces coating prices considerably compared to plating parts individually. As coating thickness is evenly distributed, results are equivalent to single-piece treatment but with notably less coating material used. Appropriate system technology or belt guidance allows inexpensive partial coating, e.g. by dipping only half of the substrate into the treatment solutions. The process is used, for example, in large-scale gold plating of contact plugs. Contacts here, are punched from sheet metal, but not detached completely, allowing them to pass through the system in a belt-type arrangement. Suitable substrate materials are all types of belt-like, i.e. endless sheet, materials such as metal substrates, e.g. steel, copper, or aluminium, and plastic foil, e.g. polyaniline. A significant disadvantage of continuous plating, limiting its use in many applications, is that no end user is interested in a coated coil weighing possibly more than 10 tons with a length of more than 700 metres. The material is supplied only as fabricated material for further machining and fabrication, before it is sold to end users as part of toaster hoods, as hot plates for coffee machines, electrical contacts, beer cans, clock faces, or as an engine bonnet in an automobile. Material separation is necessary for further processing steps, when producing the final product. The resulting edges from cutting are naturally uncoated and therefore subject to possible corrosive attack, disqualifying pre-coated material for many applications.

Fig. 7.10 Continuous electroplating at Hille & Müller GmbH, Corus Special Strip, Düsseldorf, Germany.

113

114

7 An Introduction to Electrodeposition and Electroless Plating Processes

7.3.4 Brush Plating

Technically speaking, brush plating (also referred to as tampon plating) is a very important technology. However, it is uncommon and with much lower throughput than the processes described above, and therefore far less known. In brush plating, a pad of porous material is soaked with treatment solutions and pressed onto the surface to be coated. During actual coating, a counter electrode is mounted on the back side of the pad, then, electrical current is applied to the counter electrode and the coating substrate. When the solution in the pad is used up, the pad is refilled either by recirculating or by re-dipping the pad into treatment solution. The process is very elaborate and the number of coated pieces is limited. However, it is ideal for repairing machine parts that are large and difficult to transport because the electrolysis cells are brought to the substrate and not vice versa. 7.3.5 Tank Plating

The name is relatively unknown but the process is more and more popular for certain products. Apart from brush plating, in all technologies mentioned above, the substrate to be coated is passed from one electrolyte bath to the next, and treated accordingly. In tank plating, the substrate is placed in a tank, and the desired treatment solutions are successively pumped to the tank. The process is applicable for large-scale, production-integrated coating systems, e.g. hard-chrome plating of valves.

7.4 Overview of Individual Process Steps in Electroplating

Based on the deposition mechanisms and systems technology described above, this section discusses the overall process. Independent of the utilised technology, the sequence of treatments is comparable for defined substrates and coatings. 7.4.1 Degreasing

Usually, workpieces are pre-treated in some way, and therefore, surfaces are contaminated with lubricant grease, drawing compounds, rolling oil, or corrosion-protection oil, and require degreasing.

7.4 Overview of Individual Process Steps in Electroplating

For this, two processes are common in cases where no specialised process for a particular type of oil is used: degreasing using a hot alkaline solution, referred to as decoction, and degreasing using an alkaline solution assisted by electrical current, referred to as electrolytic degreasing. Frequently, both processes are used consecutively. In the decoction process, the oil detaches from the surface and dissolves at high temperature in a tenside solution, just as in everyday dish washing in a household kitchen. However, it usually does not involve mechanical means, i.e. scrubbing and brushing. This is where electrolytic degreasing helps: The electrical current decomposes water and oxygen or hydrogen is produced at the substrate that is charged as anode or cathode. The resulting gas evolution loosens contaminating particles mechanically from the substrate surface. Subsequently, they are enclosed by tensides in the solution and removed from the surface. 7.4.2 Activating or Pickling

The next step aims at removing possible oxides and inorganic components, still remaining at the surface. This is done by pickling, using an acid that matches the substrate material. Duration is crucial in this process, because, when process time is exceeded, the substrate surface is subject to corrosive attack. On the other hand, too short a treatment leaves part of the oxide layer on the surface, and therefore, coating adhesion to the substrate will be insufficient. So-called pickling inhibitors or restrainers reduce corrosive damage to the substrate material. These additives do not inhibit the effect of the pickling bath on the oxides, but rather passivate the metal surfaces and protect them from further corrosive attack. 7.4.3 Carryover

At this stage it is appropriate to briefly discuss the problem of carryover, which always represents a challenge for process control within the complete process chain: Typically, degreasing agents are alkaline, whereas activating agents are usually acid. When the workpiece designed for coating is transferred from the degreasing to the activating bath, a certain amount of residue from the previous process step remains at the surface. Part of the degreasing solution is thus carried over to the pickling bath. Most challenging, in this context, is coating of hollow objects that are closed at one end, e.g. cans, lids, or caps. Completely removing the treatment solutions from these objects is virtually impossible in practice. Thus, carryover of a subsequent treatment solution is inevitable. Therefore, electroplating system design always includes at least one rinsing bath between individual process steps. Ideally, rinsing liquid is used to compensate for loss of treatment solution volume

115

116

7 An Introduction to Electrodeposition and Electroless Plating Processes

due to evaporation in the previous bath, and thus, part of the lost active agent is fed back to the process step. 7.4.4 Coating Passivating Materials such as Stainless Steel and Aluminium

After activation, surfaces must be free of oxides in order to guarantee sufficient coating adhesion. In practice, however, materials are often coated that immediately form a new oxide layer when immersed in water, before reaching the next process step. Examples are aluminium, magnesium, and chromium-nickel stainless steels. Here, special treatment is necessary. A special nickel bath is used for coating stainless steel. The pH is extremely low, and the chloride concentration is high. The bath is referred to as Woods nickel strike solution. The chloride concentration is high enough to destroy the passivation layer on stainless steel. At spots where the passivation layer is removed, a thin nickel coating is deposited, serving as substrate for subsequent coating. Current efficiency for this electrolyte is only in the range of approx. 5%. The remaining 95% of electrical current are used to produce hydrogen. In order to reduce electrical power consumption, only a very thin layer of nickel strike is deposited, and subsequently coated with standard electrolytes. A similar technique is used for coating aluminium. Here, however, the aluminium oxide layer is corroded and destroyed by zincate pickling. A thin zinc layer cements instantaneously on the aluminium substrate material, protects it, and serves as a substrate for the following coating. 7.4.5 Summary of Pre-treatment

Pre-treatment involves cleaning and conditioning of substrate material, and prepares the surface for the actual subsequent deposition of electroplated coating material. When pre-treatment is insufficient, faulty coating is inevitable. The residue of non-conducting material at the surface prevents deposition, and therefore, leads to holes within the coating in the affected area. Although insufficiently removed oxide layers can conduct electrical energy under certain circumstances, coating adhesion is unsatisfactory. Errors in coating are often a result of insufficient pre-treatment. Particular attention should therefore be paid to this process step.

7.5 Microstructuring and Electroforming

Although both technologies have developed into independent branches, they feature similar basic process steps:

7.6 Summary

In electroforming, a base material, e.g. wax body covered with conductive silver, is charged and coated, for instance, with a gold plating. After this, the wax is melted out of the workpiece. The result is a geometry corresponding exactly to the original wax body. The technique is used, e.g., to produce gold jewellery. Alternatively, the resulting hollow body can be cut and opened in order to serve as a mould for producing further copies, absolutely identical to the original wax core. A completely different way of producing small structures is the so-called LIGA technology (an acronym of the German Lithographie, Galvanik, Abformung, meaning lithography, electroplating, and moulding). The technique involves electrically conductive substrate, coated with a photosensitive thin film (resist). The photosensitive resist is then exposed through a mask. Exposed areas of the film are dissolved in appropriate solvent, whereas the other areas are not affected. In the next step, copper or nickel is deposited on the stripped surfaces. Removing the remaining resist with a different solvent then produces a geometry that corresponds precisely to the image in the mask. Detaching the produced workpiece from the base material finally yields an electroformed, micron-scale workpiece.

7.6 Summary

The fundamentals of electroplating described in this chapter can only provide a first insight to basic techniques. Basically, all electroplating or galvanotechnical processes are similar. However, many characteristic features of particular processes cannot be covered in detail here, and study of specific technical literature is recommended. One particular aspect of electroplating, which will probably broaden possible fields of application for this coating technology in the near future, is the ability to add certain particles or substances to the electrolyte. These can either have a dramatic impact on appearance and surface feel of deposited coatings (pearlescent nickel, aluminium look), or completely change coating properties such as wear characteristics. Future development will show whether or not electroplating of aluminium and other less-noble metals (base metals) by deposition from aprotic, organic solvents will prevail against physical techniques. Basically, however, deposition of less-noble metal coatings, e.g. lithium, by means of electroplating is possible. Microstructuring is one feasible method of producing very small machine and part components economically at a time when the trend to miniaturisation is observed for nearly all produced parts. Therefore, electroplating today is still, after 150 years of development, a modern surface technology.

117

118

7 An Introduction to Electrodeposition and Electroless Plating Processes

References 1 2

3

4 5

Brugger, R.: Die galvanische Vernicklung, Leuze, Saulgau, 1984 Gaida, B., Assmann, K.: Technologie der Galvanotechnik, Leuze, Saulgau, 1st edn, 1996 Gebauer, K.: Über einige Entwicklungslinien der angewandten Galvanotechnik in Deutschland, in: Galvanotechnik 68 (1977), pp. 5–22 Jelinek, T. W.: Praktische Galvanotechnik, Leuze, Saulgau, 5th edn, 1997 Parkinson, R.: Properties and applications of elektroless nickel, Nickel Devel-

opment Institute, NiDI Techical Series No. 10 081, London, 1997 6 Riedel, W.: Funktionelle chemische Vernickelung, Leuze, Saulgau, 1989 7 Strauch, A., et al.: Galvanotechnisches Fachwissen, VEB Deutscher Verlag für Grundstoffindustrie, Leipzig, 3rd edn, 1990 8 Suchentrunk, R., et al.: Kunststoff Metallisierung – Handbuch für Theorie und Praxis, Leuze, Saulgau, 1st edn, 1991

119

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying Z. Babiak, T. Wenz, L. Engl, Institute for Materials Science, University of Hannover, Germany 8.1 Introduction

In 2000, results of an empirical investigation carried out at the Institut für Wissenschaftstransfer durch wissenschaftliche Weiterbildung (IfW, Institute for Science Transfer by Scientific Education) showed that thermal spraying is in fourth place of the top modern surface technologies, ranging behind electroplating, paint coating, and heat treatment for diffusion methods, and therefore in front of PVD, CVD, and build-up welding [1]. This ranking results from strong growth of thermal spraying technology in German small and medium-sized enterprises. One reason, certainly, is that thermal spraying is focused on the primary goals of surface technology being, according to the report, corrosion and wear protection. Results reveal that industry mainly uses flame spraying techniques (30%), followed by plasma and arc spraying (20% each), as well as highvelocity flame spraying (15%). The potential for thermal spraying to develop new markets is high, says the report, as companies rank the technology third in the priority of planned new processes.

8.2 Fundamentals of Thermal Spraying

Thermal spraying offers coating processes for parts with many different coating materials in order to yield wear and corrosion-resistant coatings as well as thermal-barrier coatings, or to produce the desired electrical or magnetic properties, etc. The basic principle and definition of the term thermal spraying is standardised in DIN EN 657 [2]. Spray material is fed to a source of heat as powder, wire, or rod, inside or outside of the spray apparatus. Here, it is melted superficially or completely, or heated until it is sufficiently soft. A gas stream accelerates the particles towards the substrate where they deposit as a coating Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

120

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Fig. 8.1 Principle of thermal spraying.

Fig. 8.2 Classification of thermal spray techniques according to the source of energy.

(Fig. 8.1). During coating, the substrate usually is subject to moderate thermal stress, but does not melt at the surface. Processes of thermal spraying can be classified by considering the type of heat source, according to DIN EN 657 (Fig. 8.2). Most common are flame spraying, high-velocity flame spraying, arc spraying, and plasma spraying. Cold gas spraying is the newest process, on the verge of being state-of-the-art. The most prominent advantages of thermal spraying are: · virtually any coating material can be used (metals, ceramics, cermets, plastics) · low thermal stress on substrate parts

8.2 Fundamentals of Thermal Spraying

· · · ·

local and reinforced coatings possible processes are available as field services nearly any substrate material can be coated high deposition rates (coating thickness from 20 lm to several millimetres)

8.2.1 Structure of Thermal Spray Coatings

Spraying conditions and the particular spray material predominantly influence the properties of spray coatings. Coating structure and configuration of sprayed coatings determine the characteristics of the coating/substrate system. Spray coatings grow when individual spray particles impact the substrate surface, deform or splatter, and solidify due to heat transfer into the workpiece (Fig. 8.3). Thermal spray coatings usually have lamellar and, depending on spray technique and material, more or less porous, micro-cracked, heterogeneous, and anisotropic structure. Furthermore, they contain partially molten particles or particles that solidified before impacting the surface or have reacted with gas from the atmosphere.

Fig. 8.3 Formation of thermal spray coating.

121

122

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Concentrated heat input to the particle as well as flight time through the surrounding atmosphere lead to the following reactions that influence composition and structure of spray coatings [3]: · selective evaporation of a component · reactions of metal compounds (e.g. decomposition of hard material in the presence of O2) · formation of non-volatile metal compounds such as oxides, nitrides, and hydrides in the presence of O2, N2, H2 (particularly for reactive metals). Superficial in-flight oxidation of metallic spray particles is particularly important. The formed oxides are jointly responsible for the lamellar spray coating structure. This can, in part, lead to increased coating hardness as well as wear resistance. Conducting the process in inert gas atmosphere, e.g. nitrogen or argon depending on spray material, particularly prevents formation of metal oxides, as described above. Careful surface pre-treatment determines the properties of the substrate surface, and is a key factor for coating adhesion. Substrates are pre-treated according to DIN EN 13507 that generally includes the three process steps pre-cleaning, blasting, and post-cleaning. Pre-cleaning focuses primarily on removing oil and grease as well as possible varnish residue from the surface. Blasting with corundum or chilled iron shot (also steel wire shot or silicon carbide) effectively activates, decontaminates, and roughens the surface to be coated in order to condition the surface for the described mechanisms of adhesion. Roughening the surface increases the free surface energy, vacancy concentration, dislocation density, and the occurrence of stacking faults, due to plastic deformation in the near-surface zone. Furthermore, the surface area increases and allows mechanical bond of the impacting spray particles [5]. Post-cleaning, e.g. by using ultrasonic cleaning in alcohol, removes residue of blasting processes from the surfaces [6, 7]. 8.2.2 Adhesion of Thermal Spray Coatings

The main quality characteristics of thermal spray coatings are bond strength (tensile adhesive strength) and cohesion, i.e. coating adhesion to the substrate on the one hand and bond strength within the coating on the other. The following phenomena determine coating bond strength and cohesion of thermal spray coatings: mechanical interlocking, adhesion, utilising surface energy, diffusion, and electrostatic forces [8–10], as well as adiabatic shear band formation in modern high-velocity technologies where solid particles impact the substrate, e.g. in cold gas spraying [12]. These mechanisms generally superimpose each other. The weight of individual factors strongly depends on particle properties, mainly particle velocity and temperature, as well as substrate characteristics such as material, roughness, temperature, etc.

8.3 Flame Spraying Table 8.1 Standard values for adhesion (in MPa) of thermal sprayed coatings with respect to spray technique [11]. Processes

Spray materials Ferrous metals Non-ferrous metals Self fluxing alloys Ceramics Cermets

Wire flame spraying (WFS)

Powder flame spraying (PFS)

Arc spraying (AS)

Plasma spraying (PS)

High-velocity oxygen fuel flame spraying (HVOF)

14 21 – – –

28 21 > 69 14–34 34–48

41 > 41 – – –

> 34 > 34 – > 21 55–69

62 70 62 – > 83

Values for bond strength of thermal coatings, therefore, are generally given as guidelines. The actual bond strength depends on the particular circumstances in the coating process. Table 8.1 gives an overview of commonly obtainable minimum bond strength values for a number of coating technologies and materials [11]. The bond strength of a coating is measured using a tensile adhesive strength test according to DIN EN 582. This test method determines the bond strength of thermal spray coatings under tensile stress perpendicular to the interface surface. Traction-adhesive strength tests reveal the influence of substrate material, coating material, pre-treatment of workpiece surface, and coating conditions on tensile adhesive strength and, furthermore, can control metal spray work [8, 14, 15].

8.3 Flame Spraying

Schoop is said to have invented metal spraying. In 1912, while working with Herkenrath, he produced adhesive metal coatings using a spray gun [16]. 8.3.1 Flame Spraying Process

Flame spraying processes use a coating material that melts in a flame of fuel gas and oxygen. Process variants are a result of different spray materials used. In wire flame spraying (WFS), a wire is fed using a controllable wire feeder, and melted in a surrounding, coaxial fuel gas/oxygen flame. Expanding com-

123

124

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Fig. 8.4 Principle of wire and rod flame spraying.

bustion gas and additionally added compressed air accelerate the coating material, i.e. molten particles, towards the pre-treated substrate surface (Fig. 8.4). The most common fuel gas is acetylene but propane and hydrogen are also frequently used. The temperature inside an acetylene/oxygen flame reaches 3160 8C, compared to 2850 8C for a propane/oxygen flame, and 2660 8C for a hydrogen/oxygen flame. Substrate temperature remains relatively low as the flame envelope does not reach the substrate when the spray distance (from nozzle tip to workpiece) is in the standard range of 100 to 200 mm. Additional cooling allows controllable substrate temperatures from room temperature to 250 8C. Spray materials for wire flame spraying are produced as solid wire, cored wire, or rope [13, 17]. Rod flame spraying uses the same fundamental principle as wire flame spraying (Fig. 8.4). Here, the spray material is rod shaped and often ceramic. This technique, also, has the advantage that the spray material melts completely. Particles that leave the end of the rod are liquid and usually not superheated. Powder flame spraying is a coating technique where the spray material is in powder form. It melts in a fuel gas/oxygen flame and is carried to the substrate surface by the fuel gas and additional spray gas (Fig. 8.5). This technique is the most important flame spraying method. Spray material in powder form allows multi-

Fig. 8.5 Principle of powder flame spraying.

8.3 Flame Spraying

purpose and flexible processes. The most common application for powder flame spraying is coating with self-fluxing alloys that are sprayed and heated to coalescence, either in one single or two subsequent process steps. Torch development for wire flame spraying is also heading towards higher gas and particle velocities. In so-called high-velocity wire flame spraying guns, e.g. Metatherm W 1000 or Praxair 216 type, particle velocity exceeds 250 m/s. As their basic design is completely different from HVOF guns, performance of the two cannot be compared. 8.3.2 Materials and Applications

Spray materials for flame spray processes are used in the form of solid wire, cored wire, flex wire, rods, or powder. In wire flame spraying, deposition of the following materials is possible: unalloyed, low-alloy, and high-alloy steel, aluminium and Al alloys, copper and Cu alloys, tin and Zn alloys, lead and Pb alloys, nickel and Ni alloys, zinc, and molybdenum. Wire-shaped spray materials have the advantage that they are easy to handle. However, the selection of materials available as wires is limited as certain materials are too brittle. Introduction of cored wires allowed processing of additional materials in flame spraying, e.g. hard materials. In cored wires, appropriate powder is packed into a metal sheath. Two types of cored wires are common: a tubular wire, produced by drawing a metal tube, and seamed wire, produced by rolling a sheet metal band (Fig. 8.6). For spray processes, tubular wires have several advantages with respect to handling issues due to their higher mechanical stability. Examples of spray material combinations (sheath/core) for cored wire spraying are: metal/ metal, alloys (Ni/NiCrB), intermetallic compounds (Al/Ti), metal/carbide (Co/ WC), metal/boride (Al/B4C), metal/oxide (Al/Al2O3), metal/oxide/carbide (Al/ Al2O3/Cr3C2), and metal/short fibres (Al/C short fibres) [18, 19]. Flex wires and cored wires available for flame spraying have similar potential. Flex wires have a plastic tube that burns during spraying and contains the spray powder, bound with cement [13].

Fig. 8.6 Cross sections of cored wires: (a) NiCrBSi seamed wire, (b) NiAl tubular wire.

125

126

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Powder flame spraying allows processing of pure metals (Al, Cu, Ni, Zn, . . .), alloys (steel, NiCr, SnAl, . . .), oxides (Al2O3, Cr2O3, Al2O3+TiO2, . . .), as well as coated, mixed, and agglomerated carbides (WC/Co, Cr3C2/NiCr, TiC/Ni, . . .), and mixed powders (NiCrBSi+WC/Co, . . .) [20, 23]. Examples of materials for rod flame spraying [13]: Alloys: NiCr, NiCrBSi, CoCrB Oxides: Al2O3, Al2O3+TiO2, Cr2O3, ZrO220CaO, ZrO27Y2O3 Carbides: WC-Co, WC-NiCrBSi Diameters of rods used in rod flame spraying range from 4.8 mm (3/16'') to 7.9 mm (5/16''). The main goals when applying flame sprayed coatings to new parts as well as for repairs are corrosion protection, wear protection, thermal insulation, special applications, e.g. bond coats, coatings for medical implants, decorative coatings, compound coatings, X-ray shielding coatings, and casting moulds [24–26]. Flame sprayed zinc or aluminium coatings can provide corrosion protection for steel and cast iron materials. In the pH range of 7 to 12.5, zinc forms a stable oxide layer, making it relatively resistant. In corrosive environments with pH < 7, e.g. acid waters, zinc dissolves and forms salts. When the pH is greater than 12.5 (alkaline solutions), zincates are formed [24]. Aluminium also forms a stable oxidic covering layer. It is therefore comparatively stable in the pH range from 5 to 8.5. In corrosive media with pH < 5 (non-oxidising acids) aluminium dissolves and forms salts, and in solutions with pH greater than 9, aluminates are formed [24]. Figure 8.7 shows cross sections of flame sprayed corrosion protection coatings: a zinc coating (a) and an aluminium coating (b). The main flame spraying applications for corrosion protection are bridge, metal, and offshore constructions. For example, the alloy AlMg5 is used for coating

Fig. 8.7 Cross sections of flame-sprayed protective coatings: (A) zinc coating (corrosion protection), (B) aluminium coating (corrosion protection), (C) molybdenum coating (wear protection).

8.4 Arc Spraying Table 8.2 Comparison of service life/costs of Zn-sprayed coatings and paint coatings for corrosion protection. No.

Coating

Average service life in years

Average costs in dollars per m2

Costs in dollars per m2 year

1 2 3

metal spraying paint coat 1 paint coat 2

27.5 8.9 6.8

47.9 31.1 27.6

1.74 3.49 4.06

oil rigs. Although metal spray coatings are more expensive than paint coatings their increased service life makes them more profitable (Table 8.2) [25]. Wire flame spraying started out with wear-resistant coatings made of molybdenum (Fig. 8.7 c), carbon steel, and alloyed steel. Development of cored wires and powder flame spraying processes allowed using hard materials as well as metal-ceramic spray materials and, thus, broadened the fields of applications for the technology.

8.4 Arc Spraying

In 1915, the Swiss M. U. Schoop was the first to describe arc spraying [27]. Initial problems regarding instabilities of the process were solved by developing new electrical energy sources [28]. Industrial applications, however, were only possible after scientific research focused on fundamentals of this type of thermal spray process in the 1950s and 1960s [8, 29]. Arc spraying is very economical and therefore, has prevailed in many applications in spite of newly developed high-energy techniques such as plasma, detonation gun, and high-velocity flame spraying. Fields of application are broad due to system variety and mobility, further developments in systems design, and extended range of appropriate spray materials [30, 35–37, 46]. 8.4.1 Arc Spraying Process

In arc spraying, two wires of the same or different material (pseudo-alloy) are fed to the spray gun. The wire feed rate is controlled. Electrical energy usually is transmitted via copper contacting wire feeds, arranged at a 308 angle. When the wire feeds are running, spray wires move towards each other until they touch (Fig. 8.8). This spot shows the lowest ohmic resistance of all partial resistances within the electrical circuit. The high short-circuit current produces intense heat that evaporates the metal and strikes the arc. Voltage in the electrical source usually drops to lower values while the current rises, reaching short-circuit current for a short period of time.

127

128

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Fig. 8.8 Principle of arc spraying.

Temperatures within the arc rise to 6500 8C. The material melts off and particles are atomised, accelerated, and deposited at the surface intended for coating, by gas leaving the nozzle (usually compressed air). The process can be executed similarly using three or four wires, then referred to as multi-wire arc spraying [38]. High temperature at the wire ends can cause a burning loss of alloying elements in the spray material. For elements such as silicon or manganese, loss can rise to 40%, for other elements, e.g. carbon, even up to 60% [48]. In conventional arc spraying, DC generators with constant voltage characteristics guarantee continuous burning of the arc, similar to welding processes, and allow stable processes without having to control wire feed rates [28]. Inside the electrical current source, a so-called inner control loop regulates the process (Fig. 8.9). Disturbance and faults, e.g. discontinuous wire feeding, cause minor deviations from the desired operating point (A) in terms of arc voltage and length, due to the static characteristic of the power supply. With increasing wire tip distance, arc voltage rises (condition 1) which causes electrical current to

Fig. 8.9 Inner control loop in arc spraying.

8.4 Arc Spraying

drop to I1. The rate of melt-off is lowered and the wire tips again move towards each other until the distance for the desired operating point is reached, while the electrical current rises continuously. The opposite occurs in the second condition. Wire distance and arc length are reduced, voltage rises, and electrical current increases to I2. This leads to an increase in melt-off and readjustment towards the desired operating point. The arc itself is subject to many faults and interference during spraying. Compared to a welding arc, the thermal spraying arc is a discontinuously burning DC arc. Interference and disturbance, leading to non-uniform burning and permanent readjusting, are caused by the flow of compressed air, elastic wire effects (stress relaxation), as well as wire melt-off and resulting spray particles. When these leave the wire, the arc root area, arc diameter, and arc length on the wire tips decrease. Voltages are in the range of 18 to 40 V, depending on spray material. Spray materials should always be processed using the lowest possible voltage (see above). The current can be adjusted between approx. 50 and 150 A for routine applications, in special systems up to 2000 A, determined by the desired deposition rates, material, spray wire diameter, and available arc spraying systems. Drawing or pushing drives are used for wire feed. Combinations (push/pull systems) yield reduced influence of wire geometry when the wire is clamped (squeezing or slipping), particularly for spraying soft or cored wires. Process interference is thus minimised. Apart from energy and wire supply as well as carefully adjusted spray parameters, the deployed nozzle system has a considerable effect on atomising and acceleration of molten material. A number of nozzle concepts focus on high particle atomisation (e.g. OSU LD/U type) on the one hand and/or very fast and nearly laminar gas flow (e.g. Sulzer Metco SmartArc type) on the other. These nozzle designs allow narrow particle-size distribution and/or high particle velocities for good coating quality. Using a secondary gas flow, e.g. in Tafa 9000 TWEA, allows higher gas and particle velocities, and the spray jet is protected from environment gases, respectively [39]. Virtually all modern arc-spray systems include interchangeable nozzles for appropriate spray-jet modification. Obtainable velocities are in the region of the speed of sound. Particle speeds for well-bonding particle sizes at standard spray distances were determined to lie between approx. 50 and 150 m/s [29, 40, 41]. The development of closed atomising gas systems optimised atomising results considerably. A conversion kit allows spray guns to be simply modified from an open to a closed nozzle system (Fig. 8.10). The difference in the technical design of the two nozzle systems is that the closed system features an additional nozzle disk (1) on top of the wire electrodes in front of which a radial nozzle (2) is mounted. The distance (A2) between the radial nozzle and the nozzle disk determines the amount of additional radial atomising gas flow, which yields finer spray particles and a spray particle jet that is easier to focus [42]. Primarily, this leads to better atomisation of the molten metal and, thus, spray coating structure is more even, but also includes more oxides [43]. In arc spray-

129

130

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

Fig. 8.10 Open (top) and closed (bottom) nozzle system in arc spraying guns, OSU, LD/U2 type (cross sections top view, front views).

ing, the metallic spray particles oxidise exothermally on their way to the substrate [44]. The particle formation and kinetic behaviour of spray particles during inflight period depend on a large number of specific properties of the atomising gas and the metal spray material. The main influences are process parameters such as the spray system, nozzle system, atomising gas pressure p, voltage U, electrical current I, and spray distance s. Particle sizes vary considerably between 2 and 200 lm [29]. OSU introduced a newly developed system for arc technology, the VISU ARC 350 spray system. The arc-spray system is equipped with a new type of electronic power supply that includes a control loop delivering more stable spray voltage compared to conventional transformer/rectifier/impedance systems used previously. It provides a more stable arc formation and better spray-jet focus. One disadvantage of arc spraying is that spray materials are limited to electrically conductive wires. Furthermore, particular operational safety precautions are necessary due to the produced dust, high sound emissions (110 dB), as well as high ultraviolet and infrared radiation [45]. Technical advantages of arc spraying are high deposition rates, coating thickness of up to several millimetres, relatively simple operation and, compared to other spray processes except for wire and powder flame spraying, low systems costs (initial and operational costs). An additional advantage in terms of safety is that non-combustible gas is used. Finally, arc spraying is the most economical

8.4 Arc Spraying

thermal spray process with respect to energy efficiency in melting of spray material [26, 48]. 8.4.2 Special Arc Spraying Processes

In addition to many industrial applications, arc spraying with its relatively simple process technology has a potential for further development aimed at improved coating quality. A number of special arc spraying process variants are available. The most important group are processes conducted within chambers, i.e. under controlled atmosphere. Single-wire arc spraying is an additional process variant. Processes under controlled atmosphere are protective gas arc spraying (SAS, shrouded arc spraying), operating nearly at atmospheric pressure, and vacuum arc spraying (VAS), conducted under low-pressure conditions. Protective gas arc spraying yields better coating quality, particularly with respect to spray material oxidation. This method allows processing of reactive materials in arc spraying [31, 32] and at the same time offers simple process handling due to nearly atmospheric pressure conditions. VAS processes further reduce the influence of possible contamination in process gas. Higher coating purity is achieved, but at the expense of reduced coating density. This is caused by the different energy density of the arc at low pressure, i.e. the energy provided for melting by the arc is at lower levels under low-pressure chamber conditions than under higher chamber pressure. Higher chamber pressure yields increased charge carrier transition per unit time, i.e. the electrical resistance of the arc is reduced and the rate of melt-off increases in spite of constant voltage and wire feed rate. Then, particle temperature rises under higher pressure yielding lower coating porosity. Furthermore, strongly deviating melt-off rates of the two spray wires, depending on chamber pressure and material, must be considered and compensated by using appropriate independent wire-feed drives. Gas consumption is five times lower in single-wire vacuum arc spraying (SWVAS) compared to two-wire processes. Very fine atomisation of the spray material, here, is due to symmetric gas flow and the centred wire position, when the wire is charged as the anode [33, 34]. 8.4.3 Materials and Applications

Classic materials for arc spraying are pyro- or powder metallurgically produced solid wires made of carbon steel, chromium and high-alloy steel, bronze for bearings, and non-ferrous metals such as aluminium, nickel-base alloys, or zinc. Typical process applications include corrosion protection of offshore and steel constructions, wear-resistant coatings (e.g. on roller axles, mould surfaces in mould production), as well as repair of wear damages and metallisation of plas-

131

132

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

tic housings. Deposition rates for aluminium reach 8 kg/h, for zinc up to 40 kg/h. Further examples of the large number of arc-sprayed coating applications are deposition of adhesive layers (NiCr, NiAl, etc.) for subsequently deposited ceramic coatings as well as coatings on automotive parts. Sophisticated systems technology and reproducible (usually automated) processing allows production of arc-sprayed metal coatings complying with quality standards that, so far, were produced via plasma spraying. The following metals and alloys can be processed in arc spraying: · Al, AlMg for repairs on Al parts, silencers (mass production), heat exchangers, offshore technology: pipelines and oil rigs, boilers, slab markings, and also food processing industry (butcher’s machines) · Cu resistors, capacitors, conductor paths on plastic and ceramic substrates, continuous-casting moulds · CuAl roller-bearing seats, fittings · Mo adhesive layers on steel, wear-protection coatings · NiCr, NiAl, NiCrMo radiation shielding (X-rays), adhesive layers for plasma spray coatings, oxidation barrier layer below plasma-sprayed ceramic coatings · NiCuMn valve housings, fitting parts, pump shafts · Sn alloys friction bearings, capacitors · Steel wear resistant and repair coatings: plunger, transmission cases, aluminium parts (automobile valves) · Zn bridges and steel constructions, boilers, gates, pipelines, casting moulds, carbody joints. In addition to the materials listed above, cored wires also allow hard materials (oxides, nitrides, carbides, borides, carbonitrides, and carboxynitrides) to be processed in arc-spraying technology. The multitude of possible combinations provides large potential for future material development and new applications. Already, a number of nickel- and iron-based cored wires are used to protect parts subject to complex load conditions in chemical industry and power-plant environments, e.g. chemical attack, thermal stress, and erosion [46]. Atomising gas pressure is one of the key factors influencing particle size and resulting coating structure. Increasing atomising gas pressure produces smaller and faster particles that form a more homogeneous coating [47]. Figure 8.11 shows possibilities of influencing arc-sprayed coating structure by appropriate parameter selection.

8.4 Arc Spraying

Fig. 8.11 Cross sections of arc-sprayed steel coatings (X46Cr13): (a) OSU-LD/U2, open nozzle system (3 bar, 300 A, 30 V): coarse particle lamella, low oxide content, (b) OSU-LD/U2, closed nozzle system (4.5 bar, 200 A, 25 V): fine particle lamella, high oxide content.

Fig. 8.12 Cross sections of arc-sprayed coatings: (a) steel (1% C steel), etched (Miller/Praxair BP 400, HV nozzle), (b) zinc (Sulzer Metco Smart Arc, HV nozzle).

Newer system-design developments focus on maximising the velocity of gas flow in order to improve coating structure. So-called high-velocity nozzles allow deposition of very dense arc-sprayed coatings (Fig. 8.12). Due to remarkable economical benefits, arc spraying has prevailed in many applications of thermal spray coatings, in spite of newly developed high-energy processes. Modern arc-spray systems produce high-quality coatings that, in part, allow substitution of expensive plasma-sprayed coatings [49].

133

134

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

8.5 Summary and Conclusions

Thermal spray processes are important surface technologies and will continue to develop their share of the market. The main reasons for the increasing significance and the impact on consumer product industry of this technology are: ongoing, continuous development, assisted by diagnostics and simulations, automation and process control, as well as access to new technologies such as cold gas spraying or nano-technology [50]. Flame and arc spraying are amongst the oldest thermal spray technologies. Although competition from new spray processes, e.g. plasma or high-velocity flame spraying, is high, versatility and flexibility guarantee a large number of applications. New process design and matched material developments open new fields of application and make flame and arc spraying attractive.

References 1 Dietz, T., Eritt, H. J., Novello-von

2

3

4

5

6

7

Bescherer, W.: Anwendungsfelder moderner Oberflächentechnik bei Anbietern oberflächentechnologischer Leistungen. Wissenstransfer Oberflächentechnik Series, Vol. 8. VDI, Düsseldorf, 2000. N. N.: EN 657: Thermisches Spritzen, Beuth, Berlin, 1994-06 (under revision: prEN 657: 2003-04). Thermal spraying – Terminology, classification Haefer, R. A.: Oberflächen- und Dünnschichttechnologien. Springer, Berlin, 1987 Herman, H.: Plasmagespritzte Beschichtungen. Spektrum der Wissenschaft, 11, 1988 Drozak, J.: Teilcheneigenschaften und Haftung beim thermischen Spritzen von Metall und Keramik. Doctoral Thesis, University of Dortmund, 1992 N. N.: EN 13507: Thermisches Spritzen – Vorbehandlung von Oberflächen metallischer Werkstücke und Bauteile für das thermische Spritzen, 2001-07. Thermal spraying – Pre-treatment of surfaces of metallic parts and components for thermal spraying N. N.: DVS Merkblatt/Instructions 23071: Arbeitsschutz beim Entfetten und Strahlen von Oberflächen zum thermischen Spritzen. DVS, Düsseldorf, 1999-01

8 Steffens, H.-D.: Haftung und Schichtauf-

9

10

11

12

13

bau beim Lichtbogen- und Flammspritzen. Doctoral Thesis, Technical University of Hannover, 1963 Steffens, H.-D., Dammer, R., Fischer, U.: Einfluß der Oberflächenvorbereitung auf die Haftzugfestigkeit. Oberflächentechnik, 3rd SURTEC Congress in co-operation with INTERFINISH Europe 1985, Eds.: H. Czichos, et al., VDE, Berlin, 1985, pp. 75–82 Sepold, D.: Die Haftung von Schichten, DGM Symposium, Bad Nauheim, 1977, pp. 1 et seqq. N. N.: Guide to Engineered Materials. In: Advanced Materials & Processes, 159 (12), Dec 2001 Halter, K., Sickinger, A., Zysset, L., Siegmann, St.: Low Pressure Wire Arc and Vacuum Plasma Spraying of NiTi Shape Memory Alloys. Thermal Spray 2003: Advancing the Science and Applying the Technology, Eds. B. R. Marple, C. Moreau. ASM International, Materials Park, OH, USA, 2003 N. N.: DIN EN 14919: Thermisches Spritzen – Drähte, Stäbe und Schnüre zum Flammspritzen und Lichtbogenspritzen – Einteilung – Technische Lieferbedingungen. Thermal spraying – Wires, rods and cords for flame and arc

References

14

15

16

17

18

19

20

21

22

23

24

25

spraying – Classification; Technical supply conditions. DVS, Düsseldorf, 2001-10 Smolka, K.: Thermisches Spritzen, Ein Leitfaden für den Praktiker, DVS, Düsseldorf, 1985 N. N.: DIN EN 582: Ermittlung der Haftzugfestigkeit. Thermal spraying; determination of tensile adhesive strength. Beuth, Berlin, 1994-01 Schoop, M. U., Daeschle, C. H.: Handbuch der Metallspritztechnik. Rascher Cie A. G., Zürich, Leipzig, Stuttgart, Vienna, 1935 N. N.: DVS Merkblatt/Instructions 2301: Thermische Spritzverfahren für metallische und nichtmetallische Werkstoffe, DVS, Düsseldorf, 2001-06 Steffens, H.-D., Babiak, Z., Kaczmarek, R.: Thermisches Spritzen – Verfahren, Anwendungen, Tendenzen. DVS Report 130, Thermische Spritzkonferenz TS ’90, Essen, 1990 Borisov, Y., Voropaj, N., Netesa, I., Korzhyk, V., Kozjakov, I.: Composite Flux Cored Wires for Thermal Spraying. Thermal Spray 1995: Science & Technology, Eds. C. C. Berndt, S. Sampath. ASM International, Materials Park, OH, USA, 1995 N.N.: EN 1274: Thermisches Spritzen: Pulver – Zusammensetzung, Technische Lieferbedingungen. Thermal spraying – Powders – Composition, technical supply conditions. Beuth, Berlin, 1996-08 (under revision: prEN 1274:2003-05) Kulik, A., Porison, J. J., Mnuchin, A. S., Nikitin, M. L.: Thermal Spraying with Composite Powders. Maschinostrojenije, Leningrad, 1985 (in Russian) Borisov, Y. S., Charlamov, J. A., Sidorenko, S. L., Ardatovskaja, E. N.: Thermal Spray Coatings from Powder-Shaped Materials. Handbook, Naukovaja Dumka (in Russian), Kijev 1987 Kretschmar, E.: Metall-, Keramik- und Plastspritzen. VEB Verlag Technik, Berlin 1970 N. N.: DVS Merkblatt/Instructions 2302: Korrosionsschutz von Stählen und Gusseisenwerkstoffen durch thermisch gespritzte Überzüge. DVS, Düsseldorf, 2003-11 Unger, R. H.: Thermal Spraying of Bridges. Thermal Spray 1987: Advances

26

27

28

29

30

31

32

33

34

35

in Coatings Technology, Ed. D. L. Houck. ASM International, Materials Park, OH, USA, 1987 N. N.: Thermal Spraying – Practice, Theory and Application. American Welding Society (AWS), 1985 Schoop, M. U.: Method of Plating or Coating with Metallic Coatings. US Patent 11228059, February 9th, 1915 Wamser, W.: Der Einfluß der Stromquellencharakteristik auf das Lichtbogenmetallspritzen. Schweißtechnik, 13 (1963) 9, pp. 405–408 Busse, K. H.: Das Verhalten von Spritzteilchen beim atmosphärischen Lichtbogenspritzen. Doctoral Thesis, Dortmund, 1989 Hock, V. F., Benary, R., Ganertz, R., Herman, H.: Automated Thermal Spray Technology for Rehabilitation and Maintenance of Civil Works Infrastructure. In Thermal Spray 1997: A United Forum for Scientific and Technological Advances, Ed. C. C. Berndt. ASM International, Materials Park, OH, USA, 1997 Chen, H. C., Duan, Z., Heberlein, J., Pfender E.: Influence of Shroud Gas Flow and Swirl Magnitude on Arc Jet Stability and Coating Quality in Plasma Spray. Thermal Spray: Practical Solutions for Engineering Problems, Ed. C. C. Berndt, ASM International, Materials Park, OH, USA, 1996, pp. 553 et seqq. Bach, Fr.-W., Babiak, Z., Tegeder, G.: Arc Sprayed Ti-N-Coatings for Protection against Corrosion and Wear. Thermal Spray 2001: New Surfaces For A New Millennium, Eds. C. C. Berndt, K. A. Khor, E. Lugscheider. ASM International, Materials Park, OH, USA, 2001, pp. 1179–1184 Steffens, H. D., Wewel, M., Nassenstein, K.: Ein neues Spritzverfahren: EindrahtVakuum-Lichtbogenspritzen. DVS Report, Vol. 152, TS ’93, pp. 234–237 Carlson, R. R., Heberlein, J. V. R., Hussary, N., Shi, K.: High Definition SingleWire-Arc-Spray. Thermal Spray 2000: Surface Engineering via Applied Research, Ed. C. C. Berndt. ASM International, Materials Park, OH, USA, 2000, pp. 709–716 Marantz, D. R.: State of the Arc Spray Technology. Thermal Spray 1990:

135

136

8 Fundamentals of Thermal Spraying, Flame and Arc Spraying

36

37

38

39

40

41

42

Research and Applications, Ed. T. F. Bernecki. ASM International, Materials Park, OH, USA, 1991, pp. 113–118 Grant, L. J.: How Recent Advances in Arc Spraying Broaden the Range of Applications, International Thermal Spray Conference, UK, 1989, Paper 111, pp. 1–17 Steffens, H. D., Wilden, J., Duda, T.: Thermal Spraying. High Temp Chem Processes, 3, 1994, pp. 653–664 Dietzschold, D., Hellwig, J., Weichbrodt, K. H.: Mehrdraht-Lichtbogenspritzen von Verbundwerkstoffen. DVS Report, Vol. 152, Thermische Spritzkonferenz TS ’93, 1993, pp. 93–96 Varacalle, D. J., et al.: An SDE Study of Twin Wire Electric Arc Sprayed Nickel Aluminium Coatings. Thermal Spray Science & Technology, Eds. C. C. Berndt, S. Sampath. ASM International, Materials Park, OH, USA, 1995, pp. 373–386 Wang, X., Heberlein, J., Pfender, E., Gerberich, W.: Effect of Gas Velocity and Particle Velocity on Coating Adhesion in Wire Arc Spraying. Thermal Spray 1996: Practical Solutions for Engineering Problems, Ed. C. C. Berndt. ASM International, Materials Park, OH, USA, 1996, pp. 807–811 Bach, Fr.-W., Duda, T., Babiak, Z., Tegeder, G.: Hochgeschwindigkeitslichtbogenspritzen. Conference Proceedings Thermal Spray Conference, Eds. E. Lugscheider, P. A. Kammer. DVS Deutscher Verband für Schweißen, Germany, 1999, pp. 240–241 Schmidt, H., Matthäus, D.: Stage of Development of the Arc Metal Spraying Systems – New Experiences and Data. Conference Proceedings Thermal Spray Conference, Netherlands, 1980, pp. 225– 231

43 Brennek, J., Milewski, W.: The Effect of

44

45

46

47

48

49

50

So-called Closed Atomising Chamber on the Structure and Quality of Sprayed Coatings. Conference Proceedings, Thermal Spray Conference, Netherlands, 1980, pp. 239–243 Steffens, H.-D.: Metallurgische Vorgänge beim Verspritzen von Stählen an der Luft und unter Schutzgas. Professorial Thesis, Hanover, 1967 N. N.: DVS Merkblatt/Instructions 23073: Arbeitsschutz beim Lichtbogenspritzen. DVS, Düsseldorf, 1996-10 Sampson, E. R.: Cored Wire Application using Electric Arc Spray. In Thermal Spray 1998: Meeting the Challenges of the 21st Century, Ed. C. Coddet. ASM International, Materials Park, OH, USA, 1998, pp. 133–137 Steffens, H.-D., Dvorak, M., Wewel, M.: Einfluß der Prozeßparameter beim Lichtbogenspritzen – Ein Leitfaden für den Praktiker. DVS Report, Vol. 130, TS ’90, Düsseldorf, pp. 23–26 Lugscheider, E. Ed.: Handbuch der thermischen Spritztechnik; Technologien – Werkstoffe – Fertigung. Fachbuchreihe Schweißtechnik: 139. DVS, Düsseldorf, 2002 Sampson, E. R., Sahoo, P.: New Arc Wire Approvals for Aircraft Power Plant Overhaul. Thermal Spray 2000: Surface Engineering via Applied Research, Ed. C. C. Berndt. ASM International, Materials Park, OH, USA, 2,000, pp. 717–720 Fauchais, P., Vardelle, A., Dussoubs, B.: Quo vadis Thermal Spraying? In Thermal Spray 2003: Advancing the Science and Applying the Technology, Eds. B. R. Marple, C. Moreau. ASM International, Materials Park, OH, USA, 2003

137

9 Spray Materials J. Beczkowiak, H. C. Starck GmbH, Laufenburg, Germany

9.1 Introduction

Within the scope of surface technologies, thermal spraying is one of the coating technologies that is highly adaptable to the demands of individual parts requiring protective coating. This property is mainly due to the fact that thermal spraying, particularly during the last three decades, has met the challenges of modern technology with respect to deployed spray techniques and processable materials. Innovations have opened up sophisticated branches of technology. The development produced a large variety of materials that require professional selection in order to solve coating problems by means of optimal material technology. This chapter, therefore, will assist the selection process by focusing on materials, production process, and application descriptions.

9.2 Spray Material Properties Determined by Production Issues

Generally, the following forms of materials for industrial thermal spray applications are distinguished: wires or rod-shaped materials and powder materials. Wire and powder materials show distinct advantages and disadvantages for materials selection and, therefore, for the user also (Fig. 9.1). The main advantage of wires, compared to powders, is simple handling, but material variety is limited. Ultimately, choosing a form of coating material is determined not only by material supply but also by availability of appropriate spray systems and required coating properties. For powders particularly, processing characteristics are determined by powder-production processes.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

138

9 Spray Materials

Fig. 9.1 Spray material shapes, advantages and disadvantages.

The following considerations shall therefore focus on the most common production processes. 9.2.1 Powder-production Processes

The group of powder-production processes can be subdivided into the following processes (cf. Fig. 9.2): Figure 9.3 schematically illustrates the atomising process in a protective gas atomising system. It is notable that, here, melting of raw material can occur under protective gas or in air, depending on the desired oxygen content of the produced spray powder. Additionally, either gas (argon, nitrogen, etc.) or water can be used to atomise the molten material, which has an influence on the gas content as well as grain geometry of the powder.

abradables, TBCs, metal carbides, oxides

Fig. 9.2 Powder production (Praxair Services).

9.2 Spray Material Properties Determined by Production Issues

Fig. 9.3 Atomising.

Fig. 9.4 Properties of atomised spray powders (Praxair Services).

Figure 9.4 shows properties of atomised powders depending on the deployed atomising process, with respect to handling properties, and lists material examples. The distribution of powder sizes determines flowability and melting behaviour. Sintering (Figs. 9.5 and 9.6) is used primarily for production of composite materials made from carbides and metals, alloys, or oxides. However, other than molten materials, oxides are becoming less important. Fragmented particles, caused by breaking and grinding during size reduction, should be avoided in spray powders for sintering. These splintered particles can prevent and disturb the desired continuous powder flow to the spray flame con-

139

140

9 Spray Materials

Fig. 9.5 Powder production by means of sintering.

Fig. 9.6 Properties of sintered spray powders (Praxair Services).

siderably. Here again, appropriate processing methods reduce the content of chips and fragments to a minimum. Production of molten and milled spray powder is derived from abrasive (corundum) production. Here, batch sizes of 8–10 tons are quite common. Figures 9.7 and 9.8 schematically illustrate the melting process inside an arc furnace as well as subsequent size reduction. Agglomeration, also referred to as spray drying, is one of the most modern production methods for spray powders. Nowadays, this production process is usually followed by sintering or spheroidising in order to compact and densify

9.2 Spray Material Properties Determined by Production Issues

Fig. 9.7 Powder production by means of melting and breaking.

Fig. 9.8 Properties of molten, broken spray materials (Praxair Services).

the powder in order to prevent the agglomerated powder from being destroyed during powder transport or in the flame (schematic representation in Figs. 9.9 and 9.10). Apart from agglomeration, coating of so-called primary grains with fine secondary grains and additional organic binder is a further method for producing composite powders. Also, relatively rough primary grains can be electroplated with nickel or cobalt. Nickel graphite is typical for this group of materials. Here, the core of the powder particle is graphite that is protected from burning during spraying by the nickel shell. Powder production by means of powder mixing is not covered in this chapter.

141

142

9 Spray Materials

Fig. 9.9 Powder production by means of agglomerating and post-treatment.

Fig. 9.10 Properties of agglomerated spray powders.

9.3 Material Selection for Coating Applications

Generally, when selecting spray materials, attention should be paid to the fact that properties of sprayed coatings can be completely different from the properties of cast, forged bulk material. Thus, density and porosity, but also intrinsic stresses and oxide content of sprayed coatings differ from, e.g., cast bulk material. Publications by powder producing companies, e.g. Praxair, Sulzer Metco, and H. C. Starck, provide guidelines for material selection and are recommendable, as standard solutions have been available in many fields of application for some time.

9.3 Material Selection for Coating Applications

9.3.1 Materials for Wear Protection

Wear resistance is a system characteristic. Apart from the material combination of parts involved, it is influenced by many other factors including relative motion, strain, roughness, atmosphere, temperature, etc. The main wear mechanisms that can also occur simultaneously in certain applications are: Adhesive wear: In order to avoid adhesive wear, metallurgical interactions between friction partners have to be prevented as they promote micro-welding. In engine construction, for instance, this is realised by producing spray coatings with a porous structure that picks up and holds lubricant, e.g. plasma-sprayed molybdenum mixed with NiCrBSi. Abrasive wear: Abrasive counterparts require spray coatings with high and uniform hardness. For this, homogeneous and dense coating structure is necessary. High-velocity oxygen fuel flame spraying (HVOF) is preferable for producing such coatings. Typical materials are WC/Co and NiCrBSi+WC/Co mixtures. Fatigue: Fatigue often emanates from defects in the coating structure, e.g. pores, cracks, oxide lamella, etc. Here, dense coatings, produced under protective gas, in vacuum, or by means of HVOF from metal alloys, are best suited. Corrosive wear: Hard Co- or Ni-base alloys or sealed Cr2O3 and Al2O3 oxide/ ceramic coatings have frequently proven to protect against corrosive wear. 9.3.2 Materials for Corrosion Protection

In cathodic corrosion protection, the part is charged as cathode, whereas the wire-sprayed zinc or aluminium coating is consumed as anode in the electrochemical circuit. Anodic protection involves opposite charge of coating and substrate, and the formation of an electrochemical circuit must be prevented. The only way to guarantee this is to deposit dense, metallic Ni-, Fe-, or Co-base coatings or sealed ceramic coatings, which frequently require additional dense NiCror NiAl adhesive layers in order to seal the substrate. Corrosion protection at elevated and higher temperatures with metal alloys is only feasible by using relatively complex systems that form Al2O3, SiO2, or Cr2O3 surface oxide layers. If necessary, the content of these alloying elements in the coating has to be sufficiently high in order to form the protective oxidic layer several times. So-called MCrAlY alloys are spray materials developed specifically for hot-gas corrosion-resistant coatings on turbine blades. The basis for these alloys is an iron, nickel, or cobalt alloy that is adjusted for particular applications by adding a large number of additional alloying elements. These materials, combined with ceramic top coats, are also used for high-temperature corrosion protection, and as adhesive layers for very high temperatures in the range of 1000 8C.

143

144

9 Spray Materials

9.3.3 Materials for Biotechnology

Biomedicine presents a new field of application for thermal spray processes. An example are metal coatings, made from titanium and its alloys, which are sprayed onto metal prostheses under vacuum conditions. They produce an inert surface with sufficient roughness allowing mechanical interlocking of bone tissue. Hydroxyapatite (HA) or fluorapatite coatings with bone-similar phosphate structure show even better performance, supporting growth of bone tissue on the prosthesis without requiring any disturbing interface layer. 9.3.4 Materials for Special Applications

Ni-graphite-abradable coatings, also referred to as mechanical seals, have been used successfully in turbine and engine construction for many years. After grinding-in during start-up, the coatings prevent performance loss due to friction of expanding components against the housing of the turbine at operating temperature. Today, ceramic zirconium-based materials are also used here, as the temperature exceeds 1000 8C. Further examples of special thermal spray coating applications are electrically insulating Al2O3, superconductive YBaCu oxide with perovskite-similar structure, oxygen ion conduction in fuel cells, molybdenum and tungsten coatings for high-vacuum technology, and decorative coatings on household products that are often Al2O3-TiO2 coatings. Available applications for spray materials are virtually endless and cannot be covered comprehensively in this publication. Additional information is given in the German Handbuch des Thermischen Spritzens (Handbook of Thermal Spraying, published soon by DVS) and conference proceedings of spray conferences in the TS and URSC series of past years, also available from DVS, Düsseldorf, Germany.

145

10 High-velocity Oxygen Fuel Flame Spraying O. Brandt, Becon Technologies GmbH, Thun, Switzerland 10.1 Introduction

After a quarter of a decade, high-velocity oxygen fuel flame spraying (HVOF) is now on a par with classic spray processes such as flame spraying, plasma spraying, or arc spraying. Compared to other processes, HVOF introduces less heat to the coating material. Additionally, the particle velocity on substrate impact is considerably higher than in other spray processes. Therefore, the produced coatings generally feature higher bond strength and more compact structure with less porosity. Reduced heat input to the material is used in order to effectively suppress thermally activated transition processes [1]. Introduced in the early 1980s, HVOF spraying is one of the younger process techniques of thermal spraying. The basic idea was developed from the discontinuously operating detonation gun (D-gun) principle. In the D-gun process, reaction gases acetylene and oxygen are mixed with coating material and nitrogen, serving as carrier gas, and detonated by ignition. Powder particles then accelerate, reaching velocities of more than 750 m/s. The distance from the substrate surface to the nozzle tip is in the range of 100–120 mm. Fuel gas temperature can rise up to 4000 8C. After each detonation process, the combustion chamber is rinsed with nitrogen. The average operating frequency is 4–8 ignitions per second [2]. The advantage of the D-gun, compared to other spray processes, is a considerably higher particle velocity. Coatings, therefore, show reduced porosity, higher bond strength, and thus qualify for many applications. HVOF development has focused on designing a continuous process that would produce coating quality comparable to the D-gun. James A. Browning, an American engineer, is one of the pioneers in this field. The first process variant was the Jet Kote system, introduced in 1982. In this process, a fuel gas (usually propane, propylene, or hydrogen) and oxygen burn in the burning chamber under high pressure. Burning pressure pulses at a frequency of 500 to 1000 Hz. The temperature of the burning gas reaches approx. 3000 8C. The gas leaves the burning chamber towards a cylindrical nozzle where it expands and accelerates. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

146

10 High-velocity Oxygen Fuel Flame Spraying

After passing through the nozzle, the combustion gas mixture develops average flame velocities of approx. 1800 m/s, and noise reaches approx. 120 dB [A]. The expansion nozzle injects the carrier gas stream and powdered spray material into the centre of the flame. Powder particles melt partially or completely and accelerate to velocities of up to 500 m/s as they leave the expansion nozzle. The burning chamber, the mixing head, and the expansion nozzle, all made of a copper alloy, form the water-cooled inner part of the burner. A hydrogen and oxygen pilot flame ignites the main flame [3, 4]. Partially or completely molten powder particles impact the substrate surface with high kinetic energy, splatter due to high velocity, and spread into a starshaped splat. The new surface forms after multiple passes of the traversing spray gun, creating adjacent coating strips and superimposed coating layers with new material properties. Typical coating thicknesses for HVOF processes are in the range of 0.01 to 0.5 mm. Generally, HVOF coatings are applied when extremely dense and well-bonded coatings with low porosity are desired. Today, approximately ten different system types are available for processing powdered spray materials, using propane, propylene, hydrogen, ethylene, or kerosene as fuel gases. In the 1980s and 1990s, development first focused on design of burners and spray guns suitable for spray powders available at the time. Particularly for HVOF processes, powdered spray materials were then designed and tested, with respect to particle-size distribution and morphology, which now are state-of-the-art. Nowadays, development aims at modern process control and integration with automated production processes as well as qualitycontrol concepts [5].

10.2 Characteristics 10.2.1 HVOF Gun

Compared to conventional flame spraying guns, HVOF guns can be distinguished by their typical expansion nozzle which is considerably longer. Attainable gas speed depends on individual gun type and design, nozzle geometry, and type of fuel. In addition to a classification according to the year of market introduction, guns are classified according to general design (cf. Fig. 10.1), not considering components such as controllers and powder feeders. Thermal spraying guns are classified according to the type and arrangement of burning chambers, powder injection, and cooling. Open system guns do not include a separate combustion chamber and are cooled with air as well as water. Closed system guns feature an additional expansion nozzle. Particle speed, here, is higher than in open systems, however, there is a danger that molten particles stick to the expansion nozzle and may even clog it entirely.

10.2 Characteristics

Fig. 10.1 Design and classification of HVOF systems [7].

10.2.2 Fuel Gases and Process Parameters

The kinetic energy of spray particles determines properties and microstructure of sprayed coatings. Kinetic particle energy, however, is greatly influenced by spray gun design. An additional way of controlling heat input and acceleration of spray powder is selecting the appropriate fuel gas. Apart from the type of

147

148

10 High-velocity Oxygen Fuel Flame Spraying

Table 10.1 Properties of typical fuels for high-velocity oxygen fuel flame spraying [1]. Fuel, fuel gas

Propane Propylene Hydrogen Ethene Acetylene Kerosene a)

Maximum flame temperature in 8C

Calorific value in MJ/m3

2828 2896 2856 2924 3160 approx. 2900

93.2 87.6 10.8 59.5 56.4 37.3 MJ/l

Oxygen/fuel ratio for Maximum surface temperature

Stoichiometry

HVOF applications

4.5 3.7 0.42 2.4 1.5 1.9 a)

5.0 4.5 0.5 3.0 2.5 3.4 a)

3–8 3.5–7 0.3–0.6 2–5 1.3–4 2.8–4.8 a)

mass mixture ratio oxygen/kerosene

gas, the mixture ratio of fuel gas and oxygen is the key parameter. The type of fuel gas determines the maximum flame temperature, whereas system performance is determined by the amount and net calorific value of the supplied fuel. Flame temperature can be controlled to a certain extent by varying mass flow rates of fuel gas and oxygen. HVOF systems usually operate with overstoichiometric flame parameters yielding complete combustion of the fuel gas. Furthermore, overstoichiometric flame parameters lead to reduced flame temperature and, at the same time, increased gas and particle speed. Reduced maximum flame temperature is often desired in order to minimise thermally induced phase transitions in carbide powders. Certain systems even allow feeding a cooling gas, e.g. compressed air or nitrogen, into the flame for reduced flame temperature. Table 10.1 shows the most important and common fuel gases [1]. 10.2.3 Spray Materials

The main applications for HVOF spraying are wear-protecting carbide/metal coatings. Most commonly used carbides are tungsten carbides (WC) and chromium carbides (Cr3C2) with a cobalt (Co), nickel (Ni), or combined (NiCr or CoCr) binder matrix. TiC, ZrC, HfC, SiC, or TaC hard alloys are hardly ever used in thermal spraying. Recent developments aimed at processing WB, CrB, MoB with Co or Ni as binding matrix in HVOF spraying. During the past 20 years, powders with maximum particle diameters of 45 lm have proven to be of value in HVOF spraying of metal carbide coatings, independent of the spray system used. Minimum particle size varies depending on powder and/or system supplier. Lower sieve sizes with diameters of 37, 22.5, 15, 10, and 5.6 lm are common. Particle size fractions are always specified by maximum (e.g. –45 lm) and minimum (e.g. +15 lm) mesh sizes of the sieves. Larger powder particles remain in the top sieve whereas smaller particles pass through the lowest sieve. The particle size fraction then contains all particle sizes within the sieve limits.

10.2 Characteristics

Fig. 10.2 HVOF coating of a driving shaft with Jet-Kote system (first commercial HVOF system), company photograph Stellba Schweisstechnik AG, Birrhard, Switzerland.

In commercially available particle size fractions, a certain proportion of too large and too small particles is tolerated. The low particle size limit strongly influences powder flowability. Powders including higher proportions of fine particles require more precise powder feeders and tend to stick more when passing through the expansion nozzle. Closer tolerances for powder fractions are preferable for spray processes, however, their price is higher due to the additional effort of sieving and control. Today, commonly used hard alloy powder materials are, nearly all, agglomerated, sintered powders with spherical geometry (Fig. 10.2). According to supplier data, the carbide grain size for these powders is approx. 3–5 lm, a market typical value for HVOF spray technology [6, 7]. Ultimate coating hardness and wear resistance of HVOF-sprayed WC coatings is determined not only by spray powder composition but also by spray parameters. Current research focuses on production and spray processing of powders with finer carbides, down to the nanometre scale. Additionally, new matrix materials based on highly corrosion resistant nickel alloys are introduced [7]. Considering metals and alloys, so-called MCrAlY coatings for high-temperature oxidation protection are used in turbine blade repairs. HVOF is also suitable for materials such as stainless steel grade 316L or nickel-based alloys of Hastelloy or Inconel type. The reason for this is that the producible coating structure is very compact, nearly non-porous, and the spray material takes up very little oxygen, keeping oxidation low. HVOF coatings, subsequently heat treated, are known to show properties comparable to equivalent forged materials [8].

149

150

10 High-velocity Oxygen Fuel Flame Spraying

10.3 Technical Considerations

Generally, HVOF spraying is one of the high-energy process technologies of thermal spraying. Systems available today operate at nominal power ranges of approx. 50–250 kW, determined by the energy content of the supplied fuel gas per unit time. The high performance allows processing of high powder feed rates up to 9 kg/h, depending on the particular material [9, 10], which is advantageous for coating large parts. On the other hand, substrate material of small parts can be subject to overheating. Apart from processable powder feed rates, economical coating also requires sufficient process and deposition efficiency, which is determined considerably by the particular process and material, e.g. 50–85% for spraying of metal carbides [9, 10]. The high-energy HVOF process generates noise levels of up to approx. 135 dB [A]. Additionally, the emanating combustion gas produces considerable recoil force on the gun. Apart from general dust and heat development, these are further reasons for commonly used robotic handling in HVOF systems. However, for on-site coating, hand-held guns are available, but appropriate work place and safety measures are necessary (Fig. 10.3) [11, 12]. High gun performance and powder feed rates cause a considerable heat input to the coating surface. Small parts or parts with low wall thickness therefore require cooling. Often, a flow of compressed air is sufficient. In certain applications, however, liquid CO2 cooling is necessary, e.g. for substrate materials with high coefficients of thermal expansion such as Al or Cu [12]. Additionally, intensive cooling can reduce changes in part geometry after coating. Maximum part temperatures should be kept at an average of 60–100 8C.

Fig. 10.3 Morphology of a typical tungsten carbide powder, scanning electron microscopic image, secondary electron mode, magnification ´1000.

10.4 Applications

The high kinetic energy of spray particles yields very smooth surfaces with considerably lower roughness values compared to conventional flame or plasma spraying. Necessary oversize for machining is thus minimised, which is particularly important for hard carbide coatings as these require machining with diamond tools or grinding. For coating inside diameters, typical spray distances of approx. lC = 200– 350 mm limit applications, however, coating of hidden or difficult to reach surfaces is possible. This is due to the fact that HVOF coatings can be deposited at angles of up to 308 without any significant loss of quality.

10.4 Applications

Published results of applications and coating examples show that mechanical engineering is the main field of application. Here, the majority of applications involves wear resistance for abrasive wear conditions, partially superimposed with corrosion. Here, WC-Co, WC-Co-Cr, and NiCrBSi-WC coatings are appropriate for operating temperatures of up to approx. 500 8C. HVOF coatings are being used increasingly for protection against liquid erosion and cavitation (Fig. 10.5). Typical examples are [13]: · ball and plate valves · components for plastics extruders · rotors and pistons for compressors and pumps · treads of large-scale engines · wire-drawing rolls · torch nozzles · circumferencial and running surfaces of hydraulic cylinders · paper and foil rollers · sieves in salt industry · sliding surfaces for railway switches · surfaces of water turbines and large-scale ventilators · different spindles. In aerospace, named applications mostly include examples of wear-protection coatings on running surfaces of hydraulic cylinders for landing-gear components or diverse wing-adjusting systems. Here also, WC-Co and WC-Co-Cr hard alloys dominate. The most important selection criteria are said to be favourable intrinsic stresses after coating and excellent fatigue resistance. This also favours applications on different aerospace turbine blade surfaces. An additional reason for aerospace applications is that such coatings allow removal and re-coating after a certain period of operation [13]. Apparatus engineering for the petrochemical and gas industry is a further large field of applications for thermal spray coatings. Experience has been gained with HVOF coatings in boilers, heat exchangers, pipes, valves, sliders,

151

152

10 High-velocity Oxygen Fuel Flame Spraying

Fig. 10.4 Typical coating of a roller with kerosene-fuelled GTV K I system, guidance with axial advance and rotating roller, company photograph Buser Oberflächentechnik AG, Wiler, Switzerland.

Fig. 10.5 Setup for flat seal coating with GTV K II system, one of the most sophisticated HVOF systems, company photograph Stellba Schweisstechnik AG, Birrhard, Switzerland.

and similar components in combustion engineering. Predominant materials are Cr3C2-NiCr and NiCrBSi. In certain cases, WC containing coatings are used for lower operating temperatures. The main wear conditions involve gas erosion superimposed with corrosion and oxidation [13]. A further field of applications is high-temperature oxidation preventing coatings on blades of stationary gas or steam turbines. Here, so-called MCrAlY coat-

10.5 Process Monitoring and Control

ings (M = Fe, Ni, or Co) are used, not only to prevent oxidation but also as bond coats for ceramic thermal-barrier coatings, usually deposited by vacuum plasma spraying (VPS). Publications focusing on application examples compare VPS and HVOF coatings in terms of properties and costs. Results show that VPS is advantageous for coatings on new parts, whereas HVOF coatings are better suitable for repairs. Several fields of application are presented in the automotive industry, e.g. brake disk coatings. Experience has been gained from a number of long-term tests, but real series applications have not yet been published [13]. Different hydraulic cylinders for rudder controls in ships have been repaired successfully using HVOF coatings. This also applies to cylinder running surfaces, pistons, and piston rings of marine diesel engines as well as different bearing seats for driving shafts [13]. A prominent example is the use of a metal 316L steel or Inconel 718 coating as a primary housing structure for the combustion chamber in a hypersonic jet engine. HVOF technology was used in this development primarily due to low oxygen content as well as negligible coating porosity. Both criteria are prerequisites for subsequent heat treatment of the coating, which then shows mechanical properties comparable to the corresponding bulk material [13].

10.5 Process Monitoring and Control

Generally, stored program controls (SPC) have the potential to integrate system technology of HVOF systems into modern manufacturing chains. At present, however, coating systems are frequently isolated within manufacturing processes and features linking to process optimising, monitoring, or data exchange for quality control are incorporated insufficiently (Fig. 10.6). Modern spray systems, equipped with measuring and control systems, can be used to guarantee reproducible process parameters that are independent of surrounding conditions. Additionally, process visualisation is continuously becoming more important, providing the user with error messages of possible variations from predefined parameters, and to allow counteractive measures at an early stage. A number of different diagnostic systems are available, but broad acceptance is still limited due to high costs of purchase. However, modern manufacturing chains increasingly require exact reproducibility of coating quality, and wide implementation of process monitoring and control systems can be expected soon (Fig. 10.7) [5].

153

154

10 High-velocity Oxygen Fuel Flame Spraying

Fig. 10.6 Principle of integrating an online process control as system control, company photograph GTV mbH, Luckenbach, Germany.

Fig. 10.7 Installation of GTV K II gun with diagnostic system, company photograph GTV mbH, Luckenbach, Germany.

10.6 Development Trends

10.6 Development Trends

Presented and published examples of applications impressively show the economic significance of high-velocity oxygen fuel flame spraying technology that, today, has reached a share of 25% of the total thermal spray market [15]. This is attributed not only to fast-paced method and process developments but also, and even more, to industrial demands on technical coatings. Increasing quality requirements and improved profitability produce a constant pressure that continuously initiates new developments aiming at optimised coatings or reduced process costs. Furthermore, the intensive search for new fields of HVOF applications has led to a certain degree of competition between HVOF and other coating technologies such as electroplating or build-up welding. 10.6.1 Application Technology

HVOF coatings have captured a considerable market share by replacing electroplated hard chrome coatings. The impetus to this development was given by the US Environmental Protection Agency (EPA), when chrome VI compounds, which can occur in hard chrome plating, were classified as highly toxic and carcinogenic. Rising costs, e.g. for waste water disposal, triggered a search for possible alternatives in several branches of industry such as aerospace and defense. In thermal spraying, this search was focused mainly on HVOF. Several research and development activities gave evidence that HVOF coatings can be superior to hard chrome coatings in terms of wear and corrosion resistance. However, considering economical aspects, HVOF is limited in terms of providing an alternative to hard chrome coatings, especially when part geometry and coating thickness are predominant criteria for technology selection. Further limitations are due to the fact that machining of hard alloy coatings requires diamond tools. Several research groups aim at these and additional aspects in order to increase coating quality and profitability and, therefore, to open new fields of applications and new markets for HVOF technology [16, 17]. A typical example of a coating replacing chrome is a carbide coating on a calendar roller for the paper industry. 10.6.2 Coating Materials

Investigations show that the wear resistance of coatings can be increased further by using powder with a fine carbide dispersion. Therefore, current research focuses on spray powders with carbide particle size down to the nanometre scale [18]. An additional advantage of fine powders is reduced surface roughness of the sprayed coating, prior to machining. Thus, the necessary oversize for subsequent machining can be reduced, which again cuts coating costs.

155

156

10 High-velocity Oxygen Fuel Flame Spraying

A further aspect of current basic research and development work is processing of different plastics, sprayed with or without dispersed fibres or particles. This applies likewise to projects where silicon carbide particles are added to metal powders for increased wear resistance of HVOF coatings that are compared to conventional carbide coatings. This work is fundamental and will, in the future, broaden the fields of applications for HVOF technology [19]. Combination coatings that include a subsequent plastic coating, sealing an HVOF hard alloy coating, are of particular interest. For example, high abrasive wear resistance of a tungsten carbide coating can be combined with excellent anti-stick properties of a PTFE coating. Material combinations here are virtually unlimited. Furthermore, plastic sealing can increase corrosion resistance of HVOF coatings considerably [20]. 10.6.3 Process Technology

Developments in process technology focus on spraying with reduced flame temperatures in order to prevent phase transitions and oxidation of spray material. One way to do this is to use air instead of oxygen for combustion. For hard alloy coatings, this method improves impact resistance. Another way to reduce spray particle temperature is to adjust to much higher combustion chamber pressures that also allow processing of coarser powder. Such systems are expected to increase coating quality considerably in terms of hardness, density, bond strength, and toughness [1]. 10.6.4 Techniques and Methods

High-velocity wire flame spraying is an important innovation in HVOF technology. The technique is currently being evaluated for spraying aluminium, copper, molybdenum, or different steel alloys. Coating porosity and the tendency to oxidise is reduced notably compared to conventional wire flame-sprayed coatings. However, this technique requires coating material in the form of wires and, therefore, material variety is limited compared to process variants using powder. Only cored wires allow hard-phase dispersion in the coating for increased hardness, e.g. carbides [21].

10.7 Summary

Today, autogenous HVOF is a well-established technique for producing protective coatings, primarily providing preventive wear protection for technical surfaces. Investigations focus on highly wear-resistant tungsten carbide coatings with hardness in the range of approx. 1000–1400 HV0.3, for operating tempera-

References

tures of up to approx. 540 8C, and chromium carbide coatings with hardness of approx. 750–900 HV0.3 for operating temperatures up to approx. 900 8C. In addition, many applications combine metal coating materials due to comparatively high particle velocity and moderate temperatures in HVOF. Both process characteristics are prerequisites for dense and well-bonded coatings with low porosity.

References 1 Kreye, Heinrich, et al.: Hochgeschwin-

2

3

4

5

6

7

8

digkeitsflammspritzen – Stand der Technik, neue Entwicklungen und Alternativen. In: Conference Proceedings 6th Kolloquium Hochgeschwindigkeitsflammspritzen, Erding, 2003, pp. 5–17 Gill, B. J.: Detonations- und Plasmaspritzen gegen Korrosion und Verschleiss. Technische Rundschau 79, 1987, pp. 46– 51 Buss, Stephan: Anwendungsvergleich thermisch gespritzter Schichten im Automobilbau. Student Research Project, University of the Federal Armed Forces, Hamburg, 1993, unpublished Heinrich, Peter: Neue Entwicklungen und Anwendungsmöglichkeiten beim Hochgeschwindigkeitsflammspritzen. Linde AG – TG Sonderdruck 154. DVS Report Vol. 123 Nassenstein, Klaus; et al.: Zukunftsweisende Steuerungskonzepte für HVOF Applikationen. In: Conference Proceedings 6th Kolloquium Hochgeschwindigkeitsflammspritzen. Erding, 2003, pp. 101–107 Zimmermann, Stephan, et al.: Verbesserte Schichteigenschaften durch optimierte Karbidpulver für moderne HVOF-Systeme. In: Conference Proceedings 6th Kolloquium Hochgeschwindigkeitsflammspritzen. Erding, 2003, pp. 31–38 Brandt, Oliver: Aspekte zur Herstellung und Charakterisierung von Wolframkarbidschichten, aufgetragen durch Hochgeschwindigkeitsflammspritzen. Papierflieger, Clausthal-Zellerfeld, 1. 1998 Voggenreiter, Heinz, et al.: Atmosphärisch Rheo-gespritzte Brennkammer für Hyperschallflugzeuge. Deutscher Verlag für Schweißtechnik DVS, Düsseldorf, 1996, pp. 37–41

9 Nestler, M., Henriksen, R.: Charakteristi-

10

11

12

13

14

15

ka und fortschrittliche industrielle Anwendungen der Diamond Jet Hybrid – der neuen Generation gasbetriebener HVOF-Anlagen. Characteristics and Progressive Industrial Applications of the Diamond Jet Hybrid – The New Generation of Gas-Powered HVOF-Systems. Conference Proceedings 3rd Kolloquium – Hochgeschwindigkeitsflammspritzen 1997, pp. 104–117 Matthäus, Götz, Stevens, G.: OSU Super Jet System SJS und Carbide Jet System CJS/OSU Super Jet System SJS und Carbide Jet System CJS. Conference Proceedings 3rd Kolloquium Hochgeschwindigkeitsflammspritzen 1997, pp. 127–141 Nassenstein, Klaus, Isch, Hans-Peter: Vorteile und Möglichkeiten durch GTVSteuerungen beim HVOF-Spritzen mit TOP-GUN Pistolen. Advantages and Possibilities Using the GTV-Control for HVOF Spraying with TOP-GUN Torches. Conference Proceedings 3rd Kolloquium – Hochgeschwindigkeitsflammspritzen 1997, pp. 93–102 Heath, G. R., Dumola, R. J.: Practical Experience with the New Generation of Low-Cost. Portable HVOFs. Proceedings of 15th International Thermal Spray Conference – Thermal Spray: Meeting the Challenges of the 21st Century 2 1998, pp. 1495–1500 Brandt, Oliver, Siegmann, Stephan: Anwendungsbeispiele für HVOF-Schichten. In: Moderne Beschichtungsverfahren, Wiley, Weinheim, 2,000 pp. 68–79 Brandt, Oliver: Hochgeschwindigkeitsflammspritzen – Anwendungsbeispiele, Trends. Giesel, Isernhagen, Vol. 58, 2002 Keye, Heinrich, et al.: Hochgeschwindigkeitsflammspritzen, Stand der Technik, Perspektiven und Alternativen. Confer-

157

158

10 High-velocity Oxygen Fuel Flame Spraying

16

17

18

19

ence Proceedings 5th Kolloquium Hochgeschwindigkeitsflammspritzen, 2000, pp. 5–18 Wassermann, C., et al.: Ersatz von Hartchrom durch HVOF-Schichten. Conference Proceedings 5th Kolloquium Hochgeschwindigkeitsflammspritzen, 2,000, pp. 109–114 Wassermann, C., et al.: Replacement for Hard Chrome Plating in Printing Machinery. Proceedings of International Thermal Spray Conference – New Surfaces for a New Millennium, 2001, pp. 69–74 Fischer, F., et al.: Development of Ultra Thin Coatings for Wear and Corrosion Resistance. Proceedings of International Thermal Spray Conference – New Surfaces for a New Millennium, 2001, pp. 1131–1135 Wielage, Bernhard: Manufacture of SiC Composite Coatings by HVOF. Proceed-

ings of International Thermal Spray Conference – New Surfaces for a New Millennium, 2001, pp. 251–258 20 Brandt, Oliver, Siegmann, Stephan: Verschleissfeste Antihaftschichten durch die Kombination von thermisch gespritzten Metallen und Keramiken mit Kunststoffen. Conference Proceedings 5th Industrial Symposium “Oberflächen- und Wärmebehandlungstechnik and 6th Werkstofftechnisches Kolloquium”, Chemnitz University of Technology, Vol. 016, Chemnitz, 2003, part 1, pp. 75–81 21 Kreye, Heinrich, et al.: High Velocity Combustion Wire Spraying – Systems and Coatings. Proceedings of International Thermal Spray Conference – New Surfaces for a New Millennium, 2001, pp. 461–466

159

11 Triplex II – Development of an Economical High-performance Plasma Spray System for Highest-quality Demands even under Challenging Production Conditions H. Zimmermann, Sulzer Metco AG (Switzerland), Wohlen, Switzerland H.-M. Höhle, Sulzer Metco Europe GmbH, Hattersheim, Germany

11.1 Introduction

In virtually all areas of technology, experts have come to the conclusion that rising demands on modern constructional elements are best met by separating the function of actual bulk material of a part from its surface functions. While the body material determines, for instance, shape, dimensioning, supporting points, and areas of contact, the surface is responsible for contact conditions and part protection. For both body material and surface, careful selection of the appropriate material and corresponding optimal manufacturing processes is inevitable. Modern coating technologies provide numerous economically reasonable, and in many respects technically interesting alternatives for the design of functionally optimised part surfaces. Even simple and inexpensive base materials can be combined with high-grade coatings yielding modern material composites, matched precisely to particular operating conditions. The full economic, technological, and environmental potential of coated material composites has not yet been tapped. However, it has paved the way for coating technology to become a key technology that will continue to open up new vistas of applications. The multitude of services and offerings around coating technology, though, is more of an obstacle for its spread. Selection of appropriate coatings and processes for a particular application requires thorough knowledge and, therefore, intensive guidance from experts. They can also help with cost/benefit analysis and with the decision whether to integrate a coating system into one’s own production, or rather outsource the task to a coating service provider. Development work in the field of coating technology is widespread. It includes development of new applications as well as new and further development of coating processes and systems. Figure 11.1 shows a selection of surface coating technologies commonly used today, and individual limitations with respect to achievable coating thickness and thermal load onto the base material (substrate). In terms of coating materials, work is focused on an extended selection Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

160

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.1 Coating thickness and process temperatures of selected coating technologies.

of materials as well as improved matching of coating/substrate composites and applications. Possibilities seem virtually unlimited as new coating materials on the one hand and process parameters on the other, as well as multi-layer and graded coating structures, yield new properties. Additionally, considering economical manufacturing and optimal material utilisation of often expensive materials is advisable. This applies equally to powder and wire materials for thermal spraying as well as to target materials for PVD processes. Hence, TiNTiAlN multi-layer coatings were developed mainly because titanium targets were far less expensive than titanium/aluminium targets. Figure 11.1 shows that thermal spraying is the coating technology with the largest range of coating thicknesses, which is one reason for its popularity. Additional advantages of thermal spraying include a practically unlimited choice of materials, low thermal stress to the substrate, independence of substrate geometry in spite of line-of-sight characteristic, and comparatively low specific costs. Thermal spraying is defined in DIN EN 657 as a method of applying coatings to a cleaned and pre-treated part surface with special apparatuses and systems by partial or complete melting, atomising, and deposition of a spray material at high velocity. The same standard distinguishes thermal spray technologies using a gas flame for melting of the coating material, supplied as powder or wire, from spray technologies using electrical energy. One of the latter is plasma spraying which will be described here in more detail. After a brief, general overview of plasma spraying processes and its variants, a technical innovation is discussed, the three-cathode plasma torch, representing a quantum leap of performance and reliability.

11.2 Fundamentals of Plasma Spraying

11.2 Fundamentals of Plasma Spraying

Arc plasmas were first used as heat sources for production of coatings in 1939. Since then, plasma spraying technology developed continuously, evolving to the spray technique with the largest field of applications today. One reason for this development is the broad selection of coating materials, e.g. metals, alloys, oxides, carbides, as well as mixtures of these. Deposition of the coating material is possible on virtually every substrate material, provided that cooling is sufficient. Part geometry in plasma spraying is subject to only a few restrictions as appropriate plasma torches are available for coating inside diameters down to 40 mm. Plasma spraying has many process variants, each developed and specifically advantageous for individual applications. A selection of process variants is listed here with common abbreviations and characterised: · APS: atmospheric plasma spraying · VPS: vacuum plasma spraying – This actually is an inert gas plasma spray process (IPS) conducted in a vacuum chamber that is evacuated, and then flooded with inert gas in order to avoid reactions between coating material and atmospheric oxygen. Sulzer Metco favours the term CAPS (controlled atmosphere plasma spraying) that also includes LPPS (low-pressure plasma spraying) and LVPS · UPS: underwater plasma spraying · SPS: shrouded plasma spraying, which uses a protective gas envelope shielding the particle jet · RPS: reactive plasma spraying, in which the coating material reacts with a reactive gas and new material combinations are formed. RPS is usually conducted in a vacuum chamber flooded with reactive gas. In all types of plasma spraying, computers can be used in order to acquire and control process-relevant data and, thus, achieve a high degree of automation in manufacturing processes for coated parts. This in turn guarantees continuously high coating quality. Coatings are used for a multitude of applications, e.g. repairs as well as wear, oxidation, or corrosion protection. Also, particular thermal, electrical, or tribological properties are obtainable by coating. In plasma spraying, typical coating thickness is in the range of 50 to 500 lm. Materials, process type, and application determine the maximum coating thickness. For special applications, e.g. thermal-barrier coatings, abradable coatings, or repair coatings, coating thickness can rise up to 2 mm. Figure 11.2 illustrates the basic principle of plasma spraying. A high-voltage impulse ignites an arc that burns between a finger-shaped, thorium-doped tungsten cathode and the nozzle-shaped copper anode. Both electrodes are water cooled. The working gas, which can be a mixture of primary and secondary gas, is fed to the nozzle through a gas-distribution ring, and flows in between the electrodes. Here, due to the arc, it forms a plasma, i.e. electrons, emitted by the cathode and accelerated by the potential difference, collide with working gas

161

162

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.2 Principle of plasma spraying.

atoms that get partially ionised. Molecular gases dissociate into individual atoms prior to ionisation. Therefore, a thermal plasma is a hot gas with a considerable proportion of ionised atoms and molecules. Such plasmas behave distinctly different from regular gas as charges are no longer attached to atoms but move freely as electrons and ions within the gas which thus becomes electrically conductive. It should be noted that the plasma is neutral as a whole and can be referred to as a cold plasma. On the other hand, the term hot plasma refers to a plasma with excess positive or negative charge carriers. For simple heating, Fig. 11.3 shows at which temperatures the typical plasma gases argon, helium, nitrogen, and hydrogen ionise, or dissociate and ionise. It appears that the energy content of the gas increases rapidly in dissociation (separation of atoms) and ionisation (removal of electrons) although the increase in temperature is rather low. Figure 11.3 allows a number of conclusions for plasma spray processes. At constant electrical power input, use of argon as a plasma gas yields high plasma temperature due to comparatively low enthalpy. The thermal expansion and thus high pressure lead to high plasma velocity. Adding helium further increases the velocity of the plasma. On the other hand, molecular gases absorb and carry high energy contents even at low temperatures. Low temperatures, however, reduce temperature-determined heat loss so that a molecular gas plasma jet cools much slower than a noble gas plasma jet, and transfers more heat to the injected spray powder. Therefore, argon/helium plasma is often referred to as cold and fast-moving (high temperature, low enthalpy) whereas nitrogen/hydrogen plasma is referred to as hot. After passing the arc, usually in the immediate vicinity of the nozzle tip, ions and electrons of the plasma recombine to atoms and molecules, and release

11.2 Fundamentals of Plasma Spraying

Fig. 11.3 Enthalpy against plasma temperature for Ar, He, N2, and H2 [1].

their energy as heat. This recombination heat is responsible for temperatures of up to 20 000 K in the centre of the formed plasma jet. The plasma jet is defined as the currentless plasma that leaves the nozzle. The surrounding atmosphere reduces the velocity of the plasma jet. This causes turbulent flow and heat loss of the plasma jet. By controlling the injection of plasma gas into the nozzle, the characteristics of the plasma jet can be adjusted. Usually, plasma gas is introduced to the cathode coaxially, and is forced into a rotary motion by means of a distribution ring with outlets arranged at an angle. Practice shows that rotating plasma gas, along with electromagnetic effects, increases rotary motion of the anodic arc root and, therefore, prevents local overheating of the anode and selective, erosive wear. On the other hand, however, a rotating plasma jet is subject to increased turbulence and heat loss. In order to guarantee the highest possible degree of melting, powder spray material is injected into the high-energetic zone of the plasma jet by means of an appropriately positioned injector. After injection, the plasma gas flow picks up and accelerates the powder particles. At the same time the particles melt due to the energy transferred by the recombining plasma gas. The longer the particle dwell time in the plasma jet, the more energy is absorbed by the particles. The plasma jet velocity determines the axial velocity of the powder particles. Higher plasma gas flow and temperature produce higher pressure and gas velocity. Particles at the point of injection have nearly zero velocity and initially accelerate strongly in the direction of flow. Particle velocities can rise up to 400 m/s during acceleration in atmospheric plasma spraying, depending on gas composition, particle size, etc. Due to the high particle velocity, the flight time of powder particles until impact onto the substrate surface is only several milliseconds [2].

163

164

11 Triplex II – Development of an Economical High-performance Plasma Spray System

The plasma jet traverses across the surface, depositing the transported molten spray particles onto the workpiece as a coating. Coating thickness is controlled by powder feed rate, traversing speed of the torch, and the number of passes. Particle behaviour on substrate impact is determined by the velocity and viscosity of the molten powder particles. Only fully molten particles spread flat during impact and adapt to the surface topography. Adjacent and superimposed spray particles form a dense and covering layer. Even particles only partially molten at impact, are embedded in the coating. In this case, however, pores form due to insufficient particle deformation on the surface contour. Unmolten particles can reduce powder efficiency due to possible particle rebound. The so-called deposition rate describes the ratio of sprayed powder mass to coating mass on the workpiece, and represents a characteristic value for spray powder and the spraying process. In thermal spray coatings, the predominant bond mechanism is mechanical interlocking of impacting particles with the roughened substrate surface (adhesion) as well as bonding of contacting particles (cohesion). Metallurgical interaction, diffusion processes, micro-welding, or similar are less significant for the bond strength of coatings.

11.3 Standard Plasma Gun Design

The plasma torch (or gun) is the main element of every plasma spray system. Additional elements include controls, heat exchangers, powder feeders, as well as elements for gas, water, and compressed air supply. A coating system further includes handling systems for torches and workpieces as well as auxiliary equipment such as an acoustically insulated chamber and filters. This section focuses mainly on the spray gun and means of optimising spray guns. The spray gun converts the electrical energy input into heat. However, only part of the thermal energy is used for dissociation, ionisation, and heating of the plasma gas mixture. The other part is lost due to water cooling in the spray gun as well as convection of heat and thermal radiation from the housing. Initially, a high-voltage or high-frequency ignition impulse creates an ionised discharge channel in a plasma gas flow between the electrodes in order to ignite the actual arc. The direct-current voltage between cathode and anode causes formation of a direct-current arc that considerably widens the previously thin discharge channel. Continuous plasma gas flow moves the arc root towards the nozzle tip where it reaches an equilibrium condition. After ignition of the plasma gun, gas flow and electric current are increased and adjusted according to the application. Additionally, a second gas type (secondary gas, usually helium or hydrogen) is added to the first plasma gas (primary gas, usually argon or nitrogen), creating the desired temperature, enthalpy, and velocity in the plasma jet for coating. The resulting plasma characteristics affect the melting behaviour of the powder particles and thereby determine coating properties. Thus, the following factors are crucial for plasma gun performance:

11.3 Standard Plasma Gun Design

· · · ·

type of plasma gas or gas composition electric current between electrodes torch geometry heat transfer (loss) by torch cooling.

Figure 11.2 illustrates the basic design of a plasma gun, with three supply connectors in the system. The gas supply is responsible for feeding the plasma gas. When passing through the arc in between the electrodes, the gas heats, expands, and leaves the nozzle with high velocity. Here, the powder to be melted is injected by means of a carrier gas flow, is picked up by the high-speed plasma flow, accelerated, and melted. The current flows from the positive pole of the source over the anode to the gas discharge gap and via an electric arc into the cathode and back to the negative pole. Water flows through the nozzle’s coolant loop, starting at the positive pole and exiting again at the negative pole after having reached the electrode tip. The type of powder injection has a considerable influence on the melting behaviour of powder particles. Generally, powder injection should place the particles in the centre of the plasma jet. During coating, the radial injection velocity and axial plasma velocity are superimposed. This leads to different flight paths through the plasma, also affected by a number of spray-powder properties, e.g. particle-size distribution and density. Extreme temperature gradients in the temperature field of the plasma jet cause variations in the degree of particle melting. The superimposed velocities have the effect that the symmetry axes of plasma and powder cone diverge at an angle a. The angle a is a function of powder properties and injection velocity. Considering powder-injection methods, two main principles can be differentiated: axial injection, coaxial to the plasma jet, and radial injection, perpendicular to the plasma jet. Axial injection is hardly ever used as it causes a number of problems. For example, placing a powder injection tube close enough to the centre of the plasma jet in a nozzle charged as anode is impossible as no material is available that would withstand the temperatures found there. Furthermore, the demands on the spray powder are very high (close tolerances with respect to particle size fraction) in order to prevent powder evaporation inside the spray gun or clogging of the nozzle. The only commercially available spray gun with axial powder injection is a three-cathode/three-anode gun, i.e. it produces three independent arcs, and the three plasma flows join while initially passing through a special first nozzle in the divergent part of a second nozzle. Here, the powder is injected. The produced plasma powder mixture then expands in the second nozzle. The complicated design of the gun and the associated high maintenance and spare parts costs as well as special demands on the spray powder have so far prevented widespread use of this technology. In radial powder injection, internal powder injection within the nozzle (anode) and external injection outside the nozzle are differentiated. Sulzer Metco uses internal powder injection for vacuum spray guns (e.g. F4-VB or O3CP) in order to minimise vacuum influence on injection behaviour. In APS, external

165

166

11 Triplex II – Development of an Economical High-performance Plasma Spray System

injection is preferred due to easier accessibility. Additionally, positive, negative, and neutral powder injection angles are distinguished. The injection angle to the plasma jet can be varied according to the powder melting point or melting range. For processing of high-melting ceramics, powder is injected against the direction of the plasma jet (positive injection angle), increasing particle dwell time under high plasma temperature. In contrast, low-melting metals or plastics are injected in the direction of the plasma jet (negative injection angle) to reduce particle dwell time in the plasma. When different materials are used in a single process, plasma-injection positions can be adjusted, taking into consideration the particular physical properties of the different materials. Thus, material segregation during the coating process is prevented and different particles impact the surface at the same time and in the same area. The variety of available spray gun variants is due to the fact that, until now, a special spray gun has been designed for each application, substrate geometry, and performance range (necessary electrical power for melting a powder material). Spray gun engineering is far from being standardised. All systems have in common an anode/cathode arrangement that determines the formation of the arc. The longer an arc, the higher are the plasma ionisation and plasma power. The insight into this interrelationship has led to a large variety of developments aimed at increasing effective arc length and stabilising the arc, e.g. a Metco-Perkin-Elmer torch with movable cathode, schematically illustrated in Fig. 11.4. However, this torch is less common too. Since plasma spraying was introduced in 1955, a large number of spray gun designs have been developed, aimed at improving arc-travelling behaviour, e.g. changes in nozzle geometry, superimposed fields, etc. Designs are summarised in Fig. 11.5. The two principle designs are: a one-cathode plasma spray gun with a single-piece nozzle charged as anode, and a one-cathode plasma spray gun with a cascaded nozzle and anode ring. The former is widely accepted in industry. The design principle allows construction of torches with an electrical power range of several kilowatt to over 200 kW. The cathode tip is rod shaped, cone shaped, or blunted. It is made of tungsten, doped with 1 to 2% of thorium, and is embedded in a copper heat sink. The cathode is aligned to the axis

Fig. 11.4 Schematic representation of 3APG-II plasma torch (advanced plasma gun) with movable cathode.

11.3 Standard Plasma Gun Design

of the rotationally symmetric nozzle. Precise alignment (less than 0.1 mm tolerance) is crucial for uniform wear behaviour. While the arc is fixed to the tip of the finger-shaped cathode with its cathodic root, it travels across the inner diameter of the nozzle, serving as anode, with its anodic root. Axial and azimuthal (traversing radially) root movement are differentiated when investigating the motion of the anodic arc root. Axial movement, in particular, changes arc length and affects the electrical energy consumption of the torch, and has a number of disadvantages. The following list summarises the reasons for redesign of conventional spray-gun concepts: · arc root traverses randomly across the anode (axially and tangentially) · fluctuating voltage and power consumption · alternating plasma fluctuation (time dependent) · fluctuating efficiency as well as heating and acceleration of powder injected at the nozzle orifice · inhomogeneous coatings · increased power input reduces service life of cathode–anode combination considerably · limited plasma diameter · increased plasma enthalpy requires molecular gases (N2, H2) · high noise emission.

Fig. 11.5 Nozzle designs of single-cathode plasma guns. (a) Single-piece anodic charged nozzle, (b) cascaded nozzle with anode ring.

167

168

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.6 Sulzer Metco F4 plasma gun power against plasma gas composition [4].

Figure 11.6 shows the influence of plasma gas composition on the arc formation and, therefore, on the power consumption of a standard plasma gun. Inert plasma gas produces a wide-stretched arc with a large anodic arc root, exposing the anode to low thermal strain. The resulting plasma jet is stable and homogeneous but its enthalpy is too low to melt ceramic powders at economical powder feed rates. Usually, mixtures of inert and molecular gases are used in order to increase enthalpy. However, this leads to a contraction of the arc and the anodic root moves across the inner nozzle diameter rapidly in a criss-cross pattern. Thus, performance fluctuates [3].

11.4 Development of the High-performance Three-cathode Plasma Gun Triplex

In the mid-1990s, the disadvantages of standard plasma guns prompted Sulzer Metco to develop a new plasma gun concept in co-operation with the Universität der Bundeswehr, Munich (University of the Federal Armed Forces). The result was a three-cathode plasma spray gun with cascading nozzle and anode end ring, the Triplex gun. The gun was designed to operate in atmosphere as well as vacuum. The spray gun design is illustrated in Fig. 11.7. Three rodshaped cathodes are arranged on a circle around a central axis in a cathode carrier, made of electrically insulating boron nitride. The active tips stick out several millimetres from the cathode carrier. Plasma gas can be fed to the cathode rods through an annular gap to the cathode carrier. The diameter of the three cathodes is matched to the electric current load. The plasma channel, consisting of a stack of toroidal neutrodes that are electrically insulated against each other,

11.4 Development of the High-performance Three-cathode Plasma Gun Triplex

Fig. 11.7 Triplex gun with cascaded nozzle and anode ring [5].

extends from the cathodes to the anode end ring. The plasma gas is fed towards the entry zone at the neutrode closest to the cathode. Behind the entry zone, the nozzle contour initially narrows, then widens, forming a cylindrical plasmaguiding channel, leading to the terminating anode end ring. Figure 11.8 shows a different illustration of the three cathodes and cascading nozzle with the anode ring. The top right image visualises the three discrete partial arcs, beginning at the three cathode tips, and ending separately on the anode end ring. If the nozzle narrowing is too tight, the partial arcs will join to form a single arc with only one anodic arc root. This single arc would melt the anode locally and destroy it due to excessive heat input. The position of the arc roots is also fixed on the anode side. In the axial direction, the anode length allows hardly any freedom of motion. In the azimuthal direction, the arc follows the shortest path from the cathode tip to the anode, as the power consumption of the arc reaches a minimum (principle of energy minimisation) here. In single-cathode guns, due to the electrode centred on the gun axis, every point F1, F2, . . ., Fi on the inner diameter of the anode has an equal distance to the electrode tip. This is different for the three-cathode gun where the electrodes are arranged eccentrically on a circle around the gun axis. In this case, there is a uniquely defined shortest path from the electrode tip S1 to a specific point F1 on the anode ring. The arc follows this path to reach the energetically optimal energy minimum. This stabilising effect is essential for Triplex gun operation and has been confirmed by experimental data [6]. The three stable partial arcs and corresponding anodic roots lead to a plasma jet with triple symmetry that is particular apparent under low-pressure conditions. Ignition of the arcs is not covered in detail, here. Generally, plasma guns use different ignition methods such as contacting ignition, cathode–anode combines, or flashover ignition. For the Triplex, an incremental ignition method or electrical pilot arc ignition are suitable.

169

170

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.8 Principle of three-cathode plasma gun with cascaded nozzle.

Very high service life is characteristic for the Triplex gun. The Triplex can be operated for long periods of time without suffering any erosion on the anode. It has been shown that pure argon is preferable as the plasma gas, however, other gases are also possible. Long durability of the Triplex anode is a result of modified heat input into the anode walls. Local thermal load is reduced considerably compared to standard plasma guns as the total current is divided among three partial arcs. Thus, the maximum heat flow density that has to be eliminated is reduced to one ninth and melting of the anode walls is prevented. Standard plasma guns use injection causing rotating plasma gas flow in the nozzle in order to increase the velocity of the anodic arc root. The Triplex also features a gas-distribution ring with bores arranged at an angle to the gun axis to produce a swirling gas flow. Depending on the electric current and plasma gas flow, the arcs are twisted helically and stabilised. Furthermore, swirling increases heat transfer to the plasma. In principle, however, gas can also be fed to the Triplex without any rotating motion. The Triplex offers several possibilities for radial powder injection, including different angles, positions inside or outside the nozzle and different numbers of injectors. To benefit from the triple symmetry, however, it is preferable to select three injectors arranged in a way that maximum energy of the three arcs is used for melting and accelerating the powder. Appropriate powder feeding thus allows carefully controlled and stationary utilisation of the three partial arcs in this type of spray gun.

11.5 Triplex II – A New Era in Plasma Spraying Technology

Much progress has been made since the beginning of Triplex gun development. Developments were aimed at: · substituting out-of-date plasma spray technology · increasing reliability and availability · reducing materials and labour costs · increasing production capacity · reducing process and processing time · cost reduction due to higher spraying efficiency · guaranteeing constant quality while increasing productivity (higher deposition efficiency) · increasing process stability (extremely stabile arcs) · assuring constant long-term coating quality · increased coating quality (no electrode and nozzle residue within the coating). Development of the Triplex II originated here and brought the following improvements: · same basic principle as Triplex I (performance of Triplex I: 24 kW, powder feed rate for yttria-stabilised zirconia, YSZ: 100 g/min, deposition efficiency: 60%) · improved gun design and cooling for higher power levels (up to 55 kW) · larger anode rings, allowing voltages above 100 V · two nozzle diameters available, producing different gas velocities – 9 mm for dense coatings (e.g. Cr2O3) – 11 mm for porous coatings (e.g. thermal-barrier coatings, TBC). Today, a Triplex II is offered as a sealed gun together with a service contract for professional maintenance. Thus, operating times of up to 200 h at constant performance can be guaranteed.

11.5 Triplex II – A New Era in Plasma Spraying Technology

In 1998, Triplex II, a new revolutionary plasma spraying technology was introduced to the market. More than 25 Triplex II systems have been installed in Europe, the United States, and Japan, since. Industrial feedback reflects considerable cost reduction due to less downtime, high availability, high deposition efficiency and, thus, increased productivity. The Triplex II plasma spraying technology is aimed at increasing service life of wearing parts, reducing porosity of coatings if desired, and considerably increasing production output. Furthermore, noise emission is reduced to less than 90 dB(A) and, thus, employees are exposed to considerably less noise. Unlike the arcs in standard plasma guns, the arc roots remain stationary. This reduces plasma fluctuation and turbulence within the plasma jet. Usually, this turbulence causes inhomogeneous melting and processing of injected powder, result-

171

172

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.9 Cross section of new Triplex II plasma gun.

ing in reduced coating quality and coating process efficiency. In contrast, minimised arc fluctuation increases plasma jet stability, improves melting behaviour and, finally, yields considerably higher quality of deposited coatings. Figure 11.9 shows a cross section of the new Triplex II plasma gun. The redesigned basic principle meets the requirements in terms of high performance due to several improved features. As in the Triplex, direct current is divided and fed to three water-cooled tungsten cathodes. But compared to Triplex I, their design is larger. Two nozzle kits are available for Triplex II. Each includes a neutrode serving as an entry zone, seven neutrode rings, insulated against each other with boron nitride rings, and an anode ring with a tungsten bush. The two types of nozzles differ in the diameter of the narrowing. Diameters are varied in order to control the velocity of the plasma flow. The nozzle kit with a diameter of 9 mm is used for dense coatings, e.g. chromium oxide coatings (Cr2O3) for anilox rollers, when high plasma velocity is desired. Highly porous yttrium-stabilised zirconium oxide coatings (ZrO 8Y2O3) for thermal-barrier applications are deposited using the 11-mm nozzle diameter to reduce plasma velocity. Apart from nozzle diameters, other coating parameters are essential, e.g. electric current, gas flow, powder feed rate, spray distance, and traversing speed of the torch. Also, the type of powder injection should be considered as Triplex II offers two different injector carriers. Both can carry three injectors, arranged at 1208 angles, but differ in the injection angle. 908 and 1058 injectors are available. Both injectors are water cooled in order to prevent powder particles from sticking due to the high-temperature plasma jet. Compared to Triplex, the Triplex II nozzle is longer, thus increasing the distance between anode and cathodes. The arcs, therefore, span a greater distance, which results in higher plasma jet enthalpy for Triplex II. A common overall electrical current for the system is in the range of 250 to 600 A. Here too, dividing the current among three separate arcs reduces current density and increases

11.5 Triplex II – A New Era in Plasma Spraying Technology

Fig. 11.10 Triplex II – cutting coating time in half by increasing powder feed rate and deposition efficiency (DE) compared to conventional plasma guns.

anode and cathode service life. Exclusive use of the noble gases argon and helium that, in the Triplex II, are also fed to the nozzle via a gas-distribution ring with outlets arranged at an angle, further increases the service life of the corresponding wearing parts. High productivity of the Triplex II is a result of high adjustable powder feed rates, good deposition efficiency, as well as associated high deposition rates. Depending on the industrial application, Triplex II offers two modes of operation. For coating of many small parts it is advisable to use only a single powder injector for optimal targeting. For metal bond coats, one injector is usually sufficient too. When large parts are coated, all three injectors provide a higher feed rate and produce a larger deposition area so that the processed area is maximised. Figure 11.10 compares production-related parameters of a standard plasma gun and Triplex II. The advantages of the Triplex II are obvious. Spraying time is usually reduced by 50% compared to standard guns, in certain applications only 20 to 30% of standard technology spraying time is necessary. These exceptional characteristics allow cost reduction at constantly high quality. Figure 11.11 illustrates the Triplex II plasma spraying system, as offered since 2003. It includes the spray gun, a controller, a JAM box (junction and monitoring), an electric power supply, the powder feeder, heat exchanger, and cooler. Controllers and powder feeders can be selected according to individual demands. Apart from the described basic components, Sulzer Metco provides individually matched handling systems for spray guns and parts to be coated.

173

174

11 Triplex II – Development of an Economical High-performance Plasma Spray System

Fig. 11.11 Components of Triplex II plasma spray system (allows selection of controller and powder feeder).

11.6 Positive Feedback from Industry 11.6.1 Chromium Oxide Coating of Anilox Rollers for Printing Industry

Tightened environmental requirements have pressed the printing industry to use more and more water-soluble paints and reduce quickly drying solvents, even for cleaning. Printing units, therefore, have to cope with new demands, and at the same time continuously increase production output and quality of printed products. Modern flexographic printing facilities fulfil these requirements. The central element, here, are so-called anilox rollers that pick up paint from a pressurised container and dose it precisely to the printing cylinders. Anilox rollers feature a structured surface with many small cavities, referred to as ink cells. The cells are arranged on the roller in a linear array. Every cell corresponds to an individual colour dot and, thus, the sharpness of the printed image is determined by the number of cell lines per inch (lpi) on the roller. The number of lines is limited to approx. 500 lpi for chromium-plated, mechanically knurled anilox rollers. Therefore, use of laser-engraved rollers with ceramic coatings is spreading. Plasma-sprayed, subsequently polished and laser-engraved chromium oxide coatings can yield up to 1200 lpi. Furthermore, plasma-sprayed chromium oxide coatings feature higher wear and corrosion resistance than electroplated chromium coatings and show optimal wetting behaviour, thus

11.6 Positive Feedback from Industry Fig. 11.12 Polished cross section of typical thermal spray Anilox roller coating with NiCr bond coat and Cr2O3 top coat.

minimising the amount of colour necessary for printing. Less colour requires less drying time. Successful laser engraving requires extremely dense coatings as the diameter of the produced ink cells is only approx. 70 lm. Additionally, coating impurities can scatter the laser beam and cause faulty ink cells, which ultimately lead to a defective roller. Sulzer Metco has developed several solutions in order to increase the economic efficiency of chromium oxide coating on anilox rollers. On the one hand, these include a new chromium oxide spray powder with optimised powder flowability, higher density, and increased deposition efficiency. On the other hand, corresponding spray parameters were determined for Triplex II, allowing a threefold rise in performance for thermal sprayed anilox rollers compared to treatment with standard plasma guns and, furthermore, producing less costs per unit. Figure 11.12 shows a polished cross section of a Triplex II sprayed coating for anilox rollers with the typical structure of an approx. 80 lm NiCr bond coat (Metco 43F-NS) and an approx. 300-lm Cr2O3 layer (AMDRY 6417). The chromium oxide layer has less than 2% porosity and hardness of 1300 HV0.3. After machining, the coating had surface roughness values of Ra = 0.09 lm. 11.6.2 Abradable Coatings

Abradable coatings are used as seals in stationary gas turbines and aircraft jet engines. They reduce the clearance between blade tips and coated compressor or turbine housings to a minimum. Usually, the inner diameters of housings are coated with relatively soft abradable materials, whereas abrasive material is applied to the blade tips. When the turbine is started for the first time, the blades expand due to rising temperature and centrifugal force, and cut sealing

175

176

11 Triplex II – Development of an Economical High-performance Plasma Spray System

grooves into the coating. During operation, these minimum sealing grooves prevail and reduce leakage flow. They allow controlled leakage of the gas flow inside the turbine and keep pressures and temperatures at optimal levels. Use of abradable coatings has significant advantages, e.g. increased efficiency and reduced fuel consumption. Abradable coating materials for low (350 8C) to moderate (600 8C) operating temperatures usually include three components, a metal matrix, a solid lubricant, and polyester. The metal matrix yields oxidation and corrosion resistance and determines the maximum operating temperature of the coating. Polyester is used to control porosity. Resistance against abrasion and erosion is determined by the size and the amount of plastic particles embedded in the coating. Increasing polyester content and porosity improves the abrasive behaviour and reduces erosion resistance. Graphite or hexagonal boron nitride, often referred to as white graphite, are embedded in the coating as solid lubricants and facilitate cutting of the blade tip into the coating. To prevent damage to the turbine, particles released from the coating should not be too large. During coating, a segregation and formation of layers within the coating can occur due to the varying densities and particle sizes of typical abradable coating material components. During injection, large and heavy particles pass through the plasma jet, while small and light particles do not reach the plasma. The result is that the plasma jet separates individual material components and, depending on the torch motion, impact the part surface with a time-dependent offset, and form several layers during a single pass of the torch. This problem is avoided when using the Triplex II. It is assumed that the three arcs initially form three independent plasma jets that do not join until they reach a certain distance from the spray gun orifice. The distance is determined by the swirl magnitude of the gas flow and the electrical current. Usually, spray material

Fig. 11.13 Formation of three plasma flames and quasi-axial powder injection in Triplex II.

11.6 Positive Feedback from Industry

should be injected directly into the plasma jet. For abradable coatings, however, injecting the material between the jets guarantees uniform material distribution in the plasma which is also improved by the swirling gas flow. The method therefore features a quasi-axial injection with the Triplex II, as shown in Fig. 11.13. 11.6.3 Thermal-barrier Coatings

Ceramic thermal-barrier coatings with extremely low thermal conductivity below 1 W/m K reduce the thermal load on metallic base materials. To date, such coatings serve mainly as thermal-barrier linings in combustion chambers as well as for protection of stators and rotating blades in turbine engineering. They prevent direct contact between base materials and hot combustion gas with temperatures of up to 1500 8C. Typically, thermal-barrier coatings combine the actual thermal-barrier coating with a hot-gas corrosion and oxidation preventing coating. Also applicable are thermally grown oxide coatings and diffusion barrier coatings. Thermal sprayed thermal-barrier coatings usually involve partially stabilised zirconia as coating material, i.e. a stabilising agent is added to zirconium oxide in its tetragonal, high-temperature modification, preventing or reducing the formation of the monoclinic phase with higher specific volume below 1170 8C. Yttrium oxide (Y2O3) is a typical stabilising agent, yielding partial stabilising at contents of 7%, and complete stabilising of the high-temperature phase at contents of 20%. When part cooling is sufficient, thermal-barrier coatings can reduce the thermal load on a part by up to 400 8C, depending on coating thickness and porosity. The higher the porosity, the better for the thermal

Fig. 11.14 Thermal-barrier coating sprayed with Triplex II.

177

178

11 Triplex II – Development of an Economical High-performance Plasma Spray System

barrier effect. However, considerably reduced powder deposition efficiency (40 to 50%) must be considered when spraying highly porous coatings in the range of 15–20%. Dense zirconia coatings can be sprayed at powder deposition efficiencies of up to 70%. Figure 11.10 compares performance values of standard spray guns and Triplex II for partially stabilised zirconia. Figure 11.14 shows a cross section of a thermal-barrier coating sprayed with Triplex II and Metco 204C-NS powder, and lists the corresponding spray parameters as well as coating and process values. 11.6.4 Further Applications

Sulzer Metco, in co-operation with its customers, continuously develops new fields of applications for the Triplex II.

11.7 Summary

This chapter describes the development of an economical high-performance plasma spray system for highest-quality demands under extreme production conditions. The basic principles of plasma spraying are covered. The design of standard plasma guns is investigated, and demands for development of a threecathode plasma gun are concluded. Development stages leading to state-of-theart Triplex II gun technology are described. Application examples from industrial practice demonstrate the superior properties of the new Triplex II technology.

References Ingham, H. S., Shepard, A. P.: Flame Spray Handbook, Volume III, Plasma Flame Process. Published by Metco Inc., Westbury, Long Island, New York, 1965 2 Lugscheider, E.: Beschichtungstechnik. Lecture Notes for Materials Science Specialisation in Mechanical Engineering Studies at RWTH Aachen University, 1994 3 Coudert, J.-F., Planche, M.-P., Betoule, O., Fauchais, P.: Study of the influence of the arc root fluctuations on a DC plasma spray torch. ISPC11, Ed. J. Harry, 1993 4 Barbezat, G., Zierhut, J., Landes, K. D.: Triplex – a high performance plasma 1

torch. United Thermal Spray Conference and Exposition, UTSC99, Düsseldorf, March 17th–19th, 1999, Conference Proceedings, Eds. E. Lugscheider, P. A. Kramer. DVS, Düsseldorf, 1999, pp. 271–274 5 Haselbeck, P.: Entwicklung eines Dreikathoden-Plasmabrenners unter Anwendung adaptiver plasmadiagnostischer Methoden. Doctoral Thesis, University of the Federal Armed Forces, Munich, 1995 6 Landes, K. D., Forster, G., Zierhut, J., Dzulko, M., Hawley, D.: Computer Tomography of Plasma Jets – Applied on a Triplex II Torch. To be published on ITSC 2004, Osaka, Japan, May 10th–12th, 2004

179

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying W. Krömmer, P. Heinrich, Linde AG Business Segment Linde Gas 12.1 Introduction

The first applications have proven that cold gas spraying is a new, trend-setting technology. The ability to produce dense, oxide-free coatings opens new vistas in areas, industries, and applications, not yet tapped by thermal spraying today. In a strategic partnership, CGT Cold Gas Technology GmbH, Prof. Kreye at the University of the Federal Armed Forces Hamburg, and Linde AG Business Segment Linde Gas, designed an advanced system based on process-oriented research and development. Co-operative, practical research is progressing, and results are transferred to applications in order to maintain state-of-the-art, efficient, and application-oriented systems.

12.2 System Design 12.2.1 Pressure Tank and Nozzle

Most progress has been made in optimising the nozzle [6], the most significant part of the system. The Laval nozzle (Figs. 12.1 and 12.2) features an optimised geometry and new material. As a result, production of complicated, two-piece nozzles is dispensable, and service life increases by 200%. The new nozzles can operate at higher temperature without causing powder particles to stick to the nozzle surfaces. Higher temperature and thus increased gas and particle velocity boost deposition efficiency by 20%, e.g. from 65 to 85% for copper. The nozzle attaches to the pressure tank with a quick-release fastener. Relevant process parameters such as pressure and temperature are recorded at the pressure tank and sent to the control unit in order to guarantee a stable process. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

180

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying Fig. 12.1 Laval nozzle.

Fig. 12.2 Original cold gas gun with gas supply, pressure sensor, and temperature measurement.

12.2.2 Control Unit

The control unit (Fig. 12.3) divides into two separate parts. While the left part holds the entire electrical system, gas regulators, valves, and safety devices are arranged on the right side. The central component [3] of the unit is a Siemens Simatic S7-300 SPC including all necessary controls. The integrated TS (TeleService) adapter allows telediagnostic service, remote programming, and parameter transfer via ISDN or modem and an analogue telephone line, providing immediate assistance to the user. For example, spray parameters developed in external laboratories can easily be made available to the user via a wired network. The system presented here operates with N2, He, or any combination of the two.

Fig. 12.3 Control unit, general view.

12.2 System Design Fig. 12.4 Main view, touch screen.

Fig. 12.5 Powder feeder.

12.2.3 Touch Screen

Operators control the system with the integrated Simatic Multipanel that includes a 12-inch colour display (Fig. 12.4). The touch screen holds five integrated interfaces for communication with external PCs, a network, or printers. The main image shows a schematic diagram with all relevant system components and corresponding measured values. Additional masks allow parameter input for all attached equipment.

181

182

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying Fig. 12.6 Parameter set.

Fig. 12.7 Setting desired values.

12.2.3.1 Main Mask Parameters Heating-coil temperature gas temperature at heater output gas temperature at the nozzle pressure at heater output pressure at the nozzle valve conditions (open or closed in different colours) gas consumption N2 primary gas He primary gas carrier gas transformer power

12.2 System Design

For coating different materials, specific parameters can be stored to a programme number and archived for simple parameter change. The fully functional programmes can be recalled and executed any time to guarantee reproducible coating processes (Fig. 12.6). 12.2.4 Gas Heater LINSPRAY®

With experience gathered in other fields of gas technologies, a gas heater was developed that can heat 90 m3 of gas to 700 8C in less than 2 minutes (Fig. 12.8). A high-temperature resistant CrNi steel is used as heating coil material.

Fig. 12.8 LINSPRAY® gas heater.

Fig. 12.9 Glowing heating coil, 800 8C.

183

184

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying

High-grade insulation prevents heat loss (Fig. 12.9) and appropriate safety devices avoid overheating of the heating coil. Safe operation is thus guaranteed at all times. Extensive investigations showed no sign of fatigue after more than 1000 hours of operation. 12.2.5 Gas Supply for Cold Gas Spraying

Apart from designing the heating element, the development group also focused on gas supply. Nitrogen, helium, or mixtures of the two are the most common process gases in cold gas spraying. Cu coatings produced at deposition rates greater than 85% with nitrogen, a pressure of 28 bar, and a temperature of 420 8C are absolutely dense. Results of LDA measurements (laser Doppler anemometry) of particle velocities at Linde

Fig. 12.10 Cold gas sprayed copper coating.

Fig. 12.11 DESYTM, pressure increase.

12.2 System Design

AG [5] ranged from 500 to 700 m/s. Helium increases particle velocities to 600– 1200 m/s at a deposition efficiency (DE) of 95% (Fig. 12.10). Cold gas spraying systems are designed for pressures of up to 35 bar. Gas supply, therefore, is subject to a number of considerations. Two options are available for tank-based gas supply: Either an installation featuring a high-pressure tank combined with a highpressure evaporator or, alternatively, with a low-pressure tank followed by a pressure boost and high-pressure evaporator. Pressure is raised by using hydraulic cylinders, installed outside, next to the tank (Fig. 12.11). This technique allows for feeding high-purity products at pressures of up to 300 bar. The system operates without storage tanks, as used in cryogenic pumps, as operation is limited to periods when N2 is needed for the process. 12.2.6 Helium Recovery

Due to high particle velocities, helium cold gas spraying allows processing of critical materials [1]. For helium use, Linde AG modified a helium-recovery system, common in other applications in order to reduce resources and coating costs of cold gas spraying. The process should be sealed to prevent contamination (Fig. 12.12). Waste gas flows through a particle filter and, if necessary, a cooler. A compressor feeds the gas to the main part of the system, a highly sophisticated membrane equip-

Fig. 12.12 Gas supply for cold gas spraying.

185

186

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying

ment, which releases the gas back to the storage tank at 99% purity. New helium is required only to compensate for gas lost in the process. The entire, fully operational system is mounted inside a 20-foot container. Costs for helium and the recovery systems are high and should be considered carefully for each application. Gas loss is the main expense factor, and determined by the type and size of parts as well as batch-change intervals.

12.3 Applications

Cold gas spraying is appropriate for applications where electrical and thermal conductivity are crucial (Figs. 12.13 and 12.14). Here, cold gas sprayed copper coatings (Fig. 12.8) are ideal. Oxide-free copper coatings, usually only producible by means of VPS, provide non-porous structure with electrical conductivity reaching 95% of rolled copper. The thermal conductivity is comparable, and it is already used in certain applications. An additional advantage is that coating of aluminium substrate is simplified as blasting pre-treatment is expendable. Particles develop a high mechanical bond to the surface, producing a strong interface between substrate and coating. Cold gas spraying is also popular in applications where corrosion is an issue. Coating materials include zinc (Fig. 12.15) for active corrosion protection, nickel (Fig. 12.16), and stainless steel for passive protection. Both alternatives benefit from the oxide-free structure and low porosity. Different thermal spray coating materials can be combined. Figure 12.17 shows a plasma-sprayed Al2O3 oxide ceramic insulating layer with cold gas sprayed, electrically conductive copper top coat.

Fig. 12.13 Cu for contacting surface.

12.3 Applications Fig. 12.14 Cu for heat transfer.

Fig. 12.15 Zinc.

The focused jet in cold gas spraying reduces the need for masking, e.g. for protection of cutting edges, laser weld seams, or, as shown in Fig. 12.18, a row of spot weldings on zinc-coated sheet metal. A subsequently applied 15-lm zinc coating seals the surface in order to prevent corrosion.

187

188

12 System Technology, Gas Supply, and Potential Applications for Cold Gas Spraying Fig. 12.16 Nickel.

Fig. 12.17 Protected weld seam.

12.4 Summary

Research and development, but industry as well, depends on sophisticated system technology. Successful transfer of knowledge and technology to initial applications led to series coating systems for automotive and electrical engineering. Furthermore, results show that the produced coatings meet all requirements and that the system is effectively ready for series production.

References Fig. 12.18 Al2O3 with Cu.

Design and structure of cold gas sprayed coatings feature properties new to thermal spraying and, therefore, open new fields of applications for surface technology. The development team introduced above is continuing research and development in co-operation with coating service providers in order to broaden the range of applications and materials for this process, and make them available to industry. References Werner Krömmer, Peter Heinrich: Linde Gas AG Unterschleißheim. What influence does the purity of industrial gases have on the quality of thermal spraying? ITSC Singapore, 2001 2 Carina Rickfält, Werner Krömmer, Peter Heinrich: Linde Gas AG Unterschleißheim. Moderne Gasversorgung für das Thermische Spritzen, ITSC 2002 Essen 3 Richter, P., Ampfing, D., Krömmer, W., Heinrich, P., Unterschleißheim, ITSC 2002 Essen. Equipment Engineering and Process Control for Cold Spraying. Anlagentechnik und Prozesssteuerung beim Kaltgasspritzen 1

Voyer, J., Stoltenhoff, T., Prof. Kreye, University of the Federal Armed Forces, Hamburg. Development of Cold Gas Sprayed Coatings 5 Dipl. Ing. Alfred Reusch: Die Entwicklung eines Laser-Doppler-Meßsystems und seine Anwendung bei Verfahren des Thermischen Spritzens 6 Thorsten Stoltenhoff: Kaltgasspritzen von Kupfer. Eine strömungsmechanische und werkstoffkundliche Analyse und Optimierung des Spritzprozesses 4

189

191

13 Diagnostics in Thermal Spraying Processes J. Prehm, K. Hartz, Institute for Materials Science, University of Hannover, Germany 13.1 Introduction

Primarily, properties of thermal spray coatings are functions of the thermal and kinetic energy of spray particles. Measurement and careful control of these values\kep can yield coating properties that are precisely matched to individual applications. During the past years, a number of diagnostic devices for measuring particle velocities, sizes, and temperatures were developed. Further parameters that influence particles in many thermal spraying processes are hot gas velocity and temperature. Next to process-optimising measures, online process diagnostics are gaining importance for the production of coatings with constant quality.

13.2 Classification of Diagnostic Methods

In this chapter, not all available diagnostic methods can be covered in detail. Therefore, an overview of individual techniques and measured values is given in Table 13.1.

13.3 Methods for Particle Diagnostics 13.3.1 Laser Doppler Anemometry (LDA)

Laser Doppler anemometry is based on the fact that coherent light waves show a Doppler shift when scattered by moving boundary layers, and therefore, carry velocity information. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

192

13 Diagnostics in Thermal Spraying Processes Table 13.1 Diagnostic methods. Plasma/hot gas Measured values Enthalpy Temperature Velocity Viscosity

Processes/methods Enthalpy probe Thermography Schlieren method Laser two-focus method Emission spectroscopy Pyrometry Laser Doppler anemometry

Particle

Substrate

temperature velocity trajectories particle density melting behaviour shape

surface temperature part temperature temperature distribution

laser two-focus DPV 2000 LDA particle image velocimetry particle shape imaging phase Doppler anemometry

pyrometry thermocouples thermography

In laser Doppler anemometry, appropriate beam-splitting optics separate laser light waves into two partial waves. A convex lens focuses the two partial waves. The point of intersection in the focal point of the lens is referred to as the measurement volume. In this measurement volume, the laser beams interfere and form interference fringes. The distance x between interference fringes is a function of interference half-angle a and light wavelength k. Particles following a flow reflect or scatter the interference fringes within the measurement volume. A detector in the vicinity collects these frequencies which correspond to the particle’s velocity component perpendicular to the interference fringe pattern. For fixed half-angles (a) of crossing laser waves and fixed wavelengths of laser light (k), it is simple to determine the frequency range in 1/s, or the velocity by the measuring signal frequency (fDoppler), Figs. 13.1 and 13.2. Particle velocity is calculated by mx ˆ d
fDoppler

…1†

and 1 TD

…2†

k 2
sin a

…3†

fDoppler ˆ dˆ

The Doppler frequency is independent of the direction a particle passes through the measuring volume. Therefore, inversions of direction, occurring frequently

13.3 Methods for Particle Diagnostics Fig. 13.1 Interference-fringe model.

Fig. 13.2 LDA signal.

in turbulent flow, remain undetected. Also, particles moving parallel to the interference fringe pattern do not produce viable signals. An acousto-optical modulator (Bragg cell) integrated into the system shifts the frequency of the laser light in one of the two separate beams by fb and allows detection of the particle direction. The frequency difference of the two laser beams causes a continuous shift of the interference fringe pattern in the measuring volume. Particles moving through the measuring volume in the same direction as the interference fringe pattern produce a lower-frequency signal f = fB–fDoppler, while particles passing through in the opposite direction produce a higher-frequency signal f = fB + fDoppler. Linear proportionality between the velocity reference value, the frequency, and the velocity increases the transparency of LDA compared to other detection methods and has many advantages in practice [1, 2]. An LDA system (Fig. 13.3) comprises the laser, the optical system, a detector unit (photomultiplier), as well as a signal processing and analysing unit. Radiation from the actual process is an important factor when selecting an appropriate diagnostic system and laser power. When detecting scattered light with the collector optics, interference of background radiation must be considered. Most cases require very narrow band filters, carefully matched to the laser wavelength, in order to yield acceptable results. Use of this type of filter is advisable and usually inevitable in all laser optical techniques detecting scattered light.

193

194

13 Diagnostics in Thermal Spraying Processes Fig. 13.3 LDA measurement setup.

13.3.2 Phase Doppler Anemometry (PDA)

Many years of development went into equipment for simultaneous detection of particle size and velocity. Most methods involved enhanced interpretation of detected signals (bursts) in laser Doppler anemometry. Introduced just recently, PDA (phase Doppler anemometry) features considerably improved quality and reliability. Compared to earlier particle-measuring techniques, PDA has a number of advantages: broad dynamic range from micrometre to millimetre scale particles, high accuracy, and high immunity to optical interference. An important disadvantage, however, is that size measurements are limited to spherical particles. Phase Doppler anemometry is based on the principle of classic laser Doppler anemometry. In addition, the phase position of the scattered light representing particle size data is interpreted. The surface of non-transparent particles reflects the laser light. Detector B, and subsequently detector A, collects the scattered light. The resulting phase shift – u of the scattered light carries reliable size information of particles inside the measuring volume (Fig. 13.4). The PDA method for thermal spray technology diagnostics is not yet well established. Preconditions for the technique, e.g. spherical particles and known index of refraction, still prevent broad use in coating technology. However, PDA is used successfully in other areas of materials processing such as spray forming [3].

Fig. 13.4 Reflection at particle – phase shift.

13.3 Methods for Particle Diagnostics Fig. 13.5 Optical setup of laser two-focus system.

13.3.3 Laser Two-focus Method (L2F)

In the laser two-focus method, two laser beams are produced by means of a Rochon prism and focused at different points. The focal points act similar to a light barrier in the flow. Two photomultipliers detect the light scattered by the particles that pass these measuring points. A particle passing through the two measurement volumes creates a starting and a stop signal that, amplified and electronically enhanced, allow calculation of the time difference between the two [4, 5]. The advantages of the laser two-focus method are substantial compared to laser Doppler anemometry: · particularly good suppression of stray light · considerably smaller measuring volume compared to LDA, and therefore, strong backscattering on extremely small particles · low laser power due to reduced measurement volume · interference immunity · low wall distance attainable · larger velocity measurement range (1–2000 m/s). Figure 13.5 schematically illustrates the experimental setup for measuring velocities with an L2F system. 13.3.4 Particle Image Velocimetry (PIV)

Particle image velocimetry features a measuring area of several square centimetres rather than punctual measuring volumes characteristic to the abovementioned methods. Therefore, PIV can visualise phenomena such as turbulence in the jet [6, 7]. In particle image velocimetry, two laser flashes in quick succession illuminate travelling particles. A digital camera records the backscattered particle light. The images are recorded with one of the following two methods. Either two separate

195

196

13 Diagnostics in Thermal Spraying Processes Fig. 13.6 Basic principle of PIV technique.

Fig. 13.7 Particle image velocimetry in arc spraying (Sulzer Metco SmartArc, U = 26 V, I = 200 A, pz = 2 bar).

exposures, i.e. one per laser flash (cross correlation), or one double exposure (autocorrelation) is used. The main difference between auto- and cross correlation is that autocorrelation yields the magnitude and direction of the velocity, but not the orientation. Cross correlation, however, determines this value too. The two separate images necessary for measuring produce twice as much data and require more sophisticated camera technology. In thermal spraying, the orientation of the particle jet is a known value, and thus, considering the above, usually autocorrelation techniques are used [8]. In both cases, a single particle appears as a pair of particles. Velocity rates and, for known flight orientation, velocity vectors are calculated by measuring the distance between particles of a pair and the period of time between firing the two laser flashes (see Fig. 13.6). Figure 13.7 shows a PIV image recorded during arc spraying. Autocorrelation was used for imaging, and thus, particle pairs appear in the image [9, 10]. Typical PIV equipment usually includes two lasers. Via fibre optics, the laser light is guided to prismatic optics equipment, the flat beam illuminator (FBI), which fans out the laser spot into light areas, illuminating the flow field. A band-pass filter in the digital camera optics, matched to laser wavelength, mostly inhibits false exposure, e.g. due to combustion processes. Synchronising

13.3 Methods for Particle Diagnostics

Fig. 13.8 PIV measurement setup.

of the laser flashes, exposure, as well as image acquisition is computer controlled (see Fig. 13.8). 13.3.5 In-flight Particle Diagnostics

One of the characteristic parameters in thermal spraying is particle temperature, however, traditional diagnostic techniques are inadequate for its sufficiently precise measurement. On the one hand, high background radiation from the gas jet impedes high-resolution measurement. On the other hand, separating a particle jet for high-speed, single-particle imaging is difficult in terms of imaging and analysing technology. The Tecnar DPV-2000 system is capable of simultaneously analysing surface temperature, velocity, and size of up to 200 individual particles per second at one point in the spray jet. The surface temperature is determined using twowave-length pyrometry at 995 ± 26 lm and 787 ± 25 lm. As the particles are considered non-selective radiators, the system can detect surface temperatures in the range of 1350 8C to 4000 8C. Particle size is estimated by measuring absolute energy radiated by a particle, and is limited to a range of 10 lm to approx. 300 lm. The velocity measurement uses the light barrier principle, an analogue to the technique in L2F [11]. Figure 13.9 shows the principle of in-flight particle diagnostics.

197

198

13 Diagnostics in Thermal Spraying Processes Fig. 13.9 Measuring principle DPV2000.

13.4 Methods for Plasma and Hot Gas Diagnostics 13.4.1 Enthalpy Probe Diagnostics

The enthalpy probe method characterises the hot gas jet in terms of enthalpy, temperature, and velocity. The principle of enthalpy probe measurement is based on a two-stage energy balance of enthalpy probe cooling water. After placing the probe into the gas jet, the cooling water heats due to heat transfer from the gathered gas volume and from the gas jet flowing around the probe. Hence, a sealed probe (non-flowing, index nf) is used for the initial reference measurement. Subsequent measurements use an open probe (flowing, index f), i.e. gas is gathered. The temperature increase of the cooling water and additionally, when gathering gas from the sample flow, the gas temperature at the output of the probe (Ts) are recorded for both measurements. The specific enthalpy of the gas jet is calculated from the difference of cooling water temperature increases, using the following equations: hp ˆ hs ‡

_ w DT†f …m _ w DT†nf † Cpw ……m _g m

…4†

and hs ˆ cps
Ts

…5†

13.5 Methods for Online Process Control

_ m Cpw Cps Ts DTnf DTf _g m

= rate of cooling water flow in kg/s = specific heat capacity of the water in kJ/kg 8C = specific heat capacity of the gas in kJ/kg 8C = temperature of gathered gas after leaving the probe in 8C = temperature increase of cooling water during measurement without gathering any gas (no-flow regime) in 8C = temperature increase of cooling water during measurement with gathering gas (flow regime) in 8C = rate of gas flow during measurement in kg/s.

If necessary, the gas temperature is determined from the calculated enthalpy value and gas composition. Furthermore, the dynamic gas pressure allows calculation of the velocity of the gas jet at the measuring spot: s 2…Pnf Ps † vˆ qg

…6†

Pnf = absolute pressure, measured by the probe with gas valve shut in Pa Ps = static pressure in measuring chamber in Pa qg = gas density at the tip of the probe in kg/m3. When investigating ultrasonic flow, Eq. (6) must be modified. For plasma spraying, too, enthalpy probes yield reliable gas-temperature values in areas where the degree of ionisation has dropped so far that spectroscopic methods would, if at all, produce extremely inaccurate results of temperature distribution. At the same time, enthalpy probes allow evaluation of spectroscopic results for temperatures of up to 10 000 8C, and can be used to examine equilibrium conditions of the plasma [12, 13].

13.5 Methods for Online Process Control

Previous methods for quality control of thermal sprayed coatings involve timeconsuming and cost-intensive destructive testing of coatings. Online process control has been established in the area of thermal spraying in line with cost reduction and rising quality demands in recent years. In the meantime, a number of commercially available systems have become common to industry. The following chapter describes particle-flux imaging (PFI) as an example of a process-control system.

199

200

13 Diagnostics in Thermal Spraying Processes

13.5.1 Particle-flux Imaging (PFI)

Particle-flux imaging is applicable to online process control of plasma spraying and high velocity flame spraying processes. Here, the process control of an atmospheric plasma spray process is described. A stationary camera (Fig. 13.10) captures the area between the torch orifice and the substrate. The complete region of the plasma and particle jet spectrum is recorded. Relatively long exposures of approx. 1000 ms cause the plasma and the particle jet to appear as diffusely luminous areas of brightness. Individual particle trajectories and short-term variations of plasma and plasma jet are undetectable. Two neutral filters with different transmission values, s1 and s2, allow simultaneous image acquisition of the light-intensive plasma and the less-brilliant particle jet with the camera system. The method produces a dual image of the coating process that contains all process-relevant data (cf. Fig. 13.11). A patented analysis algorithm allows discrete interpretation of the plasma jet and the particle jet. Thus, the plasma gun condition and the properties of the depositing particle jet are analysed independently. Acquired images show the plasma jet on the left side and the particle jet on the right side (cf. Fig. 13.12). The analysis algorithm initially determines lines of constant plasma-jet and particle-jet intensity (determination of contours). In the following step, ellipses are fit to the contours. A single plasma-jet ellipse and a single-particle-jet ellipse are sufficient for precise process characterisation. Therefore, the following analysis uses only two ellipses.

Fig. 13.10 PFI camera system.

13.5 Methods for Online Process Control Fig. 13.11 Principle of image acquisition.

data reduction: ellipse approximation

Fig. 13.12 Principle of PFI image analysis.

· · · ·

Each ellipse is characterised by five parameters: ellipse centre co-ordinates sx (x) and sy (y) length of major axis a length of minor axis b angle } (phi) between axis a and x-co-ordinate.

During periods of optimal process conditions, two appropriate ellipses (reference ellipses) are calculated and stored. The actual process is monitored by

201

202

13 Diagnostics in Thermal Spraying Processes

comparing continuously calculated ellipses with the reference ellipses. When ellipse parameters exceed or fall short of predefined threshold values they indicate deviations of the actual process from the desired process. An error detection then triggers an alarm signal. The possible cause of the error can usually be suggested as most process errors create characteristic ellipse-deviation patterns [14, 15].

13.6 Summary and Conclusions

Diagnostics allow optimising complete thermal spray processes by providing fundamental knowledge of process-relevant properties and determination of energetic particle conditions. Furthermore, coating quality and process reliability are enhanced. The applied methods have partially been known for years and are well established. In recent years, new online process control systems were developed that provide competitive alternatives for quality assurance of thermal sprayed coatings.

References 1 Ruck, B.: Laser-Doppler-Anemometrie; 2

3

4

5

6

AT-Fachverlag GmbH, Stuttgart, 1987 Mayr, W., Henne, R.: Investigation of a VPS-Burner with Laval Nozzle using an Automated Laser Doppler Measuring System; Proceedings 10th Plasma-Technik-Symposium Luzern (1988) Sultan, G., Schulte, G., Bauckhage, K.: PDA-Messungen von Tropfengrößenverteilungen in optisch dichten Sprays, proceedings GALA 2000, München (2000), pp. 50.1–50.7 Thompson, D. H.: A Tracer Particle Fluid Velocity Meter Incorporating a Laser; Journal of Scientific Instruments, 2, 1 (1968) Kugler, H. P.: Two Spot Measurement in High Temperature and High Speed Flows; Proceedings of the Symposium on Long Range and Optical Velocity Measurement, ISL, German-FrenchResearch Institute, St. Louis, France (1980) Bach, Fr.-W., Copitzky, T., Prehm, J., Duda, T., Versemann, R.: Untersuchung Thermischer Spritzverfahren mit der Particle-Image-Velocimetry Lasermeß-

methoden in der Strömungsmesstechnik, Ed. A. Delgado, F. Werner, B. Ruck, A. Leder, D. Dopheide, Shaker, Aachen, pp. 40.1–40.6, 2000 7 Bach, Fr.-W., Henne, R., Borck, V., Landes, K., Streibl, T., Lugscheider, E., Fischer, A., Seemann, K., Copitzky, T., Prehm, J.: Process diagnostics at thermal spraying processes – new experiences from current projects of the DFG-sponsored research groups, International Thermal Spray Conference: Lectures and Posters, Essen 2002, Ed.: Lugscheider, E. and Berndt, C. C., DVS-Verlag, Düsseldorf, 2002, ISBN 3-87155-783-8, pp. 78– 85 8 Westerweel, J.: Digital particle image velocimetry, theory and application, Delft, Techn. Univ., Doctoral Thesis. (1993) 9 Bach, Fr.-W., Copitzky, T., Tegeder, G., Prehm, J., Duda, T.: Untersuchung des Lichtbogenspritzprozesses mit der Particle Image Velocimetry; Lasermethoden in der Strömungsmesstechnik: 9th Symposium, September 18th–20th, 2001, Winterthur, Switzerland, Presented by the Deutsche Gesellschaft für Laser-Ane-

References mometrie GALA e.V. Ed.: Zh. Zhang, ISBN: 3-8265-9214-X 10 Bach, Fr.-W, Copitzky, T., Tegeder, G., Prehm, J.: Particle Image Velocimetry (PIV) as a Tool to Investigate the Influence of Nozzle Configuration and Spray Parameters on the Arc Spray Process, International Thermal Spray Conference: Lectures and Posters, Essen 2002, Ed.: Lugscheider, E. and Berndt, C. C., DVS, Düsseldorf, 2002, ISBN 3-87155-783-8, pp. 450–452 11 Bach, Fr.-W., Henne, R., Landes, K., Lugscheider, E.: Prozessdiagnostik an thermischen Beschichtungsverfahren – eine DFG geförderte Forschungsgruppe stellt sich vor; United Thermal Spray Conference, Eds. E. Lugscheider, P. A. Kammer, pp. 750–755, Düsseldorf, 1999 12 Chen, W. L. T., Heberlein, J., Pfender, E.: Measurements of two-dimensional elec-

tron densities in turbulent argon-helium plasma jets; Therm. Plasma Applications in Mat. and Metallurgical Processing, Ed. N. El-Kaddah (1992) 13 Rahmane, M., Soucy, G., Boulos, M.: Analysis of the enthalpy probe technique for thermal plasma diagnostics; Rev. Sci. Instrum. 66, 6, pp. 3424–3431 (1995) 14 Landes, K., Streibl, T., Zierhut, J.: Particle flux imaging (PFI) and particle shape imaging (PSI) – two innovative diagnostics for thermal coating, International Thermal Spray Conference: Lectures and Posters, Essen 2002, Eds. Lugscheider, E. and Berndt, C. C., DVS, Düsseldorf, 2002, ISBN 3-87155-783-8, pp. 47–51 15 Zierhut, J.: Entwicklung von Diagnostikverfahren zur Optimierung von Plasmaspritzsystemen, University of the Federal Armed Forces, Munich, Doctoral Thesis (2000)

203

205

14 Sol-gel Coating Processes M. Kursawe, V. Hilarius, G. Pfaff, Merck KGaA, Darmstadt, Germany R. Anselmann, Degussa AG, Marl, Germany 14.1 Introduction

The sol-gel technique is a method for fabricating non-metallic, inorganic materials. In contrast to mixed-oxide processes, the synthesis produces an inorganic material via liquid and gel-type intermediate stages (sol and gel). In solutions, appropriate precursors show cross-linking reactions similar to reactions observed in the production of organic polymers. In sol-gel cross-linking, soluble precursors, unlike organic polymers, hydrolyse and subsequently condense. Alkoxides and halogenides of the desired inorganic material that dissolve in water or other solvents, are appropriate precursors for these reactions, e.g. Si(OEt)4 for fabricating SiO2. A clear solution of these precursors in alcohol or other solvents transforms into amorphous gel via colloidal sol after hydrolysis and condensation reactions are initiated. The point at which former liquid solidifies is referred to as the sol-gel transition. Subsequent transformation of gel produces the pure oxide material, usually by means of drying, heat treatment, and/or sintering. 14.1.1 Background and Origin of Sol-gel Chemistry

Ebelmen (1846) is considered the first to have published on sol-gel reactions. He attempted to synthesise a metal alkoxide, Si(OEt)4, by transforming SiCl4 with ethanol. However, Ebelmen observed that the formerly clear solution for the reaction spontaneously transformed into a gel-type mass [1]. What had happened? Si(OEt)4 and SiCl4 are also excellent precursors for solgel synthesis. They hydrolyse spontaneously with water, and the humidity of the surrounding air is sufficient to initiate the reaction. Hydroxides form and, in condensation reactions, immediately transform into a Si-O-Si network that completely fills the reaction container – the solution solidifies. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

206

14 Sol-gel Coating Processes

Around 1930, the first products became available that were produced by solgel synthesis. Materials were mostly partially condensed alkoxysilanes used for protecting earthenware products. Since 1960, more and more precursors and additives for colloidal and particulate ceramics have been produced by means of sol-gel techniques. The use of colloidal sols for coating processes spread. 14.1.2 Material Fabrication by Means of Sol-gel Techniques

Sol-gel is a multi-purpose technology for fabrication of materials. The liquid intermediate state allows different types of forming (cf. Fig. 14.1). For example, a sol can be dried to gel, and subsequently heat treated, producing a dense ceramic. Specialised drying procedures, e.g. subcritical and supercritical drying, yield highly porous solids referred to as aerogels. Nano-particles are produced with spray-drying methods. Also, substrates can be coated with sols. Today, sol-gel technology is used primarily to produce such coatings. The variety of applicable forming methods for processing sol generates a wide range of properties in final oxidic products. The only fixed property after selecting the composition of the sol is the composition of the final oxide. Possible products include pure oxides, e.g. SiO2 from Si(OEt)4 sol and TiO2 from Ti(OEt)4 sol, as well as mixed oxides and complex compositions, e.g. eight-component glass from Si(OMe)4 + Al(OsecBu)3 + P2O5 + LiOEt + Mg(OMe)2 + NaOMe + Ti(OBu)4 + Zr(OPr)4 [2, 3]. Physical material properties such as density, porosity, hardness, etc. are determined mainly by the conditions during sol-gel synthesis.

Fig. 14.1 Available forming techniques in sol-gel technology. Representation following Iler, in Brinker/Scherer, Sol-Gel Science, 1990.

14.2 Sol-gel Coating Formation for SiO2

For sol, pH, above all, controls the shape of the produced nano-particles. It causes differently formed networks when the sol transforms to solid. A typical example is synthesis of SiO2 sol for either dense or porous coatings, differentiated only by the pH of the sol. A further example of the effects of varying synthesis conditions on the produced material is found in fabrication of either highly porous or dense oxides. The temperature and pressure during gel drying predominantly determine which of the two material types is produced. The following notes cover the mechanistic processes during production of coatings by means of sol-gel technology. Descriptions focus on SiO2, the most common system in sol-gel chemistry. It should be considered that mechanistic processes in the SiO2 system cannot generally be transferred to other systems. Sol-gel synthesis frequently uses transition metals, e.g. titanium, zirconium, aluminium, or heavy metals such as lead. Electronegativity is lower for transition metals than for silicon. The result is that hydrolysis and condensation reactions involving transition metals are much faster and more difficult to control than for silicon alkoxides. Furthermore, transition metals are capable of increasing the coordination sphere, whereas silicon always has a coordination number of four. Increased coordination spheres allow transition metals, in contrast to silicone, to form aquo and oxo complexes. Complex sol-gel reactions of transition metals are covered in the corresponding technical literature [9 a, b].

14.2 Sol-gel Coating Formation for SiO2 14.2.1 Coatings with SiO2 Sol from Salts of Silicic Acid

The simplest way of producing SiO2 sol is by hydrolysis of a silicic acid salt, e.g. sodium silicate. Sodium silicate solutions produce stable, colloidal sol. Nano-particles in this sol form during the following condensation reaction:

with R = OH, ONaOK. Nano-particles in the sol can be analysed accurately with spectroscopic methods, e.g. Si-NMR. Investigations showed that 80% of nano-particles in a 0.5-molar solution of K2O-SiO2 are of the following molecule type [4]:

207

208

14 Sol-gel Coating Processes

(each corner represents a tetrahedrally co-ordinated silicon atom). The sol can be applied as a coating by means of common techniques, e.g. dip coating, spin-on deposition, roller coating, slit coating, meniscus coating, and doctor coating techniques. Independent of the specific coating technique, the solid layer forms from sol when the surface-to-volume ratio increases rapidly. This leads to spontaneous cross-linking reactions between the nano-particles. High temperature promotes the condensation. Ultimately, the gel forms a coating on the substrate. The gel coating is usually hardened by an additional heat treatment, i.e. gel is transformed to a pure, dense SiO2 coating by dehydration. Although variations are limited in this system, the physical properties of the produced dense SiO2 coatings are determined by process parameters. The key value is the pH of the colloidal solution. Parameters have to be adjusted precisely for each separate process. 14.2.2 Coatings with SiO2 Sol from Si Alkoxides

Tetraethoxysilane (TEOS, Si(OEt)4) is the most common precursor for sol-gel reactions. SiO2 sol is usually produced from TEOS in ethanol. A solution of TEOS and ethanol is predefined, then, added water initiates hydrolysis and condensation. Analysis of the individual reactions and formed nano-particles is far more complex than for sodium silicate. The main reason is that three parallel reactions occur when TEOS, or other alkoxides, are transformed:

with R = –H, –OEt, –O–Si–R3. All three are equilibrium reactions and, therefore, can react in the opposite direction, as well. Identifying nano-particles formed in hydrolysis and condensation of TEOS or other Si alkoxides is very difficult. Even spectroscopic analysis cannot identify

14.2 Sol-gel Coating Formation for SiO2

Fig. 14.2 Graphical representation of simulated structures of selected model calculations, following Schaefer and Maekin [10, 11]. D values represent the fractal structure dimensions.

the molecules [5–9]. Therefore, a precise chemical description of the formed nano-particles is not available. Theoretical calculations attempted to assess structures appearing in Si alkoxide transformations. They consider, for example, whether condensation reactions are determined by diffusion processes or rather by reaction rates [10, 11]. Figure 14.2 illustrates results of such calculations. It is apparent that structures of formed nano-particles vary significantly. In fact, experiments with dispersion photometers confirmed the structures in Si alkoxide sol. The structures can be controlled by process parameters such as temperature, dosage rates, stirring intensity and, particularly, pH. Alkaline media primarily produce dense, discrete nano-particles in the sol. Condensation reactions are controlled by reaction rates and show monomercluster growth, i.e. clusters form, and grow due to condensing monomers. In acid media, very open, branched nano-particles form in the sol. Here, condensation is controlled by diffusion as a cluster-cluster growth, i.e. monomers initially form a large number of small clusters that subsequently condense. How do different nano-particles in the sol affect subsequently produced coatings? Generally, every type of sol can be deposited to form an SiO2 coating. Physical properties of the coating, however, are determined by the type of nanoparticles in the sol. Discrete, dense nano-particles typically lead to porous coatings because, during coating processes, gaps form in between the particles.

209

210

14 Sol-gel Coating Processes

Branched, open-structured nano-particles yield denser coatings. During coating, this type of nano-particles produces tight networks without developing any gaps. Variations in porosity of coatings determine hardness, and also cause different refractive indices. These coatings are therefore interesting for optical applications, one of which is introduced in the following section.

14.3 Application Examples 14.3.1 Translating an Idea into a Product: Development of an Anti-reflection Coating for Glass

The anti-reflection coating development for glass is a good example of how features of sol-gel technology are used to invent a new product. The principle behind the development is that every transparent material, e.g. glass or plastic, also reflects light. Reflections occur when light penetrating the transparent material is subject to changing refractive indices. The same happens with light that leaves the material. Reflection losses are calculated using optical laws. For white glass, they are in the range of approx. 8% of total light intensity. Light reflections are particularly troublesome when mirror effects impair seethrough properties, e.g. on displays. Reflections are equally unwanted in applications that require the total energy of sunlight, e.g. solar collectors and photovoltaic modules. On glass, a coating with an appropriate refractive index can reduce reflections considerably. The necessary optimal refractive index can be calculated. For white glass, the value is equal to nD = 1.22. When the coating thickness is adjusted to k/4, reflections of light with a wavelength of k disappear. Figure 14.3 shows the principle of the anti-reflection coating. The problem is, however, that there is no dense, stable material available with a refractive index of nD = 1.22. The refractive index of MgF2 fluoride, nD = 1.38, comes closest to the desired value. MgF2, therefore, is a commonly used material for anti-reflection coatings on spectacle lenses. Sol-gel technology allows fabrication of nano-porous coatings, as described above. Refractive indices of porous coatings are very low. For example, the desired refractive index of nD = 1.22 for anti-reflection coatings is obtained with SiO2 and a porosity of 57%. The initial step in production of nano-porous SiO2 coatings by means of solgel is a chemical preparation of an appropriate SiO2 coating solution. As described in Section 14.2, an alkoxysilane is hydrolysed and condensed in an alkaline medium. Figure 14.4 illustrates individual stages of the synthesis. Sufficient shelf life, essential for commercial use of the produced sol, is provided by modification and stabilising steps following after hydrolysis and con-

14.3 Application Examples

Fig. 14.3 Principle of anti-reflection coating on glass. 8% of light intensity is lost due to reflection when light passes through uncoated glass (left). A coating with a refractive index of nD = 1.22 (right) reduces reflections to zero for a particular wavelength.

Fig. 14.4 Left: schematic representation of sol synthesis for anti-reflection coatings. Right: SEM image of monodisperse particles, prior to coating sol modification and stabilising.

densation of the silanes. During synthesis, control of the process parameters temperature, stirring intensity, and dosage rates is challenging. Comprehensive knowledge of parameters influencing the product is crucial for successful transfer of synthesis to large-scale production. Merck KGaA produces such coating sol on an industrial scale, based on a joint development with Flabeg Solarglas GmbH as well as ISC and ISE Fraunhofer Institutes. It is used for production of highly transmissive float glass for solar-energy applications (solar collectors and photovoltaic modules) [9 e]. Here, the coating is applied by means of dip coating, however, other application techniques are applicable as well. In dip coating, the float glass is im-

211

212

14 Sol-gel Coating Processes

(bottom graph)

Fig. 14.5 Transmission spectrum of anti-reflection coating. The lower graph represents the uncoated reference sample, the upper graph shows the behaviour of glass with an antireflection coating. Measurement: Fraunhofer Institute ISE.

Fig. 14.6 Comparison of anti-reflection coatings with porous SiO2 (upper graph) and a multi-layer coating (alternating high- and low-refracting materials, intermediate graph). The porous SiO2 anti-reflection coating cov-

ers the entire spectrum whereas the effect of the multi-layer coating only covers the visible spectrum of the sunlight. Measurement: Fraunhofer Institute ISE.

mersed in a cuvette containing the coating sol, and drawn out of the solution at a defined velocity. The sol film on the glass surface dries and forms the gel. The coating thickness is controlled precisely by adjusting the velocity at which the glass leaves the bath. After drying, the coating is hardened at temperatures close to TG of the glass. Figure 14.5 shows a typical spectrum of a coated glass pane. The produced porous SiO2 coatings have a refractive index of approx. 1.25, very close to the target value of 1.22. An important prerequisite for solar-energy

14.3 Application Examples Fig. 14.7 Application of highly transmissive anti-reflective glass in solar-energy production. Photograph: Flabeg Solarglas GmbH.

production is that the anti-reflective properties cover the complete solar spectrum. Figure 14.6 compares the porous SiO2 anti-reflection coating with coatings of alternately high and low refracting layers. These multi-layer coatings are used frequently for anti-reflection on glass, e.g. lenses. For further increase of porosity and, thereby, reduction of refractive index, mechanical stability of the SiO2 coating should be considered. Stability, of course, drops when coating porosity increases. For the presented example of an SiO2 coating, abrasive wear resistance is sufficient for application as highly transmissive covering glass on solar collectors or photovoltaic modules (tested

213

214

14 Sol-gel Coating Processes

according to DIN EN 1096-2). But certainly, it is not high enough for every type of application, e.g. automobile windscreens and side windows. Figure 14.7 shows highly transmissive anti-reflection glass used as covering glass for solar collectors or photovoltaic modules. 5% of added usable sunlight corresponds to the integrated, normalised transmission increase across the total solar spectrum. Annual gross heat production for thermal solar collectors is increased by 10%, depending on collector design. Highly transmissive glass for photovoltaic modules yields a 3.5% gain in energy efficiency. Geographical position and climatic conditions have additional effects. 14.3.2 Application of Wet Chemical Coating Techniques for a Common Product Type: Pearlescent Pigments

Pearlescent pigments by Merck KGaA are an example of how wet chemical coating methods can be used for a common and established group of products. Numerous applications of materials with pearlescent pigments are found in everyday life, e.g. automotive paint coatings, printed packaging materials, wallpaper, textiles, plastics, and cosmetic products. Here, a brief discussion of interference colours and gloss is presented prior to describing pearlescent pigment production.

14.3.2.1 Gloss and Colour Why does an object seam to have a particular colour? When visible light hits a blue object, the object absorbs part of the light (red, green, and yellow part of the spectrum) and reflects the remaining light that then appears blue to the human eye. Absorption and reflection cannot explain certain colour phenomena, e.g. iridescent colours of a thin oil film on water. Such phenomena occur at layer interfaces when the thickness of boundary layers is in the range of the light wavelength. Part of the light is reflected at the first boundary layer of the film, and is subject to a 1808 phase shift. The other part of the light is reflected at the second boundary layer. The two reflected light waves superpose and interfere. Depending on film thickness, light intensity can either increase, or the two waves can destructively interfere. The result is that certain parts of the visible light intensity decrease, whereas others are reflected stronger, determined by the coating thickness. Iridescent colours of an oil film covering water are therefore caused by variations of film thickness. Gloss, i.e. reflections towards a preferred direction, are prerequisites for formation of interference colours. However, there are different types of gloss. The human eye differentiates finest nuances from soft pearlescent shine to sparkling and twinkling. Generally, gloss is created by light reflected at a smooth surface. Rough surfaces cause diffuse reflections of light. The soft gloss of a natural pearl is created by alternately arranged transparent protein and partially re-

14.3 Application Examples

flecting calcium carbonate layers. Reflections from the surfaces of CaCO3 layers originate literally from deep inside the material.

14.3.2.2 Production of Pearlescent Pigments with Interference Colours Thin layers of non-fading, highly refractive material, e.g. TiO2, embedded more or less parallel in low refracting media such as paint are necessary for fabrication of pigments with pearl-like gloss and interference colours (Fig. 14.8). Alternatively, the highly refractive layer can also be applied to a transparent, laminar substrate material. To date, monocrystalline titanium dioxide flakes are not yet available. However, deposition of thin TiO2 coatings on laminar substrates is possible. Biotite (black mica) is an appropriate substrate, and is found in large quantities as a natural resource. It can be split into flakes, is transparent, chemically inert, and heat resistant. Use of biotite as a substrate for metal oxide coatings caused a breakthrough in the technology of pearlescent pigments. Figure 14.9 shows a photograph of the mineral and an SEM image of biotite flakes. Merck KGaA uses wet chemical hydrolysis of a titanium salt solution in biotite suspension for coating pearlescent pigments. It involves the following procedure [9 c]: An aqueous, strong acid Ti(O)Cl2 solution is continuously added to a biotite suspension. When a certain pH is reached, TiO2 · xH2O deposits on the biotite. The reaction continues until a predefined coating thickness is obtained.

Fig. 14.8 Principle of pearlescent pigments with interference colour (left): TiO2 layers within paint coat. Alternatively, TiO2 can be deposited onto laminar carriers (right).

Fig. 14.9 Photography and SEM image of natural biotite (black mica).

215

216

14 Sol-gel Coating Processes

Fig. 14.10 Coating thickness of titanium oxide determines the interference colour (top). SEM image of TiO2 coating on biotite (left).

During the described process, the initially unimpressive biotite suspension starts to show a dull gloss. As the reaction progresses, the gloss intensifies. When the reaction is interrupted at this point and the suspension is filtered, rinsed, and heat treated, it produces a silver-white glossy effect pigment. The titanium dioxide is now 40–60 lm thick. Silver-white pigments are the largest group of pearlescent pigments. Depending on the particle-size distribution of biotite used, the gloss is either satin, brilliant, or sparkling. A remarkable phenomenon is observed when coating of biotite with titanium dioxide is not interrupted but continues past the stage of silver gloss. The biotite suspension then starts to show a gold-yellow gloss. During continued deposition reaction, the gold-yellow colour disappears and the colour of the suspension changes to copper red, then to lilac, glossy blue, turquoise, and finally green. When the reaction continues further, the play of colours starts afresh. The explanation is simple: The titanium oxide layers, in each case, reached a thickness at which the observed interference colours occur (Fig. 14.10). Anhydrous, crystalline modification of the oxide coating forms during subsequent drying or heat treatment of the coated biotite. Titanium dioxide on the

14.3 Application Examples

Fig. 14.11 SEM image of biotite particle with TiO2/SiO2/TiO2 coating.

biotite flakes can crystallise as rutile or anatase. Generally, the rutile modification is preferred due to a higher refractive index and light fastness. On biotite, however, anatase forms even under conditions that would usually promote rutile formation. Pre-coating of biotite with tin dioxide, also a wet chemical process, improves crystallisation of TiO2 to rutile. Apart from titanium dioxide, other metal oxides can be deposited onto biotite. Iron oxide in hematite modification is applicable due to a very high refractive index. In this case, iron salt solution is hydrolysed in a biotite suspension and an iron hydrate coating forms on the biotite. Subsequent heat treatment dehydrates the coating. Hematite is red-brown and, thus, the effect and colour of this type of pearlescent pigment are caused by a combination of absorption and interference effects. The technique yields bronze, copper, and red tones. Multi-layer coatings on biotite produce unique colour effects. These pigments can show a so-called colour flop, i.e. an observer sees either one or the other colour, depending on the viewing angle. Production is more complex but is based on the same principles of wet chemical coating processes described above. Figure 14.11 shows a SEM image of a biotite particle with a TiO2/SiO2/TiO2 coating. 14.3.3 Effect Pigments on SiO2 Flakes

A completely new class of effect pigments is created when, apart from the oxidic coating, the substrate material of a pigment is optically active, as well. This case requires a substrate with precisely adjusted, homogenous thickness. Merck KGaA produces such substrate from SiO2 by means of sol-gel technology. Figure 14.12 illustrates the principle. A Si precursor is used to deposit a SiO2 sol coating onto a rotating belt. During drying, the sol layer transforms to

217

218

14 Sol-gel Coating Processes

Fig. 14.12 Production of artificial SiO2 flakes by means of solgel coating (web coating technique).

Fig. 14.13 SEM image of SiO2 flakes, produced by sol-gel coating technique. Left: SiO2 flakes, right: SiO2 flake coated with TiO2.

gel. Then, the solid gel is removed from the belt in order to obtain SiO2 gel flakes with constant thickness. Coating thickness can be adjusted by controlling the deposition process. Heat treatment of the gel flakes yields completely oxidic SiO2 flakes (cf. SEM image, Fig. 14.13). Coating of these SiO2 flakes with titanium oxide, as presented in Section 14.3.2.2, produces effect pigments with a strong colour change, depending on the viewing angle. The reason is that not only the substrate but also the coating

Fig. 14.14 Sample with effect pigment coating: Specimen appears blue, green, or lilac, depending on viewing angle.

14.4 Conclusions

Fig. 14.15 Principle of anti-wrinkle effect due to coated polydisperse SiO2 spheres (left), SEM image of applied SiO2 spheres.

produces interference colours. Figure 14.14 shows the results: A coated model, photographed from three different directions, appears blue, green, or lilac, depending on the viewing angle. 14.3.4 Coating of SiO2 Spheres for Cosmetic Formulations

Wet chemical deposition methods described in Section 14.3.2.2 allow applications in further branches as well. For example, polydisperse SiO2 nano-spheres are produced as additives for cosmetic formulations [9 d]. Additives of inorganic particles for cosmetic formulations are long known and common, e.g. as fillers or for controlled rheological properties. However, particles with specific size distribution and appropriate coatings can be used to create special effects. TiO2/Fe2O3-coated polydisperse SiO2 spheres provide a so-called anti-wrinkle effect. When added to cosmetic formulations, these custom SiO2 spheres cause diffuse reflection of incoming light. A small skin wrinkle, clearly appearing due to light shadowing effects, brightens and becomes less obvious as reflections are scattered. Figure 14.15 illustrates the anti-wrinkle principle. The SEM image shows corresponding polydisperse SiO2 spheres. TiO2/Fe2O3 coatings are applied by the same means as biotite coatings, described in Section 14.3.2.2.

14.4 Conclusions

The broad potential of sol-gel coatings is demonstrated, considering anti-reflection coatings for glass, wet chemical coating techniques for pearlescent pigments and effect pigments, as well as coating of SiO2 spheres for cosmetic formulations, as examples.

219

220

14 Sol-gel Coating Processes

With conceivably increasing demand for upgrading and functionalisation of all types of surfaces, coating technologies are gaining importance. Sol-gel coating techniques will play a decisive role in applications where physical coating methods are either unfeasible or uneconomical. Sol-gel technology is usually the first choice for coating small particles, for applications that require porous coatings, and for coating processes operating at room temperature.

References 1 Ebelmen, M.: Ann. Chim. Phys. 16, 2 3

4 5

6

7

8 9

1846, 129 Dislich, H.: Angew. Chemie Int. Ed. Engl. 10, 1971, 363 Dislich, H.: in: Transformation of Organometallics into Common and Exotic Materials, Ed. R. Laine. NATO ASI Series E, 141, pp. 236–249 Brinker, C. J., Scherer, G. W.: Sol-Gel Science. Academic Press 1990, 101 Kelts, L. W., Armstrong, N. J.: in: Better Ceramics Through Chemistry III, Eds. C. J. Brinker, D. E. Clark, D. R. Ulrich. Material Research Society, Pittsburgh, Pa., 1988, 519 Klemperer, W. G., Ramamurthi, S. D.: in: Better Ceramics Through Chemistry III, Eds. C. J. Brinker, D. E. Clark, D. R. Ulrich. Material Research Society, Pittsburgh, Pa., 1988, 1 Lin, C. C., Basil, J. D.: in: Better Ceramics Through Chemistry II, Eds. C. J. Brinker, D. E. Clark, D. R. Ulrich. Material Research Society, Pittsburgh, Pa., 1986, 585 Mulder, C. A., Damen, A. A. J. M.: Journal of Non-Crystalline Solids 93, 1987, 169 Brinker, C. J., Scherer, G. W.: Sol-Gel Science. Academic Press 1990, 160–191

(a) Brinker, C. J., Scherer, G. W.: Sol-Gel Science. Academic Press 1990, Chapter 2 (b) Livage, J., Henry, M., Sanchez, C.: “Sol-Gel Chemistry of Transition Metal Oxides” in Progress in Solid State Chemistry 18, 1988, 259–342 (c) Glausch, R., Kieser, M., Maisch, R., Pfaff, G., Weitzel, J.: Die Technologie des Beschichtens, Ed. V. Zorll. Vincentz, Hannover, 1995 (d) Anselmann, R., Journal of Nanoparticle Research 3, 2001, 329–336 (e) Kursawe, M., Hofmann, Th.: “Antireflective Coating on Float Glass for Solar Collectors”. Session 32 Glass Processing Days, June 18th–21st, 2001, http:// www.glassfiles.com/ 10 Schaefer, D. W.: MRS Bulletin 8, 1988, 22 or cited in: C. J. Brinker, G. W. Scherer, Sol-Gel Science. Academic Press 1990, 193–206 11 Maekin, P.: in: On Growth and Form, Eds. H. E. Stanley, N. Ostrowsky. Martinus-Nijhoff, Boston 1986, 111 or cited in: C. J. Brinker, G. W. Scherer, Sol-Gel Science. Academic Press 1990, 193–206#

221

15 Hot-dip Coating W. Bleck, Department of Ferrous Metallurgy, RWTH Aachen, Germany D. Beste, Vallourec & Mannesmann, Düsseldorf, Germany 15.1 Mechanisms of Corrosion Protection

Iron tends to release electrons in order to achieve an energetically stable condition. In aqueous solutions, this happens when iron oxides or iron hydroxides are formed. These processes cause, amongst others, corrosion effects found in everyday life, especially on older automobiles (Fig. 15.1). When a paint coating on sheet steel is damaged, red rust can form. Due to increasing volume, the rust lifts off the coating and sub-surface migration can occur. Figure 15.2 shows a zinc pre-coated, painted sheet metal. In this case, the steel surface is protected cathodically even if localised lacquer damage leads to media penetration [1]. Anodic dissolution of the zinc coating produces white rust, a porous, voluminous corrosion product that protects the surface from ongoing corrosive attack.

Fig. 15.1 Corrosion damage on powder-coated steel sheet without zinc coating, after 4 days of salt spray testing. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann„ Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

222

15 Hot-dip Coating

Fig. 15.2 Schematic representation of corrosion on zinc- and paint-coated steel sheet with local damage.

Fig. 15.3 Schematic representation of basic corrosion preventing methods in the potential model: (1) active corrosion protection, (2) passive corrosion protection.

Generally, corrosion processes can be inhibited by one of the following means: 1. Release of electrons is impeded. The potential-well model explains the principle (Fig. 15.3). The initially lower Fe2+ energy level is raised until it exceeds Fe potential. This method is referred to as active corrosion protection. 2. Activation energy for electron release is increased by impeding charge exchanges during electrochemical corrosion by means of non-conductive layers. This technique is referred to as passive corrosion protection. Corrosion in aqueous media involves anodic (Eq. (1)) and cathodic (Eqs. (2) or (3)) partial reactions: Fe ˆ Fe2‡ ‡ 2e

…1†

2H‡ ‡ 2e ˆ H2

…2†

2H2 O ‡ O2 ‡ 4e ˆ 4…OH†

…3†

15.1 Mechanisms of Corrosion Protection

Cathodic partial reaction in acids shows hydrogen gas evolution, corresponding to Eq. (2). Oxygen reduction occurs in aerated water as suggested by Eq. (3). The overall reaction reads: 2Fe ‡ 2H2 O ‡ O2 ˆ 2Fe2‡ ‡ 4…OH†

…4†

In active corrosion protection, a sacrificial anode causes iron to act as cathode, which is then protected against spontaneous electron release. All chemical elements less noble than iron, with respect to position in the electrochemical series, can act as sacrificial anode, e.g. zinc. Oxidation of zinc is described by the following equation: 2Zn ‡ O2 ‡ 2H2 O ˆ 2Zn2‡ ‡ 4…OH†

…5†

The reaction product of this equation is a white precipitate of zinc hydroxide that forms zinc oxide after dehydration. The precipitate creates a layer that reduces the transfer of oxygen to the iron surface. In acids, the layer dissolves: Zn…OH†2 ‡ H‡ ˆ Zn2‡ ‡ 2H2 O

…6†

In alkaline solutions as well: Zn…OH†2 ‡ OH ˆ Zn…OH†3

…7†

The rate of surface-layer dissolution predominantly controls the rate of zinc corrosion. The corrosion rate, therefore, increases considerably with increasing or decreasing pH. In passive corrosion protection, plastic coatings as well as dense Al2O3 or Cr2O3 coatings prevent iron from releasing electrons. Dense Al2O3 coatings can be produced by hot dipping, e.g. hot-dip aluminium or Galvalume coatings. Chromium alloys with more than 13% generate dense chromium oxide coatings in situ. Generally, corrosion protection for steel is provided by: · using passivation by appropriate alloying (e.g. adding 13% chromium) · protective coatings, often combined with organic coatings (paint), and produced by means of browning in molten salt at approx. 145 8C, phosphating, or paint coating · non-metal coatings, e.g. glass (enamel) · metal coatings with cathodic protection properties, e.g. zinc coatings · metal barrier coatings, e.g. hot-dip aluminium or roll bonding.

223

224

15 Hot-dip Coating

15.2 Phase Diagrams Fe-Zn, Fe-Al, Al-Zn, and Fe-Al-Zn

Metal coatings are used in households, the construction industry, and, increasingly in recent years, in the automotive industry. Various fields of applications and characteristic demands on coated materials have led to the development of numerous coatings, optimised to specific requirements. The most important coatings are based on zinc- and aluminium-containing coatings, produced within seconds by dipping steel band into molten material. The zinc-rich side of the Fe-Zn phase diagram [2] is of particular interest (Fig. 15.4). It exhibits three peritectic phase transformations: L + a ? C, L + C ? d, and L + d ? f, as well as the peritectoid reaction C + d ? C1. Table 15.1 lists the most important phases of the iron-zinc phase diagram, corresponding crystal structures, and several other characteristics. Pure zinc is referred to as g phase. Intermetallic phases f, d, C1, and C form prior to or after hot-dip coating, depending on the thermal cycle. The process is aimed at producing very thin intermetallic coatings with respect to subsequent forming operations. The iron-aluminium phase diagram [2] also shows a number of intermetallic compounds (Fig. 15.5). Here, Al5Fe2 is particularly important for good adhesion of zinc coatings. Up to 10% silicon initially added as alloying element to molten aluminium causes formation of FeO, Al2O3 spinels, or (AlSi)2O5 at the steel in-

Fig. 15.4 Binary alloy phase diagram of Fe-Zn.

15.2 Phase Diagrams Fe-Zn, Fe-Al, Al-Zn, and Fe-Al-Zn

225

Table 15.1 Phases in the Fe-Zn binary alloy phase diagram. Phase

Eta g Zeta f Delta d Gamma C1 Gamma C

Lattice

hexagonal monoclinic hexagonal fcc bcc

Chemical notation

Zn FeZn13 FeZn7 Fe5Zn21 Fe3Zn10

Fe Weight per cent

Atomic per cent

0 5–6.2 7–10 15.8–23.5 20.2–31

0 6.7–7.2 8.5–13.5 18.5–23.5 23.2–31

Microhardness

Characteristics

52 208 358 505 326

very tough tough brittle hard and brittle brittle

Fig. 15.5 Binary alloy phase diagram of Al-Fe.

terface. Silicon suppresses the rapid formation of the very brittle Fe2Al5 phase. The high melting temperature of aluminium, compared to zinc, permits applications requiring certain temperature resistance, e.g. sheet metal mounted in the vicinity of automobile engines. The Al-Zn phase diagram [2] (Fig. 15.6) shows a eutectic phase transformation at 5% aluminium content. This eutectic composition is referred to as Galfan. At room temperature, the alloy contains zinc and eutectic phase with lamellar arrangement of g zinc and c crystallites (Al-Zn crystallites with approx. 20% zinc). Figure 15.7 shows the most important phases of the ternary iron-zinc-aluminium system [3]. The alloy containing 55% by mass Al, 43.5% by mass Zn, and 1.5% by mass Si is referred to as Galvalume.

226

15 Hot-dip Coating

Fig. 15.6 Binary alloy phase diagram of Al-Zn.

Fig. 15.7 Isothermal ternary alloy phase diagram of Fe-Al-Zn, at 450 8C.

15.3 Metal Coatings

227

15.3 Metal Coatings

Table 15.2 summarises prominent metal coatings for hot-dip-coated steel band. Most significant, with respect to mass coating, are hot-dip zinc, Galfan, Galvalume, and type 1 hot-dip aluminium. Nowadays, type-2 hot-dip aluminium and hot-dip lead are less important in Europe. Zinc on steel yields dual corrosion protection: · Zinc forms well-bonded, shielding coatings that protect steel against decomposition (barrier effect). · Zinc provides cathodic protection at cutting edges or damaged parts of the coating.

Table 15.2 Summary of the most important hot-dip coatings [4]. Product

Abbreviation Coating structure

Composition

Typical deposit in g/m2

Applications

Hot dip zinc

Z

Zn/steel

100% Zn approx. 0.2% Al

100–600, standard 275

Galvannealed zinc

ZF

ZnFe/steel

Galfan

ZA

ZnAl/steel

Galvalume

AZ

AlZn/steel

90% Zn cf. hot dip zinc 10% Fe approx. 0.12% Al 95% Zn 95–255 5% Al 55% Al 85–160 43.5% Zn 1.5% Si

interior and external parts of car bodies roof and wall elements in building construction casings, mountings, and supporting structures in systems engineering base material for organic coatings air conditioning technology household appliances parts from shell of car body stoneguard underbody

Type 1 hot-dip AS aluminium Type 2 hot-dip A aluminium Hot-dip lead/ TE tin

AlSi/steel Al/steel PbSn/steel

90% Al 10% Si 100% Al 8–11% Sn 1–3% Sb rest Pb

50–250 200–300 75–150

roof and wall elements complex deep draw parts roof and wall elements constructional elements subject to particular corrosion attack in acid environments outlet systems household appliances roof and wall elements fuel tanks

228

15 Hot-dip Coating

Table 15.3 Comparison of different hot dip coatings in terms of corrosion protection, paint coatability, and suitability for phosphatising. Criterion

Z

ZF

ZA

AZ

AS

Cold-rolled strip

Corrosion resistance without paint coating Paint coatability (CDC) Corrosion resistance with paint coat (CDC) Suitability for phosphatising Fuel resistance

good

–

improved

excellent

excellent

–

good good

satisfactory excellent

good good

– –

– –

excellent barely good

good

excellent

satisfactory

–

–

excellent

satisfactory

–

satisfactory

–

excellent

insufficient

The term Galvannealed refers to Fe-Zn phase coatings with approx. 10% Fe that are produced by heat treatment of hot-dipped zinc coatings. The aluminium content in Galfan and Galvalume coatings increases corrosion resistance. Thus, coating thickness can be reduced without impairing corrosion protection. The characteristic advantage of these coatings is a combined cathodic protection effect of zinc and passivation effect by aluminium. Table 15.3 compares the corrosion protection of hot-dip coatings described above and cold-rolled strip subject to various boundary conditions. Selecting an appropriate coating for a particular application certainly involves considering, apart from corrosion resistance, processability (behaviour during forming, coefficients of friction, wear behaviour, weldability) and costs. The typical microstructure of a zinc coating on hot-dipped thin sheet, according to Baumgartl, shows coexistence of several phases, schematically illustrated in Fig. 15.8. Bonding is provided by a 0.02-lm Fe2Al5 phase at the interface to the substrate surface. The following layer is a less than 1-lm thick delta phase,

Fig. 15.8 Schematic cross section of zinc coating on hot-dip zinc-coated steel sheet.

15.4 Systems Technology

followed by the zinc coating that can contain small amounts of lead particles. The lead originates from lead-containing zinc baths. The outer surface layers are formed by Al2O3 and ZnO. In the production of galvannealed thin sheet, heat treatment (approx. 550 8C/ 10 s) transforms the zinc coating into an iron-zinc coating. A several-step process thereby yields a dual-layer structure with C1/C phase intermediate layer and d phase main layer. Galvannealed coatings contain approx. 10% by mass of iron.

15.4 Systems Technology 15.4.1 Design of Hot-dip-coating Systems

Parts in industrial hot-dip coating are immersed in batches into molten metal baths or continuously by processing steel band in hot-dip zinc-coating systems. Figure 15.9 [5] illustrates the principle design of a hot-dip coating system with continuous band equipment and individual, subsequent processing steps. Two reels feed strip to a welding machine that produces an endless band that initially is transferred to a band storage. The band storage is necessary in order

Fig. 15.9 Hot-dip coating system for production of surface refined steel sheet (Hot-dip zinc-coating system 2 of voestalpine – Division Steel).

229

230

15 Hot-dip Coating

to separate the discontinuously operating welding machine at the feeding section and, if required, the cutting station at the system output from the continuous-band processing unit, preferably operating at constant strip speed. Pre-treatment of the band includes cleaning by mechanical brushing, electrochemical degreasing and, in certain modern systems, etching. Cleaning removes residual rolling emulsions and rubbed-off iron particles from the steel band surface. Therefore, contamination and material accumulation on rollers, causing damages on band surface, are prevented. The next step is a heat-treatment unit for recrystallisation annealing between 700 8C and 850 8C, depending on steel grade. For annealing, protective gas mixtures of nitrogen and 2% to 7% H2 are used. Subsequent cooling is performed under protective gas, in order to reach a coating-band temperature of approx. 20 K above the temperature of the zinc bath. The strip is fed to the molten metal bath directly from the protective gas atmosphere through an immersion snout (Fig. 15.10). The so-called zinc pot contains a guide roll and one or more stabilising rolls providing smooth band travel. The band leaves the molten bath with a liquid coating, passes through an air knife blowing pressurised air to regulate coating thickness, and solidifies in an outlet tower before contacting subsequent rolls. Within the short immersion period of approx. 1–3 s, steel and molten zinc react and form an intermetallic intermediate layer of Fe2Al5, serving as a bonding layer. When leaving the bath, the strip drags along liquid zinc. The thickness of the zinc coating is controlled by the air knife as well as the coating band and molten bath temperature. The outlet tower can contain a continuous annealing zone for galvannealing treatment. The final stages of hot-dip coating systems can involve surface posttreatment of zinc-coated steel band, e.g. phosphating or chromatising. Usually, in-line quality control is integrated prior to final trimming and cutting units where the band is coiled or stacked.

Fig. 15.10 Detailed view of sheet travel in zinc pot and subsequent vertical stretch with galvannealing unit.

15.4 Systems Technology

Characteristic values and features of hot-dip coating equipment, referring to the system shown in Fig. 15.9, are: · annual capacity approx. 400 000 t, at an average of 54 t/h · strip dimensions: width 750–1600 mm, thickness 0.4–1.5 mm, coil weight max. 30 t · strip velocity: inlet/outlet max. 190 m/min, treatment max. 150 m/min · buffer content: inlet/outlet 234 m · annealing furnace: vertical furnace · steel grades: low-alloy steel, deep-drawing steel, bake-hardening steel, highstrength multi-phase steel · annealing cycles: up to 860 8C · zinc coating: 1 pot with 180 t · air-knife system: 0–1000 mbar (air or N2) · coating thickness: 50–275 g/m2 · cooling tower: optional ZF unit (2 MW/100 kHz), switchable heating/cooling zone · dressing frame: four-high stand (Æ 440–500 mm), decoupled bending stretcher level. The system is capable of applying Z (zinc) and ZF (zinc-iron) coatings of various quality, and can provide subsequent phosphating, chemical passivation, as well as oil lubrication of steel band. 15.4.2 Reacting Agents in Molten Zinc

The size of zinc flowers, i.e. visible size of zinc crystallites (spangle), is determined mainly by the lead or antimony added to the zinc bath, and by the solidification rates of zinc coatings. Certain traditionally oriented users still consider clearly apparent zinc flowers as a significant quality feature. Rising demands on surface appearance have induced steel manufacturers to produce considerably reduced zinc flowers or prevent the formation of zinc flowers at the surface by a number of carefully aimed measures. Formation of small zinc flowers is promoted by increasing nucleation sites for crystallisation of the zinc coating. For this, different types of nuclei are applied to not yet solidified metal coatings, e.g. water, steam, water-soluble salts, or zinc powder. Lead-free zinc baths are used to produce zinc surfaces without any visible zinc crystallites (spangle free). In production processes for hot-dip-zinc-coated thin strip, 0.18 to 0.20% by mass Al is added to the zinc bath. For galvannealed surfaces, Al content is considerably lower, in the range of 0.10 to 0.14% by mass. If the iron content in molten zinc rises above the solubility, which is only 0.002% by mass Fe at bath temperatures of 460 8C, Fe initially reacts with Al and forms solid FexAly phases. However, after the phases grow to particle size their density is below the density of molten zinc and the particles float to the surface. This slag or dross is fairly simple to remove from the zinc surface with-

231

232

15 Hot-dip Coating

out severely influencing band quality. The mechanism is not a difficult issue as long as Al contents are sufficiently high, as in hot-dip zinc coating of thin sheet metal. In production processes for galvannealed thin sheet, the Al content is too low for the reaction described above and, thus, increasing Fe content in the molten zinc leads to a predominant reaction with zinc. Here, according to the phase diagram of the binary Fe-Zn system, several phases form, and grow to large particles. The density of the particles is higher than the density of zinc. Thus, solid components precipitate at the bottom of the zinc pot. The amount of bottom dross increases steadily with time and considerably influences product quality. The only way to remove the bottom dross during production is by adding increased amounts of Al in order to dissolve the sediment and produce particles described above that float on the surface of the zinc bath. Al in the molten zinc bath serves two purposes in hot-dip zinc coating of thin sheet (Z): · When the steel strip is immersed in the molten zinc, aluminium promotes the formation of a bond coat and an intermediate layer that is well wetted by molten zinc, i.e. aluminium guarantees an overall well-bonded coating. · Aluminium forms a distinct barrier layer for Fe-Zn reactions, unwanted in Z, and thus guarantees a shiny metallic appearance of the product as well as, if required, improved forming behaviour. As in Z, production of ZF (galvannealed) depends on aluminium as the initial bonding or wetting agent. Subsequently, Al contents should not be any higher than necessary to allow activation of diffusion processes between Fe and Zn during the temperature increase in the galvannealing furnace. However, Al content should be sufficiently high in order to prevent a Fe-Zn reaction in the molten zinc, which would have negative effect on subsequent stripping for production of thin coatings. Furthermore, the Al content determines the surface roughness of ZF coatings. Fissured surfaces are observed primarily on steel grades that show slow al-

Fig. 15.11 Optimising of forming behaviour for galvannealed coatings [6].

15.4 Systems Technology

loying behaviour, e.g. vacuum decarburised steel (VAC), compared to titaniumalloyed IF steel (Ti-IF steel). Production of sealed coatings on these steel grades requires very low Al concentrations. Adding approx. 0.06–0.08% by mass of Si increases the adhesion of ZF coatings, particularly during deformation. In Fe-Zn diffusion, Si additionally contributes to alloying along grain boundaries in near-surface grain layers of steel. This helps subsequent forming as it promotes crack growth perpendicular to the surface, and prevents prolonged cracking parallel to the surface (Fig. 15.11). 15.4.3 Surface Post-treatment

Hot-dip refined band and sheet is factory available with a number of surface treatments, listed here with common abbreviations: C chemically passivated O oiled CO chemically passivated and oiled S sealed P phosphated steel. Today, phosphating or chromating are common surface post-treatment methods. Zinc phosphating comprises two treatment steps: 1. Surface activation by etching and brushing, according to the following reaction: Zn ‡ 2H‡ ‡ 2Ox ˆ Zn2‡ ‡ 2HOx

…8†

with Ox representing the oxidising agent, NO–2, ClO–3 or H2O2. 2. Formation of a very thin phosphate layer, according to the following reaction: 3Zn2‡ ‡ 2 ‡ H2 PO4 ‡ 4H2 O ˆ Zn3 …PO4 †2
4H2 O ‡ 4H‡

…9†

The Zn3PO4 phase is also referred to as hopeite (tertiary zinc phosphate). This phosphate layer yields corrosion protection during storage periods, serves as a lubricant in cold forging, and acts as a bonding agent for subsequent paint-coating processes. Chromating involves initial surface activation by brushing and etching as well. Coatings develop according to: Zn2‡ ‡ CrO3 ‡ H2 O ˆ ZnCrO4 ‡ 2H‡

…10†

Subsequent rinsing and drying provides corrosion protection and improved paint adhesion. The process allows fabrication of decorative coloured coatings. Organic coatings can be applied to hot-dip refined band and sheet in order to increase corrosion protection and/or for optical and decorative reasons.

233

15 Hot-dip Coating

15.5 Quality Control 15.5.1 Testing Mechanical Properties

Apart from being error free and appearing homogeneous, surface coatings have to fulfil further important testing criteria such as sufficient coating adhesion, tested, for example, by means of the ball-impact method, bending tests, or lapshear tests. Additional criteria are tribological properties, wear behaviour (powdering, flaking), weldability, and processability. 15.5.2 Testing Corrosion Properties

Zinc coatings withstand numerous chemicals. Resistance to many hydrocarbons is particularly high. Corrosion resistance in aqueous media and other liquids depends strongly on the pH of the liquids. Zinc coatings usually show high resistance to alkaline media (high pH). However, they are less resistant to acid media. Generally, zinc shows good resistance to liquids with pH in the range of approx. 5 to 13. Outside of this range, increased corrosion of zinc must be considered (Fig. 15.12). Based on research data, corrosion categories have been specified that allow at least a rough estimate of the atmospheric influence in a specified region. Table 15.4 gives an overview of corrosivity categories, their classification, and influence on average zinc wear. Corrosion resistance of parts is tested either during natural weathering in outdoor exposure tests, or in short-term tests. The latter frequently involve ageing tests in environmental chambers apart from potentiostatic and potentiodynamic

corrosiveness of zinc coatings

234

Fig. 15.12 Corrosion behaviour of zinc coatings against pH (schematic representation).

15.5 Quality Control

235

Table 15.4 Categories of corrosivity and corrosion rates of zinc coatings for selected atmospheres (DIN EN ISO 12944). Corrosivity category

Typical environment interior

C1

C2

C3

C4

C5-I

C5-M

heated buildings with indifferent atmosphere, e.g. offices, shops, schools, hotels unheated buildings with possible condensation, e.g. warehouses, sports halls production areas with high humidity and modest air purification, e.g. food processing facilities, laundries, breweries, dairies chemical plants, baths, seawater boathouses

Corrosive attack

Average corrosion rate of zinc

exterior insignificant < 0.1 lm/a

atmospheres with low pollution, usually rural areas

low

0.1 to 0.7 lm/a

urban or industrial atmosphere, moderate sulfur dioxide pollution, coastal areas with low salt contents

moderate

0.7 to 2.1 lm/a

high

2.1 to 4.2 lm/a

very high (industry)

4.2 to 8.4 lm/a

very high (seawater)

> 4.2 to 8.4 lm/a

industrial areas and coastal regions with moderate salt contents buildings or areas with nearly industrial areas with high persistent condensation and humidity and aggressive high pollution atmosphere buildings or areas with nearly coastal and offshore areas persistent condensation and with high salt content high pollution

testing. However, short-term ageing has the disadvantage that test results are influenced by intensified corrosive conditions. Environmental chambers can simulate numerous climatic and atmospheric conditions: · urban atmosphere containing SO2 and NOx · industrial atmosphere containing SO2 · maritime atmosphere containing Cl and SO2. Frequently used testing methods are: · salt spray tests according to DIN 50021 for simulation of coastal areas with 5% NaCl solution at 35 8C · condensed water test as specified in DIN 50017 for simulation of high humidity (tropical climate) with a water vapour/air mixture (100% relative humidity) at 40 8C · VDA (German Association of the Automotive Industry) cycle test. A one-week test cycle for simulating natural weathering comprises 24 h salt spray, 72 h condensed water test and 48 h standard climate.

236

15 Hot-dip Coating

15.6 Summary and Conclusions

Due to a number of integrated process steps (cleaning, recrystallisation annealing, zinc coating, skin passing, etc.) continuous hot-dip coating is a cost-efficient process for highest surface quality coating of large-quantity steel sheet. Most zinc-coating systems established in recent years are based on hot-dip developments by Tadeusz Sendzimir. Electrolytic deposition systems are less common. Today, the majority of thin sheet steel sold by steel suppliers in Germany is surface treated. Most of the steels are hot-dip coated, low-alloyed steel grades. Use of modern, high-strength steels, e.g. dual-phase or TRIP steel, is spreading in order to meet increasing materials requirements of the automotive industry. Thin, high-strength sheet metal meets today’s and future standards in terms of technology as well as safety and at the same time reduces the weight of the car body. Metallic zinc or zinc alloy coatings provide effective corrosion protection and guarantee the service life of coated parts. Alloys for high-strength steel grades often contain considerable amounts of alloying elements and show multi-phase microstructure. Heat treatments prior to coating therefore often face the problem of superficial oxides, formed by alloying elements at the surface. During hot-dip coating, this leads to modified reaction kinetics, changes in wetting behaviour of the steel surface, and therefore ultimately to varying behaviour in the coating process. Thus, the quality of coatings is often insufficient. Current research and development work therefore focuses on trying to coat such steel grades by means of hot-dip zinc coating. Investigations involve different approaches to develop a steel surface composition more appropriate for coating, particularly by adjusting alloy composition or varying annealing atmosphere [7]. For example, critical elements such as silicon, manganese, and chromium can be replaced by less-critical molybdenum. An oxidation/reduction annealing cycle initially oxidises the alloying elements within the steel structure and forms pure iron oxide at the surface. The iron oxide then reduces in a subsequent process step. The substrate surface for coating in the zinc bath is thus oxide free, consists of pure iron, and is easy to coat. Use of high-strength steel grades as car-body materials in future automobile generations therefore depends also on developments in coating technology.

References

References Béranger, G., Henry, G., Sanz, G.: The Book of Steel. Intercept LTD, Hampshire U.K., 1996 2 Massalski, Th. B.: Binary Alloy Phase Diagrams, American Society for Metals, Metals Park, Ohio, 1986 3 Perrot, P., Tissier, J.-Ch., Dauphin, J.-Y.: Stable and Metastable Equilibria in the Fe-Zn-Al System at 450 8C. Zeitschrift für Metallkunde, 83 (1992) 11, pp. 786– 790 4 Stahl-Informations-Zentrum: Charakteristische Merkmale 095, Schmelztauchveredeltes Band und Blech, Düsseldorf, 2001 1

voestalpine Stahl GmbH, System Brochure of Hot-dip zinc-coating system 2 6 Brisberger, R., Berndsen, H., Etzold, U., Mail, O., Warnecke, W.: Laboratory investigations on the morphology of the coating and forming behaviour of galvannealed steel sheet. Galvatech ’95 – 3rd International Conference on Zinc and Zinc Alloy Coated Steels, Proceedings, Association for Iron & Steel, 2004, pp. 553– 759 th 7 Galvatech ’04 – 6 International Conference on Zinc and Zinc Alloy Coated Sheet Steels, Proceedings, Association for Iron & Steel, 2004 5

237

239

16 Build-up Brazed Wear-protection Coatings H. Krappitz, Innobraze GmbH, Esslingen, Germany 16.1 Introduction

Soldering and brazing are often associated with bonding in electrical circuits. Soldering, as a joining technology, is in fact one of the most commonly found applications of soldering and brazing, and at the same time, considering the total number of produced joints, probably the most frequently used joining technology. Brazing or high-temperature brazing as a joining technology for production of highly sophisticated joints with substrate strength, however, is less known. The reader may wonder why brazing is an issue in the realm of modern coating technologies. This chapter focuses on brazing methods used for coating and presents typical examples of applications.

16.2 Brazing and Soldering 16.2.1 Fundamentals

Following the DIN 8505 [1] standard, brazing and soldering are thermal processes that either create a metallurgical bond, or serve as coating processes for materials. They involve a liquid phase of molten filler metal. Processes operate below the solidus temperature of the substrate materials. This fact, in particular, distinguishes brazing and soldering from joining and coating technologies based on welding. In soldering, the liquidus temperature of filler metals is below 450 8C. Above 450 8C, the process is referred to as brazing. High-temperature brazing covers techniques where the liquidus temperature of filler metals lies above 900 8C, and filler metals are applied without any flux on exclusion of air. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

240

16 Build-up Brazed Wear-protection Coatings Fig. 16.1 Filler metal groups for brazing and high-temperature brazing.

The following investigations focus on the field of brazing and high-temperature brazing; soldering is not covered in any more detail, here. Filler metals used in brazing and high-temperature brazing are usually silver- and copperbased brazing filler metals, including brass and bronze filler metals. Nickelbased and precious-metal fillers such as gold-, palladium-, or platinum-based filler metals, are common for applications with higher operating temperatures. Figure 16.1 shows frequently used filler metal groups for brazing as well as high-temperature brazing and assigns corresponding brazing temperatures, as applied in practice. In mass production by means of furnace brazing under protective gas or vacuum, primarily Al, Cu, and Ni filler metals are used. Aluminium filler metals are applied to aluminium and aluminium alloys only, because these filler metals are subject to undesired metallurgical reactions in contact with steel or non-ferrous metals and thus would have negative impact on the usability of the brazed joints. Material costs of filler metal groups vary considerably, particularly regarding precious-metal fillers. However, economical assessments should consider that silver-based filler metals melt at relatively low temperatures, and therefore, are widespread for air brazing, e.g. manual flame brazing, in spite of filler metal costs. An important prerequisite for molten filler metal to be able to wet the substrate material is that direct metallic contact is established between the two components. Technical metal surfaces always carry more or less distinct oxide layers and surface contaminants that initially have to be removed. They are decomposed in furnace brazing by proper protective gas (e.g. argon, hydrogen, or

16.2 Brazing and Soldering

Fig. 16.2 Oxide decomposition on surfaces under protective gas.

Fig. 16.3 Wetting during brazing.

nitrogen-hydrogen mixtures) (Fig. 16.2) [2] or by vacuum in the pressure range of 10–2 to 10–5 mbar. Actual wetting occurs due to alloying of filler metal and substrate material, initially only within a few atomic layers of the joint area. Subsequent diffusion processes create metallurgical bond between the filler metal and the substrate, yielding strength properties close to substrate strength (Fig. 16.3). 16.2.2 Repair Brazing

Repair of parts made of high-performance materials in an important application for brazed coatings. Stators and rotary blades of turbines in aircraft jet engines as well as stationary gas turbines are subject to high mechanical and thermal loads, and at the same time, corrosive attack. Materials with carefully controlled microstructure, e.g. directionally solidified or dispersion-strengthened materials,

241

242

16 Build-up Brazed Wear-protection Coatings

Fig. 16.4 (a) Repair brazing of turbine blades, damaged blade. (b) Repair brazing of turbine blades, build-up brazed blade.

are used in order to withstand operating conditions. After a certain time of operation, wear causes damages such as material loss and pitting (Fig. 16.4 a). During maintenance of these components, compensation of material loss is necessary in order to spare resources and reuse valuable parts economically. Arc welding of such material would melt the substrate surface and change its microstructure. The part would thus be unsuitable for further use. Build-up brazing is the ideal solution to the problem as brazing, by definition, does not involve any substrate melting. Repair by means of build-up brazing uses mixtures of substrate material and special filler-metal powders. From this, either initially pre-sintered preforms (PSP) are fabricated and subsequently applied to the part by build-up brazing, or repair is performed in a single step with plastic-bound substrate/filler metal mixtures (tapes) or pastes of corresponding powder mixtures. Vacuum furnace brazing produces a restored part, with properties close to those of new parts, after final mechanical processing (Fig. 16.4 b). 16.2.3 Coating by Build-up Brazing of Sintered Hard Metals

Due to high hardness and wear resistance, sintered hard metals are used in applications that impose the highest wear loads on parts. Fusion welding is inappropriate for applying hard-metal preforms sintered from tungsten carbide and a cobalt binder phase, as the microstructure, and therefore, desired hard metal properties would be destroyed. Here, brazing again provides a means of joining,

16.2 Brazing and Soldering

Fig. 16.5 Brazing of hard-metal tips on tools.

yielding high-strength, thermally stable bonds. Figure 16.5 shows a selection of parts with brazed hard-metal preforms. Wetting of hard metals is generally difficult. Therefore, special filler metals for these materials are silver- or copper-based and additionally contain manganese, nickel, or cobalt. Depending on the cobalt content of the hard metal, the special filler metals produce joints with shear strength values exceeding 300 N/ mm2. The thermal expansion coefficient ratio of hard metals and steel substrates is in the range of approx. 1 : 2. For hard metal/steel joints, the disparity leads to internal stresses in the compound due to unequal shrinkage of joining components during cooling of the joint, after solidification of the filler metal. Possible distorting and cracking due to developing stresses in hard metals thus limit the dimensions of coated parts. However, possible workarounds are to divide the coated area into segments or to use appropriate layered filler metals. The principle of layered filler metals for reduced stress is illustrated in Fig. 16.6 [3]. Plastic deformation of filler metal and an optional intermediate layer compensate for the disparity in coefficients of thermal expansion of both materials. Apart from filler-metal properties, the mechanical strength of the joint is certainly determined considerably by the material properties of the steel substrate and the hard metal. The highest strength is produced by rigid supporting constructions and a high yield limit of the steel. In practice, unalloyed or low-alloyed tool steels with carbon contents of 0.5–0.7% and tensile strength between 700 and 1000 N/mm2 are common.

243

244

16 Build-up Brazed Wear-protection Coatings

Fig. 16.6 Stress reduction in hard-metal/steel joints.

The chief applications for hard metal coatings are tools industry, deep drilling, mining industry, logging, wood processing, as well as minerals processing. 16.2.4 Brazing of Ceramics

Ceramic materials are used for applications with very high requirements in terms of hardness, wear resistance, and often combined corrosion resistance. Prominent ceramic types for wear protection are aluminium oxide, zirconium oxide, silicon oxide, and silicon nitride. Figure 16.7 shows an application in engine construction [4]. A ceramic sliding block made of silicon nitride is brazed to the slide face of a rocker arm in order to reduce wear of the camshaft and rocker arm. Brazing is the dominant joining technology for ceramic/metal joints. As described above for hard metals, different thermal expansion behaviour of ceramic and metal materials presents a problem. Therefore, special metals that offer matched coefficients of thermal expansion, and particularly ductile filler metals are selected. A second peculiarity of joining ceramics is that these materials are exceptionally difficult to wet. The common adhesion mechanism for brazing is not applicable here as the material, by definition, is non-metallic. Instead, special filler metals are applied that react with the ceramic due to alloying elements such as titanium, zirconium, or hafnium, and allow wetting by forming predominantly ceramic phases. Table 16.1 lists industrially common filler metals

16.2 Brazing and Soldering

245

Fig. 16.7 Ceramic applied to a rocker arm.

for brazing of ceramics and ceramic/metal joints, referred to as active brazing filler metals. Apart from the ceramic materials introduced above, so-called super-hard materials are used for wear protection, e.g. natural diamond, synthetic diamond, or cubic boron nitride. Here, brazing is the most commonly used joining technology as well. The main applications are cutting tools, components in deep-drilling systems, cutting discs for stone machining, as well as wear surfaces on measuring tools and rulers. For further information see [5]. Economic drilling

Table 16.1 Active brazing filler metals for brazing of ceramics. Composition in weight per cent Ag

Cu

In

other

Degussa CB 1 Degussa CB 2 Degussa CB 4

75 100 72.5

20 – 27.5

5 – –

– – –

Degussa CB 5 Degussa CB 6 Degussa CS 1

65 99 10.5

35 – –

– 1 –

Degussa CS 2

–

–

4

– – 89.5 Sn 96 Pb

a) b)

titanium activated

Filler metal

Melting range in 8C

Brazing temperature in 8C

Substrate materials

730–760 970 780–805

850–950 a) 1000–1050 b) 850–950 a)

770–810 948–959 221–300

850–950 b) 1000–1050 b) 850–950 b)

320–325

850–950b)

ceramics, ceramic compounds, graphite, and diamond – silicon nitride ceramics, graphite glass

Allows brazing at 850 8C, higher brazing temperatures improve wetting behaviour. Due to relatively high vapour pressure of silver, brazing temperature in vacuum brazing processes should not exceed 1000 8C. Under argon, brazing temperatures of up to 1050 8C are tolerable.

246

16 Build-up Brazed Wear-protection Coatings Fig. 16.8 Drill with brazed diamond insert.

of highly abrasive composite materials in series production requires drills with long service life, as shown in Fig. 16.8. Here, a polycrystalline diamond insert was brazed to a shaft made of sintered hard metal. The strength of ceramic/ceramic and ceramic/metal brazed joints is usually measured in a four-point bending test. Depending on the material combination, part geometry, and process parameters, strength values can exceed 200 N/mm2. 16.2.5 Brazing of Hard-material Particles

So far, plating-type build-up brazing of wear-resistant material has been introduced. Now, an additional variant will be discussed. In this technique, hard-material particles are embedded in a metal matrix by brazing and are applied to the part surface. Surface structure varies considerably and depends on grain geometry as well as grain size of the hard-material particles, etc. The method can be used, e.g., to produce very rough surfaces with coarse tungsten carbide chips that collect large amounts of abraded material. The abraded material itself thereby takes over part of the wear-protecting function of the coating. Surface structures with coarse, sharp-edged hard-material particles are also used for abrasive tools. Figure 16.9 gives an application example of a milling cutter used in hard-rubber machining. While hard metal particles can be applied to a surface with standard filler metals such as CuNiZn (Alpacca silver, sometimes referred to as German silver or new silver) or nickel-based filler metals, similar coatings made of ceramic hard material again require special filler metals. Ceramic hard-material grains pro-

16.2 Brazing and Soldering Fig. 16.9 Milling tool with brazed hard-metal chips.

Fig. 16.10 Wire saw segments with brazed diamond grains.

247

248

16 Build-up Brazed Wear-protection Coatings

vide the highest hardness values, and therefore, are used in machining of highly abrasive substrate materials. Wire saws, for example, are often used to separate large blocks of rock. Here, diamond-coated steel pellets with elastic spacers are threaded onto a steel wire, which is mounted in a bandsaw. Apart from applications for machining natural rocks, the wire saws are also used for precisely separating whole buildings. Figure 16.10 shows pellets with build-up brazed diamond grains.

16.3 BrazeCoat Technology

Build-up brazing of hard-material grains, as described above, creates more or less rough surfaces in which hard-material particles are embedded and applied to the substrate by the filler metal. When grains of the hard material are reduced to a certain size the deposited coating appears homogeneous on a macroscopic scale. The product can then be referred to as a hard-material/hard-alloy composite material. This idea presented the starting point for developing the so-called BrazeCoat technology [6]. Here, polymer-bound filler metals and hard-material powders are applied that can be processed by two different methods. Coatings with thicknesses of several millimetres are produced by a technique that uses mats of flexible filler metal and hard material (BrazeCoat M technology). In contrast, a suspension of carbide powder, filler metal, and binder is applied to the surface prior to brazing for fabrication of thin coatings up to 0.5 mm thickness. 16.3.1 Coating with Mats of Filler Metal and Hard Material (BrazeCoat M)

The BrazeCoat M process allows deposition of wear-resistant coatings for protecting surfaces of components in high-load applications of mechanical and systems engineering. Here, mats of polymer-bound hard-material powders are fabricated, which are easily cut or punched to blanks or preforms with the desired size, geometry, and thickness between 0.7 and several millimetres. The mats are placed onto the surface that requires protection. A second layer of nickelbased filler-metal powder, polymer-bound as well, is aligned precisely and placed onto the first hard material mat. Typical hard materials are powders of WC, Cr3C2, or compounds of the two, filler metals are of NiCrBSi alloy type. A subsequent furnace process at approx. 1100 8C creates the final coating under hydrogen atmosphere. The molten nickel hard alloy infiltrates the applied carbide layer and at the same time brazes it to the substrate material. Technical data of the process is listed in Table 16.2. Compared to coating materials, steel behaves notably different, depending on the specific coating. The mean coefficient of thermal expansion is particularly important when considering possible stresses created during cooling. The WC/NiCrBSi (W/1002) material combination shows low values, and therefore, similar problems related to stress can oc-

16.3 BrazeCoat Technology Table 16.2 Properties of coatings produced by means of BrazeCoat. BrazeCoat

Macrohardness in HV10/HRA

Density in g/cm3

Coefficient of thermal expansion a in 10–6/K

W/1002 C/1002 CW/1002 W/21–80

1240/88 1150/86 1180/87 –/64

13.0 7.0 9.9 13.1

8.1 11.4 9.5 9.6

Fig. 16.11 Polished cross section of BrazeCoat coating.

cur for large parts as described for hard-metal compounds. Cr3C2/L-Ni2 compounds, however, show a thermal expansion behaviour very similar to that of constructional steel, yielding considerably reduced stresses in coated parts. The wear resistance of produced composite coatings is high and allows significantly longer service life of the coated components. Typical applications of BrazeCoat build-up brazed coatings are areas with high abrasive wear or combinations of abrasive wear and corrosive attack. Depending on particular application conditions, service life was increased dramatically for parts such as pump housings, mixing machines, extruders, mixer blades, deflectors, and tube bends. Coating and substrate are metallurgically bonded and thus show high adhesive strength. High hard-material contents (approx. 70% by volume) allow carbide particles to support each other and to create a stable structure. Therefore, settling of the heavy carbide particles is prevented and the coating microstructure is homogeneous (Fig. 16.11).

249

250

16 Build-up Brazed Wear-protection Coatings

Fig. 16.12 Mixer blades, coated with BrazeCoat.

A special advantage of the BrazeCoat process is that contours are accurate and sharp, which allows manufacturing close to the final contour. The technique is thus economical as costly hard-material machining is reduced to a minimum or avoided completely. Furthermore, the coatings can be applied to complex-shaped surfaces. Mixer blades (see Fig. 16.12) of chemical reaction tanks for processing of ceramic materials at 400 8C in corrosive atmospheres have proven successful under production conditions. Coated with a 3-mm thick Cr3C2/L-Ni2 wear-protection coatings, they suffered only a few tenths of a millimetre wear after 750 operating hours. Blades used earlier with a build-up weld coating of CoCrWC were worn down to the stub after the same period of operation. The BrazeCoattreated blade allowed 3000 operating hours under the same operating conditions. 16.3.2 Coating with Suspensions of Filler Metal and Hard Material (BrazeCoat S)

The BrazeCoat S technology produces coatings of 0.05 to 0.3 mm thickness by applying a hard-material/hard-alloy suspension and subsequently brazing it in a furnace process. The suspension can be applied to complex geometries as well as using simple deposition techniques, e.g. dipping, spraying, or brushing. Subsequent heat treatment at 1040 8C bonds the hard materials to the nickel alloy to form a compound, and at the same time, wets the substrate surface. The result is a hardmaterial/hard-alloy coating with a metallurgical bond to the substrate.

16.3 BrazeCoat Technology

Fig. 16.13 Channelled plate of grinding track, coated with BrazeCoat.

The produced coatings are dense, even, and nearly free of pores (< 1%). Mechanical post-treatment is unnecessary for most applications. High hard-material contents above 60% by volume yield high hardness values of approx. 65 HRC. Investigations of wear behaviour under laboratory conditions and on parts under operating conditions revealed notably increased wear resistance compared to nitrided, boronated, or thermal sprayed coatings. Impact crushers with corrugated grinding tracks are used for fine-powder grinding of substances, e.g. graphite for toner. Grains constantly impact rotors and stators of the mill and are broken down to fine particles. However, the milling tools are subject to considerable wear. BrazeCoat S allows precisely contoured well-bonded coatings on these structures that otherwise impede coating due to unfavourable geometry. Figure 16.13 shows a segment of a grinding track, protected with a 0.2-mm thick BrazeSkin coating. For many years, the application has proven successful in limiting wear to an economically justifiable range during mass production (flour, graphite), or for grinding of high-purity materials. Low material loss caused by wear yields high purity of ground stock in electronics material production. Further practical applications of successful BrazeSkin coatings include coatings on housings, rotors, or tube bends that carry abrasive solid particles pneumatically or in liquid flow (e.g. blowers, pumps, sluices, whirl gates). The ability to coat inner diameters of slender tubes and tube bends promotes new technical solutions.

251

252

16 Build-up Brazed Wear-protection Coatings

16.4 Summary

Brazing is in fact an alternative to well-established coating technologies such as build-up welding, thermal spraying, electroplating, and thermochemical methods, and is worth considering during coating technology selection. Although less known in the field of coating technologies, build-up brazing offers solutions to special coating tasks, not obtainable by any other coating technology.

References DIN 8505, Teil 1: Löten; Allgemeines; Begriffe. Soldering and Brazing; Part 1: General, Terms, Beuth, Berlin, 1979 2 Reardon, D., Feldbauer, S. L.: Stainless steel brazing in continuous belt furnaces, Paper Presented at the International Brazing and Soldering Conference IBSC 2003, AWS/ASM, San Diego, 2003 3 Mahler, W., Zimmermann, K.-F.: Löten von Hartmetallen, Technik die verbindet, Issue 30, Degussa AG, Hanau, 1985 4 Krappitz, H., Thiemann, K. H., Weise, W.: Herstellung und Betriebsverhalten 1

gelöteter Keramik-Metall-Verbunde für den Ventiltrieb von Verbrennungskraftmaschinen, DVS Report, Vol. 125, DVS, Düsseldorf, 1989 5 Kübler-Tesch, G.: Polykristalliner Diamant für Verschleißanwendungen, Diamant-Information no. 43/Verschleißtechnik, De Beers Industriediamanten GmbH, Düsseldorf 6 Lugscheider, E., Schmoor, H., Krappitz, H.: Verschleißschutz durch Auftraglöten, DVS Report, Vol. 166, DVS, Düsseldorf, 1995

253

17 Applications of Coating Processes in Brazing Technology K. Möhwald, U. Holländer, A. Laarmann, Institute for Materials Science, University of Hannover, Germany 17.1 Introduction

In the past decades, brazing has evolved into a high-performance, economical alternative to other joining technologies, e.g. welding, gluing, or riveting, for fabrication of complex parts, particularly in compounds including dissimilar materials. Here, brazing often presents the only feasible method to produce a metallurgical bond, and thus provides joints with high mechanical and thermal stability. Furnace brazing generally is the technique of choice for mass-production brazing of complex parts. Furnace processes operate either in (reducing) protective gas atmospheres, with or without application of brazing fluxes, or fluxless, under high-vacuum conditions. Wherever applicable, automation is used to apply brazing filler metals. Except for roll bonding of aluminium sheet metal with AlSi filler metals, coating technologies for brazing filler-metal deposition onto semi-finished products for subsequent production of brazed part components (e.g. heat exchangers) are still uncommon. In fact, filler metal is applied directly to the parts in nearly all applications, usually by screen printing of filler-metal paste or spraying of fillermetal suspension. Manual application of filler-metal paste, wire, or foil, is labour intensive and time consuming, hence is becoming less important for economical reasons. Nowadays, these techniques are limited to areas where part and joining geometries preclude automated processes. Conventional automated methods of filler-metal application have obvious disadvantages: 1. Parts preloaded with filler-metal suspension or paste require time-consuming drying prior to assembly and charging. 2. Paste or spray must be applied shortly before brazing. Longer storage periods of preloaded components, particularly in humid air, lead to corrosion-related quality reduction of applied filler metal due to binders and open-pored fillermetal deposits. Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

254

17 Applications of Coating Processes in Brazing Technology

3. The mechanical stability of parts loaded with filler metal (cohesion and adhesion of filler metal) is poor in spite of binders added to suspensions and pastes. Therefore, machining by cutting and forming, or even transport of parts with filler metal applied, is limited if not impossible. These methods are thus inappropriate for pre-placing brazing filler metal onto semi-finished products. 4. Binders decompose to gaseous products during the brazing process. They cause defects (pores) especially in the interior or large lap joints. Considering the disadvantages of conventionally applied filler metal, coating technologies for filler-metal application prior to brazing have a number of advantages. Deposited coatings are composed of pure filler metal, processes yield comparatively high adhesion strength, and the coatings show either low or no porosity at all. Furthermore, modern coating processes allow high degrees of automation and reproducibility in terms of coating thickness and homogeneity. The amount of filler metal can thus be controlled precisely. However, not every coating technology is equally suitable for deposition of any desired filler metal. This chapter focuses on the suitability and individual limitations of thermal spraying, electroplating, and PVD technology for preplaced filler-metal deposition onto parts and semi-finished products. The capabilities of processes are illustrated by practical applications of filler metal.

17.2 Brazing Filler-metal Application by Thermal Spraying

Thermal spraying technologies, as described in several sections of this book, are generally capable of depositing virtually any metal alloy. Selecting an appropriate spraying technology and process parameters for deposition of a given filler metal, must consider that partial oxidation of filler metal during spray processes should be prevented. Oxidic phases in the filler-metal coating can possibly reduce wetting ability and flowability of the brazing filler metal dramatically [1–3]. Filler metals susceptible to oxidation should be sprayed using high-velocity techniques (HVOF, D-Gun, CGS) with low thermal load and a short period of exposition to the atmosphere. Processes in which spray particles are not exposed to atmospheric oxygen (CAPS, VPS, UPS, etc.) are particularly suitable. However, these are sophisticated and expensive methods. A large group of brazing filler metals suitable for spraying with the particularly economical atmospheric spraying processes are nickel-based filler metals, provided that processes are carefully controlled. The composition of these filler metals is similar to that of self-fluxing thermal spray alloys. They are commonly used in mass production of stainless steel and super-alloy parts (heat exchangers, exhaust gas coolers, etc.), and have reached the highest market share among high-temperature brazing filler metals (filler metals for brazing temperatures above 950 8C), apart from copper-based filler metals.

17.2 Brazing Filler-metal Application by Thermal Spraying

255

Table 17.1 Investigated nickel-based filler metals. Filler metal (ISO 3677)

Material code

Ni

Cr

Si

B

Fe

P

B-Ni74CrFeSiB-980/1070 B-Ni82CrSiBFe-970/1000 B-Ni71CrSi-1080/1135 B-Ni89P-875 B-Ni76CrP-890

NI1A1 NI102 NI105 NI106 NI107

73.9 82.4 71.0 89.0 76.0

14.0 7.0 19.0 – 14.0

4.5 4.5 10.0 – –

3.1 3.1 – – –

4.5 3.0 – – –

– – – 11.0 10.0

Table 17.1 lists representatively tested brazing filler metals from the group of nickel-based filler metals, applied by APS, three-cathode APS, and HVOF. Experimental investigations assessed a number of parameters during the process phases of surface pre-treatment, pre-placing of filler metal by thermal spraying, and brazing, taking the following considerations into account: 1. Best practice surface pre-treatment depending on substrate material and brazing filler metal. 2. Selection of appropriate spraying technology and optimal spray parameters with respect to filler-metal type and spray coating suitability for brazing (grain-size distribution of filler-metal powder, coating thickness, porosity, oxide content, etc.). 3. Selection of best-practice brazing process and parameters with respect to filler-metal type. 4. Mechanical-technological properties of produced brazed joints. 5. Adaptation of thermal spray filler-metal deposition to special geometrical features of semi-finished products or parts. 6. Economic assessment of thermal spray filler-metal deposition compared to conventional methods of filler-metal application. Figure 17.1 shows the general process optimisation approach, initially performed on samples with simple geometry. Essential criteria for characterising the produced brazing filler-metal coatings are coating composition and microstructure, adhesion to the substrate, as well as coating porosity. Similar to conventional metallic spray coatings, acceptable filler-metal coatings show minimum oxide contents and are free of defects at the interface to the substrate. Pores in the filler-metal coating are of less importance as most usually disappear during melting in the brazing process. Brazed joints fabricated from samples with pre-placed filler metal were also investigated. Acceptable brazed joints show a metallurgical bond to the substrate, free of defects, as well as oxide-free and mostly non-porous filler metal structure. After process optimisation, plasma-sprayed (single-cathode APS, three-cathode APS) and HVOF-sprayed coatings of the nickel-based filler metals yielded brazing results equivalent to those of brazed parts with conventionally pre-placed filler metal.

256

17 Applications of Coating Processes in Brazing Technology

Fig. 17.1 Experimental procedure.

Transferring the technology from samples with simple geometry to practical parts that usually show significantly more complex joint geometry, requires further part-specific investigations. The following questions, which also relate to design aspects, are considered: 1. What is the optimal joint geometry with respect to the desired joint clearance? 2. What thickness of filler-metal coating is necessary in order to guarantee sufficient filler-metal amount for the desired joining task? 3. Is coating of every joining component necessary or is partial filler-metal coating of selected areas or individual components sufficient? 4. If necessary, is machining of the filler-metal-coated part possible (milling, drilling, forming)? 5. Do brazed parts meet requirements such as sealing functions, strength, or corrosion resistance? These topics were investigated thoroughly for simple demonstrators, as shown in Fig. 17.2, as well as for industrially manufactured parts of various stainless steel heat exchangers. Best-practice part-specific coating and brazing strategies were determined in order to meet technical demands of industrial applications. Apart from deposition of nickel-based filler metals, the techniques presented here are applicable to other brazing filler metals. Today, thermally sprayed AlSi filler-metal coatings are used successfully, for example, in fabrication of aluminium parts suitable for brazing in a Nocolok process. Currently, economical aspects of brazing filler-metal deposition by means of thermal spraying are subject to investigations in several applications. Such economic considerations cannot yield generally applicable results but require specific analysis of the individual application. However, distinct cost factors of conventional filler-metal placing techniques and thermal spraying for pre-placing of filler metal can be categorised according to qualitative classification as shown in

17.3 Electroplating and Electroless Plating Methods for Brazing Filler-metal Application

257

Fig. 17.2 Stainless steel demonstrators coated and brazed with nickel-based filler metals.

Table 17.2 Profitability of application techniques for brazing filler metals. Cost factors

Investment Process Expenditure of labour Material Output Process stability

Application techniques Manual paste application

Melt-spin foils

Filler-metal suspensions

Screen printing

Thermal spraying

none none high low low low

none none high high low low

low low moderate low moderate moderate

moderate low low low moderate moderate

high moderate low low high high

Table 17.2. Hence, economical thermal spraying for brazing filler-metal deposition is given in applications where the technique demonstrates its particular advantages such as low personnel costs, high degree of automation, process stability, and high coating rates. This applies particularly to mass production of parts that require large-area pre-placing of filler metal without needing immediate nearby brazing.

17.3 Electroplating and Electroless Plating Methods for Brazing Filler-metal Application

As reported elsewhere in this book, electroplating and electroless (chemical) plating technologies allow fabrication of many types of functional metal coatings on a large number of different substrates [4]. Electroplated and electroless coatings are characterised as precisely adjustable thin coatings with high reproducibility and purity. Bonding mechanisms to metal substrates are similar to wetting if the substrate was pre-treated correctly. Coatings thus show high adhesive strength.

258

chromium cobalt copper iron manganese nickel tin lead zinc silver gold etc.

17 Applications of Coating Processes in Brazing Technology

tin/copper zinc/copper tin/lead etc.

nickel, copper, or chromium matrix with embedded particles silver copper gold

e.g. carbides, nitrides, oxides, but also silicon

Ni-phosphorous (3–3%) Ni-boron (1–7%) copper silver gold tin etc.

Fig. 17.3 Overview of electroplating and electroless plating techniques.

Figure 17.3 summarises technologies and associated metallic coating materials. A large part of the metals in this overview are typically found in filler metals for soldering, brazing, and high-temperature brazing. Copper, silver, and gold are well-established, precious filler metals. Copper is particularly relevant as a commercial high-temperature filler metal for all types of steel grades. Electroless nickel, deposited from hypophosphite containing electrolytes, is an example of a coating that deposits directly as an alloy. Due to process characteristics, the coating usually contains approx. 10 to 11% by weight phosphorous [5]. This electroless nickel coating features a composition equivalent to the nickel-based brazing filler metal B-Ni89P-875 (NI107), and thus can serve directly as a brazing filler-metal coating. Other filler-metal alloys with common compositions can be produced by means of multi-layer electroplating and electroless plating of significant metal components. During brazing, individual layers form the actual filler-metal alloy in situ by contact reaction. An example of this is the B-Ni76CrP-890 brazing filler-metal alloy. A triple-layer coating with two layers of NiP11 and an electroplated chromium intermediate layer was deposited in order to produce a fillermetal coating with appropriate composition (Fig. 17.4). The thickness of the individual layers was adjusted to yield the desired filler-metal alloy (76% Ni, 14% Cr, 10% P) during brazing. Stainless steel coated in the manner described shows excellent behaviour in brazing due to the existing adhesion between filler metal and substrate. Use as semi-finished parts is possible as the coating tolerates virtually any forming and mechanical processing without developing defects (Fig. 17.5).

17.4 Brazing Filler-metal Application by PVD Fig. 17.4 B-Ni76CrP-890 brazing filler-metal coating, produced by combined electroplating and electroless plating.

Fig. 17.5 Stainless steel sheet coated with brazing filler metal and folded.

Electroplated and electroless plated coatings are not limited to applications on semi-finished parts and components. Techniques also allow fabrication of, e.g., large-sized filler-metal foils or wires. These applications are particularly relevant for Ni-based alloys because lack of ductility impedes rolling of filler-metal foils and drawing of filler-metal wires. Current research projects are focusing on this field.

17.4 Brazing Filler-metal Application by PVD

PVD technology, being one of the modern thin-film technologies, offers a wide range of depositable metal coatings suitable for brazing due to various processes available today (cf. corresponding chapters in this book). PVD allows deposition of brazing filler-metal alloys, metal thin films as wetting surfaces, as well as reactive intermediate layers for materials that are difficult to braze by conventional means (see below). Metallic PVD coatings show extremely high purity and thus guarantee high-quality filler-metal coatings and metallisations with excellent substrate adhesion. Depositable metals and alloys are subject to only a few restrictions. Vaporising of very low melting metals (zinc, indium) requires careful control of process pa-

259

260

17 Applications of Coating Processes in Brazing Technology

Fig. 17.6 Ion plated multi-layer brazing filler metals.

rameters as too high sputtering power can melt the target in spite of cooling. Also, metals with high vapour pressure (Zn, Mg) are difficult to process in PVD and are thus inappropriate for filler-metal deposition. Additionally, ferromagnetic metals and alloys, including nickel, cannot be evaporated in magnetron sputter processes. Deposition of these materials requires other PVD techniques such as arc PVD, EB-PVD, etc. PVD technology resolves challenging joining problems. The following example discusses fabrication of high-strength joints of dissimilar joining components [6]: Arc PVD coatings, composed of a reactive Ti, Zr, Hf, or Cr bond coat and a Ni top coat, are suitable for metallisation of soldering, brazing, and high-temperature brazing filler metals. This applies particularly to metallic and ceramic substrates that restrict or prevent wetting of conventional filler metal. Here, composition and structure of the coatings can be matched to the substrate in order to produce a diffusion-barrier effect during brazing. This effect prevents interdiffusion of foreign elements. During cooling stages of the brazing process the filler metal solidifies, developing a comparatively low content of brittle phases. The technique thus allows production of dissimilar metallic and metalceramic brazed joints with high strength, either not attainable in conventional processes or with insufficient strength for technical applications. PVD filler-metal coatings, not available by means of other technologies, can be designed specifically for dissimilar joining components such as titanium/ steel joints. TiAl6V4/X5CrNi18 10 brazed joints with ion-plated (arc PVD) CrNiCu multi-layer filler metal showed a tensile strength as high as the yield strength of X5CrNi18 10. Furthermore, the multi-layer brazing filler metal solidifies isothermally and, therefore, features a re-heating melting point that is actually higher than the brazing temperature. The thermal stability of the produced brazed joint is thus comparably high (Fig. 17.6). Future applications of PVD processes in joining technology will use hybrid processes that will allow filler-metal deposition and brazing within a single PVD chamber. Current research work shows that, for filigree parts, PVD not only allows deposition of filler-metal coatings. Furthermore, glow-discharge plasmas,

17.5 Summary and Conclusions

Fig. 17.7 Temperature measurements in glow-discharge plasma for selected levels of generator power and bias voltages in a magnetron sputter system (cathode material Cu, tip of thermocouple centred at 65 mm distance from cathode); solid graph data: high-frequency plasma; interrupted graph data: direct current plasma with operating cathode.

usually used in PVD for part cleaning (ion etching) and ion plating, can serve as a source of heat for subsequent brazing (Fig. 17.7). The method has already been used for successful production of microbrazed joints.

17.5 Summary and Conclusions

Joining technologies are an important factor when considering increasing cost pressure in the metal-processing industry. Developing new technologies is essential for further reduction of material and process costs. For thermal spraying, electroplating and electroless plating, as well as PVD technology, the results presented here show that modern coating technologies are particularly suitable for deposition of brazing filler metals. However, special demands of brazing fillermetal deposition require technology modifications as these techniques are usually used for wear- and corrosion-protective coatings. Development of stainless steel sheet coated with brazing filler metal opens up encouraging economic and technological outlooks. Considerable demand for these semi-finished products, as for roll-bonded brazing filler metal on aluminium sheet, is found, for example, in the fabrication of heat exchangers. Additionally, automated pre-placing of filler metal, flexibly adaptable to part geome-

261

262

17 Applications of Coating Processes in Brazing Technology

try, broadens the market potentials for a number of applications, e.g. thermal spraying systems adapted as end effectors on robots.

References Bach, Fr.-W., Möhwald, K., Bach, C., Holländer, U.: Thermally sprayed filler-metal coatings for high temperature brazing. In: Conference Proceedings, ITSC 2004 – International Thermal Spray Conference & Exposition, Osaka, Japan, May 10th–12th, 2004: Thermal Spray Solutions – Advances in Technology and Application. DVS, Düsseldorf, 2004 2 Bach, Fr.-W., Möhwald, K., Demmler, A., Bach, C.: Neue Lotapplikationstechniken mittels Thermischer Spritzverfahren. In: Hart- und Hochtemperaturlöten und Diffusionsschweißen. Conference Proceedings, Löt 2004, June 15th–17th, 2004, Aachen, DVS Report Vol. 231. DVS, Düsseldorf, 2004 3 Füssel, U., Eckhart, G., Knepper, P., Scheffler, O.: Eignung thermisch gespritzter Schichten aus Nickel-Basisle1

gierungen zum Hartlöten von ChromNickel-Stählen in Abhängigkeit vom Spritzverfahren und den gewählten Spritzparametern. DVS Report Vol. 166, Deutscher Verlag für Schweißtechnik, Düsseldorf, 1995, pp. 40–45 4 Durney, Lawrence J.: Electrochemical and Chemical Deposition. Ullmann’s Encyclopedia of Industrial Chemistry, Vol. A9, pp. 125–181 5 Burkhardt, W: Über galvanisch und chemisch-reduktiv abgeschiedene Schichten für funktionelle Anwendungen. Part 6: Chemisch-reduktive Abscheidungsverfahren für Ni-P und Ni-B-Schichten. Galvanotechnik, Vol. 85 (1994) Issue 1, pp. 82–88 6 Möhwald, K.: Einsatz des Ionenplattierens beim Löten. Doctoral Thesis, University of Dortmund, 1996

263

18 Surface Protection by Means of Build-up Welding A. Gebert, B. Bouaifi, CeWOTec gGmbH, Chemnitz, Germany 18.1 Introduction

Build-up welding is a technique where a coating is applied via a fully or partially molten phase. A metallurgical bond between coating and substrate material is created when substrate as well as coating materials are melted. In a special process variant – build-up brazing – only one of the joining components is partially or fully melted so that diffusion processes play a decisive role in adhesion. However, transition between both process variants is smooth, and they are often generally referred to as build-up welding. Compared to other coatings, build-up welded coatings feature higher thickness and/or higher adhesion. In principle, every welding technique is appropriate for build-up welding (Table 18.1). During the historical development of welding technology, each welding technique was also used for build-up welding. Nowadays, due to process characteristics, however, certain popular welding processes or variants stand out. Benchmark values include obtainable coating quality, property transition to the substrate material, and deposition rate. Protective coatings by means of build-up welding can be used locally, or for large areas. Similar to other coating technologies, build-up welding allows fabrication of coatings designed specifically to withstand surface attack in terms of corrosion, wear, or combined loads of corrosion and wear. The desired part strength is attained by an appropriate selection of substrate material and, if necessary, heat treatment of the part prior to build-up welding. Coating materials are thus applicable that would not be suitable for bulk material in terms of strength or brittleness.

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

264

18 Surface Protection by Means of Build-up Welding

Table 18.1 Classification of welding technologies. Build-up welding technologies

Pressure welding

Fusion welding

Induction welding

Gas welding

Arc welding

Manual arc welding

Inert gas shielded arc welding

Beam welding

Submerged arc welding

Consumable electrode

MIG

MAG

Roll seam welding

Wire powder welding

Laser welding

Electron beam welding

Friction welding

Resistance fusion welding

Electroslag welding (RES)

Non-consumable electrode

TIG

Plasma welding

18.2 Process Variants

With respect to applications, DIN 1910 classifies build-up welding and distinguishes the following techniques: Armour plating: for increased wear resistance Plating: for increased corrosion resistance Buffering: production of application-oriented bond characteristics for dissimilar materials. Alloys containing hard phases and pseudo-alloys (composites of matrix material and hard phases) are used in armour plating. Also, in situ formation of hard phases during build-up welding is possible. Alloy composition and microstructural morphology (matrix hardness, arrangement, size, and amount of hard phases) of coating material are matched to the friction pairing of the particular application. Plating involves highly corrosion resistant metals or alloys. Atmospheric corrosion at room temperature is distinguished from corrosion caused by the attack of chemically active agents, considering potentially elevated operating temperature.

18.3 Characterisation of Build-up Welded Coatings

Industrial practice often requires combining the two techniques. However, only single-phase materials yield the highest corrosion resistance. Thus, the corrosion resistance of alloys and pseudo-alloys with embedded hard phases is limited. Buffering is a special technique that produces multi-layer coatings with gradients of properties. It is applicable for systems where substrate and coating materials interact to form brittle phases, which would lead to adhesion problems and could ultimately cause spalling of the coating. An additional reason for buffering arises when partial warming and subsequent contraction in build-up welding creates high internal stresses within the coating. These stresses are particularly high for dissimilar coating materials with different coefficients of thermal expansion. The resulting hot cracks and contraction cracks are often tolerated. If this is not the case, however, ductile intermediate layers can reduce internal stresses and prevent the formation of cracks. Buffer coatings should be limited to applications where they are absolutely necessary. Measures should rather include alloy and process optimisation in order to guarantee sufficient coating adhesion. This applies particularly to wear protection applications where high forces can promote a shifting of the functional top coat on the buffering layer and, therefore, cause geometrical changes that would impair part function.

18.3 Characterisation of Build-up Welded Coatings

Build-up welding, compared to other deposition techniques, generally produces coatings with higher adhesion due to metallurgical bond created by partially or fully molten materials. The joint between coating and substrate is never the weakest area of the compound as long as adhesion-reducing hard phases, produced by inadequate combinations of coating and substrate material, are prevented. These coatings are thus particularly appropriate for applications with heavy wear conditions. Additionally, they show high edge strength. Coating structures for corrosion protection are single-phased or with only few phases as phase boundaries are subject to potential corrosive attack. Wear-protective coatings have a multi-phase structure that includes matrix material and different hard phases. Microstructure and hard phases, in particular, form according to fusion metallurgical laws. Phases with melting points higher than process temperatures in build-up welding, usually carbides in wear-protection materials, allow control of coating morphology (carbide grain size) by adjusting the configuration of the added material. The size and shape of hard phases should be matched to the particular load spectrum. Hard eutectic matrix material has the advantage of preventing matrix erosion caused by contacting abrasive material. However, it also leads to increased brittleness of the coating material. In-depth knowledge of binary and polynary systems, as well as the temperature influence on phase formation, is thus an important prerequisite for development and use of coating materials.

265

266

18 Surface Protection by Means of Build-up Welding

Fig. 18.1 Stress cracking in build-up welding of fourth layer (shape welding).

Depending on the process variant, coatings produced by build-up welding show a thickness range of 1 to 6 mm. Multi-layer production can even yield considerably higher coating thickness. A special process variant is shape welding. Multi-layer build-up welding can be used to deposit desired 3D geometries onto a base plate in order to reduce complicated cutting processes in machining. Coating thickness is limited by internal stresses that rise with increasing coating thickness and, particularly, with growing number of subsequent passes, due to multiple warming and cooling. At worst, these internal stresses can cause coating delamination (Fig. 18.1). Coatings are self-supporting due to their thickness. Thus, a substrate material with appropriate basic strength is adequate. Constructional steel is often sufficient. Under heavy load conditions, higher-strength tempered or tool steels are certainly applicable as well. Even thermally sensitive or low-melting materials such as magnesium can be coated by means of build-up welding. The ductility and strong adhesion of coatings for corrosion protection allow subsequent forming of coated parts. This, however, usually does not apply to wear-protection coatings due to rising brittleness caused by the increasing content of hard material. A fusion metallurgical bond always creates a mixture of coating and substrate materials. The ratio of molten substrate material to the total volume of molten material is a characteristic value, referred to as degree of fusion: Aˆ

Vsubstrate
100 Vtotal

…A in per cent†

…1†

Practical measurements of the degree of fusion involve either planimetric methods on cross sections, or spectrometric analysis of substrate and coating material. For enhanced accuracy, spectrometric calculations focus on the alloying element with the greatest variation between the two materials:   qc mc 1 qsubstrate mfc   Aˆ
100 …A in per cent† …2† qc mc 1 ‡ 1 mfc q substrate

– density of coating material qc qsubstrate – density of substrate material

18.3 Characterisation of Build-up Welded Coatings

Fig. 18.2 Polished cross section of build-up welded machine knife (plasma powder technique).

mc mfc

– percentage by mass of alloying element in coating material – percentage by mass of alloying element in fused coating

If the difference in density of substrate and coating material is negligible, calculation of the degree of fusion simplifies to:  Aˆ 1

 mfc
100 mc

…A in per cent†

…3†

Fusion dilutes the coating material and changes its properties. The degree of fusion is characteristic of an applied build-up welding technique and welding parameters. Processes with high energy density in the operating beam produce the lowest degrees of fusion. Therefore, apart from energy-beam processes, plasma powder build-up welding is particularly appropriate for build-up welding. For processes with higher degrees of fusion, the expected fusion of the coating material must be considered in advance, and balanced by over-alloying during fabrication of the coating material. The appropriate welding technique is selected from the wide range of process variants with respect to geometrical requirements and quality demands. Simple manual welding techniques require skilful welders in order to produce coatings with sufficient quality. However, manual techniques are common in repair welding. Mechanised or automated welding technologies guarantee high coating quality as well as reproducibility. Without any subsequent machining, geometries produced by fusion build-up welding are adequate for applications with low requirements in terms of geometrical accuracy. Applications with precisely defined geometry, e.g. build-up welded machine blades, require mechanical post-processing of deposited coatings. Apart from milling, eroding is applicable for highly wear-resistant coatings. Technologies using water jets or laser beams have proven successful for subsequent cutting of build-up welded parts.

267

268

18 Surface Protection by Means of Build-up Welding

18.4 Build-up Welding Techniques 18.4.1 Distinguishing Features

Torch dimensions and modes of operation suggest a preferential purpose for each welding technique. However, choosing an appropriate build-up welding process, compared with other coating technologies, generally is less influenced by part geometry. Build-up welding processes are distinguished by the operating media, modes of operation, shielding of the molten pool to protect against oxygen penetration, melting performance, and specific fields of applications (Table 18.2). The following paragraphs focus on individual processes and associated characteristics. Selection of applicable build-up welding processes is subject to required part and coating properties. Selection criteria are: · coating quality · coating thickness · degree of fusion · coating formation · reproducibility and homogeneity of coating properties across the coating profile · systems availability · economic considerations (process costs, deposition rate). Deposition costs, in particular, have promoted the use of simple welding technologies in spite of poor reproducibility of coating geometry and properties. Additionally, welding technologies can be distinguished according to the means of protection of the hot deposition area. Categories include open processes (gas flame), processes that include covering of the molten pool (manual arc welding, RES, submerged arc welding), and protective gas technologies (MIG, TIG, plasma). This criterion is particularly important for assessing consumables that tend to show decomposition of hard materials when the molten pool is exposed to oxygen. Production of high-quality coatings should generally consider covering or protective gas shielding of the hot, molten pool against oxygen penetration in order to reduce porosity. As for the coating material, consumables for highly wear-resistant coatings are often available as powder, only. Here, processes that require consumables as wires or strips are restricted to cored wires or sintered strips as workarounds. However, hard-material contents are limited to maximum technical fill factors.

1.6–5 mm a)

1 kg/h

low

small areas, edges, repairs

difficult to control, 15–30%

2–4 mm a)

Real melting rate

Equipment costs

Applications

Degree of fusion

Coating thickness (single pass)

a)

10–30%

manual

Operation

4–8 mm

high, 15–25%

small to large areas, new parts, repairs

5–8 mm

1–6 mm

low, 5–20%

0.1–2 mm

low

smallest to small areas, new parts, repairs

small to large areas, new parts, repairs

large areas, new parts

high, 13–40%

high

1–2 kg/h a)

(manual), automated

powder

Beam

raised

2–30 kg/h a)

(manual), mechanised, automated

powder (wire)

Plasma

raised

10–40 kg/h a)

8–9 kg/h a) raised

mechanised, automated

strip/wire

Submerged arc

manual, mechanised, automated

wire/cored wire

MIG/MAG

According to Wahl (symposium on build-up welding, June 4th–5th, 1996, Halle/Saale, Germany)

small to medium-sized areas, edges, repairs, new parts

low

2 kg/h

manual, mechanised

rod/wire

electrode/rod

Consumables

TIG

Manual arc welding

Welding technique

Table 18.2 Characteristics of build-up welding techniques.

4–5 mm

high, 10–15%

large parts, new parts

raised

18 kg/h a)

mechanised, automated

strip/wire

RES

18.4 Build-up Welding Techniques 269

270

18 Surface Protection by Means of Build-up Welding

18.4.2 Shop Welding (Manual Arc Welding, Gas Flame)

Simple manual techniques using build-up welding electrodes or rod-shaped consumables in welding torches are still commonly employed, particularly in repair work. This is due mainly to the fact that their use does not require any particularly system technology or specialised knowledge. Coating quality is rather determined by experience and the quality of the individual workmanship of the welder. In terms of reproducibility, the degree of fusion is hardly controllable. The deposition rate is low and materials are limited to rod shape. However, sophisticated production methods also allow carbide-containing materials. Fluxes in melting electrodes that cause formation of slag provide a means of protecting the hot molten pool against oxygen penetration in manual arc welding. Decarburisation and oxidation effects are thus prevented. Flame spraying with subsequent fusion is a special type of build-up welding technique. The process uses a welding torch with an integrated powder feeder (Fig. 18.3). The kinetic energy of the gas jet accelerates the powdery coating material, which is simultaneously heated in the flame. The principle is similar to metal spraying. Coating material is heated on the substrate surface by the gas flame until a fusion bond is created. Here, self-fluxing alloys are used with melting points well below the melting point of the substrate material. The process is conducted either in a single step or by pre-placing the material and subsequently melting the deposit by means of high-performance torches. The attainable coating thickness is below 1 mm. Only excellent workmanship of welders yields acceptable coating quality.

Fig. 18.3 Torch for flame spraying and fusion (company photograph Castolin GmbH).

18.4 Build-up Welding Techniques Fig. 18.4 Schematic representation of RES technique.

18.4.3 Processes with Protective Slag 18.4.3.1 Electroslag Build-up Welding (RES – Resistance Electroslag) Figure 18.4 illustrates the basic principle of the process. As the molten pool defies direct observation, electroslag build-up welding is generally mechanised. Heat input is created by resistance heating of a strip, fed continuously to, and molten by the molten slag. Welding currents can rise to 3000 A. The system technology is robust and simple. Commonly used slag material is a composite of calcium fluoride (50–90%) and aluminium oxide (10–40%). Due to process characteristics, high melting rates and heat input are characteristic, limiting applications to large parts. Frequently used consumables are strips with a width of up to 180 mm. The technique is used primarily for plating. Predominant consumable materials are Ni and Ni alloys. The high degree of fusion often requires multi-pass operation. Recent developments of sintered strips are aimed at broadening applications in order to produce wear-resistant coatings as well.

18.4.3.2 Submerged Arc Build-up Welding Similar to RES, submerged arc build-up welding uses slag in order to shield the build-up welding deposit. Material is heated by an arc ignited between the continuously fed, melting electrode and the workpiece. The arc burns inside a cavern of partially molten slag, formed by the generated gases. The liquid part of the slag covers the weld pool and has to be removed after cooling. Ideally, the slag chips off easily during the process. Unmolten slag can be recycled. Welding powders differ in terms of composition, grain size, and production methods. Active slag influences the welding processes as well as coating properties. Electrode material is available as wires, bands, and wide strips (> 200 mm). Magnetic arc movement along the edge of the strip promotes a uniform meltoff when using wide strips. Certain process variants feature several electrodes

271

272

18 Surface Protection by Means of Build-up Welding Table 18.3 Variants of submerged arc welding. Variant

Schematic illustration

Characteristics

Single-wire welding

· single wire electrode · single power supply · single control unit

Twin-wire welding

· two wire electrodes · single power supply · single control unit

Tandem welding

· two wire electrodes · two power supplies · two control units

Series arc welding

· two wire electrodes · single power supply · arc between electrodes

Submerged arc welding with iron powder or cut wire

· · · ·

Submerged arc welding with strip electrode

· strip electrodes up to approx. 200 mm · single power supply

single single single single

electrode powder or cut wire feeder power supply control unit

(Table 18.3). The process is frequently used for plating with Ni coating materials. However, due to high degrees of fusion, more than three subsequent layers are required in order to yield high corrosion protection. It is characteristic to the process that lower degrees of fusion (< 10%) produce considerable porosity. Wear-resistant coatings can be deposited from sintered strips or cored wires. A relatively high degree of fusion though is disturbing. Multi-pass operation for higher coating thickness results in an expensive process, particularly when processing high-priced consumables.

18.4 Build-up Welding Techniques

An additional process characteristic is high melting loss of alloying elements and pick-up from the slag. Production of high corrosion-resistant coatings is impeded particularly by Si pick-up from the slag. Due to considerable heat input from consumable materials and slag, this process also is appropriate for large parts only. 18.4.4 Inert-gas-shielded Arc Welding

While the protective gas shield in processes with protective slag is generated within the slag, processes with inert gas use a shielding gas, usually argon, fed by the welding torch. Formation of the welding bead can be promoted by using protective gas with small amounts of active substances, e.g. carbon dioxide or oxygen. Certain extremely sensitive consumables or substrate materials (e.g. titanium) require special measures in order to guarantee reliable shielding with protective gas. Apart from additional trailing nozzles that follow the torch and provide shielding until the welding bead has cooled sufficiently, use of protective gas chambers can be necessary.

18.4.4.1 Tungsten Inert Gas Build-up Welding (TIG Process) In TIG welding, the arc burns between a non-consumable, negatively charged tungsten electrode and the workpiece. A concentric gas nozzle guarantees continuous supply of a protective gas envelope (Fig. 18.5). Positive electrical charge on the workpiece intensifies the heat input to the substrate part. Heating of consumables is adjusted independently of workpiece heating by controlling the wire feeding rates or by shifting the point of wire immersion into the molten pool. A broad range of variation in temperature control is thus attained. Therefore, the technique also allows processing of thermally sensitive coating materials. Consumables are added as welding rods or wires manually or, for large surfaces, by mechanised equipment. Common wire diameters are 0.8 and 1.2 mm. Preheating wires in hot wire power sources increases melt-off rates up to a factor of three (reaching 3 kg/h). In order to prevent wire oxidation, contacting tubes for wire heating can be flooded with protective gas. Compact torch design and good observability of the molten pool allow experienced welders to control the process thoroughly. Mechanised work is generally possible but limited to special applications due to low deposition rates. For build-up welding of light metals, cathodic cleaning effects in a process variant with alternating current can be used in order to remove the oxide layer found on all light metal alloys. Short-time switching of polarity prevents overheating that would ruin the TIG torch thermally when using general polarity reversal.

273

274

18 Surface Protection by Means of Build-up Welding

Fig. 18.5 Schematic representation of TIG technique.

18.4.4.2 Gas-shielded Metal Arc Welding Gas-shielded metal arc welding (GMAW, SMAW) operates with a consumable wire electrode of coating material (Fig. 18.6). Positive charge is applied to the electrode in order to yield sufficient wire melt-off. Consumables are fed to the torch by wire-feeding equipment via a bundle of tubes from a wire coil. Certain sensitive materials with reduced strength or cored wires require an additional feeding system at the torch (push-pull systems). Electrical current is conducted through a contacting tube. The arc burns between the wire and the electrode, heats the workpiece, and melts off the wire. If inert gases are used as protective gases the process is referred to as inert gas metal arc welding (MIG). Processes using inert gases with active additives for coating formation, e.g. CO2, O2, or N2, are referred to as metal active gas welding (MAG). Additional protective gas is not needed when cored wires contain gas-producing elements (open arc welding). Coatings are deposited with mechanised forward feed as stringer beads or, by transverse oscillation of the torch, as weave beads. Manual operation is possible but coating quality, however, is usually poor. Fusion penetration of the technique is high and can reach up to 30%. Special torch placement (arc root set onto previous bead [1]) and additional cold wire feeding is aimed at limiting fusion penetration. These measures can reduce fusion penetration to 5%. Application of powdery consumables is under current investigation as well. A modern process variant is GMAW pulse welding. Here, a low base current and high-current pulses are superimposed. The method improves detaching of molten drops from the electrode and reduces fusion as well as welding spatter. Modern power sources for welding are inverter based and equipped with programmable control units. Parameters for standard procedures can be stored and

18.4 Build-up Welding Techniques Fig. 18.6 Schematic representation of GMAW technique.

Fig. 18.7 State-of-the-art power supply for GMAW with mounted wire-feed unit (company photograph Rehm Schweißtechnik GmbH).

reloaded at any time. Figure 18.7 shows an example of an industrial power source.

275

276

18 Surface Protection by Means of Build-up Welding

18.4.4.3 Plasma-transferred Arc Process (PTA) Due to process characteristics, the plasma process yields highest-grade coatings and is discussed in more detail here. Build-up welding by means of a plasma jet is distinguished from plasma spraying by the arc that burns between the torch and the workpiece (transferred arc). It directly contributes to workpiece heating (Fig. 18.8). In contrast to TIG, plasma gas flow (argon) and plasma nozzle constrict the arc that shows little divergence down to the substrate. Thus, the torch distance can vary within certain limits without changing the heat input to the workpiece. DIN 1910 distinguishes two process variants: Plasma arc welding (PAW): The constricted arc (main arc) forms the argon plasma and simultaneously heats the workpiece.

Plasma jet/plasma arc welding: In addition to the main arc, plasma formation is promoted by a pilot arc between the electrode and the nozzle on the inside of the torch. Both process variants yield high-energy deposition. Power density is the highest of all conventional welding processes (Fig. 18.9) and leads to concentrated heat input to the workpiece. Fusion penetration can thus be limited to 10% for steel. But even materials with high thermal conductivity such as copper or aluminium can be partially melted and, therefore, coated by means of plasma jet. The twin arc technique, which is more sophisticated in terms of system and controller design, is only relevant in applications with low welding currents

Fig. 18.8 Schematic process of PTA welding.

18.4 Build-up Welding Techniques Fig. 18.9 Comparison of power densities of selected welding techniques.

Fig. 18.10 Schematic representation of PTA build-up welding torch with constricted plasma arc.

(< 80 A) where heat input from the main arc is insufficient for continuous plasma formation. A pilot arc is not required in the common operating range between 100 A and 500 (600) A. Consumables are applied in powdery form via feeding units (powder feeders). Thus, apart from plasma and protective gas flow, a third gas flow, the powder carrier gas, is characteristic of the process. Separate feeding of two different powders (matrix material and hard material) into the plasma jet is possible and has advantages in certain applications. Here, complex plasma torches are necessary (Fig. 18.10). Torch design has considerable, direct impact on the results of build-up welding. Torches are water-cooled in order to prevent overheating. Furthermore, combination processes have been developed that use cold or hot wires as consumables (Fig. 18.11). Powdery consumables have the advantage that materials selection is virtually unlimited. Carbide-containing consumables, in particular, are either not available as wires and strips, or are limited in terms of carbide content when fabri-

277

278

18 Surface Protection by Means of Build-up Welding

Fig. 18.11 Schematic representation of plasma technique with welding filler wire.

Fig. 18.12 PTA alternating polarity power supply (company photograph Castolin GmbH).

cated as cored wires and strips. Therefore, highly carbide containing, powdery consumables can be deposited only by means of plasma technology. Positive electrical charge of the workpiece in plasma processes produces intensive heat input to the workpiece, but has the disadvantage that it impedes the cathodic cleaning effects of the arc, especially for light metals. Thus, highmelting oxide layers, found on all light metals, are not removed. This gave rea-

18.4 Build-up Welding Techniques

Fig. 18.13 PTA coating on wearing plate (company photograph Valco Edelstahl und Schweißtechnik GmbH).

son to develop alternating-polarity plasma techniques. Going further than alternating current techniques known from other welding processes, the technique features individual phase parts that can be shifted independently, yielding additional degrees of freedom for optimising build-up welding parameters. Figure 18.12 shows an example of an alternating-polarity plasma power supply unit. Figure 18.13 shows a typical PTA coating. 18.4.4.4 Plasma MIG Process Plasma MIG is an enhanced plasma process that combines the advantages of plasma processes with the benefits of MIG welding (Fig. 18.14). The arc at the end of a fusible wire is surrounded by a concentric plasma arc. Positive charge is applied to the fusible wire. The plasma anode is ring-shaped and thus simple to cool. This leads to a high admissible current load for the torch. System design is complex as two power sources with different control characteristics are combined. Plasma processes require constant current. MIG, in contrast, operates at constant voltage. Furthermore, up to three process gases are used in the process. The positive charge on the torch produces a cathodic cleaning effect that is important for processing of light metals. The combined process concentrates the heat input to the workpiece. Therefore, fusion behaviour is well controllable, even for highly heat conducting materials (light metals, copper). Independent control of the two arcs allows carefully aimed temperature control and prevents overheating of consumables. The technique features the following advantages: · process reliable cathodic cleaning · individually controllable melt-off performance and energy input · well-controllable geometry of welding bead · reduced welding spatter in spite of high melt-off performance.

279

280

18 Surface Protection by Means of Build-up Welding Fig. 18.14 Schematic representation of plasma MIG technique.

Plasma MIG processes have been known for years. However, processing of light metals in welding techniques is an ongoing issue for lightweight constructions not only in automotive industry. Plasma MIG is thus again a focus of attention in new developments. Applications for build-up welding can be expected. 18.4.5 Resistance Roll Seam Technique

Roll seam build-up welding with powder consumables (Fig. 18.15) can be regarded as a transition from metal spray coating to build-up welding. Coating material is poured in front of electrode rollers by a feeding unit, and melted by means of resistance heating. At the same time, electrode rollers compact the coating and produce a high-temperature sintered compound with a fusion com-

Fig. 18.15 Schematic representation of resistance roll seam build-up welding technique, following [2].

18.4 Build-up Welding Techniques

pound fraction. Coating thickness is extremely low (0.5 mm [2]) for a coating produced by build-up welding. Compared to metal spray coatings, adhesion is higher. The technique is suitable for processing rotationally symmetric parts as well as parts with plane geometry. 18.4.6 Laser Cladding

Lasers are an increasingly used power source in build-up welding. The high-energy beam creates intense heat input and thus allows low degrees of fusion. Heat input to the part is minimised and distortion remains low. Available lasers are: · gas lasers (CO2) with a wavelength of 10.6 lm · Nd : YAG solid-state lasers with a wavelength of 1064 nm · diode lasers with a wavelength of 808 and/or 940 nm. Small beam focus and available beam-travelling precision particularly qualify lasers for small parts and local precision build-up welds. This is due to the fact that, in spite of high-speed motion, deposition rates are low compared to conventional welding techniques. The laser beam is transmitted to the weld by means of encapsulated beam lead systems (CO2 laser) or along fibre optic cables and focusing optics (Fig. 18.16). Due to high power density, out-of-focus operation is frequently necessary in order to prevent overheating of alloying elements in the coating material. While CO2 lasers allow operation in continuouswave mode, solid state and diode lasers are operated in pulsed mode. Consumables can be applied as wires, cored wires, or as powders via powder feeders (Fig. 18.17). A dual-step process variant uses pre-placed powders. External or in-

Fig. 18.16 Principle of laser alloying (CO2 laser) with pre-placed coating material (following [3]), example of titanium materials.

281

282

18 Surface Protection by Means of Build-up Welding Fig. 18.17 Setup of laser welding head with powder feeder and shielding gas chamber for magnesium coating.

Fig. 18.18 Encapsulated laser for manual build-up welding [4].

ternal protective gas feeding is necessary in order to protect laser optics against contamination, and to shield the hot molten pool against penetration of oxygen. Surfaces are blackened in order to provide good CO2 laser beam absorption. Solid state and diode lasers only require this measure when processing highly reflective materials, e.g. aluminium or copper.

18.5 Coating Materials for Build-up Welding

Coating thickness can reach up to 2 mm. Current developments in the field of laser cladding are hand-operated laser processing heads (Fig. 18.18) [4] that allow, for instance, refurbishing of slightest abrasion marks on forming or cutting tools without depositing too much excess material at the surface. Systems can be self-controlled to compensate for unsteady motion during manual operation [5]. The multitude of research activities in laser technology suggests ongoing development of laser-assisted build-up welding. Today, high system costs of laser equipment limit the technology to niche applications such as small volumes with minimal post-treatment, and/or special materials that impede build-up welding with conventional techniques.

18.5 Coating Materials for Build-up Welding

Four groups of coating materials can be distinguished according to the desired application: · materials with high corrosion resistance · materials with high wear resistance · materials with high wear and corrosion resistance · materials for special applications. Metal-matrix material is always part of the coating. It guarantees a metallurgical fusion bond and embeds other phases, usually hard phases, that are not necessarily molten during build-up welding, also due to the danger of thermal decomposition. Alloying elements that improve wetting and flow of coating material are added in order to yield desired weld bead formation. Slag-forming elements bind to residual contaminants and carry them to the surface of the weld bead. Hence, porosity and contamination within the coating are kept low. In processes without protective gas, slag completely covers the molten pool and prevents oxidation due to penetrating oxygen. While slag-forming elements are integrated into consumables in manual arc welding and, partially, in MIG cored wire welding, the slag in submerged arc and RES welding is pre-placed as powder and molten during the build-up welding process. 18.5.1 Materials for Corrosion Protection

Corrosion resistance plays a key role particularly in apparatus and chemical engineering, environmental technology, as well as offshore industry (salt-water applications). Plated constructional steel is frequently used for economical and manufacturing reasons. Platings are produced by rolling or build-up welding (Fig. 18.19).

283

284

18 Surface Protection by Means of Build-up Welding Fig. 18.19 Resistance electroslag welding [7].

The latter is common for processing of complicated geometries where plating is inappropriate. Coating materials have to withstand contacting agents that would cause different types of corrosion. A large number of special materials has thus been developed, and can be deposited by build-up welding. Over-alloying of consumables is only necessary to compensate for fusion with the substrate material. Post-plating of weld beads on welded parts is an important field of application, e.g. large pipes or containers. The same type of materials are used as in roll-bonding, or higher alloyed grades that yield sufficient corrosion protection in spite of fusion with the substrate. Ni alloys are often used as post-platings of Fe alloys for safety reasons. Coating materials can be distinguished according to base materials (nickel and iron materials). Typical material groups for corrosion protection are: · FeNiCr alloys · NiCr alloys.

18.5.1.1 Corrosion-resistant Iron-based Materials Corrosion-resistant iron-based materials are derived from the widely known austenitic steel grades with ³ 18% chromium and ³ 8% nickel (1.4301 or 1.4306). In build-up welding they tend to form carbide precipitations that present potential weak points for pitting corrosion. The problem is solved to a large extent by

18.5 Coating Materials for Build-up Welding

reducing the carbon content to £ 0.03% [7]. As coatings produced by build-up welding can always face a rise in carbon concentration due to fusion with the substrate material, Ti- or Mo-stabilised material grades are used in build-up welding (1.4541 or 1.4571). Heavier corrosion attack calls for higher alloy concentrations and added niobium. According to [7], individual alloy constituents have the following, specific effects: · Mo: increases pitting and crevice corrosion protection as well as erosion protection · Ni: increases resistance against stress corrosion cracking delays precipitation of sigma phase, increases strength · N 2: · Ti, Nb: increases resistance against intergranular corrosion · C: reduction of C increases resistance against intergranular corrosion. A second important group are austenite-ferrite steels (duplex steel, e.g. 1.4462). These steels have higher strength in addition to excellent corrosion resistance.

18.5.1.2 Nickel Alloys Nickel alloys are used in applications where corrosion resistance of steel is insufficient. Pure nickel materials usually do not meet the high requirements. Therefore, Cr- and/or Mo-alloyed nickel-based materials with higher corrosion resistance are applied. Chromium additionally increases high-temperature stability [8]. Nitrogen and sulfur are an issue and, thus, have to be considered in build-up welding as they can cause porosity and intergranular corrosion [8]. For post-plated large pipes, Table 18.4 shows typical materials used in rollbonding and weld bead post-plating. 18.5.2 Materials for Wear Protection

Coatings for wear protection present the larger field of applications for build-up welding. Here, alloys containing hard phases and pseudo-alloys with added carbides are used. Appropriate alloy composition additionally provides corrosion and/or hightemperature resistance for these materials. Four groups are distinguished: · nickel hard alloys · iron hard alloys · cobalt hard alloys (stellites) · aluminium pseudo-alloys.

285

17

22

23

22

Alloy 316 L 1.4435 X2CrNiMo 18 14 3

Alloy 625 2.4856 NiCr22Mo9Nb

Alloy 59 2.4605 NiCr23Mo16A

Alloy 825 2.4858 NiCr21Mo

Cr

40

59

61

12

Ni

3.5

16

9

2.4

Mo

0.2 Ti 2 Cu 31 Fe

0.3 Al 1 Fe

3.8 Nb 3 Fe

0.03 N rest Fe

others

Alloy 625 2.4856 NiCr22Mo9Nb

Alloy 59 2.4607 NiCr23Mo16

Alloy 59 2.4607 NiCr23Mo16

Alloy 21 13 3L strip electrode

Alloy type Material no. DIN/EN code

Alloy type Material no. DIN/EN code

Chemical composition (guidelines) in %

Post-plating

Plating

Table 18.4 Selected materials for roll plating and post plating, according to [6].

0.01

0.015

0.015

0.01

C

0.5

0.1

0.1

0.2

Si

0.85

0.15

0.15

1.8

Mn

21.5

23.5

23.5

20.5

Cr

Chemical composition (guidelines) in %

rest

rest

rest

13.5

Ni

8.9

16

16

2.9

Mo

3.8 Nb

0.45 Fe

0.45 Fe

0.08 N

others

Plasma

RES

RES

RES

Welding technique

286

18 Surface Protection by Means of Build-up Welding

18.5 Coating Materials for Build-up Welding

18.5.2.1 Nickel Hard Alloys Typical nickel materials for wear protection are self-fluxing, i.e. boron added by alloying reduces the melting point to 1000–1100 8C. Precise matching of parameters allows build-up brazing on steel with hardly any fusion. Apart from boron, these materials contain chromium, responsible for hard phase formation, as well as silicon in order to improve weldability. The carbon content is matched to the amount of carbide forming constituents. The content of chromium carbide as well as other hard phases of borides and silicides determine macroscopic hardness of the different materials. The majority of chromium carbides dissolves in build-up welding and precipitates during cooling as needles or flakes (Fig. 18.20). Increasing carbide content makes the material susceptible to stress cracking. Preheating and gradual cooling prevents cracking to a great extent. Table 18.5 lists examples of materials with different hardness. These materials are highly corrosion resistant, however, heterogeneous microstructure does limit stability. A further increase in wear resistance is provided by adding tungsten carbide. In welding technology, a special molten tungsten carbide with feather-like microstructure, composed of tungsten monocarbide, WC, and ditungsten carbide, W2C (eutectic), is used (Fig. 18.21). Strength and ductility, in particular, are higher than in monocarbide. The high density of tungsten carbide (12 to 17 g/ cm3, depending on configuration) leads to intense gravity segregation in buildup welding. The tungsten carbide content, therefore, should exceed 50% in order to yield sufficiently homogeneous properties across the thickness of the coatings. A new variant contains spherical carbides (Fig. 18.22) produced by a special post-treatment of the carbides in the production process. The material is

Fig. 18.20 Coarse dendritic chromium carbide precipitations (Fe-Cr-C alloy, superheated).

287

288

18 Surface Protection by Means of Build-up Welding Table 18.5 Standard analyses of self-fluxing NiCrBSi alloys as well as a chromium-free variant. Material

NiCrBSi 1 NiCrBSi 2 NiCrBSi 3 NiBSi

Standard hardness in HRC

Alloy content (standard analysis) in % C

Cr

B

Si

Al

Fe

Ni

30 42 58 n. s.

0.20 0.4 0.7 0.03

4.0 10.0 16.0 –

1.0 1.8 3.3 3.0

2.5 2.7 4.2 3.0

1.0 – – –

< 2.0 < 2.0 < 3.0 < 3.0

rest rest rest rest

Fig. 18.21 Polished cross section of fine feather-like structure of molten tungsten carbide.

harder and suggests additional benefits in terms of wear behaviour, similar to macrocrystalline tungsten carbide (Fig. 18.23). Tungsten carbide is sensitive to high temperature and tends to show decomposition effects (formation of brittle phases) due to overheating during build-up welding. Vanadium carbide, here, is an alternative hard material [8]. It features similar properties, is less sensitive to heat, and precipitates as dendritic monocarbide after melting during the build-up welding process (Fig. 18.24). Properties of vanadium carbide are discussed further in the chapter on iron-based materials.

18.5.2.2 Iron Hard Alloys Every hardenable steel that shows high wear resistance due to martensitic microstructure and carbide content, can generally be deposited by build-up welding. Microstructure forms according to fusion metallurgical laws, i.e. a cast-type

18.5 Coating Materials for Build-up Welding

Fig. 18.22 Polished cross section of wear protection coating with 60% molten tungsten carbide, spherical carbides.

Fig. 18.23 Macrocrystalline tungsten carbide.

structure with eutectic and eutectoid networks is created. For high carbide contents, in particular, this microstructure is more brittle than rolled or forged material. Application is thus advisable only for production of composite parts with a particular combination of property requirements. This can apply to large parts where machining of bulk material is too expensive. More important are wear-protection materials that either feature higher wear resistance than steel due to high carbide contents, or show a particular combination of properties that is not producible by any other means.

289

290

18 Surface Protection by Means of Build-up Welding

Fig. 18.24 VC carbides in NiBSi (60% VC, polished cross section).

Fig. 18.25 FeV18Cr5Mo2C alloy with approx. 30 per cent by volume VC.

Commonly used materials for wear protection are Fe-Cr alloys with Cr contents up to approx. 30%. In certain cases, niobium is added to the alloy for increased hardness. Macrohardness values are in the range of 30 to 56 HRC, depending on carbon content. Chromium passivation layers protect the material from corrosion, even at high temperatures. Coatings with higher chromium contents show cracks. Typical applications are wear-protection sheets, manufactured by fully automated build-up welding of large sheet metal that is subsequently flame cut to yield a preformed, coated part.

18.5 Coating Materials for Build-up Welding

Fig. 18.26 Wear intensity (mass loss in abrasive paper test) against VC content for alloyed fillers and pseudo-alloys with added VC [10].

Further increased wear resistance is obtained by adding vanadium to the alloy. A new group of wear-resistant materials has been developed with a variety of wear and corrosion properties. The materials have the advantage that, by melting and atomising, consumables can be produced with vanadium carbide (VC) contents of up to approx. 30% by volume. In contrast to pseudo-alloys produced by adding and mixing carbides, a fine-grained carbide distribution is obtained (Fig. 18.25). Precisely adjusted carbon content in the coating creates high operating hardness and leads to good edge-holding properties. Thus, typical applications are cutting edges of machine blades (see Fig. 18.2). Wear resistance is considerably higher than for the highest alloy steel and can be increased further by adding more carbide. The carbide content can increase up to 70% by volume, and coating formation is still sufficient. Wear resistance is determined by the vanadium content (Fig. 18.26). The danger of cracking is considerably lower than in high chromium coating materials. The matrix alloy consists of cold work steel with 5% chromium and approx. 2% Mo. During cooling from welding temperature, martensitic microstructure with residual austenite and ledeburitic precipitation is formed, but brittle networks are not observed. Adding further alloying elements creates the following material variations: · Cr: corrosion-resistant materials (chromium martensitic structure) · Cr and Ni: corrosion-resistant materials (austenitic structure) · W and Co: increased high-temperature stability (structure similar to highspeed steel). The wear resistance of these materials is considerably higher than for hardened tool and high speed steel (Fig. 18.27).

291

292

18 Surface Protection by Means of Build-up Welding

Fig. 18.27 Wear resistance of selected VC alloys with different VC contents and different matrix microstructure compared to steel and hard metal.

18.5.2.3 Cobalt Hard Alloys Cobalt-based materials, or stellites, contain chromium, tungsten, and carbon as the most important alloying elements, and sometimes also nickel, molybdenum and niobium. The materials are numbered following their US American origin (Table 18.6). Iron content is limited to a maximum of 2 to 3% in order to assure high corrosion resistance. In certain cases, iron contents of up to 20% are included for special properties. Boron and silicon as alloying elements yield self-fluxing variants with melting points well below the melting point of steel. Build-up brazing with minimised fusion is thus applicable.

18.5 Coating Materials for Build-up Welding Table 18.6 Standard analyses of selected stellites (according to Deloro Stellite GmbH). Material

Stellite Stellite Stellite Stellite Stellite Stellite Stellite Stellite

1 SF 1 6 SF 6 12 SF 12 20 21

Standard composition in % Cr

W

C

Ni

B

Si

others

Co

33 19 26 19 29 19 33 27

13 13 5 8 9 9 18 –

2.5 1.3 1 0.7 1.8 0.9 2.5 0.25

– 13 – 13 – 13 – 2.8

– 2.5 – 2.5 – 1.8 – –

– 3 – 3 – 2.5 – 2

– – – – – – – 5.5 Mo

rest rest rest rest rest rest rest rest

Hardness in HRC

Melting temperature in 8C

51–58 54–58 39–43 43–46 47–51 48–50 55–59 28

1255–1290 1069–1180 1285–1395 1085–1150 1280–1315 1061–1104 1260–1265 –

Fig. 18.28 Different hard phases in special stellite for machine knifes (comparable to Stellite SF 6).

The cobalt matrix is responsible for superior corrosion resistance and hightemperature stability. Intermetallic hard phases (in materials with low carbon content) or carbidic hard phases provide wear resistance against erosion and abrasion (Fig. 18.28). A load applied via an external force additionally promotes hardening effects in the adjacent surface area. Wear behaviour is thus comparatively high, even for materials with relatively low hardness. A broad selection of alloys is available, meeting the demands of a wide range of applications. The main field of application is heat-resistant parts (gas turbine blades, extrusion dies, valves, hot-shear blades, etc.). Stellites are also common in cold wear applications with the highest demands in terms of corrosion resistance (blades, shears). Crack-free deposition of stellites with high hardness by means of build-up welding requires high preheating temperature and gentle cooling. Overheating can lead

293

294

18 Surface Protection by Means of Build-up Welding

to the formation of large gas pores. Thus, processing with hand torches is difficult and requires welders with high levels of manual dexterity and experience. Frequently, a compromise between favourable wear behaviour and mechanical machinability for post-treatment is necessary in stellite applications. Complex convex geometries of forging dies, for instance, can be produced cost-effectively only by milling. In spite of newly developed, special coated hard-metal grades, economical machining is limited to intermediate operating hardness values up to 40 HRC. Stellite 1 and Stellite 6 are thus either inappropriate or restricted to special applications that allow grinding processes.

18.5.2.4 Aluminium Pseudo-alloys In high-wear applications, use of light metals requires a sophisticated combination with steel. Current investigations, therefore, focus on build-up welding on Al and Mg substrate materials [9, 10]. Local heating of aluminium requires high power density in the operating beam or jet in order to compensate for high thermal conductivity. It must be considered that high-melting oxide layers, present on all aluminium surfaces, disturb wetting behaviour considerably. Selection of appropriate welding processes is thus limited to techniques with not only high power density but also a cathodic cleaning effect (MIG, PTA with alternating polarity). An additional characteristic must be considered. Common build-up welding materials, including added hard phases, feature considerably higher density than light metal substrate materials. By coating, the total mass would increase dramatically and nearly negate the weight-saving effect. Furthermore, adhesion is reduced due to formation of brittle phases (e.g. Fe-Al) in many cases. Application of aluminium-matrix coating materials (AlSi or AlMgMn) with added carbides featuring matched density (B4C, SiC, in certain cases TiC) leads

Fig. 18.29 Wear intensity (volume wear) of AlSi-SiC coatings compared to steel.

18.6 Summary and Conclusions

Fig. 18.30 Worn surface of AlSi-B4C coating with protruding, firmly bound carbides.

to a substantial increase in wear resistance (Fig. 18.29). Abrasive wear tests reveal that performance equal to that of hardened tool steels is achieved. Carbides are well-embedded (Fig. 18.30). For SiC, featuring heteropolar bonds, added Si provides good carbide embedding. On the whole, there still is a demand for developments in the field of reliable production technology for light-metal coatings. Current research work is investigating applications of hardenable materials for improved matrix stability.

18.6 Summary and Conclusions

Surface protection by means of build-up welding is particularly necessary in applications operating under heavy wear conditions or edge loads. This is due to the fact that coatings produced by build-up welding, compared to other coatings, feature high adhesion and are self-supporting. Many applications, therefore, do not require high-strength substrate materials. From the large selection of welding processes, RES as well as submerged arc build-up welding and, recently, due to high coating quality, plasma powder build-up welding are used for corrosion protection. Wear-protection coatings are produced by virtually any technique, from simple manual arc welding to plasma MIG. Apart from economical reasons, coating quality and availability of consumables as wires, strips, or powders are key factors in technology selection. Build-up welding is a cross sectional technology, appearing in nearly every branch of industry, either directly or as coated parts. Highest-quality coatings

295

296

18 Surface Protection by Means of Build-up Welding

are needed, for instance, in industrial facilities operating under chemical or salt-water conditions. The food-processing industry also has special requirements in terms of coating quality and coating materials. Due to the multitude of applications for coated parts, the range of coating materials is wide. This chapter thus focuses on the most important materials. German development activities in the realm of welding processes, welding technology, and consumables are merged within the Deutscher Verband für Schweißen und verwandte Verfahren e. V. (DVS – German Welding Society) where developers and users cooperate closely. Build-up welding is not a static technology. Instead, it is characterised by a high potential for innovation in the field of developing processes and new consumables.

References 1 Herrmann, J., Ott, T.: Untersuchungen

2

3

4

5

6

7

zum MSG-Auftragschweißen mit Flachdrahtelektroden. 5th Symposium “Verschleißschutz von Bauteilen durch Auftragschweißen”, Halle/Saale, June 14th– 15th, 2004 Müller, S., Balter, K.: Fortschritte beim Widerstandsrollennaht-Auftragschweißen. 4th Symposium “Verschleißschutz von Bauteilen durch Auftragschweißen”, Halle/Saale, May 28th–29th, 2002 Matthes, K.-J., Kolbe, G., Wielage, B., Wank, A., Podlesak, H.: Laserstrahldispergieren zur Herstellung boridverstärkter verschleißfester Oberflächen an Titanlegierungen, Schweißen und Schneiden. Issue 11 (2003), pp. 610–615 Vollrath, K.: Laserstrahl hoffähig für die Instandsetzung von Werkzeugen. Report in “Der Praktiker” Issue 9 (2003), pp. 276–281 Kimme, T., Tuband, E.: Laserauftragschweißen – Dienstleisterbericht zu Technik, Ergebnissen und Anwendungsbeispielen. 2nd Symposium “Verschleißschutz von Bauteilen durch Auftragschweißen” Halle/Saale, May 13th– 14th, 1998 Wende, U., Bouaifi, B., Bock, A.: Metallurgische und korrosionschemische Eigenschaften von Nickelbasis-Anschlussbzw. -Nachplattierungen. DVS Report No. 216, pp. 355–364 Reichmann, B., Bouaifi, B., Deppe, P.: Einsatz der Sinterbandelektrode CORO-

8

9

10

11

12

13

SINT 625 für das Einlagen-RES-Auftragschweißen in Behältern der Ölindustrie. 5th Symposium “Verschleißschutz von Bauteilen durch Auftragschweißen”, Halle/Saale, June 14th– 15th, 2004 Strassburg: Eigenschaften von hochlegierten Stählen und Nickelbasislegierungen. Publication Nickel Development Institute (Germany) 1997 Kabatnik, L.: Plasma-Pulver-Schweißen verschleißbeständiger Schichten auf Aluminiumwerkstoffe. German Doctoral Thesis, Aachen (2002), Shaker, pp. 1–114 Gebert, A., Duitsch, U., Schubert, T., Müller, U.: Zum Verschleißschutz komplizierter ebener Konturen. Praktiker 51 (1999) 1, pp. 24–28 Heinze, H., Gebert, A., Bouaifi, B., AitMekideche, A.: Korrosionsbeständige Auftragschweißschichten auf Eisenbasis mit hoher Verschleißbeständigkeit. Schweißen und Schneiden 51 (1999) Issue 9, pp. 550–555 Bouaifi, B., Ait-Mekideche, A., Gebert, A., Wocilka, D.: Nutzung von stickstoffhaltigen Hochtemperaturplasmen zum reaktiven Beschichten mittels Plasma-Auftragschweißen. Schweißen und Schneiden 53 (2001) Issue 8, pp. 478–482 Gebert, A., Heinze, H., Bouaifi, B., Teupke, E., Reichmann, B.: Festigkeit hochverschleißbeständiger Schichten optimiert. Der Praktiker 55 (2003) 2, pp. 52–54

297

19 Non-destructive Testing and Assessment of Coatings W. Reimche, R. Duhm, Institute for Materials Science, Non-Destructive Testing, University of Hannover, Germany 19.1 Introduction

Quality management of coating technologies and processes is gaining importance due to narrowing markets and growing competition. The development requires realising and standardising destructive and, increasingly, non-destructive testing methods for characterisation of thermal spray coating properties. This chapter provides an introduction and overview of common coatings, typical coating thickness ranges, and properties in order to present and discuss a number of applicable testing methods. In this context, values for individual coating/substrate combinations as well as process characteristics and obtainable measurement accuracy are investigated.

19.2 Coatings 19.2.1 Processes

Classes of surface coatings are distinguished by the type of fabricating process. Coating material is deposited onto a substrate material from a gas, vapour, liquid or solid phase, or from solutions. Examples of coating processes that deposit from vapour or gas phase are CVD and PVD technologies. Coatings deposited from solutions are produced by electroless and electroplating techniques etc. Dip coating, thermal spraying, as well as build-up brazing and welding use pasty or liquid coating material. Solids are utilised as coating materials in cold spraying or plating. Apart from ceramics and plastics, primary base materials (substrates) are metals such as cast iron, steel, titanium, nickel materials, or light metals. DevelopModern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

298

19 Non-destructive Testing and Assessment of Coatings

ment of coating composites aims at separating functions of substrate and coating. Considering coating thickness, thermal spray coatings are generally not self-supporting but serve as functional surfaces that protect parts against corrosion, wear, or thermal loads. Part substrates provide the supporting structure and can be fabricated from less-expensive appropriate standard materials [1, 2]. 19.2.2 Coating Properties

In the realm of quality assessment, the evaluation of coating conditions and properties by non-destructive testing is of growing interest. A general distinction is made between detection and characterisation of coating properties on the one hand, and testing of coatings for defects within the coating and in the interface between coating and substrate on the other. Thin films and coatings in the thickness range of 0.001 to 0.1 mm are usually deposited from gas or vapour phase, e.g. by PVD and CVD (physical and chemical vapour deposition, respectively). On average, the thickness of CVD coatings (0.01 to 0.1 mm) is slightly higher than for coatings produced by PVD. Electroplated coatings are often nickel-, titanium-, or chromium-based, e.g. neutron absorbers with Ni-B4C coatings on nuclear reactor steel, with coating thickness between 0.01 and 5 mm. Thermal spray coatings show a thickness range from 0.1 to 1 mm. Thicker coatings (2 to 20 mm, on average) are produced by means of roll-bonding as well as build-up brazing and welding. Table 19.1 shows a general classification of properties based on underlying active principles. Here, mechanical, electromagnetic, and chemical coating properties are distinguished.

Table 19.1 Classification of relevant coating properties. Mechanical properties

Physical properties

Chemical properties

Adhesion Hardness Coating thickness and weight of coating Internal stress Microstructure Tribological properties Resistance against abrasive wear Ductility Porosity

thermal conductivity electrical conductivity magnetic properties acoustic properties

chemical composition corrosion behaviour behaviour in fire erosion endurance

19.3 Thickness-measurement Techniques

19.2.3 Test Planning

Oxide ceramic coatings are electrical insulators. At room temperature, the specific electrical resistance of oxide ceramics is generally in the range of 107 to 1015 Xm, and falls rapidly at higher temperatures. In contrast, the specific electrical resistance of metals is several orders of magnitude lower (10–5 to 10–7 Xm) and far less temperature dependent. Therefore, testing methods that determine coating properties by measuring differences in electrical conductivity of the coating and the substrate, or of coating and substrate defects are particularly applicable. Non-destructive testing of characteristic coating properties has a wide selection of testing methods at hand. Depending on the properties subject to investigation, certain testing methods qualify due to process-specific advantages. VDI standard guideline 3824 [3] (pp. 2–4) distinguishes the following testing occasions: · Receiving inspections prevent semi-finished parts that are not prepared for coating of being accepted for coating (according to VDI standard guideline 3824, p. 2). · In-process testing assures early detection of quality defects and guarantees faultless parts in further processing (according to VDI standard guideline 3824, p. 3). · Final testing of finished products guarantees that products leaving the coating process to serve their purpose are free of defects (according to VDI standard guideline 3824, p. 4). Tests usually involve an initial visual examination of semi-finished parts and products, either with or without optical equipment (e.g. optical microscopes). Generally, tests are conducted as sampling investigations, however, certain applications require 100-per cent inspections. Inspections focus on identifying defects and damages in the functional zone between the tool and the semi-finished part or product. Particular attention is directed towards: · specimen condition appropriate for coating · damaged coatings, insufficient cleaning · uncoated areas, delamination, uneven coating colour.

19.3 Thickness-measurement Techniques

A wide selection of techniques is available for measuring coating thickness in single or multi-layer coatings. DIN EN ISO 2064, ISO 1463, and ISO 3882 standards include general definitions [4–6]. This section summarises most common methods and discusses specific advantages and disadvantages (cf. Table 19.2).

299

Application

Beta-backscatter analysis X-ray diffractometry

coating-thickness measurement coating-thickness measurement

coating-thickness Magnetic induction measurement analysis Eddy-current analysis coating-thickness measurement X-ray fluorescent coating-thickness analysis measurement

coating-thickness measurement coating-thickness measurement Pull-off force analysis coating-thickness measurement Acoustic emission coating-thickness analysis measurement Ultrasonic impulse coating-thickness echo analysis measurement Ultrasonic impulse delamination/lamiecho analysis nar separation

Geometric part measurement Gravimetric analysis

Technique

N, E, F F N, E, F

N, E, F N, E N, E, F

F

DIN 50988-2, ISO 4524-1 DIN 50981, DIN EN ISO 2178

Standards

DIN 50981, DIN EN ISO 2178 N, E, F N, E, F DIN 50984, DIN EN ISO 2360 N, E, F N, E, F DIN 50987, DIN EN ISO 3497, DIN 50977 DIN 50983, N, E, F, minimum 20% difference in atomic numbers DIN EN ISO 3543 N, E, F N, E, F

N, E

N, E, F, minimum 20% difference in speed of sound N, E, F N, E, F

N, E, F

Substrate

N, E, F

Coating

Table 19.2 Process comparison for coating assessment with non-destructive testing.

3.7

1% to 10%

3% on average

0.2 to 800 lm internal stress in coatings above approx. 2 lm thickness

3.6

4

3.8

3.5

delamination diameter above approx. 2 mm 10 % of measured value lm range

coating thickness from 100 lm to several millimetres 2 lm to several centimetres 1 lm to approx. 5 mm 0.1 to 200 lm

5.2

3.4

3.3

3.2

3.1

Section

3.4

< 10% between 20 and 70 lm approx. 100 lm

1 lm

Accuracy

> 0.5 mm

not applicable for thin coatings

above approx. 1 lm

Measuring range

300

19 Non-destructive Testing and Assessment of Coatings

N – non-metal, E – electrically conductive, F – ferromagnetic

N, E, F, ratio of products q(r ´ lrel) > 1.5

DIN EN ISO 21968

DIN EN 571-1

Standards

Eddy-current analysis adhesion

delamination/ laminar separation

Impulse and lock-in thermography

N, E, F N, E, F, for surfaces with closed pores or smooth surfaces N, E, F N, E, F

Substrate

DIN EN ISO 21968

surface defects

Dye-penetration test

Coating

Eddy-current analysis near-surface defects N, E, F, ratio of products q(r ´ lrel) > 1.5

Application

Technique

Table 19.2 (continued)

defect width above approx. 0.1 mm

Accuracy

coating thickness up delamination diameto several millimetres ter above approx. 50 lm maximum defect 2% of coating thickdepth approx. 5 mm ness, minimum 5 lm separations orthogo- delamination diamenal to eddy-current ter in lm range propagation

defects reaching to the surface

Measuring range

5.5

5.4.2

5.3

5.1

Section

19.3 Thickness-measurement Techniques 301

302

19 Non-destructive Testing and Assessment of Coatings

19.3.1 Geometric Measurement of Parts

Geometric measurement of parts is a simple mechanical measuring technique to determine coating thickness. A geometrically defined substrate is coated consecutively. After each deposition, the complete part is measured, and local or averaged total coating thickness is determined by subtraction. Disadvantages are continued interruptions of the coating process after each partial deposition, leading to non-productive time. Furthermore, thermal part expansion under processing conditions must be considered in coating-thickness measurements. An additional variant is single- or double-sided, non-contacting, optical-geometrical measurement of part geometry by means of laser-based measuring processes. In practice, however, online application is highly restricted due to varying surface roughness, vibrations of part and sensor mounting, eccentricity and unevenness of parts, unbalanced parts caused by insufficient fixing or centring, as well as thermal expansion of parts. 19.3.2 Differential Weight Analysis Before and After Coating

In gravimetric analysis (DIN 50988-2, ISO 4524 part 1), coating material is removed from the substrate in a defined area of the specimen. The resulting difference in mass is determined by weighing before and after coating removal [7, 8]. The mean coating thickness is then calculated from the weight difference, the area of removal, and the coating density. Other methods determine the part weight prior to and after the coating process. The difference in weight yields the coating weight. The mean coating thickness is then calculated by taking into account the average density of the coating. The obtainable coating thickness and accuracy of analysis are controlled by the mass of removed coating, the accuracy of determined area, and the accuracy of weighing. Sufficient accuracy of gravimetric analysis relies on measuring relatively large masses. The technique is thus not suitable for determining the thickness of thin coatings on heavy parts. 19.3.3 Coating-thickness Measurements Based on Magnetic Pull-off

Magnetic pull-off force measurement according to DIN 50981/DIN EN ISO 2178 is used in non-destructive testing, e.g. for ceramic coatings on ferromagnetic substrates. The technique uses a spring balance with attached permanent magnet, or a balance arm with a permanent magnet on one end of the arm [9]. The magnet rests on the surface of the coated specimen. Coating thickness is characterised by comparing the pull-off strength of coated and uncoated samples.

19.3 Thickness-measurement Techniques

Fig. 19.1 Coating-thickness measurements by means of magnetic pull-off.

The procedure is calibrated for particular substrate/coating material combinations (Fig. 19.1). Magnetic pull-off force measurement is a highly sensitive technique. Measurement errors are below 10% for coating thicknesses between 20 lm and 70 lm. Below 15 lm coating thickness, the measurement uncertainty is considered constant at 1.5 lm. Therefore, the technique is suitable for analysis of thin coatings on ferromagnetic substrates. However, diffusion zones defy thickness measurement by means of magnetic pull-off force measurement [10]. 19.3.4 Coating-thickness Measurements Based on Acoustic Principles

Acoustic techniques such as ultrasonic testing and acoustic emission analysis are generally suitable for coating-thickness measurements. Offline, acoustic emission analysis featuring active part excitation with a steel ball has been used for thermal spray coated metal parts [11]. Here, vibration signals recorded on the specimen after impact of a steel test ball are converted into a frequency response. The desired range of the spectrum (approx. 1 to 10 kHz) is analysed for primary amplitudes. The frequency and the amplitude of the impulse with the lowest acquired frequency in the analysed region of the spectrum are interpreted in order to calculate local coating thickness. In offline use, measurement accuracy of the technique, to date, is approx. 100 lm [11]. The impulse echo method is an appropriate ultrasonic technique for analyses of thick coatings (Fig. 19.2). It uses the physical principle of partial ultrasonic signal reflection at material interfaces. Reflections occur at the coating surface, and at boundaries between the coating and the substrate. The known speed of sound as well as the running time between echoes at interfaces and backwall

303

304

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.2 Measurement principle of ultrasonic impulse echo method.

Fig. 19.3 Coating-thickness measurement with ultrasonic impulse echo technique.

echoes allow calculation of coating thickness (Figs. 19.2 and 19.3). However, coating-thickness measurements of thermal spray coatings are limited. Thicknesses in the area of 100 lm require very high frequency ultrasonic signals. Such signals are subject to substantial damping, and are scattered considerably in porous and partially inhomogeneous materials such as thermal spray coat-

19.3 Thickness-measurement Techniques

ings. Furthermore, probe-to-specimen contact of ultrasonic sensors requires an ultrasonic coupling medium (couplant), e.g. oil or water. Couplants present possible disadvantages in post-treatment, e.g. sealing, or for subsequent deposition of additional top coats. Under optimal conditions (low surface roughness and porosity), impulse echo methods allow coating-thickness measurements above 0.5–1 mm. 19.3.5 Coating-thickness Measurements with Magnetic-induction Techniques

Magnetic-induction measurements of coating thickness according to DIN 50981/DIN EN ISO 2178 use a magnetic circuit formed by a yoke and the specimen [9]. A magnetic flux is created by permanent or electromagnets. The flow of magnetic flux is undisturbed in the magnetic circle that includes a ferromagnetic yoke and ferromagnetic substrate material. Non-ferromagnetic coatings interrupt the magnetic flux in the coating zone. Depending on coating thickness, magnetic flux decreases strongly. Following Ohm’s law for magnetism, magnetic resistance of the arrangement is approximated by adding the individual magnetic resistances of the yoke, the substrate material, and the coating. Inductance changes in an inductor are particularly suitable for detecting changes in the magnetic flux through a yoke. The technical design of measuring heads aims at creating a magnetic circuit based on the transformer principle. In the yoke, a primary coil induces a low-frequency magnetic flux, which passes through a secondary coil. Bridge circuits or detuned resonant circuits al-

Fig. 19.4 Coating-thickness measurement with magnetic-induction technique.

305

306

19 Non-destructive Testing and Assessment of Coatings

low high-accuracy measurement of inductance changes in the secondary coil. The testing frequency is in the range of 10 to 100 Hz in order to minimise frequency-dependent eddy-current loss. Results with an average uncertainty of 10% are obtained for coating thickness between 2 lm and several centimetres. A special case, according to DIN EN ISO 2178, are coating-thickness measurements of nickel coatings on steel substrates, as long as the magnetic permeability of coating and substrate material differ sufficiently and variations in the test batch are not too high (Fig. 19.4).

19.3.6 Coating-thickness Measurements with Eddy-current Techniques

Apart from magnetic-induction measurements, electromagnetic testing procedures additionally include eddy-current techniques according to DIN 50984/DIN EN ISO 2360. Eddy-current investigations detect changes in physical material properties by means of alternating magnetic fields [12]. The fields are generated by alternating current passing through coils. Specimens or testing materials are placed in the working area of a coil. An advantage of the technique is that it allows testing of non-ferromagnetic samples, as long as the substrate or coating are electrically conductive. Depending on magnetic permeability lrel, electrical conductivity, workpiece geometry, and coil arrangement on the workpiece, eddy currents and material magnetisation occur, causing a secondary magnetic field inverse to the primary field. Detuning of the coil due to changes in the complex induction by the workpiece can be detected directly at the transmitter coil. Commonly, however, a receiving coil, separate from the excitation coil, is used, allowing higher field intensities as well as more sensitive measuring signals (Fig. 19.5). Three fundamental cases are distinguished in applications with electromagnetic testing procedures for coating-thickness analyses of workpieces: · electrically non-conductive coating on electrically conductive substrate · electrically conductive coating on electrically non-conductive substrate · electrically conductive coating on electrically conductive substrate. In the analysis of electrically non-conductive coatings on electrically conductive substrates by means of eddy-current investigations, the distance between coil and workpiece surface predominantly determines the amplitude of the resulting magnetic flux through a coil (distance effect or lift-off effect). On approach to the workpiece, induction changes in the coil alter the operating point in the impedance plane, offering an appropriate means of characterising the change (Fig. 19.5). Increasing thickness of electrically conductive coatings on electrically non-conductive substrates causes a rise in the amplitude of the eddy-current signal due to the expanding interaction volume. Combined with calibrated measuring equipment, this effect allows quantitative coating-thickness measurement.

19.3 Thickness-measurement Techniques

Fig. 19.5 Coating-thickness measurement with eddy-current technique.

Magnetic permeability of coating and/or substrate, i.e. ferromagnetic behaviour, if present, provides an additional parameter for characterising coating properties. Considering an electrically conductive and magnetically permeable coating on an electrically conductive, magnetically permeable substrate, eddycurrent fields are subject to significant phase difference in material induction due to differences in material conductivity and permeability (cf. distance locus of steel materials in Fig. 19.5). Provided that conductivity and permeability of the two materials differ sufficiently, coating thickness is assigned to different phase positions of individual locus curves. Multi-layer coatings defy a simplified description of field conditions below the coil. Compared to electromagnetically homogeneous materials, varying electrical conductivity and permeability in coating layers and substrate material lead to considerable deviations of eddy-current distribution. Harmonic analysis of eddy-current signals is a modern variant of this technique, dedicated to acquisition and analysis of electrical conductivity and magnetic permeability. Multi-parameter applications allow analysis and assessment of multi-layer coatings with respect to penetration depth and, therefore, coating thickness (Fig. 19.6). In practice, the technique is applicable for metal and non-metal coatings on metal substrates (Fig. 19.7). For the analysis of corrosion protection coatings on

307

308

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.6 Coating assessment with harmonic analysis of eddy-current signals.

Fig. 19.7 Coating-thickness measurement of conductive coatings on conductive substrate material with eddy-current technique.

19.3 Thickness-measurement Techniques

Fig. 19.8 Analysis and assessment of coatings on turbine blades.

aircraft turbine blades, the method provides a means of assessing the quality of new coatings as well as controlling coating condition after use. For this, local variations in electrical conductivity are measured in order to yield thickness values of top coats and diffusion layers. The measurement accuracy of the technique is in the lm range (Fig. 19.8). 19.3.7 Coating-thickness Measurement by Means of X-ray Fluorescent Analysis

In X-ray fluorescent analysis according to DIN 50987/ISO 3497, specimens are exposed to X-ray radiation. Specimens then emit a line spectrum of fluorescent radiation with energy values corresponding to the atomic numbers of the chemical elements [13, 14]. The intensity of the fluorescent radiation is determined by the measuring area, intensity and energy of the exciting radiation, atomic numbers of coating and substrate materials, density of the coating material, as well as coating thickness. The intensity thus represents the mass per unit area and, for known density, represents the coating thickness. Fluorescent radiation is detected by measuring equipment, capable of detecting either both types of radiation or one of the two (coating or substrate radiation) Fig. 19.9. Thickness measurements can be performed by means of the emission method, the absorption method, or combination techniques determining the intensity ratio. Provided that measuring signal analysis is appropriate, the technique allows analysis of single, dual, and triple coatings, as well as binary and ternary alloy coatings. The measurement inaccuracy generally is in the range of 1% to 10%. Depending on the particular combination of materials, the technique is suitable

309

310

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.9 Coating analysis with X-ray fluorescent analysis.

for coating thicknesses between 0.1 and approx. 200 lm. Non-contact and continuous measurements are described in DIN 50977 [15]. 19.3.8 Coating-thickness Analysis by Means of Beta-backscatter Technique

The beta-backscatter technique is a non-destructive measurement method for metal and non-metal coatings on metal and non-metal substrate materials. A sufficient difference of coating and substrate in terms of atomic numbers is necessary, i.e. at least in the region of 20%. The procedure is standardised in DIN EN ISO 3543 (follow-up of DIN 50983) [16]. The standard also lists atomic numbers for typical substrate and coating materials. A radioactive source emits an electron beam (beta radiation), directed towards the specimen surface. The backscattered proportion of electrons is detected by a particle detector. The intensity of backscattered electrons is interpreted, and the mass per unit area of coating and, therefore, the coating thickness is calculated. The analysis requires the precise composition and density of coating and substrate as additional input data (Fig. 19.10). A few square millimetres of even coating surface are sufficient for analysis. Surfaces that are curved, or difficult to reach from the outside, restrict application of the method. Coating thickness is analysed in the range of 0.2 lm to approx. 800 lm with an average accuracy of approx. 3%. In practice, the method is used for single, uniform coatings only [13, 17].

19.4 Internal Stresses in Coatings

Fig. 19.10 Coating analysis with beta-backscatter technique – operating principle.

19.4 Internal Stresses in Coatings

Internal stresses are multi-axial elastic stresses in mechanical equilibrium, appearing without presence of any external forces and moments [13]. They can extend across macroscopic, microscopic, or sub-microscopic areas, and are classified accordingly into class I (homogeneous across larger areas of material), class II (homogeneous within grain areas or single grains), and class III internal stresses (homogeneous within interatomic distances). 19.4.1 Roentgenographic Assessment of Internal Stresses – X-ray Diffractometry

In this method, the coating surface is exposed to X-radiation. X-rays penetrate the coating, are scattered, and partially re-emitted. Depending on the angle of incidence and the detecting angle, interference occurs due to scattered rays originating from different depths of the coating. X-ray diffractometry is used for measuring elastic lattice dilatation along the diffraction vector (Fig. 19.11). Interpretation of interference fringe pattern allows calculation of a simplified plane strain condition, considered constant across the coating thickness down to the maximum penetration depth of the primary radiation. If, however, stress gradients do occur across the coating thickness, the penetration depth is varied by adjusting the intensity of the primary radiation, or successive stripping of the coating is used to analyse stresses for individual penetration depths. X-ray diffraction analysis utilises a multi-axis rotating and tilting device referred to as a four-circle diffractometer in order to adjust the angle of incidence and the detecting angle for internal stress analysis (Fig. 19.12). The equipment allows rotating motion in the direction of the above angles, and principal stress directions are acquired by adjusting these angles.

311

312

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.11 Assessment of internal stresses in coating systems by means of X-ray diffractometry.

Fig. 19.12 Four-circle diffractometer.

19.5 Detecting Coating Defects

Coating defects are classified into surface defects, near-surface defects, and internal defects. Apart from these coating defects in the strictest sense of the term, additional errors and defects occur during coating production and processing such as uneven coating thickness (cf. Section 19.3), insufficient adhesion between coating and substrate, or uneven distribution of coating components within the coating.

19.5 Detecting Coating Defects

19.5.1 Detection of Open Defects in Coatings – Dye-penetration Test

Dye-penetration tests are capable of detecting surface defects such as open cracks, pores, wrinkles, overlaps, and open bond defects in coatings. In terms of defect types, utilisation is thus analogous to that of visual inspection. However, the resolution of defect analysis is considerably higher. Additionally, the employed equipment allows photographic documentation of findings. Generally, all types of coatings can be tested. However, the technique is commonly used for non-porous surfaces. The two basic requirements for dye-penetration tests are: · Defects start at the surface and are free of residue. · Coating material is compatible with the testing agent. The principle of dye-penetration testing is best described by studying the test procedure, involving the following steps: 1. Cleaning of coating surface and drying. 2. Contrasting penetrating agent is applied to the surface and penetrates surface defects (capillary action). 3. Intermediate cleaning of coating surface in order to remove excess penetrating agent. 4. Drying. 5. Developer is applied – developing phase. 6. Inspection of coating surface – defect detection. 7. Post-cleaning in case of repair or further use. The process is fairly time consuming. However, it allows detection of defects that reach the surface and are invisible to the naked eye. Testing equipment and agents are standardised in DIN EN 571-1 [18]. 19.5.2 Detection of Laminar Separation/Coating Delamination – Ultrasonic Testing

A key aspect of coatings is a firm bond between coating and substrate. Sufficient adhesion is required and coating delamination must be prevented. Usually, errors during coating processes, e.g. contaminated or unactivated surfaces, can be ruled out. If they do occur they reduce or prevent adhesion of the coating to the substrate material. Therefore, quality control for high-quality, highly loaded coatings demands non-destructive testing of coatings with respect to adhesion and bond to the substrate. The aim of non-destructive testing is thus to provide information on whether adhesion is sufficient, based on unambiguous measured data [19]. Sound damping and reflections at the interface between the coating and the substrate change when coatings delaminate. Ultrasonic analysis is capable of reliably detecting such coating delamination provided that areas of delamination

313

314

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.13 Detection of local coating delamination with ultrasonic impulse echo technique.

Fig. 19.14 Detection of coating delamination in thin coatings with ultrasonic impulse echo technique.

are larger than 1 mm2. Figure 19.13 shows the detection of delaminated areas in an electroplated nickel coating with a thickness of approx. 1 mm. The coating contains embedded boron carbide particles and protects an austenitic base plate (material no. 1.4301) for an application in nuclear technology. Delaminations with diameters of 2 to 20 mm were reliably detected.

19.5 Detecting Coating Defects

Thin coatings are frequently tested by using a high-frequency (HF) probe with a delay tip (testing frequency 20 to 50 kHz), or by means of straight HF probes with spherical delay tips. The application example in Fig. 19.14 shows the detection of nickel coating delamination (200 lm thickness) on a cylindrical substrate (material no. 1.0254, St 37, corresponds to Fe360). 19.5.3 Detection of Laminar Separation/Coating Delamination – Lock-in Thermography

Based on changes in heat flux, thermography is a further technique for detecting coating defects such as delamination, defective spots, and variations of coating thickness. Until recently, thermographic investigations frequently used light pulses for excitation. However, the heat input from short pulses of light can cause high temperatures at the workpiece surface. Lock-in thermography presents a solution to the problem using modulated workpiece excitation. Here, the energy input is stretched over a longer period of time thus reducing the thermal load of the workpiece surface. The principle of lock-in thermography is based on detecting a temperature modulation that: · is created either by the specimen surface, and propagates as a thermal wave, · or is generated directly within the specimen (internal excitation, e.g. by ultrasonic sonotrodes, microwaves, or electrical heating) Fig. 19.15.

Fig. 19.15 Detection of coating defects with thermographic techniques.

315

316

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.16 Detection of delamination in corrosion protection coatings on turbine blades by means of thermography.

External excitation allows detection of boundaries reflecting the thermal wave. These methods are thus sensitive towards all types of changes in thermal properties of a specimen [20]. Internal excitation, however, is defect sensitive, i.e. defective areas are actually the origin of the thermal wave. The principle is often also referred to as a dark-field method, reacting very sensitively to areas with increased mechanic loss angles. Heat introduced into the specimen propagates to the specimen surface as a wave, characterising the local and temporal temperature modulation. A very sensitive analysis method is provided by assessing phase shifts between the source of energy and the measured thermal response signal. In pixel imaging of temperature distribution during modulated heat input, pixel-wise Fourier analysis yields the amplitude and phase shift of the thermal specimen response. The two values are interpreted as follows. The amplitude image is determined by non-homogeneities of optical absorption, infrared emission, as well as distribution and excitation. The phase image is independent of these effects. The phase image allows characterisation of defect depth as phase angles are determined by the path of the thermal wave. Sufficiently high levels of energy absorption yield good detectability of coating delamination and internal cavities in the size range of 50 lm (Fig. 19.16).

19.5 Detecting Coating Defects

Fig. 19.17 Detection of defects in electrically conductive coatings with eddy-current technique.

19.5.4 Detection of Internal Coating Defects – Eddy-current Testing

An important application of eddy-current testing, apart from material assessments and coating-thickness measurements in particular (see Section 19.3.6), is detection of surface and internal defects (e.g. hidden cracks, pores, delamination). The technique is applicable for electrically conductive, non- or slightly ferromagnetic coatings on metal and non-metal substrate materials. The testing procedure is based on a preliminary outline for standardisation according to DIN EN ISO 21968 (August 2003 standard outline [21]). In this field of defect testing, a differential sensor with two interconnected measuring coils is preferred. The sensor is thus insensitive to lift-off effects, but highly sensitive to defects listed above (Fig. 19.17). A prerequisite for utilising eddy-current testing, stated in DIN EN ISO 21968, is: ratio of the products of conductivity and permeability of coating and substrate must exceed q(r ´ lrel)min = 1.5. As lrel = 1 in non-ferromagnetic materials, electrical conductivity is the only influencing factor. Changes in electrical conductivity due to material separation affect eddy-current formation in the specimen, i.e. eddy-current density is reduced locally in the defect area. At the same time, the resulting magnetic field changes and modifies the complex impedance of the measuring coil, which is observed on the testing gauge by changing measured values (Fig. 19.17).

317

318

19 Non-destructive Testing and Assessment of Coatings

Fig. 19.18 Detection of defects with dye-penetration methods and eddy-current analysis.

Minor changes in material properties, e.g. material non-homogeneities at or near the surface, are detected with an absolute coil by using the lift-off effect. A differential coil would not detect lift-off as both measuring coils are equally affected, and individual effects cancel each other out. Generally, the maximum measurable coating thickness is 80% of the standard penetration depth, defined in DIN EN ISO 21968. The standard specifies a constant measuring inaccuracy of umin = 0.5 lm for individual measurements of thin coatings with less than dlim = 25 lm coating thickness [20]. Above dlim, measuring inaccuracy is approx. 2% of the coating thickness. Unintended lift-off or edge effects during testing distort the measuring results. Roughness and curves in the coating surface as well as in the interface between coating and substrate have the same effect. Intermediate layers can be investigated by using eddy-current equipment allowing several different frequencies. Above the minimum coating thickness dmin, intermediate layers are treated as substrate material. Defect testing of thick conductive coatings and of substrate material requires higher penetration depths. Typical techniques are eddy-current, far-zone eddycurrent, and impulse eddy-current methods. Compared to dye-penetration tests, eddy-current techniques allow detection of hidden coating defects, undetectable by liquid-penetration methods (Fig. 19.18). 19.5.5 Assessment of Coating Adhesion by Means of Electromagnetic Testing

The principle of electromagnetic testing methods is described above. A specimen, in this case, a ferromagnetic nickel coating on a ferromagnetic substrate,

19.6 Summary and Conclusions

is exposed to an alternating electromagnetic field (primary field) emitted by a probe. The electrical conductivity and magnetic permeability of the specimen material determine the generated magnetic reversal and eddy-current distribution in the specimen, detected by analysing the produced secondary field with a measuring coil. Here, measured signals are assessed, considering the ferromagnetic and electrical behaviour of the coating and substrate. Separation of coating and substrate, and thus the resulting gaps can lead to a disturbed eddy-current distribution in the specimen, causing changes in the measured signals. It is crucial, however, that the eddy currents progress fairly perpendicular to the gap. This distribution requires special sensor design. The fundamentals of adhesion testing by means of electromagnetic waves are described above in Section 19.5.4, and are standardised to a great extent in DIN EN ISO 21968 standard outline [21].

19.6 Summary and Conclusions

Ceramic workpiece coatings are required to meet a large number of demands in terms of coating thickness, compliance with maximum porosity requirements, absence of surface and internal cracks or pores, as well as the ability of a coating to bond to subjacent layers or the substrate, and to prevent delamination. Further important characteristic parameters are necessary in order to assess appropriate applications for coatings. Apart from mechanical-technological coating and substrate behaviour, these include physical properties such as electrical conductivity, magnetic permeability, as well as thermal and chemical properties, providing an indirect means of detecting defects and coating characteristics. Here, modern techniques such as impulse and lock-in thermography, electromagnetic testing methods, radiation and acoustic techniques offer a broad range of possibilities. They are utilised in order to accurately detect or measure the wide spectrum of coated-part properties with high certainty. At the same time, costs and time spent on non-destructive testing, and thus the expense of quality assessment, remain relatively low. Modern testing methods, therefore, allow series testing. In line with tolerable measurement costs, testing can cover sampling inspections or even 100-per cent inspections. However, the development potential of the presented testing techniques and methods remains high in terms of increased efficiency and accuracy.

319

320

19 Non-destructive Testing and Assessment of Coatings

References 1 Siegmann, S.: Thermisch gespritzte

2 3

4

5

6

7

8

Schichten: Charakterisierungen und Applikationen. In: Seminar „Angewandte Oberflächenanalytik“, pp. 105–109, Dübendorf, Switzerland, 1995 IAG Industrie-Anlagen-Bau Georgsmarienhütte, website http://www.iag-gmbh.de/ VDI Guideline 3824, Qualitätssicherung bei der PVD- und CVD-Hartstoffbeschichtung, PVD and CVD hard coatings – Quality assurance, pp. 2–4, 1997–2002 DIN EN ISO 2064 Publication date: 2000-06, Metallische und andere anorganische Schichten – Definitionen und Festlegungen, die die Messung der Schichtdicke betreffen. Metallic and other nonorganic coatings – Definitions and conventions concerning the measurement of thickness ISO 1463, Metall- und Oxidschichten – Schichtdickenmessung – Mikroskopisches Verfahren. Metallic and oxide coatings – Measurement of coating thickness – Microscopical method DIN EN ISO 3882, Publication date: 2003-10 Metallische und andere anorganische Überzüge – Übersicht über Verfahren zur Schichtdickenmessung. Metallic and other inorganic coatings – Review of methods of measurement of thickness DIN 50988-2, Publication date: 1988-04 Messung von Schichtdicken; Bestimmung der flächenbezogenen Masse von Zink- und Zinnschichten auf Eisenwerkstoffen durch Ablösen des Schichtwerkstoffes; Maßanalytische Verfahren. Measurement of coating thickness; determination of the mass per unit area of zinc and tin-coatings on ferrous materials by dissolution of the coating material; volumetric method ISO 4524-1, Publication date: 1985-03 Metallische Überzüge; Prüfmethoden für elektrolytische Überzüge aus Gold und Goldlegierungen; Teil 1: Bestimmung der Überzugsdicke. Metallic coatings; Test methods for electrodeposited gold and gold alloy coatings; Part 1: Determination of coating thickness

9 DIN 50981/DIN EN ISO 2178, Publica-

10

11

12

13

14

15

16

tion date: 1995-04 Nichtmagnetische Überzüge auf magnetischen Grundmetallen – Messen der Schichtdicke – Magnetverfahren. Non-magnetic coatings on magnetic substrates – Measurement of coating thickness – Magnetic method Tutzschky, G., et al.: Zerstörungsfreie Schichtdickenmessung auf der Grundlage des Haftkraftverfahrens. Neue Hütte 29 (1984) 12, pp. 468–471 Zhidong, X.: Untersuchung thermisch gespritzter Beschichtungen unter Anwendung der akustischen Schlag-Methode. Insight, 41 (1999) 8, pp. 517–519, ISSN 0007-1137, 1999 DIN 50984/DIN EN ISO 2360, Publication date: 2004-04 Nichtleitende Überzüge auf nichtmagnetischen metallischen Grundwerkstoffen – Messen der Schichtdicke – Wirbelstromverfahren. Non-conductive coatings on nonmagnetic electrically conductive basis materials – Measurement of coating thickness – Amplitude-sensitive eddy current method Charakterisierung dünner Schichten, DIN Technical Report 39, 1st edn, 1993. ISSN 0179-275X, Beuth, Berlin DIN 50987/DIN EN ISO 3497, Publication date: 2001-12 Metallische Schichten – Schichtdickenmessung – Röntgenfluoreszenz-Verfahren. Metallic coatings – Measurement of coating thickness – Xray spectrometric methods DIN 50977, Publication date: 1993-09 Messung von Schichtdicken; Berührungslose Messung der Dicke von Schichten am kontinuierlich bewegten Meßgut. Measurement of coating thickness – Non contact measurement of coating thickness on continuously moving objects DIN 50983/DIN EN ISO 3543, Publication date: 2001-12 Metallische und nichtmetallische Schichten – Dickenmessung – Betarückstreu-Verfahren. Metallic and nonmetallic coatings – Measurement of thickness – Beta backscatter method

References 17 Faculty of Physics, Munich University,

website http://www.physik.uni-muenchen. de/ 18 DIN EN 571-1, Publication date: 1997-03 Zerstörungsfreie Prüfung – Eindringprüfung – Teil 1: Allgemeine Grundlagen. Non-destructive testing – Penetrant testing – Part 1: General principles 19 Zerstörungsfreie Prüfungen zur Charakterisierung von NiBSorb-Beschichtungen auf Grundplatten. Work Report, November 12th, 2001, IW-ZfP, Hannover University 20 Dvorak, M., Rupp, A., Florin, C.: Zerstörungsfreie Prüfung von thermisch gespritzten Schichten durch Puls-Thermo-

graphie. Conference Report “2nd United Spray Conference”, March 17th–19th, 1999, pp. 345–349, ISBN 3-87155-653-X, DVS, Düsseldorf, 1999 21 DIN EN ISO 21968 (preliminary outline), Publication date: 2003-08 Nichtmagnetische metallische Überzüge auf metallischen und nichtmetallischen Grundwerkstoffen – Messung der Schichtdicke – Wirbelstromphasenwechselverfahren. Non-magnetic metallic coatings on metallic and nonmetallic basis materials – Measurement of coating thickness – Phase-sensitive eddy-current method

321

323

Subject Index a abradable coatings 144, 161, 175 ff. acoustic techniques 303, 319 aluminium pseudo-alloys 285, 294 anti-reflection coatings 210 ff, 219 arc spraying 119 ff., 145, 196 assessment 9, 65, 240, 255, 297 ff.

b barrel plating 109 ff. beta-backscatter technique 310 ff. biotechnology 144 BrazeCoat process 250 brazing filler metal application 211, 257 brazing of ceramics 244 f. brazing technology 253 ff. brazing 109, 239 ff., 253 ff, 263, 297 f. brazing, soldering 239 ff, 253 ff. brush plating 109, 114 build-up welding techniques 268 build-up welding 8, 119, 155, 252, 263 ff.

c chemical vapour deposition 3, 8, 35 f., 51, 65, 298 coating adhesion 8, 25, 318, 33, 35, 73 f., 96, 115 f, 122, 234, 265 coating deposition 67, 91 coating microstructure 39, 104, 249 coating processes 3, 119, 183, 205 ff., 253 ff, 297, 313 coating properties 7, 36, 73 ff., 82, 99, 117, 137, 164, 191, 268 coating technologies 6 ff, 220, 239, 253 ff. coating thickness measurement 300 ff. cobalt hard alloy 285, 292 cold gas spraying 120, 122, 134, 179 ff. combination coatings 70, 156 continuous plating 109, 112 f.

corrosion protection 7 f., 221 ff., 265 ff., 283 ff., 295, 316 corrosion rates 223, 235 corrosion resistance 51, 56, 155, 176, 234, 244, 256, 264, 283, 293 ff. corrosivity categories 234 cosmetic formulations 219 costs per piece 9 current-density distribution 106 cutting 28, 53 ff., 87 ff., 113, 173, 176, 187, 266 CVD diamond thin films 87 ff. CVD systems 67 ff.

d defects 103, 143, 254 f., 298 ff, 312 ff. degreasing 45, 114, 230 delamination 40, 266, 299 ff, 313 ff. deposition mechanism 67, 105, 114 deposition techniques 250, 265 diagnostic techniques 197 diagnostics 134, 191 ff. diamond tools 88, 94, 151, 155 diamond-coated tools 88, 94 ff. differential weight analysis 302 dye penetration test 301, 313, 318

e economic assessment 9, 255 eddy-current method 318 eddy-current testing 317 effect pigments 217 f. electrode potential 40 ff. electrode reaction 40 ff. electrode 40 ff. electroforming 47, 116 f. electroless metal plating 46 electroless plating 36 ff., 101 ff., 257 ff. electrolysis 42, 102 f., 114,

Modern Surface Technology. Edited by Friedrich-Wilhelm Bach, Andreas Laarmann, Thomas Wenz Copyright © 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim ISBN: 3-527-31532-2

324

Subject Index electromagnetic testing 306, 318 f. electroplating and electroless plating 257 ff. enthalpy probe technique 203

f Faraday’s laws 42 fatigue strength 12, 21 ff. flame spraying 119 ff., 143 ff., 200, 270 fuels 60, 148

g gas supply 165, 179 ff. gas-shielded metal arc welding 274 geometric measurement of parts 302

h hard-material coating 24, 65 ff. heat treatment 11 ff., 67 ff, 108 f. helium recovery 185 high-performance applications 51 high-performance three-cathode plasma gun 168 ff. high-temperature applications 54, 60 high-temperature CVD 65 high-velocity oxygen fuel flame spraying 123, 145, 155 hot filament technique 91, 93 hot-dip coatings 227 f. hot-dip-coating system 229 HVOF guns 125, 146

market situation 52 material selection 9, 142 f. metal coatings 40, 117, 123, 132, 144, 255 ff., 307, 310 microstructuring 116 f. moderate-temperature CVD 65, 80

n nickel alloys 149, 285 nickel deposition 101 ff. nickel hard alloys 285, 287 non-destructive testing 297 ff. non-metal diffusion processes 12

o online process control 154, 199 ff. overpotential 44

p

Kolsterising 11, 13, 19 ff.

part coating 60 ff. part surfaces 1 ff, 79, 159, particle diagnostics 191 ff. particle flux imaging (PFI) 199 ff. particle image velocimetry 192 ff. pearlescent pigments 214 ff. phase diagrams 224 ff. phase Doppler anemometry (PDA) 192 ff. physical vapour deposition 3, 8, 32, 51 pickling 115 f. pin-on-disc test 21, 26 f. plasma activated CVD 65, 82 plasma gun 164 ff. plasma MIG process 279 f. plasma spraying 120, 123, 145, 151, 153, 171 ff., 199 f., 276 plasma techniques 279 powder-production processes 137 f. pressure tank 179, 185 printing industry 174 process parameters 13, 39, 57, 89, 92, 107, 147, 153, 160, 179, 208 ff., 246, 254 PVD coating 12, 24, 32, 35, 55 ff., 83, 259 f. PVD techniques 260

l

q

laser cladding 281 ff. laser Doppler anemometry 184, 191 ff. laser two-focus method 192, 195 lock-in thermography 301, 315, 319

quality control 153, 199, 233 ff., 313

i inert-gas-shielded arc welding 273 in-flight particle diagnostics 197 interference colours 214 ff. internal stress 6, 243, 265 f., 298, 300, 311 f. iron hard alloys 285, 288 iron-based materials 284, 288

k

m magnetic induction techniques 305 magnetic pull-off technique 302 f.

r rack plating 109, 111 ff. repair brazing 241 f. resistance electroslag build-up welding (RES) 271, 284 resistance roll seam technique 280

Subject Index

s scratch test 25 f. shop welding 270 Si alkoxides 208 sintered hard metal 242, 246 SiO2 flakes 217 f. soldering 239 ff., 258, 260 sol-gel chemistry 205 ff. sol-gel process 205 ff. spray materials 123 ff., 137 ff., 146, 148 sputtering 8, 25, 32 ff., 56, 260 stainless austenitic steels 15 submerged arc build-up welding 271, 295 surface hardening 4, 11 ff. surface loads 2 surface modification 3 ff. surface post-treatment 233 surface protection 263 ff. surface treatment 1 ff., 233 systems design 127

thermal spraying 119 ff., 137 ff., 145 ff., 160, 191 ff. thermal-barrier coatings 32, 61, 119, 153, 161, 171, 177 thin-film techniques 32 thin-film technology 32 tool coating 51, 54 f., 87, 90 Triplex II 159 ff. tungsten inert gas build-up welding 273

u ultrasonic technique 303

w wear protection coatings 132, 151, 239, 250, 266, 285, 295, wear protection materials 265, 289 wear resistance 2, 11 ff., 54, 61, 143, 149 ff., 213, 242 ff., 264, 283, 287 ff.

x t tampon plating 114 tank plating 109, 114 test planning 299

X-ray diffractometry 300, 311 f. X-ray fluorescent analysis 300, 309 ff.

z zinc, molten 230 ff.

325